Limits of radiomic-based entropy as a surrogate of tumor heterogeneity: ROI-area, acquisition protocol and tissue site exert substantial influence

Neuroimaging characterization of multiple sclerosis lesions in pediatric patients: an exploratory radiomics approach

Introduction

Multiple sclerosis (MS), a chronic inflammatory immune-mediated disease of the central nervous system (CNS), is a common condition in young adults, but it can also affect children. The aim of this study was to construct radiomic models of lesions based on magnetic resonance imaging (MRI, T2-weighted-Fluid-Attenuated Inversion Recovery), to understand the correlation between extracted radiomic features, brain and lesion volumetry, demographic, clinical and laboratorial data.

Methods

The neuroimaging data extracted from eleven scans of pediatric MS patients were analyzed. A total of 60 radiomic features based on MR T2-FLAIR images were extracted and used to calculate gray level co-occurrence matrix (GLCM). The principal component analysis and ROC analysis were performed to select the radiomic features, respectively. The realized classification task by the logistic regression models was performed according to these radiomic features.

Results

Ten most relevant features were selected from data extracted. The logistic regression applied to T2-FLAIR radiomic features revealed significant predictor for multiple sclerosis (MS) lesion detection. Only the variable "contrast" was statistically significant, indicating that only this variable played a significant role in the model. This approach enhances the classification of lesions from normal tissue.

Discussion and conclusion

Our exploratory results suggest that the radiomic models based on MR imaging (T2-FLAIR) may have a potential contribution to characterization of brain tissues and classification of lesions in pediatric MS.

Radiogenomics: Contemporary Applications in the Management of Rectal Cancer

Radiogenomics, a sub-domain of radiomics, refers to the prediction of underlying tumour biology using non-invasive imaging markers. This novel technology intends to reduce the high costs, workload and invasiveness associated with traditional genetic testing via the development of 'imaging biomarkers' that have the potential to serve as an alternative 'liquid-biopsy' in the determination of tumour biological characteristics. Radiogenomics also harnesses the potential to unlock aspects of tumour biology which are not possible to assess by conventional biopsy-based methods, such as full tumour burden, intra-/inter-lesion heterogeneity and the possibility of providing the information of tumour biology longitudinally. Several studies have shown the feasibility of developing a radiogenomic-based signature to predict treatment outcomes and tumour characteristics; however, many lack prospective, external validation. We performed a systematic review of the current literature surrounding the use of radiogenomics in rectal cancer to predict underlying tumour biology.

Radiomics analysis with three-dimensional and two-dimensional segmentation to predict survival outcomes in pancreatic cancer

BACKGROUND

Radiomics can assess prognostic factors in several types of tumors, but considering its prognostic ability in pancreatic cancer has been lacking.

AIM

To evaluate the performance of two different radiomics software in assessing survival outcomes in pancreatic cancer patients.

METHODS

We retrospectively reviewed pretreatment contrast-enhanced dual-energy computed tomography images from 48 patients with biopsy-confirmed pancreatic ductal adenocarcinoma who later underwent neoadjuvant chemoradiation and surgery. Tumors were segmented using TexRad software for 2-dimensional (2D) analysis and MIM software for 3D analysis, followed by radiomic feature extraction. Cox proportional hazard modeling correlated texture features with overall survival (OS) and progression-free survival (PFS). Cox regression was used to detect differences in OS related to pretreatment tumor size and residual tumor following treatment. The Wilcoxon test was used to show the relationship between tumor volume and the percent of residual tumor. Kaplan-Meier analysis was used to compare survival in patients with different tumor densities in Hounsfield units for both 2D and 3D analysis.

RESULTS

3D analysis showed that higher mean tumor density [hazard ratio (HR) = 0.971, P = 0.041)] and higher median tumor density (HR = 0.970, P = 0.037) correlated with better OS. 2D analysis showed that higher mean tumor density (HR = 0.963, P = 0.014) and higher mean positive pixels (HR = 0.962, P = 0.014) correlated with better OS; higher skewness (HR = 3.067, P = 0.008) and higher kurtosis (HR = 1.176, P = 0.029) correlated with worse OS. Higher entropy correlated with better PFS (HR = 0.056, P = 0.036). Models determined that patients with increased tumor size greater than 1.35 cm were likely to have a higher percentage of residual tumors of over 10%.

CONCLUSION

Several radiomics features can be used as prognostic tools for pancreatic cancer. However, results vary between 2D and 3D analyses. Mean tumor density was the only variable that could reliably predict OS, irrespective of the analysis used.

Predicting the recurrence risk of renal cell carcinoma after nephrectomy: potential role of CT-radiomics for adjuvant treatment decisions

OBJECTIVES

Previous trial results suggest that only a small number of patients with non-metastatic renal cell carcinoma (RCC) benefit from adjuvant therapy. We assessed whether the addition of CT-based radiomics to established clinico-pathological biomarkers improves recurrence risk prediction for adjuvant treatment decisions.

METHODS

This retrospective study included 453 patients with non-metastatic RCC undergoing nephrectomy. Cox models were trained to predict disease-free survival (DFS) using post-operative biomarkers (age, stage, tumor size and grade) with and without radiomics selected on pre-operative CT. Models were assessed using C-statistic, calibration, and decision curve analyses (repeated tenfold cross-validation).

RESULTS

At multivariable analysis, one of four selected radiomic features (wavelet-HHL_glcm_ClusterShade) was prognostic for DFS with an adjusted hazard ratio (HR) of 0.44 (p = 0.02), along with American Joint Committee on Cancer (AJCC) stage group (III versus I, HR 2.90; p = 0.002), grade 4 (versus grade 1, HR 8.90; p = 0.001), age (per 10 years HR 1.29; p = 0.03), and tumor size (per cm HR 1.13; p = 0.003). The discriminatory ability of the combined clinical-radiomic model (C = 0.80) was superior to that of the clinical model (C = 0.78; p < 0.001). Decision curve analysis revealed a net benefit of the combined model when used for adjuvant treatment decisions. At an exemplary threshold probability of ≥ 25% for disease recurrence within 5 years, using the combined versus the clinical model was equivalent to treating 9 additional patients (per 1000 assessed) who would recur without treatment (i.e., true-positive predictions) with no increase in false-positive predictions.

CONCLUSION

Adding CT-based radiomic features to established prognostic biomarkers improved post-operative recurrence risk assessment in our internal validation study and may help guide decisions regarding adjuvant therapy.

KEY POINTS

In patients with non-metastatic renal cell carcinoma undergoing nephrectomy, CT-based radiomics combined with established clinical and pathological biomarkers improved recurrence risk assessment. Compared to a clinical base model, the combined risk model enabled superior clinical utility if used to guide decisions on adjuvant treatment.

Development and validation of a postoperative pulmonary infection prediction model for patients with primary hepatic carcinoma

BACKGROUND

There are factors that significantly increase the risk of postoperative pulmonary infections in patients with primary hepatic carcinoma (PHC). Previous reports have shown that over 10% of patients with PHC experience postoperative pulmonary infections. Thus, it is crucial to prioritize the prevention and treatment of postoperative pulmonary infections in patients with PHC.

AIM

To identify the risk factors for postoperative pulmonary infection in patients with PHC and develop a prediction model to aid in postoperative management.

METHODS

We retrospectively collected data from 505 patients who underwent hepatobiliary surgery between January 2015 and February 2023 in the Department of Hepatobiliary and Pancreaticospleen Surgery. Radiomics data were selected for statistical analysis, and clinical pathological parameters and imaging data were included in the screening database as candidate predictive variables. We then developed a pulmonary infection prediction model using three different models: An artificial neural network model; a random forest model; and a generalized linear regression model. Finally, we evaluated the accuracy and robustness of the prediction model using the receiver operating characteristic curve and decision curve analyses.

RESULTS

Among the 505 patients, 86 developed a postoperative pulmonary infection, resulting in an incidence rate of 17.03%. Based on the gray-level co-occurrence matrix, we identified 14 categories of radiomic data for variable screening of pulmonary infection prediction models. Among these, energy, contrast, the sum of squares (SOS), the inverse difference (IND), mean sum (MES), sum variance (SUV), sum entropy (SUE), and entropy were independent risk factors for pulmonary infection after hepatectomy and were listed as candidate variables of machine learning prediction models. The random forest model algorithm, in combination with IND, SOS, MES, SUE, SUV, and entropy, demonstrated the highest prediction efficiency in both the training and internal verification sets, with areas under the curve of 0.823 and 0.801 and a 95% confidence interval of 0.766-0.880 and 0.744-0.858, respectively. The other two types of prediction models had prediction efficiencies between areas under the curve of 0.734 and 0.815 and 95% confidence intervals of 0.677-0.791 and 0.766-0.864, respectively.

CONCLUSION

Postoperative pulmonary infection in patients undergoing hepatectomy may be related to risk factors such as IND, SOS, MES, SUE, SUV, energy, and entropy. The prediction model in this study based on diffusion-weighted images, especially the random forest model algorithm, can better predict and estimate the risk of pulmonary infection in patients undergoing hepatectomy, providing valuable guidance for postoperative management.

Evaluation of entropy and fractal dimension as biomarkers for tumor growth and treatment response using cellular automata

Cell-based models provide a helpful approach for simulating complex systems that exhibit adaptive, resilient qualities, such as cancer. Their focus on individual cell interactions makes them a particularly appropriate strategy to study cancer therapies' effects, which are often designed to disrupt single-cell dynamics. In this work, we propose them as viable methods for studying the time evolution of cancer imaging biomarkers (IBM). We propose a cellular automata model for tumor growth and three different therapies: chemotherapy, radiotherapy, and immunotherapy, following well-established modeling procedures documented in the literature. The model generates a sequence of tumor images, from which a time series of two biomarkers: entropy and fractal dimension, is obtained. Our model shows that the fractal dimension increased faster at the onset of cancer cell dissemination. At the same time, entropy was more responsive to changes induced in the tumor by the different therapy modalities. These observations suggest that the prognostic value of the proposed biomarkers could vary considerably with time. Thus, it is essential to assess their use at different stages of cancer and for different imaging modalities. Another observation derived from the results was that both biomarkers varied slowly when the applied therapy attacked cancer cells scattered along the automatons' area, leaving multiple independent clusters of cells at the end of the treatment. Thus, patterns of change of simulated biomarkers time series could reflect on essential qualities of the spatial action of a given cancer intervention.

Impact of deep learning image reconstruction algorithms on CT radiomic features in patients with liver tumors

Objective

To evaluate the impact of deep learning image reconstruction (DLIR) and adaptive statistical iterative reconstruction-Veo (ASIR-V) on abdominal CT radiomic features acquired in portal venous phase in liver tumor patients.

Methods

Sixty patients with liver tumors who underwent contrast-enhanced abdominal CT were retrospectively enrolled. Six groups including filtered back projection (FBP), ASIR-V (30%, 70%) and DLIR at low (DLIR-L), medium (DLIR-M and high (DLIR-H), were reconstructed using portal venous phase data. CT-based radiomic features (first-order, texture and wavelet features) were extracted from 2D and 3D liver tumors, peritumor and liver parenchyma. All features were analyzed for comparison. P &lt; 0.05 indicated statistically different. The consistency of 3D lesion feature extraction was assessed by calculating intraclass correlation coefficient (ICC).

Results

Different reconstruction algorithms influenced most radiomic features. The percentages of first-order, texture and wavelet features without statistical difference among 2D and 3D lesions, peritumor and liver parenchyma for all six groups were 27.78% (5/18), 5.33% (4/75) and 5.56% (1/18), respectively (all p &gt; 0.05), and they decreased while the level of reconstruction strengthened for both ASIR-V and DLIR. Compared with FBP, the features of ASIR-V30% and 70% without statistical difference decreased from 71.31% to 23.95%, and DLIR-L, DLIR-M, and DLIR-H decreased from 31.65% to 27.11% and 23.73%. Among texture features, unaffected features of peritumor were larger than those of lesions and liver parenchyma, and unaffected 3D lesions features were larger than those of 2D lesions. The consistency of 3D lesion first-order features was excellent, with intra- and inter-observer ICCs ranging from 0.891 to 0.999 and 0.880 to 0.998.

Conclusions

Both ASIR-V and DLIR algorithms with different strengths influenced the radiomic features of abdominal CT images in portal venous phase, and the influences aggravated as reconstruction strength increased.

Radiogenomics Reveals Correlation between Quantitative Texture Radiomic Features of Biparametric MRI and Hypoxia-Related Gene Expression in Men with Localised Prostate Cancer

OBJECTIVES

To perform multiscale correlation analysis between quantitative texture feature phenotypes of pre-biopsy biparametric MRI (bpMRI) and targeted sequence-based RNA expression for hypoxia-related genes.

MATERIALS AND METHODS

Images from pre-biopsy 3T bpMRI scans in clinically localised PCa patients of various risk categories (n = 15) were used to extract textural features. The genomic landscape of hypoxia-related gene expression was obtained using post-radical prostatectomy tissue for targeted RNA expression profiling using the TempO-sequence method. The nonparametric Games Howell test was used to correlate the differential expression of the important hypoxia-related genes with 28 radiomic texture features. Then, cBioportal was accessed, and a gene-specific query was executed to extract the Oncoprint genomic output graph of the selected hypoxia-related genes from The Cancer Genome Atlas (TCGA). Based on each selected gene profile, correlation analysis using Pearson's coefficients and survival analysis using Kaplan-Meier estimators were performed.

RESULTS

The quantitative bpMR imaging textural features, including the histogram and grey level co-occurrence matrix (GLCM), correlated with three hypoxia-related genes (ANGPTL4, VEGFA, and P4HA1) based on RNA sequencing using the TempO-Seq method. Further radiogenomic analysis, including data accessed from the cBioportal genomic database, confirmed that overexpressed hypoxia-related genes significantly correlated with a poor survival outcomes, with a median survival ratio of 81.11:133.00 months in those with and without alterations in genes, respectively.

CONCLUSION

This study found that there is a correlation between the radiomic texture features extracted from bpMRI in localised prostate cancer and the hypoxia-related genes that are differentially expressed. The analysis of expression data based on cBioportal revealed that these hypoxia-related genes, which were the focus of the study, are linked to an unfavourable survival outcomes in prostate cancer patients.

Development and external validation of a radiomics combined with clinical nomogram for preoperative prediction prognosis of resectable pancreatic ductal adenocarcinoma patients

PURPOSE

To develop and externally validate a prognosis nomogram based on contrast-enhanced computed tomography (CECT) combined clinical for preoperative prognosis prediction of patients with pancreatic ductal adenocarcinoma (PDAC).

METHODS

184 patients from Center A with histopathologically confirmed PDAC who underwent CECT were included and allocated to training cohort (n=111) and internal validation cohort (n=28). The radiomic score (Rad - score) for predicting overall survival (OS) was constructed by using the least absolute shrinkage and selection operator (LASSO). Univariate and multivariable Cox regression analysis was used to construct clinic-pathologic features. Finally, a radiomics nomogram incorporating the Rad - score and clinical features was established. External validation was performed using Center B dataset (n = 45). The validation of nomogram was evaluated by calibration curve, Harrell's concordance index (C-index) and decision curve analysis (DCA). The Kaplan-Meier (K-M) method was used for OS analysis.

RESULTS

Univariate and multivariate analysis indicated that Rad - score, preoperative CA 19-9 and postoperative American Joint Committee on Cancer (AJCC) TNM stage were significant prognostic factors. The nomogram based on Rad - score and preoperative CA19-9 was found to exhibit excellent prediction ability: in the training cohort, C-index was superior to that of the preoperative CA19-9 (0.713 vs 0.616, P< 0.001) and AJCC TNM stage (0.713 vs 0.614, P< 0.001); the C-index was also had good performance in the validation cohort compared with CA19-9 (internal validation cohort: 0.694 vs 0.555, P< 0.001; external validation cohort: 0.684 vs 0.607, P< 0.001) and AJCC TNM stage (internal validation cohort: 0.694 vs 0.563, P< 0.001; external validation cohort: 0.684 vs 0.596, P< 0.001). The calibration plot and DCA showed excellent predictive accuracy in the validation cohort.

CONCLUSION

We established a well-designed nomogram to accurately predict OS of PDAC preoperatively. The nomogram showed a satisfactory prediction effect and was worthy of further evaluation in the future.

Multiparametric MRI enables for differentiation of different degrees of malignancy in two murine models of breast cancer

Objective

The objective of this study was to non-invasively differentiate the degree of malignancy in two murine breast cancer models based on identification of distinct tissue characteristics in a metastatic and non-metastatic tumor model using a multiparametric Magnetic Resonance Imaging (MRI) approach.

Methods

The highly metastatic 4T1 breast cancer model was compared to the non-metastatic 67NR model. Imaging was conducted on a 9.4 T small animal MRI. The protocol was used to characterize tumors regarding their structural composition, including heterogeneity, intratumoral edema and hemorrhage, as well as endothelial permeability using apparent diffusion coefficient (ADC), T1/T2 mapping and dynamic contrast-enhanced (DCE) imaging. Mice were assessed on either day three, six or nine, with an i.v. injection of the albumin-binding contrast agent gadofosveset. Ex vivo validation of the results was performed with laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS), histology, immunhistochemistry and electron microscopy.

Results

Significant differences in tumor composition were observed over time and between 4T1 and 67NR tumors. 4T1 tumors showed distorted blood vessels with a thin endothelial layer, resulting in a slower increase in signal intensity after injection of the contrast agent. Higher permeability was further reflected in higher K_trans values, with consecutive retention of gadolinium in the tumor interstitium visible in MRI. 67NR tumors exhibited blood vessels with a thicker and more intact endothelial layer, resulting in higher peak enhancement, as well as higher maximum slope and area under the curve, but also a visible wash-out of the contrast agent and thus lower K_trans values. A decreasing accumulation of gadolinium during tumor progression was also visible in both models in LA-ICP-MS. Tissue composition of 4T1 tumors was more heterogeneous, with intratumoral hemorrhage and necrosis and corresponding higher T1 and T2 relaxation times, while 67NR tumors mainly consisted of densely packed tumor cells. Histogram analysis of ADC showed higher values of mean ADC, histogram kurtosis, range and the 90^th percentile (p90), as markers for the heterogenous structural composition of 4T1 tumors. Principal component analysis (PCA) discriminated well between the two tumor models.

Conclusions

Multiparametric MRI as presented in this study enables for the estimation of malignant potential in the two studied tumor models via the assessment of certain tumor features over time.

Prognostic value of texture analysis of the primary tumour in high-risk neuroblastoma: An 18 F-DOPA PET study

PURPOSE

To evaluate the prognostic value of texture analysis of the primary tumour with ¹⁸ fluorine-dihydroxyphenylalanine positron emission tomography/X-ray computed tomography (¹⁸ F-DOPA PET/CT) in patients affected by high-risk neuroblastoma (HR-NBL).

METHODS

We retrospectively analysed 18 patients with HR-NBL, which had been prospectively enrolled in the course of a previous trial investigating the diagnostic role of ¹⁸ F-DOPA PET/CT at the time of the first onset. Texture analysis of the primary tumour was carried out on the PET images using LifeX. Conventional indices, histogram parameters, grey level co-occurrence (GLCM), run-length (GLRLM), neighbouring difference (NGLDM) and zone-length (GLZLM) matrices parameter were extracted; their values were compared with the overall metastatic load, expressed by means of whole-body metabolic burden (WBMB) score and the progression-free/overall survival (PFS and OS).

RESULTS

There was a direct correlation between WBMB and radiomics parameter describing uptake intensity (SUV_mean : p = .004) and voxel heterogeneity (entropy: p = .026; GLCM_Contrast: p = .001). Conversely, texture indices of homogeneity showed an inverse correlation with WBMB (energy: p = .026; GLCM_Homogeneity: p = .006). On the multivariate model, WBMB (p < .01) and the first standardised uptake value (SUV) quartile (p < .001) predicted PFS; OS was predicted by WBMB and the N-myc proto-oncogene protein (MYCN) amplification (p < .05) for both.

CONCLUSIONS

Textural parameters describing heterogeneity and metabolic intensity of the primary HR-NBL are closely associated with its overall metastatic burden. In turn, the whole-body tumour load appears to be one of the most relevant predictors of progression-free and overall survival.

A Framework for Harmonization of Radiomics Data for Multicenter Studies and Clinical Trials

PURPOSE

Variability in computed tomography images intrinsic to individual scanners limits the application of radiomics in clinical and research settings. The development of reproducible and generalizable radiomics-based models to assess lesions requires harmonization of data. The purpose of this study was to develop, test, and analyze the efficacy of a radiomics data harmonization model.

MATERIALS AND METHODS

Radiomic features from biopsy-proven untreated hepatic metastasis (N = 380) acquired from 167 unique patients with pancreatic, colon, and breast cancers were analyzed. Radiomic features from volume-match 551 samples of normal liver tissue and 188 hepatic cysts were included as references. A novel linear mixed effect model was used to identify effects associated with lesion size, tissue type, and scanner model. Six separate machine learning models were then used to test the effectiveness of radiomic feature harmonization using multivariate analysis.

RESULTS

Proposed model identifies and removes scanner-associated effects while preserving cancer-specific functional dependence of radiomic features on the tumor size. Data harmonization improves the performance of classification models by reducing the scanner-associated variability. For example, the multiclass logistic regression model, LogitBoost, demonstrated the improvement in sensitivity in the range from 15% to 40% for each type of liver metastasis, whereas the overall model accuracy and the kappa coefficient increased by 5% and 8% accordingly.

CONCLUSION

The model removed scanner-associated effects while preserving cancer-specific functional dependence of radiomic features.

Potential advantages of FDG-PET radiomic feature map for target volume delineation in lung cancer radiotherapy

PURPOSE

To investigate the potential benefits of FDG PET radiomic feature maps (RFMs) for target delineation in non-small cell lung cancer (NSCLC) radiotherapy.

METHODS

Thirty-two NSCLC patients undergoing FDG PET/CT imaging were included. For each patient, nine grey-level co-occurrence matrix (GLCM) RFMs were generated. gross target volume (GTV) and clinical target volume (CTV) were contoured on CT (GTV_CT , CTV_CT ), PET (GTV_PET40 , CTV_PET40 ), and RFMs (GTV_RFM , CTV_RFM ,). Intratumoral heterogeneity areas were segmented as GTV_PET50-Boost and radiomic boost target volume (RTV_Boost ) on PET and RFMs, respectively. GTV_CT in homogenous tumors and GTV_PET40 in heterogeneous tumors were considered as GTV_{gold standard} (GTV_GS ). One-way analysis of variance was conducted to determine the threshold that finds the best conformity for GTV_RFM with GTV_GS . Dice similarity coefficient (DSC) and mean absolute percent error (MAPE) were calculated. Linear regression analysis was employed to report the correlations between the gold standard and RFM-derived target volumes.

RESULTS

Entropy, contrast, and Haralick correlation (H-correlation) were selected for tumor segmentation. The threshold values of 80%, 50%, and 10% have the best conformity of GTV_RFM-entropy , GTV_RFM-contrast , and GTV_{RFM-H-correlation} with GTV_GS , respectively. The linear regression results showed a positive correlation between GTV_GS and GTV_RFM-entropy (r = 0.98, p < 0.001), between GTV_GS and GTV_RFM-contrast (r = 0.93, p < 0.001), and between GTV_GS and GTV_{RFM-H-correlation} (r = 0.91, p < 0.001). The average threshold values of 45% and 15% were resulted in the best segmentation matching between CTV_RFM-entropy and CTV_RFM-contrast with CTV_GS , respectively. Moreover, we used RFM to determine RTV_Boost in the heterogeneous tumors. Comparison of RTV_Boost with GTV_PET50-Boost MAPE showed the volume error differences of 31.7%, 36%, and 34.7% in RTV_{Boost-entropy} , RTV_{Boost-contrast} , and RTV_{Boost-H-correlation} , respectively.

CONCLUSIONS

FDG PET-based radiomics features in NSCLC demonstrated a promising potential for decision support in radiotherapy, helping radiation oncologists delineate tumors and generate accurate segmentation for heterogeneous region of tumors.

Explainability of radiomics through formal methods

BACKGROUND AND OBJECTIVE

Artificial Intelligence has proven to be effective in radiomics. The main problem in using Artificial Intelligence is that researchers and practitioners are not able to know how the predictions are generated. This is currently an open issue because results' explainability is advantageous in understanding the reasoning behind the model, both for patients than for implementing a feedback mechanism for medical specialists using decision support systems.

METHODS

Addressing transparency issues related to the Artificial Intelligence field, the innovative technique of Formal methods use a mathematical logic reasoning to produce an automatic, quick and reliable diagnosis. In this paper we analyze results given by the adoption of Formal methods for the diagnosis of the Coronavirus disease: specifically, we want to analyse and understand, in a more medical way, the meaning of some radiomic features to connect them with clinical or radiological evidences.

RESULTS

In particular, the usage of Formal methods allows the authors to do statistical analysis on the feature value distributions, to do pattern recognition on disease models, to generalize the model of a disease and to reach high performances of results and interpretation of them. A further step for explainability can be accounted by the localization and selection of the most important slices in a multi-slice approach.

CONCLUSIONS

In conclusion, we confirmed the clinical significance of some First order features as Skewness and Kurtosis. On the other hand, we suggest to decline the use of the Minimum feature because of its intrinsic connection with the Computational Tomography exam of the lung.

Radiomics for the Prediction of Response to Antifibrotic Treatment in Patients with Idiopathic Pulmonary Fibrosis: A Pilot Study

Antifibrotic therapy has changed the treatment paradigm for idiopathic pulmonary fibrosis (IPF); however, a subset of patients still experienced rapid disease progression despite treatment. This study aimed to determine whether CT-based radiomic features can predict therapeutic response to antifibrotic agents. In this retrospective study, 35 patients with IPF on antifibrotic treatment enrolled from two centers were divided into training (n = 26) and external validation (n = 9) sets. Clinical and pulmonary function data were collected. The patients were categorized into stable disease (SD) and progressive disease (PD) groups based on functional or radiologic criteria. From pretreatment non-enhanced high-resolution CT (HRCT) images, twenty-six radiomic features were extracted through whole-lung texture analysis, and six parenchymal patterns were quantified using dedicated imaging platforms. The predictive factors for PD were determined via univariate and multivariate logistic regression analyses. In the training set (SD/PD: 12/14), univariate analysis identified eight radiomic features and ground-glass opacity percentage (GGO%) as potential predicators of PD. However, multivariate analysis found that the single independent predictor was the sum entropy (accuracy, 80.77%; AUC, 0.75). The combined sum entropy-GGO% model improved the predictive performance in the training set (accuracy, 88.46%; AUC, 0.77). The overall accuracy of the combined model in the validation set (SD/PD: 7/2) was 66.67%. Our preliminary results demonstrated that radiomic features based on pretreatment HRCT could predict the response of patients with IPF to antifibrotic treatment.

Early Readout on Overall Survival of Patients With Melanoma Treated With Immunotherapy Using a Novel Imaging Analysis

IMPORTANCE

Existing criteria to estimate the benefit of a therapy in patients with cancer rely almost exclusively on tumor size, an approach that was not designed to estimate survival benefit and is challenged by the unique properties of immunotherapy. More accurate prediction of survival by treatment could enhance treatment decisions.

OBJECTIVE

To validate, using radiomics and machine learning, the performance of a signature of quantitative computed tomography (CT) imaging features for estimating overall survival (OS) in patients with advanced melanoma treated with immunotherapy.

DESIGN, SETTING, AND PARTICIPANTS

This prognostic study used radiomics and machine learning to retrospectively analyze CT images obtained at baseline and first follow-up and their associated clinical metadata. Data were prospectively collected in the KEYNOTE-002 (Study of Pembrolizumab [MK-3475] Versus Chemotherapy in Participants With Advanced Melanoma; 2017 analysis) and KEYNOTE-006 (Study to Evaluate the Safety and Efficacy of Two Different Dosing Schedules of Pembrolizumab [MK-3475] Compared to Ipilimumab in Participants With Advanced Melanoma; 2016 analysis) multicenter clinical trials. Participants included 575 patients with a diagnosis of advanced melanoma who were randomly assigned to training and validation sets. Data for the present study were collected from November 20, 2012, to June 3, 2019, and analyzed from July 1, 2019, to September 15, 2021.

INTERVENTIONS

KEYNOTE-002 featured trial groups testing intravenous pembrolizumab, 2 mg/kg or 10 mg/kg every 2 or every 3 weeks based on randomization, or investigator-choice chemotherapy; KEYNOTE-006 featured trial groups testing intravenous ipilimumab, 3 mg/kg every 3 weeks and intravenous pembrolizumab, 10 mg/kg every 2 or 3 weeks based on randomization.

MAIN OUTCOMES AND MEASURES

The performance of the signature CT imaging features for estimating OS at the month 6 posttreatment landmark in patients who received pembrolizumab was measured using an area under the time-dependent receiver operating characteristics curve (AUC).

RESULTS

A random forest model combined 25 imaging features extracted from tumors segmented on CT images to identify the combination (signature) that best estimated OS with pembrolizumab in 575 patients. The signature combined 4 imaging features, 2 related to tumor size and 2 reflecting changes in tumor imaging phenotype. In the validation set (287 patients treated with pembrolizumab), the signature reached an AUC for estimation of OS status of 0.92 (95% CI, 0.89-0.95). The standard method, Response Evaluation Criteria in Solid Tumors 1.1, achieved an AUC of 0.80 (95% CI, 0.75-0.84) and classified tumor outcomes as partial or complete response (93 of 287 [32.4%]), stable disease (90 of 287 [31.3%]), or progressive disease (104 of 287 [36.2%]).

CONCLUSIONS AND RELEVANCE

The findings of this prognostic study suggest that the radiomic signature discerned from conventional CT images at baseline and on first follow-up may be used in clinical settings to provide an accurate early readout of future OS probability in patients with melanoma treated with single-agent programmed cell death 1 blockade.

Role of hepatic metastatic lesion size on inter-reader reproducibility of CT-based radiomics features

OBJECTIVES

To evaluate the effect of hepatic metastatic lesion size on inter-reader reproducibility of CT-based 2D radiomics imaging features.

METHODS

Computerized tomography (CT) scans of 59 liver metastases from 34 patients with colorectal cancer were evaluated. Image segmentation was performed manually by three readers blinded to each other's results. For each radiomics feature, we created two datasets by sorting measurements according to size, i.e., (i) from the smallest to the largest lesion and (ii) from the largest to the smallest lesion. The Lin concordance correlation coefficient (CCC) was employed to analyze the reproducibility of radiomics features. In particular, the CCC was computed as a function of a number of elements in the dataset, by gradually adding lesions from each sorted dataset. To evaluate the effect of lesion size, we analyzed the difference between these two functions thus assessing the contribution of small and large lesions into the reproducibility of radiomics features.

RESULTS

Inter-reader reproducibility of CT-based 2D radiomics features assessed using Lin's CCC demonstrates tumor-size dependence. For example, the Lin CCC for GLCM contrast equals 0.88 (95% C.I. 0.84 to 0.92, p < 0.003) and could change by an additional + / - 0.06 depending on the presence of large or small lesions.

CONCLUSIONS

Groups of "large" and "small" lesions show different inter-reader reproducibility. The inter-reader reproducibility from the mixed group consisting of "large" and "small" lesions depends on the lesion-size distribution and can vary widely. This finding could partially explain variability in reproducibility of radiomics features in the literature.

KEY POINTS

• Groups of "large" and "small" lesions show different inter-reader reproducibility. • The inter-reader reproducibility from the mixed group consisting of "large" and "small" lesions depends on the lesion-size distribution.

An imaging signature to predict outcome in metastatic colorectal cancer using routine computed tomography scans

BACKGROUND & AIMS

Quantitative analysis of computed tomography (CT) scans of patients with metastatic colorectal cancer (mCRC) can identify imaging signatures that predict overall survival (OS).

METHODS

We retrospectively analysed CT images from 1584 mCRC patients on two phase III trials evaluating FOLFOX ± panitumumab (n = 331, 350) and FOLFIRI ± aflibercept (n = 437, 466). In the training set (n = 720), an algorithm was trained to predict OS landmarked from month 2; the output was a signature value on a scale from 0 to 1 (most to least favourable predicted OS). In the validation set (n = 864), hazard ratios (HRs) evaluated the association of the signature with OS using RECIST1.1 as a benchmark of comparison.

RESULTS

In the training set, the selected signature combined three features - change in tumour volume, change in tumour spatial heterogeneity, and tumour volume - to predict OS. In the validation set, RECIST1.1 classified patients in three categories: response (n = 166, 19.2%), stable disease (n = 636, 73.6%), and progression (n = 62, 7.2%). The HR was 3.93 (2.79-5.54). Using the same distribution for the signature, the HR was 21.04 (14.88-30.58), showing an incremental prognostic separation. Stable disease by RECIST1.1 was reclassified by the signature along a continuum where patients belonging to the most and least favourable signature quartiles had a median OS of 40.73 (28.49 to NA) months (n = 94) and 7.03 (5.66-7.89) months (n = 166), respectively.

CONCLUSIONS

A signature combining three imaging features provides early prognostic information that can improve treatment decisions for individual patients and clinical trial analyses.

Decoding intra-tumoral spatial heterogeneity on radiological images using the Hilbert curve

Background

Current intra-tumoral heterogeneous feature extraction in radiology is limited to the use of a single slice or the region of interest within a few context-associated slices, and the decoding of intra-tumoral spatial heterogeneity using whole tumor samples is rare. We aim to propose a mathematical model of space-filling curve-based spatial correspondence mapping to interpret intra-tumoral spatial locality and heterogeneity.

Methods

A Hilbert curve-based approach was employed to decode and visualize intra-tumoral spatial heterogeneity by expanding the tumor volume to a two-dimensional (2D) matrix in voxels while preserving the spatial locality of the neighboring voxels. The proposed method was validated using three-dimensional (3D) volumes constructed from lung nodules from the LIDC-IDRI dataset, regular axial plane images, and 3D blocks.

Results

Dimensionality reduction of the Hilbert volume with a single regular axial plane image showed a sparse and scattered pixel distribution on the corresponding 2D matrix. However, for 3D blocks and lung tumor inside the volume, the dimensionality reduction to the 2D matrix indicated regular and concentrated squares and rectangles. For classification into benign and malignant masses using lung nodules from the LIDC-IDRI dataset, the Inception-V4 indicated that the Hilbert matrix images improved accuracy (85.54% vs. 73.22%, p < 0.001) compared to the original CT images of the test dataset.

Conclusions

Our study indicates that Hilbert curve-based spatial correspondence mapping is promising for decoding intra-tumoral spatial heterogeneity of partial or whole tumor samples on radiological images. This spatial-locality-preserving approach for voxel expansion enables existing radiomics and convolution neural networks to filter structured and spatially correlated high-dimensional intra-tumoral heterogeneity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-021-01100-8.

Intra-scan inter-tissue variability can help harmonize radiomics features in CT

OBJECTIVE

We studied the repeatability and the relative intra-scan variability across acquisition protocols in CT using phantom and unenhanced abdominal series.

METHODS

We used 17 CT scans from the Credence Cartridge Radiomics Phantom database and 20 unenhanced multi-site non-pathologic abdominal patient series for which we measured spleen and liver tissues. We performed multiple measurements in extracting 9 radiomics features. We defined a "tandem" as the measurement of a given tissue (or material) by a given radiomics. For each tandem, we assessed the proportion of the variability attributable to repetitions, acquisition protocols, material, or patient. We analyzed the distribution of the intra-scan correlation between pairs of tandems and checked the impact of correlation coefficient greater than 0.90 in comparing paired and unpaired differences.

RESULTS

The repeatability of radiomics features depends on the measured material; 56% of tandems were highly repeatable. Histogram-derived radiomics were generally less repeatable. Nearly 60% of relative radiomics measurements had a correlation coefficient higher than 0.90 allowing paired measurements to improve reliability in detecting the difference between two materials. The analysis of liver and spleen tissues showed that measurement variability was negligible with respect to other variabilities. As for phantom data, we found that gray level zone length matrix (GLZLM)-derived radiomics and gray level co-occurrence matrix (GLCM)-derived radiomics were the most correlating features. For these features, relative intra-scan measurements improved the detection of different materials or tissues.

CONCLUSIONS

We identified radiomics features for which the intra-scan measurements between tissues are linearly correlated. This property represents an opportunity to improve tissue characterization and inter-site harmonization.

KEY POINTS

• The repeatability of radiomics features on CT depends on the measured material or tissue. • Some tandems of radiomics features/tissues are linearly affected by the variability of acquisition protocols on CT. • Relative intra-scan measurements are an opportunity for improving quantitative imaging on CT.

A subregion-based positron emission tomography/computed tomography (PET/CT) radiomics model for the classification of non-small cell lung cancer histopathological subtypes

Background

This study classifies lung adenocarcinoma (ADC) and squamous cell carcinoma (SCC) using subregion-based radiomics features extracted from positron emission tomography/computed tomography (PET/CT) images.

Methods

In this study, the standard ¹⁸F-fluorodeoxyglucose (FDG) PET/CT images of 150 patients with lung ADC and 100 patients with SCC were retrospectively collected from the PET Center of the First Affiliated Hospital, College of Medicine, Zhejiang University. First, the 3D feature vector of each tumor voxel (whose basis is PET value, CT value, and CT local dominant orientation) was extracted. Using K-means individual clustering and population clustering, each tumor was divided into 4 subregions that reflect intratumoral regional heterogeneity. Next, based on each subregion, 385 radiomics features were extracted. Clinical features including age, gender, and smoking history were included. Thus, there were a total of 1,543 features extracted from PET/CT images and clinical reports. Statistical tests were then used to eliminate irrelevant and redundant features, and the recursive feature elimination (RFE) algorithm was used to select the best feature subset to classify SCC and ADC. Finally, 7 types of classifiers were tested to achieve the optimized model for the classification: support vector machine (SVM) with linear kernel, SVM with radial basis function kernel (SVM-RBF), random forest, logistic regression, Gaussian process classifier, linear discriminant analysis, and the AdaBoost classifier. Furthermore, 5-fold cross-validation was applied to obtain the sensitivity, specificity, accuracy, and area under the curve (AUC) for performance evaluation.

Results

Our model exhibited the best performance with the subregion radiomics features and SVM-RBF classifier, with a 5-fold cross-validation sensitivity, specificity, accuracy, and AUC of 0.8538, 0.8758, 0.8623, and 0.9155, respectively. The interquartile range feature from subregion 2 of CT and the gender feature from the clinical reports are the 2 optimized features that achieved the highest comprehensive score.

Conclusions

Our proposed model showed that SCC and ADC could be classified successfully using PET/CT images, which could be a promising tool to assist radiologists or medical physicists during diagnosis. The subregion-based method illustrated that non-small cell lung cancer (NSCLC) depicts intratumoral regional heterogeneity on both CT and PET images. By defining these heterogeneities through a subregion-based method, the diagnostic performance was improved. The 3D feature vector (whose basis is PET value, CT value, and CT local dominant orientation) showed superiority in reflecting NSCLC intratumoral regional heterogeneity.

Recent Radiomics Advancements in Breast Cancer: Lessons and Pitfalls for the Next Future

Radiomics is an emerging translational field of medicine based on the extraction of high-dimensional data from radiological images, with the purpose to reach reliable models to be applied into clinical practice for the purposes of diagnosis, prognosis and evaluation of disease response to treatment. We aim to provide the basic information on radiomics to radiologists and clinicians who are focused on breast cancer care, encouraging cooperation with scientists to mine data for a better application in clinical practice. We investigate the workflow and clinical application of radiomics in breast cancer care, as well as the outlook and challenges based on recent studies. Currently, radiomics has the potential ability to distinguish between benign and malignant breast lesions, to predict breast cancer’s molecular subtypes, the response to neoadjuvant chemotherapy and the lymph node metastases. Even though radiomics has been used in tumor diagnosis and prognosis, it is still in the research phase and some challenges need to be faced to obtain a clinical translation. In this review, we discuss the current limitations and promises of radiomics for improvement in further research.

Texture Analysis of Fractional Water Content Images Acquired during PET/MRI: Initial Evidence for an Association with Total Lesion Glycolysis, Survival and Gene Mutation Profile in Primary Colorectal Cancer

To assess the capability of fractional water content (FWC) texture analysis (TA) to generate biologically relevant information from routine PET/MRI acquisitions for colorectal cancer (CRC) patients. Thirty consecutive primary CRC patients (mean age 63.9, range 42-83 years) prospectively underwent FDG-PET/MRI. FWC tumor parametric images generated from Dixon MR sequences underwent TA using commercially available research software (TexRAD). Data analysis comprised (1) identification of functional imaging correlates for texture features (TF) with low inter-observer variability (intraclass correlation coefficient: ICC > 0.75), (2) evaluation of prognostic performance for FWC-TF, and (3) correlation of prognostic imaging signatures with gene mutation (GM) profile. Of 32 FWC-TF with ICC > 0.75, 18 correlated with total lesion glycolysis (TLG, highest: rs = -0.547, p = 0.002). Using optimized cut-off values, five MR FWC-TF identified a good prognostic group with zero mortality (lowest: p = 0.017). For the most statistically significant prognostic marker, favorable prognosis was significantly associated with a higher number of GM per patient (medians: 7 vs. 1.5, p = 0.009). FWC-TA derived from routine PET/MRI Dixon acquisitions shows good inter-operator agreement, generates biological relevant information related to TLG, GM count, and provides prognostic information that can unlock new clinical applications for CRC patients.

Radiomics Response Signature for Identification of Metastatic Colorectal Cancer Sensitive to Therapies Targeting EGFR Pathway

BACKGROUND

The authors sought to forecast survival and enhance treatment decisions for patients with liver metastatic colorectal cancer by using on-treatment radiomics signature to predict tumor sensitiveness to irinotecan, 5-fluorouracil, and leucovorin (FOLFIRI) alone (F) or in combination with cetuximab (FC).

METHODS

We retrospectively analyzed 667 metastatic colorectal cancer patients treated with F or FC. Computed tomography quality was classified as high (HQ) or standard (SD). Four datasets were created using the nomenclature (treatment) - (quality). Patients were randomly assigned (2:1) to training or validation sets: FCHQ: 78:38, FCSD: 124:62, FHQ: 78:51, FSD: 158:78. Four tumor-imaging biomarkers measured quantitative radiomics changes between standard of care computed tomography scans at baseline and 8 weeks. Using machine learning, the performance of the signature to classify tumors as treatment sensitive or treatment insensitive was trained and validated using receiver operating characteristic (ROC) curves. Hazard ratio and Cox regression models evaluated association with overall survival (OS).

RESULTS

The signature (area under the ROC curve [95% confidence interval (CI)]) used temporal decrease in tumor spatial heterogeneity plus boundary infiltration to successfully predict sensitivity to antiepidermal growth factor receptor therapy (FCHQ: 0.80 [95% CI = 0.69 to 0.94], FCSD: 0.72 [95% CI = 0.59 to 0.83]) but failed with chemotherapy (FHQ: 0.59 [95% CI = 0.44 to 0.72], FSD: 0.55 [95% CI = 0.43 to 0.66]). In cetuximab-containing sets, radiomics signature outperformed existing biomarkers (KRAS-mutational status, and tumor shrinkage by RECIST 1.1) for detection of treatment sensitivity and was strongly associated with OS (two-sided P < .005).

CONCLUSIONS

Radiomics response signature can serve as an intermediate surrogate marker of OS. The signature outperformed known biomarkers in providing an early prediction of treatment sensitivity and could be used to guide cetuximab treatment continuation decisions.

Association Between the Size and 3D CT-Based Radiomic Features of Breast Cancer Hepatic Metastasis

PURPOSE

To evaluate the effect of the anatomic size on 3D radiomic imaging features of the breast cancer hepatic metastases.

MATERIALS AND METHODS

CT scans of 81 liver metastases from 54 patients with breast cancer were evaluated. Ten most common 3D radiomic features from the histogram and gray level co-occurrence matrix (GLCM) categories were calculated for the hepatic metastases (HM) and compared to normal liver (NL). The effect of size was evaluated by using linear mixed-effects regression models. The effect of size on different radiomic features was analyzed for both liver lesions and background liver.

RESULTS

Three-dimensional radiomic features from GLCM demonstrate an important size dependence. The texture-feature size dependence was found to be different among feature categories and between the HM and NL, thus demonstrating a discriminatory power for the tissue type. Significant difference in the slope was found for GLCM homogeneity (NL slope = 0.004, slope difference 95% confidence interval [CI] 0.06-0.1, p <0.001), contrast (NL slope = 45, slope difference 95% CI 205-305, p <0.001), correlation (NL slope = 0.04, slope difference 95% CI 0.11-0.21, p <0.001), and dissimilarity (NL slope = 0.7, slope difference 95% CI 3.6-5.4, p <0.001). The GLCM energy (NL slope = 0.002, slope difference 95% CI -0.0005 to -0.0003, p <0.007), and entropy (NL slope = 1.49, slope difference 95% CI 0.07-0.52, p <0.009) exhibited size-dependence for both NL and HM, although demonstrating a difference in the slope between themselves.

CONCLUSION

Radiomic features of breast cancer hepatic metastasis exhibited significant correlation with tumor size. This finding demonstrates the complex behavior of imaging features and the need to include feature-specific properties into radiomic models.

Use of Texture Analysis on Noncontrast MRI in Classification of Early Stage of Liver Fibrosis

Purpose. To compare the diagnostic value of texture analysis- (TA-) derived parameters from out-of-phase T1W, in-phase T1W, and T2W images in the classification of the early stage of liver fibrosis. Methods. Patients clinically diagnosed with hepatitis B infection, who underwent liver biopsy and noncontrast MRI scans, were enrolled. TA parameters were extracted from out-of-phase T1-weighted (T1W), in-phase T1W, and T2-weighted (T2W) images and calculated using Artificial Intelligent Kit (AK). Features were extracted including first-order, shape, gray-level cooccurrence matrix, gray-level run-length matrix, neighboring gray one tone difference matrix, and gray-level differential matrix. After statistical analyses, final diagnostic models were constructed. Receiver operating curves (ROCs) and areas under the ROC (AUCs) were used to assess the diagnostic value of each final model and 100-time repeated cross-validation was applied to assess the stability of the logistic regression models. Results. A total of 57 patients were enrolled in this study, with 27 in the fibrosis stage < 2 and 30 in stages ≥ 2. Overall, 851 features were extracted per ROI. Eight features with high correlation were selected by the maximum relevance method in each sequence, and all had a good diagnostic performance. ROC analysis of the final models showed that all sequences had a preferable performance with AUCs of 0.87, 0.90, and 0.96 in T2W and in-phase and out-of-phase T1W, respectively. Cross-validation results reported the following values of mean accuracy, specificity, and sensitivity: 0.98 each for out-of-phase T1W; 0.90, 0.89, and 0.90 for in-phase T1W; and 0.86, 0.88, 0.84 for T2W in the training set, and 0.76, 0.81, and 0.72 for out-of-phase T1W; 0.74, 0.72, and 0.75 for in-phase T1W; and 0.63, 0.64, and 0.63 for T2W for the test group, respectively. Conclusion. Noncontrast MRI scans with texture analysis are viable for classifying the early stages of liver fibrosis, exhibiting excellent diagnostic performance.

Influence of Magnetic Field Strength on Magnetic Resonance Imaging Radiomics Features in Brain Imaging, an In Vitro and In Vivo Study

BACKGROUND

The development and clinical adoption of quantitative imaging biomarkers (radiomics) has established the need for the identification of parameters altering radiomics reproducibility. The aim of this study was to assess the impact of magnetic field strength on magnetic resonance imaging (MRI) radiomics features in neuroradiology clinical practice.

METHODS

T1 3D SPGR sequence was acquired on two phantoms and 10 healthy volunteers with two clinical MR devices from the same manufacturer using two different magnetic fields (1.5 and 3T). Phantoms varied in terms of gadolinium concentrations and textural heterogeneity. 27 regions of interest were segmented (phantom: 21, volunteers: 6) using the LIFEX software. 34 features were analyzed.

RESULTS

In the phantom dataset, 10 (67%) out of 15 radiomics features were significantly different when measured at 1.5T or 3T (student's t-test, p < 0.05). Gray levels resampling, and pixel size also influence part of texture features. These findings were validated in healthy volunteers.

CONCLUSIONS

According to daily used protocols for clinical examinations, radiomic features extracted on 1.5T should not be used interchangeably with 3T when evaluating texture features. Such confounding factor should be adjusted when adapting the results of a study to a different platform, or when designing a multicentric trial.

Radiomics analysis of 18F-Choline PET/CT in the prediction of disease outcome in high-risk prostate cancer: an explorative study on machine learning feature classification in 94 patients

OBJECTIVE

The aim of this study was (1) to investigate the application of texture analysis of choline PET/CT images in prostate cancer (PCa) patients and (2) to propose a machine-learning radiomics model able to select PET features predictive of disease progression in PCa patients with a same high-risk class at restaging.

MATERIAL AND METHODS

Ninety-four high-risk PCa patients who underwent restaging Cho-PET/CT were analyzed. Follow-up data were recorded for a minimum of 13 months after the PET/CT scan. PET images were imported in LIFEx toolbox to extract 51 features from each lesion. A statistical system based on correlation matrix and point-biserial-correlation coefficient has been implemented for features reduction and selection, while Discriminant analysis (DA) was used as a method for features classification in a whole sample and sub-groups for primary tumor or local relapse (T), nodal disease (N), and metastatic disease (M).

RESULTS

In the whole group, 2 feature (HISTO_Entropy_log10; HISTO_Energy_Uniformity) results were able to discriminate the occurrence of disease progression at follow-up, obtaining the best performance in DA classification (sensitivity 47.1%, specificity 76.5%, positive predictive value (PPV) 46.7%, and accuracy 67.6%). In the sub-group analysis, the best performance in DA classification for T was obtained by selecting 3 features (SUVmin; SHAPE_Sphericity; GLCM_Correlation) with a sensitivity of 91.6%, specificity 84.1%, PPV 79.1%, and accuracy 87%; for N by selecting 2 features (HISTO = _Energy Uniformity; GLZLM_SZLGE) with a sensitivity of 68.1%, specificity 91.4%, PPV 83%, and accuracy 82.6%; and for M by selecting 2 features (HISTO_Entropy_log10 - HISTO_Entropy_log2) with a sensitivity 64.4%, specificity 74.6%, PPV 40.6%, and accuracy 72.5%.

CONCLUSION

This machine learning model demonstrated to be feasible and useful to select Cho-PET features for T, N, and M with valuable association with high-risk PCa patients' outcomes.

KEY POINTS

• Artificial intelligence applications are feasible and useful to select Cho-PET features. • Our model demonstrated the presence of specific features for T, N, and M with valuable association with high-risk PCa patients' outcomes. • Further prospective studies are necessary to confirm our results and to develop the application of artificial intelligence in PET imaging of PCa.

Deep Convolutional Neural Network-Assisted Feature Extraction for Diagnostic Discrimination and Feature Visualization in Pancreatic Ductal Adenocarcinoma (PDAC) versus Autoimmune Pancreatitis (AIP)

The differentiation of autoimmune pancreatitis (AIP) and pancreatic ductal adenocarcinoma (PDAC) poses a relevant diagnostic challenge and can lead to misdiagnosis and consequently poor patient outcome. Recent studies have shown that radiomics-based models can achieve high sensitivity and specificity in predicting both entities. However, radiomic features can only capture low level representations of the input image. In contrast, convolutional neural networks (CNNs) can learn and extract more complex representations which have been used for image classification to great success. In our retrospective observational study, we performed a deep learning-based feature extraction using CT-scans of both entities and compared the predictive value against traditional radiomic features. In total, 86 patients, 44 with AIP and 42 with PDACs, were analyzed. Whole pancreas segmentation was automatically performed on CT-scans during the portal venous phase. The segmentation masks were manually checked and corrected if necessary. In total, 1411 radiomic features were extracted using PyRadiomics and 256 features (deep features) were extracted using an intermediate layer of a convolutional neural network (CNN). After feature selection and normalization, an extremely randomized trees algorithm was trained and tested using a two-fold shuffle-split cross-validation with a test sample of 20% (n = 18) to discriminate between AIP or PDAC. Feature maps were plotted and visual difference was noted. The machine learning (ML) model achieved a sensitivity, specificity, and ROC-AUC of 0.89 ± 0.11, 0.83 ± 0.06, and 0.90 ± 0.02 for the deep features and 0.72 ± 0.11, 0.78 ± 0.06, and 0.80 ± 0.01 for the radiomic features. Visualization of feature maps indicated different activation patterns for AIP and PDAC. We successfully trained a machine learning model using deep feature extraction from CT-images to differentiate between AIP and PDAC. In comparison to traditional radiomic features, deep features achieved a higher sensitivity, specificity, and ROC-AUC. Visualization of deep features could further improve the diagnostic accuracy of non-invasive differentiation of AIP and PDAC.

Radiomics Feature Activation Maps as a New Tool for Signature Interpretability

Introduction

In the field of personalized medicine, radiomics has shown its potential to support treatment decisions. However, the limited feature interpretability hampers its introduction into the clinics. Here, we propose a new methodology to create radiomics feature activation maps, which allows to identify the spatial-anatomical locations responsible for signature activation based on local radiomics. The feasibility of this technique will be studied for histological subtype differentiation (adenocarcinoma versus squamous cell carcinoma) in non-small cell lung cancer (NSCLC) using computed tomography (CT) radiomics.

Materials and Methods

Pre-treatment CT scans were collected from a multi-centric Swiss trial (training, n=73, IIIA/N2 NSCLC, SAKK 16/00) and an independent cohort (validation, n=32, IIIA/N2/IIIB NSCLC). Based on the gross tumor volume (GTV), four peritumoral region of interests (ROI) were defined: lung_exterior (expansion into the lung), iso_exterior (expansion into lung and soft tissue), gradient (GTV border region), GTV+Rim (GTV and iso_exterior). For each ROI, 154 radiomic features were extracted using an in-house developed software implementation (Z-Rad, Python v2.7.14). Features robust against delineation variability served as an input for a multivariate logistic regression analysis. Model performance was quantified using the area under the receiver operating characteristic curve (AUC) and verified using five-fold cross validation and internal validation. Local radiomic features were extracted from the GTV+Rim ROI using non-overlapping 3x3x3 voxel patches previously marked as GTV or rim. A binary activation map was created for each patient using the median global feature value from the training. The ratios of activated/non-activated patches of GTV and rim regions were compared between histological subtypes (Wilcoxon test).

Results

Iso_exterior, gradient, GTV+Rim showed good performances for histological subtype prediction (AUC_training=0.68-0.72 and AUC_validation=0.73-0.74) whereas GTV and lung_exterior models failed validation. GTV+Rim model feature activation maps showed that local texture feature distribution differed significantly between histological subtypes in the rim (p=0.0481) but not in the GTV (p=0.461).

Conclusion

In this exploratory study, radiomics-based prediction of NSCLC histological subtypes was predominantly based on the peritumoral region indicating that radiomics activation maps can be useful for tracing back the spatial location of regions responsible for signature activation.

Radiomics of Liver Metastases: A Systematic Review

Simple Summary

Patients with liver metastases can be scheduled for different therapies (e.g., chemotherapy, surgery, radiotherapy, and ablation). The choice of the most appropriate treatment should rely on adequate understanding of tumor biology and prediction of survival, but reliable biomarkers are lacking. Radiomics is an innovative approach to medical imaging: it identifies invisible-to-the-human-eye radiological patterns that can predict tumor aggressiveness and patients outcome. We reviewed the available literature to elucidate the role of radiomics in patients with liver metastases. Thirty-two papers were analyzed, mostly (56%) concerning metastases from colorectal cancer. Even if available studies are still preliminary, radiomics provided effective prediction of response to chemotherapy and of survival, allowing more accurate and earlier prediction than standard predictors. Entropy and homogeneity were the radiomic features with the strongest clinical impact. In the next few years, radiomics is expected to give a consistent contribution to the precision medicine approach to patients with liver metastases.

Abstract

Multidisciplinary management of patients with liver metastases (LM) requires a precision medicine approach, based on adequate profiling of tumor biology and robust biomarkers. Radiomics, defined as the high-throughput identification, analysis, and translational applications of radiological textural features, could fulfill this need. The present review aims to elucidate the contribution of radiomic analyses to the management of patients with LM. We performed a systematic review of the literature through the most relevant databases and web sources. English language original articles published before June 2020 and concerning radiomics of LM extracted from CT, MRI, or PET-CT were considered. Thirty-two papers were identified. Baseline higher entropy and lower homogeneity of LM were associated with better survival and higher chemotherapy response rates. A decrease in entropy and an increase in homogeneity after chemotherapy correlated with radiological tumor response. Entropy and homogeneity were also highly predictive of tumor regression grade. In comparison with RECIST criteria, radiomic features provided an earlier prediction of response to chemotherapy. Lastly, texture analyses could differentiate LM from other liver tumors. The commonest limitations of studies were small sample size, retrospective design, lack of validation datasets, and unavailability of univocal cut-off values of radiomic features. In conclusion, radiomics can potentially contribute to the precision medicine approach to patients with LM, but interdisciplinarity, standardization, and adequate software tools are needed to translate the anticipated potentialities into clinical practice.

Efficacy of ZOOMit coronal diffusion-weighted imaging and MR texture analysis for differentiating between benign and malignant distal bile duct strictures

PURPOSE

To investigate the diagnostic efficacy of ZOOMit coronal diffusion-weighted imaging (Z-DWI) and MR texture analysis (MRTA) for differentiating benign from malignant distal bile duct strictures.

METHODS

We retrospectively enrolled a total of 71 patients with distal bile duct stricture who underwent magnetic resonance cholangiopancreatography (MRCP). For quantitative analysis, the average apparent diffusion coefficient (ADC) value at suspected stricture sites was assessed on both Z-DWI and conventional DWI (C-DWI). For qualitative analysis, two reviewers independently reviewed two image sets containing different diffusion-weighted images, and receiver operating characteristic (ROC) curve analysis was performed. Several MRTA parameters were extracted from the area of the stricture on the ADC map of the ZOOMit coronal diffusion-weighted images using commercially available software.

RESULTS

Among 71 patients, 26 patients were diagnosed with malignant stricture. On quantitative analysis, the average ADC value of the malignant and benign strictures, using Z-DWI, was 1.124 × 10^-3 mm²/s and 1.522 × 10^-3 mm²/s, respectively (P < 0.001). The average ADC value of the malignant and benign strictures, using C-DWI, was 1.107 × 10^-3 mm²/s and 1.519 × 10^-3 mm²/s, respectively (P < 0.001). On qualitative analysis, for each reviewer, the area under the ROC curve (AUC) values for differentiating benign from malignant stricture was 0.928 and 0.939, respectively, for the ZOOMit diffusion set and 0.851 and 0.824, respectively, for the conventional diffusion set. Multiple MRTA parameters showed a significantly different distribution for the benign and malignant strictures, including mean, entropy, mean of positive pixels, and kurtosis at spatial filtration values of 0, 5, and 6 mm.

CONCLUSION

The addition of Z-DWI to conventional MRCP is helpful in differentiating benign from malignant bile duct strictures, and some MRTA parameters also can be helpful in differentiating benign from malignant distal bile duct strictures.

Entropy and uniformity as additional parameters to optimize the effectiveness of bone CT in the evaluation of acromegalic patients

PURPOSE

Acromegaly is considered an important cause of secondary osteoporosis. However, studies on bone mineral density (BMD) have yielded conflicting results and there are few studies that evaluate an accurate imaging method for early diagnosis of osteoporosis in these patients. The objective of this study was to assess whether entropy and uniformity on computed tomography (CT) scans are useful parameters for optimization of assessment of bone fragility in patients with acromegaly.

METHODS

We included 34 patients and 36 controls matched for age and sex in a cross-sectional study. Patients and controls underwent CT scan of the lumbosacral spine, dual-energy x-ray absorptiometry (DXA) and blood tests. A software was developed to calculate the entropy and uniformity by a region of interest (ROI) of the trabecular bone of the first lumbar vertebra (L1).

RESULTS

The acromegalic group presented higher mean bone entropy (6.87 ± 0.98 vs. 6.03 ± 1.68, p = 0.013) and lower mean bone uniformity (0.035 ± 0.704 vs. 0.113 ± 0.205, p = 0.035) than control group. Analyzing only acromegalics, mean bone entropy was higher and bone uniformity was lower in patients with hypogonadism than patients without hypogonadism (7.28 ± 0.36 vs. 6.74 ± 1.08, p = 0.038 and 0.008 ± 0.002 vs. 0.043 ± 0.079, p = 0.031) respectively. Patients with acromegaly presented higher BMD and Z-score in the femoral neck than control group (1.156 ± 0.108 vs. 0.925 ± 0.326 g/cm², p = 0.043 and 0.6 ± 0.6 vs. -0.05 ± 0.8, p = 0.041, respectively). Entropy was negatively correlated with T-score of the lumbar spine (r_p = -0.357, p = 0.033) in control group and uniformity was positively correlated with T-score of the lumbar spine, neck, and total hip, respectively (r_p = 0.371, p = 0.031; r_p = 0.348, p = 0.043 and r_p = 0.341, p = 0.049) in acromegalic group.

CONCLUSIONS

The study identified that entropy and uniformity are a relevant parameters data in bone fragility assessment in acromegalic patients.

Template Creation for High-Resolution Computed Tomography Scans of the Lung in R Software

Rationale and Objectives.

A standard lung template could improve population-level analyses for computed tomography (CT) scans of the lung. We develop a fully-automated pre-processing pipeline for image analysis of the lungs using updated methodologies and R software that results in the creation of a standard lung template. We apply this pipeline to CT scans from a sarcoidosis population, exploring the influence of registration on radiomic analyses.

Materials and Methods.

Using 65 high-resolution CT scans from healthy adults, we create a standard lung template by segmenting the left and right lungs, non-linearly registering lung masks to an initial template mask, and using an unbiased, iterative procedure to converge to a standard lung shape (Dice similarity coefficient ≥0.99). We compare three-dimensional radiomic features between control and sarcoidosis patients, before and after registration to a study-specific lung template.

Results.

The final lung template had a right lung volume of 2967 cm³ and left lung volume of 2623 cm³, with a median HU = −862. Registration significantly affected radiomic features, shifting the HU distribution to the left, decreasing variability, and increasing smoothness (p<0.0001). The registration improved detective ability of radiomics; for contrast, autocorrelation, energy and homogeneity, the group effect was significant post-registration (p<0.05), but was not significant pre-registration.

Conclusion.

The final lung template and software used for its creation are publicly available via the lungct R package to facilitate its use in practice. This study advances lung imaging by developing tools to improve population-level analyses for various lung diseases.

Reinventing radiation therapy with machine learning and imaging bio-markers (radiomics): State-of-the-art, challenges and perspectives

Radiation therapy is a pivotal cancer treatment that has significantly progressed over the last decade due to numerous technological breakthroughs. Imaging is now playing a critical role on deployment of the clinical workflow, both for treatment planning and treatment delivery. Machine-learning analysis of predefined features extracted from medical images, i.e. radiomics, has emerged as a promising clinical tool for a wide range of clinical problems addressing drug development, clinical diagnosis, treatment selection and implementation as well as prognosis. Radiomics denotes a paradigm shift redefining medical images as a quantitative asset for data-driven precision medicine. The adoption of machine-learning in a clinical setting and in particular of radiomics features requires the selection of robust, representative and clinically interpretable biomarkers that are properly evaluated on a representative clinical data set. To be clinically relevant, radiomics must not only improve patients' management with great accuracy but also be reproducible and generalizable. Hence, this review explores the existing literature and exposes its potential technical caveats, such as the lack of quality control, standardization, sufficient sample size, type of data collection, and external validation. Based upon the analysis of 165 original research studies based on PET, CT-scan, and MRI, this review provides an overview of new concepts, and hypotheses generating findings that should be validated. In particular, it describes evolving research trends to enhance several clinical tasks such as prognostication, treatment planning, response assessment, prediction of recurrence/relapse, and prediction of toxicity. Perspectives regarding the implementation of an AI-based radiotherapy workflow are presented.

MRI texture features differentiate clinicopathological characteristics of cervical carcinoma

OBJECTIVES

To evaluate MRI texture analysis in differentiating clinicopathological characteristics of cervical carcinoma (CC).

METHODS

Patients with newly diagnosed CC who underwent pre-treatment MRI were retrospectively reviewed. Texture analysis was performed using commercial software (TexRAD). Largest single-slice ROIs were manually drawn around the tumour on T2-weighted (T2W) images, apparent diffusion coefficient (ADC) maps and contrast-enhanced T1-weighted (T1c) images. First-order texture features were calculated and compared among histological subtypes, tumour grades, FIGO stages and nodal status using the Mann-Whitney U test. Feature selection was achieved by elastic net. Selected features from different sequences were used to build the multivariable support vector machine (SVM) models and the performances were assessed by ROC curves and AUC.

RESULTS

Ninety-five patients with FIGO stage IB~IVB were evaluated. A number of texture features from multiple sequences were significantly different among all the clinicopathological subgroups (p < 0.05). Texture features from different sequences were selected to build the SVM models. The AUCs of SVM models for discriminating histological subtypes, tumour grades, FIGO stages and nodal status were 0.841, 0.850, 0.898 and 0.879, respectively.

CONCLUSIONS

Texture features derived from multiple sequences were helpful in differentiating the clinicopathological signatures of CC. The SVM models with selected features from different sequences offered excellent diagnostic discrimination of the tumour characteristics in CC.

KEY POINTS

• First-order texture features are able to differentiate clinicopathological signatures of cervical carcinoma. • Combined texture features from different sequences can offer excellent diagnostic discrimination of the tumour characteristics in cervical carcinoma.

IDH mutation-specific radiomic signature in lower-grade gliomas

Unravelling the heterogeneity is the central challenge for glioma precession oncology. In this study, we extracted quantitative image features from T2-weighted MR images and revealed that the isocitrate dehydrogenase (IDH) wild type and mutant lower grade gliomas (LGGs) differed in their expression of 146 radiomic descriptors. The logistic regression model algorithm further reduced these to 86 features. The classification model could discriminate the two types in both the training and validation sets with area under the curve values of 1.0000 and 0.9932, respectively. The transcriptome-radiomic analysis revealed that these features were associated with the immune response, biological adhesion, and several malignant behaviors, all of which are consistent with biological processes that are differentially expressed in IDH wild type and IDH mutant LGGs. Finally, a prognostic signature showed an ability to stratify IDH mutant LGGs into high and low risk groups with distinctive outcomes. By extracting a large number of radiomic features, we identified an IDH mutation-specific radiomic signature with prognostic implications. This radiomic signature may provide a way to non-invasively discriminate lower-grade gliomas as with or without the IDH mutation.

Multiparametric MRI for Prostate Cancer Detection: New Insights into the Combined Use of a Radiomic Approach with Advanced Acquisition Protocol

Prostate cancer (PCa) is a disease affecting an increasing number of men worldwide. Several efforts have been made to identify imaging biomarkers to non-invasively detect and characterize PCa, with substantial improvements thanks to multiparametric Magnetic Resonance Imaging (mpMRI). In recent years, diffusion kurtosis imaging (DKI) was proposed to be directly related to tissue physiological and pathological characteristic, while the radiomic approach was proven to be a key method to study cancer imaging phenotypes. Our aim was to compare a standard radiomic model for PCa detection, built using T2-weighted (T2W) and Apparent Diffusion Coefficient (ADC), with an advanced one, including DKI and quantitative Dynamic Contrast Enhanced (DCE), while also evaluating differences in prediction performance when using 2D or 3D lesion segmentation. The obtained results in terms of diagnostic accuracy were high for all of the performed comparisons, reaching values up to 0.99 for the area under a receiver operating characteristic curve (AUC), and 0.98 for both sensitivity and specificity. In comparison, the radiomic model based on standard features led to prediction performances higher than those of the advanced model, while greater accuracy was achieved by the model extracted from 3D segmentation. These results provide new insights into active topics of discussion, such as choosing the most convenient acquisition protocol and the most appropriate postprocessing pipeline to accurately detect and characterize PCa.

Prediction of tumor grade and lymphovascular space invasion in endometrial adenocarcinoma with MR imaging-based radiomic analysis

PURPOSE

To evaluate the capabilities of two-dimensional magnetic resonance imaging (MRI)-based texture analysis features, tumor volume, tumor short axis and apparent diffusion coefficient (ADC) in predicting histopathological high-grade and lymphovascular space invasion (LVSI) in endometrial adenocarcinoma.

MATERIALS AND METHODS

Seventy-three women (mean age: 66±11.5 [SD] years; range: 45-88 years) with endometrial adenocarcinoma who underwent MRI of the pelvis at 1.5-T before hysterectomy were retrospectively included. Texture analysis was performed using TexRAD® software on T2-weighted images and ADC maps. Primary outcomes were high-grade and LVSI prediction using histopathological analysis as standard of reference. After data reduction using ascending hierarchical classification analysis, a predictive model was obtained by stepwise multivariate logistic regression and performances were assessed using cross-validated receiver operator curve (ROC).

RESULTS

A total of 72 texture features per tumor were computed. Texture model yielded 52% sensitivity and 75% specificity for the diagnosis of high-grade tumor (areas under ROC curve [AUC]=0.64) and 71% sensitivity and 59% specificity for the diagnosis of LVSI (AUC=0.59). Volumes and tumor short axis were greater for high-grade tumors (P=0.0002 and P=0.004, respectively) and for patients with LVSI (P=0.004 and P=0.0279, respectively). No differences in ADC values were found between high-grade and low-grade tumors and for LVSI. A tumor short axis≥20mm yielded 95% sensitivity and 75% specificity for the diagnosis of high-grade tumor (AUC=0.86).

CONCLUSION

MRI-based texture analysis is of limited value to predict high grade and LVSI of endometrial adenocarcinoma. A tumor short axis≥20mm is the best predictor of high grade and LVSI.

Apparent diffusion coefficient values and non-homogeneity of diffusion in brain tumors in diffusion-weighted MRI

Background

The values that have been received from apparent diffusion coefficient (ADC) maps of diffusion-weighted magnetic resonance imaging (DW-MRI) might play a vital role in evaluating tumors and their grading scale.

Purpose

To investigate the predictive role of this heterogeneity in brain tumor pathologies and its correlation with Ki-67.

Material and Methods

A total of 124 patients with brain tumors underwent brain MRI with gadolinium injection. ADC and standard deviation of each lesion have been obtained from manual localization of the region of interest on the ADC map. A receiver operating characteristic analysis was conducted to determine the minimum cut-off values of the mean ADC and mean standard deviation of ADC maps having the highest sensitivity and specificity to differentiate high-grade and low-grade tumors.

Results

Mean ADC values in the region of interest were significantly lower for malignant tumors (grade IV and metastasis) than grade I brain tumors, while a higher mean standard deviation was observed. In a more detailed comparison of tumor groups, the mean standard deviation of the ADC for glioblastoma multiform was significantly higher than meningioma grade I ( P < 0.001) and metastasis was significantly higher than grade III and IV astrocytic tumors ( P = 0.004). The analysis of Ki-67 proliferation index and mean ADC values in gliomas showed a significant inverse correlation between the parameters (r = –0.0429, P < 0.001) and direct correlation between Ki-67 and mean standard deviation of the ADC (r = 0.551, P < 0.001). As an index for the ADC to differentiate high-grade and low-grade tumors, the cut-off values of 1.40*10−3 mm2/s for mean ADC and 45*10−3 mm2/s for mean standard deviation have the highest combination of sensitivity, specificity, and area under the curve.

Conclusion

The mean value and standard deviation of the ADC could be considered for differentiating between low-grade and high-grade brain tumors, as two available non-invasive methods.

Automated Identification of Optimal Portal Venous Phase Timing with Convolutional Neural Networks

OBJECTIVES

To develop a deep learning-based algorithm to automatically identify optimal portal venous phase timing (PVP-timing) so that image analysis techniques can be accurately performed on post contrast studies.

METHODS

681 CT-scans (training: 479 CT-scans; validation: 202 CT-scans) from a multicenter clinical trial in patients with liver metastases from colorectal cancer were retrospectively analyzed for algorithm development and validation. An additional external validation was performed on a cohort of 228 CT-scans from gastroenteropancreatic neuroendocrine cancer patients. Image acquisition was performed according to each centers' standard CT protocol for single portal venous phase, portal venous acquisition. The reference gold standard for the classification of PVP-timing as either optimal or nonoptimal was based on experienced radiologists' consensus opinion. The algorithm performed automated localization (on axial slices) of the portal vein and aorta upon which a novel dual input Convolutional Neural Network calculated a probability of the optimal PVP-timing.

RESULTS

The algorithm automatically computed a PVP-timing score in 3 seconds and reached area under the curve of 0.837 (95% CI: 0.765, 0.890) in validation set and 0.844 (95% CI: 0.786, 0.889) in external validation set.

CONCLUSION

A fully automated, deep-learning derived PVP-timing algorithm was developed to classify scans' contrast-enhancement timing and identify scans with optimal PVP-timing. The rapid identification of such scans will aid in the analysis of quantitative (radiomics) features used to characterize tumors and changes in enhancement with treatment in a multitude of settings including quantitative response criteria such as Choi and MASS which rely on reproducible measurement of enhancement.

Pancreatic ductal adenocarcinoma: a radiomics nomogram outperforms clinical model and TNM staging for survival estimation after curative resection

OBJECTIVES

To identify a CT-based radiomics nomogram for survival prediction in patients with resected pancreatic ductal adenocarcinoma (PDAC).

METHODS

A total of 220 patients (training cohort n = 147; validation cohort n = 73) with PDAC were enrolled. A total of 300 radiomics features were extracted from CT images. And the least absolute shrinkage and selection operator algorithm were applied to select features and develop a radiomics score (Rad-score). The radiomics nomogram was constructed by multivariate regression analysis. Nomogram discrimination, calibration, and clinical usefulness were evaluated. The association of the Rad-score and recurrence pattern in PDAC was evaluated.

RESULTS

The Rad-score was significantly associated with PDAC patient's disease-free survival (DFS) and overall survival (OS) (both p < 0.001 in two cohorts). Incorporating the Rad-score into the radiomics nomogram resulted in better performance of the survival prediction than that of the clinical model and TNM staging system. In addition, the radiomics nomogram exhibited good discrimination, calibration, and clinical usefulness in both the training and validation cohorts. There was no association between the Rad-score and recurrence pattern.

CONCLUSIONS

The radiomics nomogram integrating the Rad-score and clinical data provided better prognostic prediction in resected PDAC patients, which may hold great potential for guiding personalized care for these patients. The Rad-score was not a predictor of the recurrence pattern in resected PDAC patients.

KEY POINTS

• The Rad-score developed by CT radiomics features was significantly associated with PDAC patients' prognosis. • The radiomics nomogram integrating the Rad-score and clinical data has value to permit non-invasive, low-cost, and personalized evaluation of prognosis in PDAC patients. • The radiomics nomogram outperformed clinical model and the TNM staging system in terms of survival estimation.

Deep Learning to Distinguish Benign from Malignant Renal Lesions Based on Routine MR Imaging

PURPOSE

With increasing incidence of renal mass, it is important to make a pretreatment differentiation between benign renal mass and malignant tumor. We aimed to develop a deep learning model that distinguishes benign renal tumors from renal cell carcinoma (RCC) by applying a residual convolutional neural network (ResNet) on routine MR imaging.

EXPERIMENTAL DESIGN

Preoperative MR images (T2-weighted and T1-postcontrast sequences) of 1,162 renal lesions definitely diagnosed on pathology or imaging in a multicenter cohort were divided into training, validation, and test sets (70:20:10 split). An ensemble model based on ResNet was built combining clinical variables and T1C and T2WI MR images using a bagging classifier to predict renal tumor pathology. Final model performance was compared with expert interpretation and the most optimized radiomics model.

RESULTS

Among the 1,162 renal lesions, 655 were malignant and 507 were benign. Compared with a baseline zero rule algorithm, the ensemble deep learning model had a statistically significant higher test accuracy (0.70 vs. 0.56, P = 0.004). Compared with all experts averaged, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.60, P = 0.053), sensitivity (0.92 vs. 0.80, P = 0.017), and specificity (0.41 vs. 0.35, P = 0.450). Compared with the radiomics model, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.62, P = 0.081), sensitivity (0.92 vs. 0.79, P = 0.012), and specificity (0.41 vs. 0.39, P = 0.770).

CONCLUSIONS

Deep learning can noninvasively distinguish benign renal tumors from RCC using conventional MR imaging in a multi-institutional dataset with good accuracy, sensitivity, and specificity comparable with experts and radiomics.

Radiomics machine-learning signature for diagnosis of hepatocellular carcinoma in cirrhotic patients with indeterminate liver nodules

PURPOSE

To enhance clinician's decision-making by diagnosing hepatocellular carcinoma (HCC) in cirrhotic patients with indeterminate liver nodules using quantitative imaging features extracted from triphasic CT scans.

MATERIAL AND METHODS

We retrospectively analyzed 178 cirrhotic patients from 27 institutions, with biopsy-proven liver nodules classified as indeterminate using the European Association for the Study of the Liver (EASL) guidelines. Patients were randomly assigned to a discovery cohort (142 patients (pts.)) and a validation cohort (36 pts.). Each liver nodule was segmented on each phase of triphasic CT scans, and 13,920 quantitative imaging features (12 sets of 1160 features each reflecting the phenotype at one single phase or its change between two phases) were extracted. Using machine-learning techniques, the signature was trained and calibrated (discovery cohort), and validated (validation cohort) to classify liver nodules as HCC vs. non-HCC. Effects of segmentation and contrast enhancement quality were also evaluated.

RESULTS

Patients were predominantly male (88%) and CHILD A (65%). Biopsy was positive for HCC in 77% of patients. LI-RADS scores were not different between HCC and non-HCC patients. The signature included a single radiomics feature quantifying changes between arterial and portal venous phases: _DeltaV-A_{_}DWT1_LL_Variance-2D and reached area under the receiver operating characteristic curve (AUC) of 0.70 (95%CI 0.61-0.80) and 0.66 (95%CI 0.64-0.84) in discovery and validation cohorts, respectively. The signature was influenced neither by segmentation nor by contrast enhancement.

CONCLUSION

A signature using a single feature was validated in a multicenter retrospective cohort to diagnose HCC in cirrhotic patients with indeterminate liver nodules. Artificial intelligence could enhance clinicians' decision by identifying a subgroup of patients with high HCC risk.

KEY POINTS

• In cirrhotic patients with visually indeterminate liver nodules, expert visual assessment using current guidelines cannot accurately differentiate HCC from differential diagnoses. Current clinical protocols do not entail biopsy due to procedural risks. Radiomics can be used to non-invasively diagnose HCC in cirrhotic patients with indeterminate liver nodules, which could be leveraged to optimize patient management. • Radiomics features contributing the most to a better characterization of visually indeterminate liver nodules include changes in nodule phenotype between arterial and portal venous phases: the "washout" pattern appraised visually using EASL and EASL guidelines. • A clinical decision algorithm using radiomics could be applied to reduce the rate of cirrhotic patients requiring liver biopsy (EASL guidelines) or wait-and-see strategy (AASLD guidelines) and therefore improve their management and outcome.

Is tumour sphericity an important prognostic factor in patients with lung cancer?

BACKGROUND AND PURPOSE

Quantitative tumour shape features extracted from radiotherapy planning scans have shown potential as prognostic markers. In this study, we investigated if sphericity of the gross tumour volume (GTV) on planning computed tomography (CT) is an independent predictor of overall survival (OS) in lung cancer patients treated with standard radiotherapy. In the analysis, we considered whether tumour sphericity is correlated with clinical prognostic factors or influenced by the inclusion of lymph nodes in the GTV.

MATERIALS AND METHODS

Sphericity of single GTV delineation was extracted for 457 lung cancer patients. Relationships between sphericity, and common patient and tumour characteristics were investigated via correlation analysis and multivariate Cox regression to assess prognostic value of GTV sphericity. A subset analysis was performed for 290 nodal stage N0 patients to determine prognostic value of primary tumour sphericity.

RESULTS

Sphericity is correlated with clinical variables: tumour volume, mean lung dose, N stage, and T stage. Sphericity is strongly associated with OS (p < 0.001, hazard ratio (HR) (95% confidence interval (CI)) = 0.13 (0.04-0.41)) in univariate analysis. However, this association did not remain significant in multivariate analysis (p = 0.826, HR (95% CI) = 0.83 (0.16-4.31), and inclusion of sphericity to a clinical model did not improve model performance. In addition, no significant relationship between sphericity and OS was detected in univariate (p = 0.072) or multivariate (p = 0.920) analysis of N0 subset.

CONCLUSION

Sphericity correlates clearly with clinical prognostic factors, which are often unaccounted for in radiomic studies. Sphericity is also influenced by the presence of nodal involvement within the GTV contour.

Prediction of Treatment Response to Neoadjuvant Chemotherapy for Breast Cancer via Early Changes in Tumor Heterogeneity Captured by DCE-MRI Registration

We analyzed DCE-MR images from 132 women with locally advanced breast cancer from the I-SPY1 trial to evaluate changes of intra-tumor heterogeneity for augmenting early prediction of pathologic complete response (pCR) and recurrence-free survival (RFS) after neoadjuvant chemotherapy (NAC). Utilizing image registration, voxel-wise changes including tumor deformations and changes in DCE-MRI kinetic features were computed to characterize heterogeneous changes within the tumor. Using five-fold cross-validation, logistic regression and Cox regression were performed to model pCR and RFS, respectively. The extracted imaging features were evaluated in augmenting established predictors, including functional tumor volume (FTV) and histopathologic and demographic factors, using the area under the curve (AUC) and the C-statistic as performance measures. The extracted voxel-wise features were also compared to analogous conventional aggregated features to evaluate the potential advantage of voxel-wise analysis. Voxel-wise features improved prediction of pCR (AUC = 0.78 (±0.03) vs 0.71 (±0.04), p < 0.05 and RFS (C-statistic = 0.76 ( ± 0.05), vs 0.63 ( ± 0.01)), p < 0.05, while models based on analogous aggregate imaging features did not show appreciable performance changes (p > 0.05). Furthermore, all selected voxel-wise features demonstrated significant association with outcome (p < 0.05). Thus, precise measures of voxel-wise changes in tumor heterogeneity extracted from registered DCE-MRI scans can improve early prediction of neoadjuvant treatment outcomes in locally advanced breast cancer.

Impact of Variability in Portal Venous Phase Acquisition Timing in Tumor Density Measurement and Treatment Response Assessment: Metastatic Colorectal Cancer as a Paradigm

PURPOSE

New response patterns to anticancer drugs have led tumor size-based response criteria to shift to also include density measurements. Choi criteria, for instance, categorize antiangiogenic therapy response as a decrease in tumor density > 15% at the portal venous phase (PVP). We studied the effect that PVP timing has on measurement of the density of liver metastases (LM) from colorectal cancer (CRC).

METHODS

Pretreatment PVP computed tomography images from 291 patients with LM-CRC from the CRYSTAL trial (Cetuximab Combined With Irinotecan in First-Line Therapy for Metastatic Colorectal Cancer; ClinicalTrials.gov identifier: NCT00154102) were included. Four radiologists independently scored the scans' timing according to a three-point scoring system: early, optimal, late PVP. Using this, we developed, by machine learning, a proprietary computer-aided quality-control algorithm to grade PVP timing. The reference standard was a computer-refined consensus. For each patient, we contoured target liver lesions and calculated their mean density.

RESULTS

Contrast-product administration data were not recorded in the digital imaging and communications in medicine headers for injection volume (94%), type (93%), and route (76%). The PVP timing was early, optimal, and late in 52, 194, and 45 patients, respectively. The mean (95% CI) accuracy of the radiologists for detection of optimal PVP timing was 81.7% (78.3 to 85.2) and was outperformed by the 88.6% (84.8 to 92.4) computer accuracy. The mean ± standard deviation of LM-CRC density was 68 ± 15 Hounsfield units (HU) overall and 59.5 ± 14.9 HU, 71.4 ± 14.1 HU, 62.4 ± 12.5 HU at early, optimal, and late PVP timing, respectively. LM-CRC density was thus decreased at nonoptimal PVP timing by 14.8%: 16.7% at early PVP ( P < .001) and 12.6% at late PVP ( P < .001).

CONCLUSION

Nonoptimal PVP timing should be identified because it significantly decreased tumor density by 14.8%. Our computer-aided quality-control system outperformed the accuracy, reproducibility, and speed of radiologists' visual scoring. PVP-timing scoring could improve the extraction of tumor quantitative imaging biomarkers and the monitoring of anticancer therapy efficacy at the patient and clinical trial levels.

The role of multimodal imaging in guiding resectability and cytoreduction in pancreatic neuroendocrine tumors: focus on PET and MRI

Pancreatic neuroendocrine tumors (pNETs) are rare neoplasms that secrete peptides and neuro-amines. pNETs can be sporadic or hereditary, syndromic or non-syndromic with different clinical presentations and prognoses. The role of medical imaging includes locating the tumor, assessing its extent, and evaluating the feasibility of curative surgery or cytoreduction. Pancreatic NETs have very distinctive phenotypes on CT, MRI, and PET. PET have been demonstrated to be very sensitive to detect either well-differentiated pNETs using ⁶⁸Gallium somatostatin receptor (SSTR) radiotracers, or more aggressive undifferentiated pNETS using ¹⁸F-FDG. A comprehensive interpretation of multimodal imaging guides resectability and cytoreduction in pNETs. The imaging phenotype provides information on the differentiation and proliferation of pNETs, as well as the spatial and temporal heterogeneity of tumors with prognostic and therapeutic implications. This review provides a structured approach for standardized reading and reporting of medical imaging studies with a focus on PET and MR techniques. It explains which imaging approach should be used for different subtypes of pNET and what a radiologist should be looking for and reporting when interpreting these studies.

NCTN Assessment on Current Applications of Radiomics in Oncology

Radiomics is a fast-growing research area based on converting standard-of-care imaging into quantitative minable data and building subsequent predictive models to personalize treatment. Radiomics has been proposed as a study objective in clinical trial concepts and a potential biomarker for stratifying patients across interventional treatment arms. In recognizing the growing importance of radiomics in oncology, a group of medical physicists and clinicians from NRG Oncology reviewed the current status of the field and identified critical issues, providing a general assessment and early recommendations for incorporation in oncology studies.