Peritumoral radiomics features predict distant metastasis in locally advanced NSCLC

The impact of inter-observer variation in delineation on robustness of radiomics features in non-small cell lung cancer

Artificial intelligence and radiomics have the potential to revolutionise cancer prognostication and personalised treatment. Manual outlining of the tumour volume for extraction of radiomics features (RF) is a subjective process. This study investigates robustness of RF to inter-observer variation (IOV) in contouring in lung cancer. We utilised two public imaging datasets: ‘NSCLC-Radiomics’ and ‘NSCLC-Radiomics-Interobserver1’ (‘Interobserver’). For ‘NSCLC-Radiomics’, we created an additional set of manual contours for 92 patients, and for ‘Interobserver’, there were five manual and five semi-automated contours available for 20 patients. Dice coefficients (DC) were calculated for contours. 1113 RF were extracted including shape, first order and texture features. Intraclass correlation coefficient (ICC) was computed to assess robustness of RF to IOV. Cox regression analysis for overall survival (OS) was performed with a previously published radiomics signature. The median DC ranged from 0.81 (‘NSCLC-Radiomics’) to 0.85 (‘Interobserver’—semi-automated). The median ICC for the ‘NSCLC-Radiomics’, ‘Interobserver’ (manual) and ‘Interobserver’ (semi-automated) were 0.90, 0.88 and 0.93 respectively. The ICC varied by feature type and was lower for first order and gray level co-occurrence matrix (GLCM) features. Shape features had a lower median ICC in the ‘NSCLC-Radiomics’ dataset compared to the ‘Interobserver’ dataset. Survival analysis showed similar separation of curves for three of four RF apart from ‘original_shape_Compactness2’, a feature with low ICC (0.61). The majority of RF are robust to IOV, with first order, GLCM and shape features being the least robust. Semi-automated contouring improves feature stability. Decreased robustness of a feature is significant as it may impact upon the features’ prognostic capability.

Radiomics for Detection of the EGFR Mutation in Liver Metastatic NSCLC

RATIONALE AND OBJECTIVES

The research aims to investigate whether MRI radiomics on hepatic metastasis from primary nonsmall cell lung cancer (NSCLC) can be used to differentiate patients with epidermal growth factor receptor (EGFR) mutations from those with EGFR wild-type, and develop a prediction model based on combination of primary tumor and the metastasis.

MATERIALS AND METHODS

A total of 130 patients were enrolled between Aug. 2017 and Dec. 2021, all pathologically confirmed harboring hepatic metastasis from primary NSCLC. The pyradiomics was used to extract radiomics features from intra- and peritumoral areas of both primary tumor and metastasis. The least absolute shrinkage and selection operator (LASSO) regression was applied to identify most predictive features and to develop radiomics signatures (RSs) for prediction of the EGFR mutation status. The receiver operating characteristic (ROC) curve analysis was performed to assess the prediction capability of the developed RSs.

RESULTS

A RS-Primary and a RS-Metastasis were derived from the primary tumor and metastasis, respectively. The RS-Combine by combination of the primary tumor and metastasis achieved the highest prediction performance in the training (AUCs, RS-Primary vs. RS-Metastasis vs. RS-Combine, 0.826 vs. 0.821 vs. 0.908) and testing (AUCs, RS-Primary vs. RS-Metastasis vs. RS-Combine, 0.760 vs. 0.791 vs. 0.884) set. The smoking status showed significant difference between EGFR mutant and wild-type groups (p < 0.05) in the training set.

CONCLUSION

The study indicates that hepatic metastasis-based radiomics can be used to detect the EGFR mutation. The developed multiorgan combined radiomics signature may be helpful to guide individual treatment strategies for patients with metastatic NSCLC.

Radiogenomic analysis reveals tumor heterogeneity of triple-negative breast cancer

Triple-negative breast cancer (TNBC) is a subset of breast cancer with an adverse prognosis and significant tumor heterogeneity. Here, we extract quantitative radiomic features from contrast-enhanced magnetic resonance images to construct a breast cancer radiomic dataset (n = 860) and a TNBC radiogenomic dataset (n = 202). We develop and validate radiomic signatures that can fairly differentiate TNBC from other breast cancer subtypes and distinguish molecular subtypes within TNBC. A radiomic feature that captures peritumoral heterogeneity is determined to be a prognostic factor for recurrence-free survival (p = 0.01) and overall survival (p = 0.004) in TNBC. Combined with the established matching TNBC transcriptomic and metabolomic data, we demonstrate that peritumoral heterogeneity is associated with immune suppression and upregulated fatty acid synthesis in tumor samples. Collectively, this multi-omic dataset serves as a useful public resource to promote precise subtyping of TNBC and helps to understand the biological significance of radiomics.

•

A radiomic signature identifies TNBC from other subtypes of breast cancer

•

Radiomic features are predictive of TNBC molecular subtypes

•

A radiomic feature reflecting peritumoral heterogeneity indicates TNBC prognosis

•

Peritumoral heterogeneity correlates with metabolic and immune abnormalities in TNBC

Deep Learning Radiomics Nomogram to Predict Lung Metastasis in Soft-Tissue Sarcoma: A Multi-Center Study

Objectives

To build and evaluate a deep learning radiomics nomogram (DLRN) for preoperative prediction of lung metastasis (LM) status in patients with soft tissue sarcoma (STS).

Methods

In total, 242 patients with STS (training set, n=116; external validation set, n=126) who underwent magnetic resonance imaging were retrospectively enrolled in this study. We identified independent predictors for LM-status and evaluated their performance. The minimum redundancy maximum relevance (mRMR) method and least absolute shrinkage and selection operator (LASSO) algorithm were adopted to screen radiomics features. Logistic regression, decision tree, random forest, support vector machine (SVM), and adaptive boosting classifiers were compared for their ability to predict LM. To overcome the imbalanced distribution of the LM data, we retrained each machine-learning classifier using the synthetic minority over-sampling technique (SMOTE). A DLRN combining the independent clinical predictors with the best performing radiomics prediction signature (mRMR+LASSO+SVM+SMOTE) was established. Area under the receiver operating characteristics curve (AUC), calibration curves, and decision curve analysis (DCA) were used to assess the performance and clinical applicability of the models.

Result

Comparisons of the AUC values applied to the external validation set revealed that the DLRN model (AUC=0.833) showed better prediction performance than the clinical model (AUC=0.664) and radiomics model (AUC=0.799). The calibration curves indicated good calibration efficiency and the DCA showed the DLRN model to have greater clinical applicability than the other two models.

Conclusion

The DLRN was shown to be an accurate and efficient tool for LM-status prediction in STS.

Radiomics-guided radiation therapy: opportunities and challenges

Radiomics is an advanced image-processing framework, which extracts image features and considers them as biomarkers towards personalized medicine. Applications include disease detection, diagnosis, prognosis, and therapy response assessment/prediction. As radiation therapy aims for further individualized treatments, radiomics could play a critical role in various steps before, during and after treatment. Elucidation of the concept of radiomics-guided radiation therapy (RGRT) is the aim of this review, attempting to highlight opportunities and challenges underlying the use of radiomics to guide clinicians and physicists towards more effective radiation treatments. This work identifies the value of RGRT in various steps of radiotherapy from patient selection to follow-up, and subsequently provides recommendations to improve future radiotherapy using quantitative imaging features.

Peritumoral radiomics features on preoperative thin-slice CT images can predict the spread through air spaces of lung adenocarcinoma

The spread through air spaces (STAS) is recognized as a negative prognostic factor in patients with early-stage lung adenocarcinoma. The present study aimed to develop a machine learning model for the prediction of STAS using peritumoral radiomics features extracted from preoperative CT imaging. A total of 339 patients who underwent lobectomy or limited resection for lung adenocarcinoma were included. The patients were randomly divided (3:2) into training and test cohorts. Two prediction models were created using the training cohort: a conventional model based on the tumor consolidation/tumor (C/T) ratio and a machine learning model based on peritumoral radiomics features. The areas under the curve for the two models in the testing cohort were 0.70 and 0.76, respectively (P = 0.045). The cumulative incidence of recurrence (CIR) was significantly higher in the STAS high-risk group when using the radiomics model than that in the low-risk group (44% vs. 4% at 5 years; P = 0.002) in patients who underwent limited resection in the testing cohort. In contrast, the 5-year CIR was not significantly different among patients who underwent lobectomy (17% vs. 11%; P = 0.469). In conclusion, the machine learning model for STAS prediction based on peritumoral radiomics features performed better than the C/T ratio model.

Prediction of Two-Year Recurrence-Free Survival in Operable NSCLC Patients Using Radiomic Features from Intra- and Size-Variant Peri-Tumoral Regions on Chest CT Images

To predict the two-year recurrence-free survival of patients with non-small cell lung cancer (NSCLC), we propose a prediction model using radiomic features of the inner and outer regions of the tumor. The intratumoral region and the peritumoral regions from the boundary to 3 cm were used to extract the radiomic features based on the intensity, texture, and shape features. Feature selection was performed to identify significant radiomic features to predict two-year recurrence-free survival, and patient classification was performed into recurrence and non-recurrence groups using SVM and random forest classifiers. The probability of two-year recurrence-free survival was estimated with the Kaplan–Meier curve. In the experiment, CT images of 217 non-small-cell lung cancer patients at stages I-IIIA who underwent surgical resection at the Veterans Health Service Medical Center (VHSMC) were used. Regarding the classification performance on whole tumors, the combined radiomic features for intratumoral and peritumoral regions of 6 mm and 9 mm showed improved performance (AUC 0.66, 0.66) compared to T stage and N stage (AUC 0.60), intratumoral (AUC 0.64) and peritumoral 6 mm and 9 mm classifiers (AUC 0.59, 0.62). In the assessment of the classification performance according to the tumor size, combined regions of 21 mm and 3 mm were significant when predicting outcomes compared to other regions of tumors under 3 cm (AUC 0.70) and 3 cm~5 cm (AUC 0.75), respectively. For tumors larger than 5 cm, the combined 3 mm region was significant in predictions compared to the other features (AUC 0.71). Through this experiment, it was confirmed that peritumoral and combined regions showed higher performance than the intratumoral region for tumors less than 5 cm in size and that intratumoral and combined regions showed more stable performance than the peritumoral region in tumors larger than 5 cm.

Predicting outcomes for locally advanced rectal cancer treated with neoadjuvant chemoradiation with CT-based radiomics

A feasibility study was performed to determine if CT-based radiomics could play an augmentative role in predicting neoadjuvant rectal score (NAR), locoregional failure free survival (LRFFS), distant metastasis free survival (DMFS), disease free survival (DFS) and overall survival (OS) in locally advanced rectal cancer (LARC). The NAR score, which takes into account the pathological tumour and nodal stage as well as clinical tumour stage, is a validated surrogate endpoint used for early determination of treatment response whereby a low NAR score (< 8) has been correlated with better outcomes and high NAR score (> 16) has been correlated with poorer outcomes. CT images of 191 patients with LARC were used in this study. Primary tumour (GTV) and mesorectum (CTV) were contoured separately and radiomics features were extracted from both segments. Two NAR models (NAR > 16 and NAR < 8) models were constructed using Least Absolute Shrinkage and Selection Operator (LASSO) and the survival models were constructed using regularized Cox regressions. Area under curve (AUC) and time-dependent AUC were used to quantify the performance of the LASSO and Cox regression respectively, using ten folds cross validations. The NAR > 16 and NAR < 8 models have an average AUCs of 0.68 ± 0.13 and 0.59 ± 0.14 respectively. There are statistically significant differences between the clinical and combined model for LRFFS (from 0.68 ± 0.04 to 0.72 ± 0.04), DMFS (from 0.68 ± 0.05 to 0.70 ± 0.05) and OS (from 0.64 ± 0.06 to 0.66 ± 0.06). CTV radiomics features were also found to be more important than GTV features in the NAR prediction model. The most important clinical features are age and CEA for NAR > 16 and NAR < 8 models respectively, while the most significant clinical features are age, surgical margin and NAR score across all the four survival models.

Radial Data Mining to Identify Density-Dose Interactions That Predict Distant Failure Following SABR

Purpose

Lower dose outside the planned treatment area in lung stereotactic radiotherapy has been linked to increased risk of distant metastasis (DM) possibly due to underdosage of microscopic disease (MDE). Independently, tumour density on pretreatment computed tomography (CT) has been linked to risk of MDE. No studies have investigated the interaction between imaging biomarkers and incidental dose. The interaction would showcase whether the impact of dose on outcome is dependent on imaging and, hence, if imaging could inform which patients require dose escalation outside the gross tumour volume (GTV). We propose an image-based data mining methodology to investigate density-dose interactions radially from the GTV to predict DM with no a priori assumption on location.

Methods

Dose and density were quantified in 1-mm annuli around the GTV for 199 patients with early-stage lung cancer treated with 60 Gy in 5 fractions. Each annulus was summarised by three density and three dose parameters. For parameter combinations, Cox regressions were performed including a dose-density interaction in independent annuli. Heatmaps were created that described improvement in DM prediction due to the interaction. Regions of significant improvement were identified and studied in overall outcome models.

Results

Dose-density interactions were identified that significantly improved prediction for over 50% of bootstrap resamples. Dose and density parameters were not significant when the interaction was omitted. Tumour density variance and high peritumour density were associated with DM for patients with more cold spots (less than 30-Gy EQD2) and non-uniform dose about 3 cm outside of the GTV. Associations identified were independent of the mean GTV dose.

Conclusions

Patients with high tumour variance and peritumour density have increased risk of DM if there is a low and non-uniform dose outside the GTV. The dose regions are independent of tumour dose, suggesting that incidental dose may play an important role in controlling occult disease. Understanding such interactions is key to identifying patients who will benefit from dose-escalation. The methodology presented allowed spatial dose-density interactions to be studied at the exploratory stage for the first time. This could accelerate the clinical implementation of imaging biomarkers by demonstrating the impact of incidental dose for tumours of varying characteristics in routine data.

Radiomic analysis for predicting prognosis of colorectal cancer from preoperative 18F-FDG PET/CT

Background

To develop and validate a survival model with clinico-biological features and ¹⁸F- FDG PET/CT radiomic features via machine learning, and for predicting the prognosis from the primary tumor of colorectal cancer.

Methods

A total of 196 pathologically confirmed patients with colorectal cancer (stage I to stage IV) were included. Preoperative clinical factors, serum tumor markers, and PET/CT radiomic features were included for the recurrence-free survival analysis. For the modeling and validation, patients were randomly divided into the training (n = 137) and validation (n = 59) set, while the 78 stage III patients [training (n = 55), and validation (n = 23)] was divided for the further experiment. After selecting features by the log-rank test and variable-hunting methods, random survival forest (RSF) models were built on the training set to analyze the prognostic value of selected features. The performance of models was measured by C-index and was tested on the validation set with bootstrapping. Feature importance and the Pearson correlation were also analyzed.

Results

Radiomics signature (containing four PET/CT features and four clinical factors) achieved the best result for prognostic prediction of 196 patients (C-index 0.780, 95% CI 0.634–0.877). Moreover, four features (including two clinical features and two radiomics features) were selected for prognostic prediction of the 78 stage III patients (C-index was 0.820, 95% CI 0.676–0.900). K–M curves of both models significantly stratified low-risk and high-risk groups (P < 0.0001). Pearson correlation analysis demonstrated that selected radiomics features were correlated with tumor metabolic factors, such as SUVmean, SUVmax.

Conclusion

This study presents integrated clinico-biological-radiological models that can accurately predict the prognosis in colorectal cancer using the preoperative ¹⁸F-FDG PET/CT radiomics in colorectal cancer. It is of potential value in assisting the management and decision making for precision treatment in colorectal cancer.

Trial registration The retrospectively registered study was approved by the Ethics Committee of Fudan University Shanghai Cancer Center (No. 1909207-14-1910) and the data were analyzed anonymously.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-022-03262-5.

Can Peritumoral Radiomics Improve the Prediction of Malignancy of Solid Pulmonary Nodule Smaller Than 2 cm?

RATIONALE AND OBJECTIVES

To compare the ability of radiomics models including the perinodular parenchyma and standard nodular radiomics model in lung cancer diagnosis of solid pulmonary nodules smaller than 2 cm.

MATERIALS AND METHODS

In this retrospective study, the computed tomography (CT) scans of 206 patients with a lung nodule from a single institution in 2012-2019 were collected. For each nodule, four volumes of interest were defined using the gross tumor volume (GTV) and peritumoral volumes (PTVs) of 5, 10, and 15 mm around the tumor.

RESULTS

Radiomics models created from GTV, GTV plus 5 mm of PTV, GTV plus 10 mm of PTV, and GTV plus 15 mm of PTV achieved AUCs of 0.89, 0.81, 0.81, and 0.73, respectively, in the validation cohort for the diagnostic classification of benign and malignant pulmonary nodules. The performance of the models gradually decreased as the PTV increased. Wavelet features were the primary features identified in optimal radiomics signatures (2/3 in R, 4/5 in GTV plus 5 mm PTV, 3/4 in GTV plus 10 mm PTV, 2/3 in GTV plus 15 mm PTV).

CONCLUSION

Our study indicated that the radiomics signatures of GTV had a good prediction ability in distinguishing benign and malignant solid pulmonary nodules smaller than 2 cm on CT. However, the radiomics feature of the surrounding parenchyma of the nodule did not enhance the effectiveness of the diagnostic model.

Predicting cancer outcomes with radiomics and artificial intelligence in radiology

The successful use of artificial intelligence (AI) for diagnostic purposes has prompted the application of AI-based cancer imaging analysis to address other, more complex, clinical needs. In this Perspective, we discuss the next generation of challenges in clinical decision-making that AI tools can solve using radiology images, such as prognostication of outcome across multiple cancers, prediction of response to various treatment modalities, discrimination of benign treatment confounders from true progression, identification of unusual response patterns and prediction of the mutational and molecular profile of tumours. We describe the evolution of and opportunities for AI in oncology imaging, focusing on hand-crafted radiomic approaches and deep learning-derived representations, with examples of their application for decision support. We also address the challenges faced on the path to clinical adoption, including data curation and annotation, interpretability, and regulatory and reimbursement issues. We hope to demystify AI in radiology for clinicians by helping them to understand its limitations and challenges, as well as the opportunities it provides as a decision-support tool in cancer management.

Optimizing the Peritumoral Region Size in Radiomics Analysis for Sentinel Lymph Node Status Prediction in Breast Cancer

RATIONALE AND OBJECTIVES

Peritumoral features have been suggested to be useful in improving the prediction performance of radiomic models. The aim of this study is to systematically investigate the prediction performance improvement for sentinel lymph node (SLN) status in breast cancer from peritumoral features in radiomic analysis by exploring the effect of peritumoral region sizes.

MATERIALS AND METHODS

This retrospective study was performed using dynamic contrast-enhanced MRI scans of 162 breast cancer patients. The effect of peritumoral features was evaluated in a radiomics pipeline for predicting SLN metastasis in breast cancer. Peritumoral regions were generated by dilating the tumor regions-of-interest (ROIs) manually annotated by two expert radiologists, with thicknesses of 2 mm, 4 mm, 6 mm, and 8 mm. The prediction models were established in the training set (∼67% of cases) using the radiomics pipeline with and without peritumoral features derived from different peritumoral thicknesses. The prediction performance was tested in an independent validation set (the remaining ∼33%).

RESULTS

For this specific application, the accuracy in the validation set when using the two radiologists' ROIs could be both improved from 0.704 to 0.796 by incorporating peritumoral features. The choice of the peritumoral size could affect the level of improvement.

CONCLUSION

This study systematically investigates the effect of peritumoral region sizes in radiomic analysis for prediction performance improvement. The choice of the peritumoral size is dependent on the ROI drawing and would affect the final prediction performance of radiomic models, suggesting that peritumoral features should be optimized in future radiomics studies.

End-to-End Non-Small-Cell Lung Cancer Prognostication Using Deep Learning Applied to Pretreatment Computed Tomography

PURPOSE

Clinical TNM staging is a key prognostic factor for patients with lung cancer and is used to inform treatment and monitoring. Computed tomography (CT) plays a central role in defining the stage of disease. Deep learning applied to pretreatment CTs may offer additional, individualized prognostic information to facilitate more precise mortality risk prediction and stratification.

METHODS

We developed a fully automated imaging-based prognostication technique (IPRO) using deep learning to predict 1-year, 2-year, and 5-year mortality from pretreatment CTs of patients with stage I-IV lung cancer. Using six publicly available data sets from The Cancer Imaging Archive, we performed a retrospective five-fold cross-validation using pretreatment CTs of 1,689 patients, of whom 1,110 were diagnosed with non-small-cell lung cancer and had available TNM staging information. We compared the association of IPRO and TNM staging with patients' survival status and assessed an Ensemble risk score that combines IPRO and TNM staging. Finally, we evaluated IPRO's ability to stratify patients within TNM stages using hazard ratios (HRs) and Kaplan-Meier curves.

RESULTS

IPRO showed similar prognostic power (concordance index [C-index] 1-year: 0.72, 2-year: 0.70, 5-year: 0.68) compared with that of TNM staging (C-index 1-year: 0.71, 2-year: 0.71, 5-year: 0.70) in predicting 1-year, 2-year, and 5-year mortality. The Ensemble risk score yielded superior performance across all time points (C-index 1-year: 0.77, 2-year: 0.77, 5-year: 0.76). IPRO stratified patients within TNM stages, discriminating between highest- and lowest-risk quintiles in stages I (HR: 8.60), II (HR: 5.03), III (HR: 3.18), and IV (HR: 1.91).

CONCLUSION

Deep learning applied to pretreatment CT combined with TNM staging enhances prognostication and risk stratification in patients with lung cancer.

Radiomics feature reliability assessed by intraclass correlation coefficient: a systematic review

Radiomics research is rapidly growing in recent years, but more concerns on radiomics reliability are also raised. This review attempts to update and overview the current status of radiomics reliability research in the ever expanding medical literature from the perspective of a single reliability metric of intraclass correlation coefficient (ICC). To conduct this systematic review, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. After literature search and selection, a total of 481 radiomics studies using CT, PET, or MRI, covering a wide range of subject and disease types, were included for review. In these highly heterogeneous studies, feature reliability to image segmentation was much more investigated than reliability to other factors, such as image acquisition, reconstruction, post-processing, and feature quantification. The reported ICCs also suggested high radiomics feature reliability to image segmentation. Image acquisition was found to introduce much more feature variability than image segmentation, in particular for MRI, based on the reported ICC values. Image post-processing and feature quantification yielded different levels of radiomics reliability and might be used to mitigate image acquisition-induced variability. Some common flaws and pitfalls in ICC use were identified, and suggestions on better ICC use were given. Due to the extremely high study heterogeneities and possible risks of bias, the degree of radiomics feature reliability that has been achieved could not yet be safely synthesized or derived in this review. More future researches on radiomics reliability are warranted.

Benchmarking Various Radiomic Toolkit Features While Applying the Image Biomarker Standardization Initiative toward Clinical Translation of Radiomic Analysis

The image biomarkers standardization initiative (IBSI) was formed to address the standardization of extraction of quantifiable imaging metrics. Despite its effort, there remains a lack of consensus or established guidelines regarding radiomic feature terminology, the underlying mathematics and their implementation across various software programs. This creates a scenario where features extracted using different toolboxes cannot be used to build or validate the same model leading to a non-generalization of radiomic results. In this study, IBSI-established phantom and benchmark values were used to compare the variation of the radiomic features while using 6 publicly available software programs and 1 in-house radiomics pipeline. All IBSI-standardized features (11 classes, 173 in total) were extracted. The relative differences between the extracted feature values from the different software programs and the IBSI benchmark values were calculated to measure the inter-software agreement. To better understand the variations, features are further grouped into 3 categories according to their properties: 1) morphology, 2) statistic/histogram and 3)texture features. While a good agreement was observed for a majority of radiomics features across the various tested programs, relatively poor agreement was observed for morphology features. Significant differences were also found in programs that use different gray-level discretization approaches. Since these software programs do not include all IBSI features, the level of quantitative assessment for each category was analyzed using Venn and UpSet diagrams and quantified using two ad hoc metrics. Morphology features earned lowest scores for both metrics, indicating that morphological features are not consistently evaluated among software programs. We conclude that radiomic features calculated using different software programs may not be interchangeable. Further studies are needed to standardize the workflow of radiomic feature extraction.

Toward High-Throughput Artificial Intelligence-Based Segmentation in Oncological PET Imaging

Artificial intelligence (AI) techniques for image-based segmentation have garnered much attention in recent years. Convolutional neural networks have shown impressive results and potential toward fully automated segmentation in medical imaging, and particularly PET imaging. To cope with the limited access to annotated data needed in supervised AI methods, given tedious and prone-to-error manual delineations, semi-supervised and unsupervised AI techniques have also been explored for segmentation of tumors or normal organs in single- and bimodality scans. This work reviews existing AI techniques for segmentation tasks and the evaluation criteria for translational AI-based segmentation efforts toward routine adoption in clinical workflows.

Gd-EOB-DTPA-enhanced MRI radiomics to predict vessels encapsulating tumor clusters (VETC) and patient prognosis in hepatocellular carcinoma

OBJECTIVES

The study was to develop a Gd-EOB-DTPA-enhanced MRI radiomics model for preoperative prediction of VETC and patient prognosis in hepatocellular cancer (HCC).

METHODS

The study included 182 (training cohort: 128; validation cohort: 54) HCC patients who underwent preoperative Gd-EOB-DTPA-enhanced MRI. Volumes of interest including intratumoral and peritumoral regions were manually delineated in the hepatobiliary phase images, from which 1316 radiomics features were extracted. The least absolute shrinkage and selection operator (LASSO) and multivariable logistic regression were used to select the useful features. Clinical, intratumoral, peritumoral, combined radiomics, and clinical radiomics models were established using machine learning algorithms. The Kaplan-Meier survival analysis was used to assess early recurrence and progression-free survival (PFS) in the VETC + and VETC- patients.

RESULTS

In the validation cohort, the area under the curves (AUCs) of radiomics models were higher than that of the clinical model using random forest (all p < 0.05). The peritumoral radiomics model (AUC = 0.972;95% confidence interval [CI]:0.887-0.998) had significantly higher AUC than intratumoral model (AUC = 0.919; 95% CI: 0.811-0.976) (p = 0.044). There were no significant differences in AUC between intratumoral or peritumoral radiomics model (PR) and combined radiomics model (p > 0.05). Early recurrence and PFS were significantly different between the PR-predicted VETC + and VETC- HCC patients (p < 0.05). PR-predicted VETC was independent predictor of early recurrence (hazard ratio [HR]: 2.08[1.31-3.28]; p = 0.002) and PFS (HR: 1.95[1.20-3.17]; p = 0.007).

CONCLUSIONS

The intratumoral or peritumoral radiomics model may be useful in predicting VETC and patient prognosis preoperatively. The peritumoral radiomics model may yield an incremental value over intratumoral model.

KEY POINTS

• Radiomics models are useful for predicting vessels encapsulating tumor clusters (VETC) and patient prognosis preoperatively. • Peritumoral radiomics model may yield an incremental value over intratumoral model in prediction of VETC. • Peritumoral radiomics-model-predicted VETC was an independent predictor of early recurrence and progression-free survival.

Improving the accuracy of prognosis for clinical stage I solid lung adenocarcinoma by radiomics models covering tumor per se and peritumoral changes on CT

OBJECTIVES

To assess methods to improve the accuracy of prognosis for clinical stage I solid lung adenocarcinoma using radiomics based on different volumes of interests (VOIs).

METHODS

This retrospective study included patients with postoperative clinical stage I solid lung adenocarcinoma from two hospitals, center 1 and center 2. Three databases were generated: dataset A (training set from center 1), dataset B (internal test set from center 1), and dataset C (external validation test from center 2). Disease-free survival (DFS) data were collected. CT radiomics models were constructed based on four VOIs: gross tumor volume (GTV), 3 mm external to the tumor border (peritumoral volume [PTV]_0~+3), 6 mm crossing tumor border (PTV_-3~+3), and 6 mm external to the tumor border (PTV_0~+6). The area under the receiver operating characteristic curve (AUC) was used to compare the model accuracies.

RESULTS

A total of 334 patients were included (204 and 130 from centers 1 and 2). The model using PTV_-3~+3 (AUC 0.81 [95% confidence interval {CI}: 0.75, 0.94], 0.81 [0.63, 0.90] for datasets B and C) outperformed the other three models, GTV (0.73 [0.58, 0.81], 0.73 [0.58, 0.83]), PTV_0~+3 (0.76 [0.52, 0.87], 0.75 [0.60, 0.83]), and PTV_0~+6 (0.72 [0.60, 0.81], 0.69 [0.59, 0.81]), in datasets B and C, all p < 0.05.

CONCLUSIONS

A radiomics model based on a VOI of 6 mm crossing tumor border more accurately predicts prognosis of clinical stage I solid lung adenocarcinoma than that based on VOIs including overall tumor or external rims of 3 mm and 6 mm.

KEY POINTS

• Radiomics is a useful approach to improve the accuracy of prognosis for stage I solid adenocarcinoma. • The radiomics model based on VOIs that includes 3 mm within and external to the tumor border (peritumoral volume [PTV]_-3~+3) outperformed models that included either only the tumor itself or those that only included the peritumoral volume.

Combining computed tomography and biologically effective dose in radiomics and deep learning improves prediction of tumor response to robotic lung stereotactic body radiation therapy

PURPOSE

The aim of this study is to improve the performance of machine learning (ML) models in predicting response of non-small cell lung cancer (NSCLC) to stereotactic body radiation therapy (SBRT) by integrating image features from pre-treatment computed tomography (CT) with features from the biologically effective dose (BED) distribution.

MATERIALS AND METHODS

Image features, consisting of crafted radiomic features or machine-learned features extracted using a convolutional neural network, were calculated from pre-treatment CT data and from dose distributions converted into BED for 80 NSCLC lesions over 76 patients treated with robotic guided SBRT. ML models using different combinations of features were trained to predict complete or partial response according to response criteria in solid tumors, including radiomics CT (Rad_CT ), radiomics CT and BED (Rad_CT,BED ), deep learning (DL) CT (DL_CT ), and DL CT and BED (DL_CT,BED ). Training of ML included feature selection by neighborhood component analysis followed by ensemble ML using robust boosting. A model was considered as acceptable when the sum of average sensitivity and specificity on test data in repeated cross validations was at least 1.5.

RESULTS

Complete or partial response occurred in 58 out of 80 lesions. The best models to predict the tumor response were those using BED variables, achieving significantly better area under curve (AUC) and accuracy than those using only features from CT, including a Rad_CT,BED model using three radiomic features from BED, which scored an accuracy of 0.799 (95% confidence intervals (0.75-0.85)) and AUC of 0.773 (0.688-0.846), and a DL_CT,BED model also using three variables with an accuracy of 0.798 (0.649-0.829) and AUC of 0.812 (0.755-0.867).

CONCLUSION

According to our results, the inclusion of BED features improves the response prediction of ML models for lung cancer patients undergoing SBRT, regardless of the use of radiomic or DL features.

Incorporating radiomics into clinical trials: expert consensus endorsed by the European Society of Radiology on considerations for data-driven compared to biologically driven quantitative biomarkers

Existing quantitative imaging biomarkers (QIBs) are associated with known biological tissue characteristics and follow a well-understood path of technical, biological and clinical validation before incorporation into clinical trials. In radiomics, novel data-driven processes extract numerous visually imperceptible statistical features from the imaging data with no a priori assumptions on their correlation with biological processes. The selection of relevant features (radiomic signature) and incorporation into clinical trials therefore requires additional considerations to ensure meaningful imaging endpoints. Also, the number of radiomic features tested means that power calculations would result in sample sizes impossible to achieve within clinical trials. This article examines how the process of standardising and validating data-driven imaging biomarkers differs from those based on biological associations. Radiomic signatures are best developed initially on datasets that represent diversity of acquisition protocols as well as diversity of disease and of normal findings, rather than within clinical trials with standardised and optimised protocols as this would risk the selection of radiomic features being linked to the imaging process rather than the pathology. Normalisation through discretisation and feature harmonisation are essential pre-processing steps. Biological correlation may be performed after the technical and clinical validity of a radiomic signature is established, but is not mandatory. Feature selection may be part of discovery within a radiomics-specific trial or represent exploratory endpoints within an established trial; a previously validated radiomic signature may even be used as a primary/secondary endpoint, particularly if associations are demonstrated with specific biological processes and pathways being targeted within clinical trials. KEY POINTS: • Data-driven processes like radiomics risk false discoveries due to high-dimensionality of the dataset compared to sample size, making adequate diversity of the data, cross-validation and external validation essential to mitigate the risks of spurious associations and overfitting. • Use of radiomic signatures within clinical trials requires multistep standardisation of image acquisition, image analysis and data mining processes. • Biological correlation may be established after clinical validation but is not mandatory.

Radiogenomics of gastroenterological cancer: The dawn of personalized medicine with artificial intelligence-based image analysis

Radiogenomics is a new field of medical science that integrates two omics, radiomics and genomics, and may bring a major paradigm shift in traditional personalized medicine strategies that require tumor tissue samples. In addition, the acquisition of the data does not require special imaging equipment or special imaging conditions, and it is possible to use image information from computed tomography, magnetic resonance imaging, positron emission tomography-computed tomography in clinical practice, so the versatility and cost-effectiveness of radiogenomics are expected. So far, the field of radiogenomics has developed, especially in the fields of brain tumors and breast cancer, but recently, reports of radiogenomic research on gastroenterological cancer are increasing. This review provides an overview of radiogenomic research methods and summarizes the current radiogenomic research in gastroenterological cancer. In addition, the application of artificial intelligence is considered to be indispensable for the integrated analysis of enormous omics information in the future, and the future direction of this research, including the latest technologies, will be discussed.

Radiomics Prediction of EGFR Status in Lung Cancer-Our Experience in Using Multiple Feature Extractors and The Cancer Imaging Archive Data

We investigated the performance of multiple radiomics feature extractors/software on predicting epidermal growth factor receptor mutation status in 228 patients with non–small cell lung cancer from publicly available data sets in The Cancer Imaging Archive. The imaging and clinical data were split into training (n = 105) and validation cohorts (n = 123). Two of the most cited open-source feature extractors, IBEX (1563 features) and Pyradiomics (1319 features), and our in-house software, Columbia Image Feature Extractor (CIFE) (1160 features), were used to extract radiomics features. Univariate and multivariate analyses were performed sequentially to predict EGFR mutation status using each individual feature extractor. Our univariate analysis integrated an unsupervised clustering method to identify nonredundant and informative candidate features for the creation of prediction models by multivariate analyses. In training, unsupervised clustering-based univariate analysis identified 5, 6, and 4 features from IBEX, Pyradiomics, and CIFE as candidate features, respectively. Multivariate prediction models using these features from IBEX, Pyradiomics, and CIFE yielded similar areas under the receiver operating characteristic curve of 0.68, 0.67, and 0.69. However, in validation, areas under the receiver operating characteristic curve of multivariate prediction models from IBEX, Pyradiomics, and CIFE decreased to 0.54, 0.56 and 0.64, respectively. Different feature extractors select different radiomics features, which leads to prediction models with varying performance. However, correlation between those selected features from different extractors may indicate these features measure similar imaging phenotypes associated with similar biological characteristics. Overall, attention should be paid to the generalizability of individual radiomics features and radiomics prediction models.

Integrative nomogram of intratumoral, peritumoral, and lymph node radiomic features for prediction of lymph node metastasis in cT1N0M0 lung adenocarcinomas

Radiomics studies to predict lymph node (LN) metastasis has only focused on either primary tumor or LN alone. However, combining radiomics features from multiple sources may reflect multiple characteristic of the lesion thereby increasing the discriminative performance of the radiomic model. Therefore, the present study intends to evaluate the efficiency of integrative nomogram, created by combining clinical parameters and radiomics features extracted from gross tumor volume (GTV), peritumoral volume (PTV) and LN, for the preoperative prediction of LN metastasis in clinical cT1N0M0 adenocarcinoma. A primary cohort of 163 patients (training cohort, 113; and internal validation cohort, 50) and an external validation cohort of 53 patients with clinical stage cT1N0M0 were retrospectively included. Features were extracted from three regions of interests (ROIs): GTV; PTV (5.0 mm around the tumor) and LN on pre-operative contrast enhanced computed tomography (CT). LASSO logistic regression method was used to build radiomic signatures. Multivariable regression analysis was used to build a nomogram. The performance of the nomogram was assessed with respect to its calibration, discrimination, and clinical usefulness. The discriminative performance of nomogram was validated both internally and externally. The radiomic signatures using the features of GTV, PTV and LN showed a good ability in predicting LN metastasis with an area under the curve (AUC) of 0.74 (95% CI 0.60–0.88), 0.72 (95% CI 0.57–0.87) and 0.64 (95% CI 0.48–0.80) respectively in external validation cohort. The integration of different signature together further increases the discriminatory ability: GTV + PTV (GPTV): AUC 0.75 (95% CI 0.61–0.89) and GPTV + LN: AUC 0.76 (95% CI 0.61–0.91) in external validation cohort. An integrative nomogram of clinical parameters and radiomic features demonstrated further increase in discriminatory ability with AUC of 0.79 (95% CI 0.66–0.93) in external validation cohort. The nomogram showed good calibration. Decision curve analysis demonstrated that the radiomic nomogram was clinically useful. The integration of information from clinical parameters along with CT radiomics information from GTV, PTV and LN was feasible and increases the predictive performance of the nomogram in predicting LN status in cT1N0M0 adenocarcinoma patients suggesting merit of information integration from multiple sources in building prediction model.

Optimising use of 4D-CT phase information for radiomics analysis in lung cancer patients treated with stereotactic body radiotherapy

Purpose. 4D-CT is routine imaging for lung cancer patients treated with stereotactic body radiotherapy. No studies have investigated optimal 4D phase selection for radiomics. We aim to determine how phase data should be used to identify prognostic biomarkers for distant failure, and test whether stability assessment is required. A phase selection approach will be developed to aid studies with different 4D protocols and account for patient differences. Methods. 186 features were extracted from the tumour and peritumour on all phases for 258 patients. Feature values were selected from phase features using four methods: (A) mean across phases, (B) median across phases, (C) 50% phase, and (D) the most stable phase (closest in value to two neighbours), coined personalised selection. Four levels of stability assessment were also analysed, with inclusion of: (1) all features, (2) stable features across all phases, (3) stable features across phase and neighbour phases, and (4) features averaged over neighbour phases. Clinical-radiomics models were built for twelve combinations of feature type and assessment method. Model performance was assessed by concordance index (c-index) and fraction of new information from radiomic features. Results. The most stable phase spanned the whole range but was most often near exhale. All radiomic signatures provided new information for distant failure prediction. The personalised model had the highest c-index (0.77), and 58% of new information was provided by radiomic features when no stability assessment was performed. Conclusion. The most stable phase varies per-patient and selecting this improves model performance compared to standard methods. We advise the single most stable phase should be determined by minimising feature differences to neighbour phases. Stability assessment over all phases decreases performance by excessively removing features. Instead, averaging of neighbour phases should be used when stability is of concern. The models suggest that higher peritumoural intensity predicts distant failure.

Dual energy CT image prediction on primary tumor of lung cancer for nodal metastasis using deep learning

Lymph node metastasis (LNM) identification is the most clinically important tasks related to survival and recurrence from lung cancer. However, the preoperative prediction of nodal metastasis remains a challenge to determine surgical plans and pretreatment decisions in patients with cancers. We proposed a novel deep prediction method with a size-related damper block for nodal metastasis (Nmet) identification from the primary tumor in lung cancer generated by gemstone spectral imaging (GSI) dual-energy computer tomography (CT). The best model is the proposed method trained by the 40 keV dataset achieves an accuracy of 86 % and a Kappa value of 72 % for Nmet prediction. In the experiment, we have 11 different monochromatic images from 40∼140 keV (the interval is 10 keV) for each patient. When we used the model of 40 keV dataset, there has significant difference in other energy levels (unit of keV). Therefore, we apply in 5-fold cross-validation to explain the lower keV is more efficient to predict Nmet of the primary tumor. The result shows that tumor heterogeneity and size contributed to the proposed model to estimate whether absence or presence of nodal metastasis from the primary tumor.

Radiomics in radiation oncology for gynecological malignancies: a review of literature

Radiomics is the extraction of a significant number of quantitative imaging features with the aim of detecting information in correlation with useful clinical outcomes. Features are extracted, after delineation of an area of interest, from a single or a combined set of imaging modalities (including X-ray, US, CT, PET/CT and MRI). Given the high dimensionality, the analytical process requires the use of artificial intelligence algorithms. Firstly developed for diagnostic performance in radiology, it has now been translated to radiation oncology mainly to predict tumor response and patient outcome but other applications have been developed such as dose painting, prediction of side-effects, and quality assurance. In gynecological cancers, most studies have focused on outcomes of cervical cancers after chemoradiation. This review highlights the role of this new tool for the radiation oncologists with particular focus on female GU oncology.

The Role of Radiomics in Lung Cancer: From Screening to Treatment and Follow-Up

PURPOSE

Lung cancer represents the first cause of cancer-related death in the world. Radiomics studies arise rapidly in this late decade. The aim of this review is to identify important recent publications to be synthesized into a comprehensive review of the current status of radiomics in lung cancer at each step of the patients' care.

METHODS

A literature review was conducted using PubMed/Medline for search of relevant peer-reviewed publications from January 2012 to June 2020.

RESULTS

We identified several studies at each point of patient's care: detection and classification of lung nodules (n=16), determination of histology and genomic (n=10) and finally treatment outcomes predictions (=23). We reported the methodology of those studies and their results and discuss the limitations and the progress to be made for clinical routine applications.

CONCLUSION

Promising perspectives arise from machine learning applications and radiomics based models in lung cancers, yet further data are necessary for their implementation in daily care. Multicentric collaboration and attention to quality and reproductivity of radiomics studies should be further consider.

Standard maintenance therapy versus local consolidative radiation therapy and standard maintenance therapy in 1-5 sites of oligometastatic non-small cell lung cancer: a study protocol of phase III randomised controlled trial

INTRODUCTION

Two-phase II randomised studies have shown a significant benefit of local consolidation therapy in oligometastatic non-small cell lung cancer (NSCLC). This phase III randomised controlled trial (RCT) will evaluate the efficacy of local consolidation radiation therapy (RT) in oligometastases (OM) NSCLC after completion of initial systemic therapy.

METHODS AND ANALYSIS

This is a single-centre phase III RCT of OM NSCLC patients. One hundred and ninety patients will undergo 1:1 randomisation to either standard maintenance therapy (control arm) or local consolidation RT and standard maintenance therapy (experimental arm). Patients will be stratified into the number of OM sites (1-2 vs 3-5), nodal metastases (N0-N1 vs N2-N3) and presence or absence of brain metastases. Stereotactic body radiation therapy to all the oligometastatic sites and definitive RT to primary disease will be given in the experimental arm. The primary endpoint is overall survival and secondary endpoints include progression-free survival, local control of OM sites, new distant metastases free survival, objective response rate, toxicity and quality of life. Translation endpoint include circulating tumour cells and radiomics using texture analysis.

ETHICS AND DISSEMINATION

All patients will be provided with a written informed consent form which needs to be signed before randomisation. The study is approved by the institutional ethics committee-II (project number 3445) and registered with Clinical Trials Registry-India, dated 21 April 2020.

TRIAL REGISTRATION NUMBER

CTRI/2020/04/024761; Pre-Results.

Protocol for a phase II randomised controlled trial of TKI alone versus TKI and local consolidative radiation therapy in patients with oncogene driver-mutated oligometastatic non-small cell lung cancer

INTRODUCTION

Tyrosine kinase inhibitors (TKIs) have significantly improved the progression-free survival (PFS) of metastatic non-small cell lung cancer (NSCLC) with oncogene mutations of epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) compared with systemic therapy alone. However, the majority eventually develop resistance with a median PFS of 8-12 months. The pattern of failure studies showed disease relapse at the original sites of the disease-harbouring resistant tumour cells.

METHODS AND ANALYSIS

This study is designed as a phase II randomised controlled trial to evaluate the efficacy of local consolidative radiation therapy (LCRT) in addition to TKI in upfront oligometastatic NSCLC. Patients will be screened at presentation for oligometastases (≤5 sites) and will start on TKI after confirmation of EGFR or ALK mutation status. After initial TKI for 2-4 months, eligible patients will be randomised in a 1:1 ratio with stratification of oligometastatic sites (1-3 vs 4-5), performance status of 0-1 versus 2 and brain metastases. The standard arm will continue to receive TKI, and the intervention arm will receive TKI plus LCRT. Stereotactic body radiation therapy will be delivered to all the oligometastatic sites.The primary end point is PFS, and secondary end points are overall survival, local control of oligometastatic sites, toxicity and patient-reported outcomes. The sample size calculation took a median PFS of 10 months in the standard arm. To detect an absolute improvement of 7 months in the interventional arm, with a one-sided alpha of 5% and 80% power, a total of 106 patients will be accrued over a period of 48 months.

ETHICS AND DISSEMINATION

The study is approved by the Institutional Ethics Committee II of Tata Memorial Centre, Mumbai, and registered with Clinical Trials Registry-India, CTRI/2019/11/021872, dated 5 November 2019. All eligible participants will be provided with a participant information sheet and will be required to provide written informed consent for participation in the study. The study results will be presented at a national/international conference and will be published in a peer-reviewed journal.

Clinical-Radiomic Analysis for Pretreatment Prediction of Objective Response to First Transarterial Chemoembolization in Hepatocellular Carcinoma

BACKGROUND

The preoperative selection of patients with intermediate-stage hepatocellular carcinoma (HCC) who are likely to have an objective response to first transarterial chemoembolization (TACE) remains challenging.

OBJECTIVE

To develop and validate a clinical-radiomic model (CR model) for preoperatively predicting treatment response to first TACE in patients with intermediate-stage HCC.

METHODS

A total of 595 patients with intermediate-stage HCC were included in this retrospective study. A tumoral and peritumoral (10 mm) radiomic signature (TPR-signature) was constructed based on 3,404 radiomic features from 4 regions of interest. A predictive CR model based on TPR-signature and clinical factors was developed using multivariate logistic regression. Calibration curves and area under the receiver operating characteristic curves (AUCs) were used to evaluate the model's performance.

RESULTS

The final CR model consisted of 5 independent predictors, including TPR-signature (p < 0.001), AFP (p = 0.004), Barcelona Clinic Liver Cancer System Stage B (BCLC B) subclassification (p = 0.01), tumor location (p = 0.039), and arterial hyperenhancement (p = 0.050). The internal and external validation results demonstrated the high-performance level of this model, with internal and external AUCs of 0.94 and 0.90, respectively. In addition, the predicted objective response via the CR model was associated with improved survival in the external validation cohort (hazard ratio: 2.43; 95% confidence interval: 1.60-3.69; p < 0.001). The predicted treatment response also allowed for significant discrimination between the Kaplan-Meier curves of each BCLC B subclassification.

CONCLUSIONS

The CR model had an excellent performance in predicting the first TACE response in patients with intermediate-stage HCC and could provide a robust predictive tool to assist with the selection of patients for TACE.

Automated gross tumor volume contour generation for large-scale analysis of early-stage lung cancer patients planned with 4D-CT

PURPOSE

Patients with early-stage lung cancer undergoing stereotactic ablative radiotherapy receive four-dimensional computed tomography (4D-CT) for treatment planning. Often, an internal gross target volume (iGTV), which approximates the motion envelope of a tumor over the breathing cycle, is delineated without defining a gross tumor volume (GTV). However, the GTV volume and shape are important parameters for prognostic and dose modelling, and there is interest in radiomic features extracted from the GTV and surrounding tissue. We demonstrate and validate a method to generate the GTV from an iGTV contour to aid retrospective analysis on routine data.

METHOD

It is possible to reconstruct the geometry of a tumor with knowledge of tumor motion and the motion envelope formed during respiration. To demonstrate this, the tumor motion path was estimated with local rigid registration, and the iGTV positioned incrementally at stations along the reverse path. It is shown that the tumor volume is the largest set common to the intersection of the iGTV at these positions, hence can be derived. This was implemented for 521 lung lesions on 4D-CT. Eleven patients with a GTV delineation performed by a radiation oncologist on a reference phase (50%) were used for validation. The generated GTV was compared to that delineated by the expert using distance-to-agreement (DTA), volume, and distance between centres of mass. An overall success rate was determined by detecting registration inaccuracy and performing a quality check on the routine iGTV. For successfully generated contours, GTV volume was compared to iGTV volume in a prognostic model for overall survival.

RESULTS

For the validation dataset, DTA mean (0.79 - 1.55 mm) and standard deviation (0.68 - 1.51 mm) were comparable to expected observer variation. Difference in volume was < 5 cm³ , and average difference in position was 1.21 mm. Deviations in shape and position were mainly caused by observer differences in iGTV and GTV interpretation as opposed to algorithm performance. For the complete dataset, an acceptable contour was generated for 94% of patients using statistical and visual assessment to detect failures. Generated GTV volumes improved prognostic model performance over iGTV volumes.

CONCLUSION

A method to generate a GTV from an iGTV and 4D-CT dataset was developed. This method facilitates data analysis of patients with early-stage lung cancer treated in the routine setting, that is, data mining, prognostic modeling, and radiomics. Generation failure detection removes the need for visual assessment of all contours, reducing a time-consuming aspect of big-data analysis. Favorable prognostic performance of generated GTV volumes over iGTV ones demonstrates opportunities to use this methodology for future study.

Value of 18F-FDG PET/CT-Based Radiomics Nomogram to Predict Survival Outcomes and Guide Personalized Targeted Therapy in Lung Adenocarcinoma With EGFR Mutations

Objectives

To investigate the development and validation of a radiomics nomogram based on PET/CT for guiding personalized targeted therapy in patients with lung adenocarcinoma mutation(s) in the EGFR gene.

Methods

A cohort of 109 (77/32 in training/validation cohort) consecutive lung adenocarcinoma patients with an EGFR mutation was enrolled in this study. A total of 1672 radiomic features were extracted from PET and CT images, respectively. The least absolute shrinkage and selection operator (LASSO) Cox regression was used to select the radiomic features and construct the radiomics nomogram for the estimation of overall survival (OS), which was then assessed with respect to calibration and clinical usefulness. Patients with an EGFR mutation were divided into high- and low- risk groups according to their nomogram score. The treatment strategy for high- and low-risk groups was analyzed using Kaplan–Meier analysis and a log-rank test.

Results

The C-index of the radiomics nomogram for the prediction of OS in lung adenocarcinoma in patients with an EGFR mutation was 0.840 and 0.803 in the training and validation cohorts, respectively. Distant metastasis [(Hazard ratio, HR),1.80], metabolic tumor volume (MTV, HR, 1.62), and rad score (HR, 17.23) were the independent risk factors for patients with an EGFR mutation. The calibration curve showed that the predicted survival time was remarkably close to the actual time. Decision curve analysis demonstrated that the radiomics nomogram was clinically useful. Targeted therapy for patients with high-risk EGFR mutations attained a greater benefit than other therapies (p < 0.0001), whereas the prognoses of the two therapies were similar in the low-risk group (p = 0.85).

Conclusions

Development and validation of a radiomics nomogram based on PET/CT radiomic features combined with clinicopathological factors may guide targeted therapy for patients with lung adenocarcinoma with EGFR mutations. This is conducive to the advancement of precision medicine.

Machine learning-based prediction of microsatellite instability and high tumor mutation burden from contrast-enhanced computed tomography in endometrial cancers

To evaluate whether radiomic features from contrast-enhanced computed tomography (CE-CT) can identify DNA mismatch repair deficient (MMR-D) and/or tumor mutational burden-high (TMB-H) endometrial cancers (ECs). Patients who underwent targeted massively parallel sequencing of primary ECs between 2014 and 2018 and preoperative CE-CT were included (n = 150). Molecular subtypes of EC were assigned using DNA polymerase epsilon (POLE) hotspot mutations and immunohistochemistry-based p53 and MMR protein expression. TMB was derived from sequencing, with > 15.5 mutations-per-megabase as a cut-point to define TMB-H tumors. After radiomic feature extraction and selection, radiomic features and clinical variables were processed with the recursive feature elimination random forest classifier. Classification models constructed using the training dataset (n = 105) were then validated on the holdout test dataset (n = 45). Integrated radiomic-clinical classification distinguished MMR-D from copy number (CN)-low-like and CN-high-like ECs with an area under the receiver operating characteristic curve (AUROC) of 0.78 (95% CI 0.58–0.91). The model further differentiated TMB-H from TMB-low (TMB-L) tumors with an AUROC of 0.87 (95% CI 0.73–0.95). Peritumoral-rim radiomic features were most relevant to both classifications (p ≤ 0.044). Radiomic analysis achieved moderate accuracy in identifying MMR-D and TMB-H ECs directly from CE-CT. Radiomics may provide an adjunct tool to molecular profiling, especially given its potential advantage in the setting of intratumor heterogeneity.

Comprehensive Analysis of Tumour Sub-Volumes for Radiomic Risk Modelling in Locally Advanced HNSCC

Simple Summary

Radiomic risk models are usually based on imaging features, which are extracted from the entire gross tumour volume (GTVentire). This approach does not explicitly consider the complex biological structure of the tumours. Therefore, in this retrospective study, we investigated the prognostic value of radiomic analyses based on different tumour sub-volumes using computed tomography imaging of patients with locally advanced head and neck squamous cell carcinoma who were treated with primary radio-chemotherapy. The GTVentire was cropped by different margins to define the rim and corresponding core sub-volumes of the tumour. Furthermore, the best performing tumour rim sub-volume was extended into surrounding tissue with different margins. As a result, the models based on the 5 mm tumour rim and on the 3 mm extended rim sub-volume showed an improved performance compared to models based on the corresponding tumour core. This indicates that the consideration of tumour sub-volumes may help to improve radiomic risk models.

Abstract

Imaging features for radiomic analyses are commonly calculated from the entire gross tumour volume (GTVentire). However, tumours are biologically complex and the consideration of different tumour regions in radiomic models may lead to an improved outcome prediction. Therefore, we investigated the prognostic value of radiomic analyses based on different tumour sub-volumes using computed tomography imaging of patients with locally advanced head and neck squamous cell carcinoma. The GTVentire was cropped by different margins to define the rim and the corresponding core sub-volumes of the tumour. Subsequently, the best performing tumour rim sub-volume was extended into surrounding tissue with different margins. Radiomic risk models were developed and validated using a retrospective cohort consisting of 291 patients in one of the six Partner Sites of the German Cancer Consortium Radiation Oncology Group treated between 2005 and 2013. The validation concordance index (C-index) averaged over all applied learning algorithms and feature selection methods using the GTVentire achieved a moderate prognostic performance for loco-regional tumour control (C-index: 0.61 ± 0.04 (mean ± std)). The models based on the 5 mm tumour rim and on the 3 mm extended rim sub-volume showed higher median performances (C-index: 0.65 ± 0.02 and 0.64 ± 0.05, respectively), while models based on the corresponding tumour core volumes performed less (C-index: 0.59 ± 0.01). The difference in C-index between the 5 mm tumour rim and the corresponding core volume showed a statistical trend (p = 0.10). After additional prospective validation, the consideration of tumour sub-volumes may be a promising way to improve prognostic radiomic risk models.

[Artificial intelligence and machine learning in oncologic imaging]

Machine learning (ML) is entering many areas of society, including medicine. This transformation has the potential to drastically change medicine and medical practice. These aspects become particularly clear when considering the different stages of oncologic patient care and the involved interdisciplinary and intermodality interactions. In recent publications, computers-in collaboration with humans or alone-have been outperforming humans regarding tumor identification, tumor classification, estimating prognoses, and evaluation of treatments. In addition, ML algorithms, e.g., artificial neural networks (ANNs), which constitute the drivers behind many of the latest achievements in ML, can deliver this level of performance in a reproducible, fast, and inexpensive manner. In the future, artificial intelligence applications will become an integral part of the medical profession and offer advantages for oncologic diagnostics and treatment.

Diagnosis of Invasive Lung Adenocarcinoma Based on Chest CT Radiomic Features of Part-Solid Pulmonary Nodules: A Multicenter Study

Background Solid components of part-solid nodules (PSNs) at CT are reflective of invasive adenocarcinoma, but studies describing radiomic features of PSNs and the perinodular region are lacking. Purpose To develop and to validate radiomic signatures diagnosing invasive lung adenocarcinoma in PSNs compared with the Brock, clinical-semantic features, and volumetric models. Materials and Methods This retrospective multicenter study (https://ClinicalTrials.gov, NCT03872362) included 291 patients (median age, 60 years; interquartile range, 55-65 years; 191 women) from January 2013 to October 2017 with 297 PSN lung adenocarcinomas split into training (n = 229) and test (n = 68) data sets. Radiomic features were extracted from the different regions (gross tumor volume [GTV], solid, ground-glass, and perinodular). Random-forest models were trained using clinical-semantic, volumetric, and radiomic features, and an online nodule calculator was used to compute the Brock model. Performances of models were evaluated using standard metrics such as area under the curve (AUC), accuracy, and calibration. The integrated discrimination improvement was applied to assess model performance changes after the addition of perinodular features. Results The radiomics model based on ground-glass and solid features yielded an AUC of 0.98 (95% confidence interval [CI]: 0.96, 1.00) on the test data set, which was significantly higher than the Brock (AUC, 0.83 [95% CI: 0.72, 0.94]; P = .007), clinical-semantic (AUC, 0.90 [95% CI: 0.83, 0.98]; P = .03), volumetric GTV (AUC, 0.87 [95% CI: 0.78, 0.96]; P = .008), and radiomics GTV (AUC, 0.88 [95% CI: 0.80, 0.96]; P = .01) models. It also achieved the best accuracy (93% [95% CI: 84%, 98%]). Both this model and the model with added perinodular features showed good calibration, whereas adding perinodular features did not improve the performance (integrated discrimination improvement, -0.02; P = .56). Conclusion Separating ground-glass and solid CT radiomic features of part-solid nodules was useful in diagnosing the invasiveness of lung adenocarcinoma, yielding a better predictive performance than the Brock, clinical-semantic, volumetric, and radiomics gross tumor volume models. Online supplemental material is available for this article. See also the editorial by Nishino in this issue. Published under a CC BY 4.0 license.

Evaluating Solid Lung Adenocarcinoma Anaplastic Lymphoma Kinase Gene Rearrangement Using Noninvasive Radiomics Biomarkers

Purpose

To develop a radiogenomics classifier to assess anaplastic lymphoma kinase (ALK) gene rearrangement status in pretreated solid lung adenocarcinoma noninvasively.

Materials and Methods

This study consisted of 140 consecutive pretreated solid lung adenocarcinoma patients with complete enhanced CT scans who were tested for both EGFR mutations and ALK status. Pre-contrast CT and standard post-contrast CT radiogenomics machine learning classifiers were designed as two separate classifiers. In each classifier, dataset was randomly split into training and independent testing group on a 7:3 ratio, accordingly subjected to a 5-fold cross-validation. After normalization, best feature subsets were selected by Pearson correlation coefficient (PCC) and analysis of variance (ANOVA) or recursive feature elimination (RFE), whereupon a radiomics classifier was built with support vector machine (SVM). The discriminating performance was assessed with the area under receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Results

In classifier one, 98 cases were selected as training data set, 42 cases as independent testing data set. In classifier two, 87 cases were selected as training data set, 37 cases as independent testing data set. Both classifiers extracted 851 radiomics features. The top 25 pre-contrast features and top 19 post-contrast features were selected to build optimal ALK+ radiogenomics classifiers accordingly. The accuracies, AUCs, sensitivity, specificity, PPV, and NPV of pre-contrast CT classifier were 78.57%, 80.10% (CI: 0.6538–0.9222), 71.43%, 82.14%, 66.67%, and 85.19%, respectively. Those results of standard post-contrast CT classifier were 81.08%, 82.85% (CI: 0.6630–0.9567), 76.92%, 83.33%, 71.43%, and 86.96%.

Conclusion

Solid lung adenocarcinoma ALK+ radiogenomics classifier of standard post-contrast CT radiomics biomarkers produced superior performance compared with that of pre-contrast one, suggesting that post-contrast CT radiomics should be recommended in the context of solid lung adenocarcinoma radiogenomics AI. Standard post-contrast CT machine learning radiogenomics classifier could help precisely identify solid adenocarcinoma ALK rearrangement status, which may act as a pragmatic and cost-efficient substitute for traditional invasive ALK status test.

Radiomics as a personalized medicine tool in lung cancer: Separating the hope from the hype

Radiomics has become a popular image analysis method in the last few years. Its key hypothesis is that medical images harbor biological, prognostic and predictive information that is not revealed upon visual inspection. In contrast to previous work with a priori defined imaging biomarkers, radiomics instead calculates image features at scale and uses statistical methods to identify those most strongly associated to outcome. This builds on years of research into computer aided diagnosis and pattern recognition. While the potential of radiomics to aid personalized medicine is widely recognized, several technical limitations exist which hinder biomarker translation. Aspects of the radiomic workflow lack repeatability or reproducibility under particular circumstances, which is a key requirement for the translation of imaging biomarkers into clinical practice. One of the most commonly studied uses of radiomics is for personalized medicine applications in Non-Small Cell Lung Cancer (NSCLC). In this review, we summarize reported methodological limitations in CT based radiomic analyses together with suggested solutions. We then evaluate the current NSCLC radiomics literature to assess the risk associated with accepting the published conclusions with respect to these limitations. We review different complementary scoring systems and initiatives that can be used to critically appraise data from radiomics studies. Wider awareness should improve the quality of ongoing and future radiomics studies and advance their potential as clinically relevant biomarkers for personalized medicine in patients with NSCLC.

18F-fluorodeoxyglucose positron emission tomography/computed tomography-based radiomic features for prediction of epidermal growth factor receptor mutation status and prognosis in patients with lung adenocarcinoma

Background

To investigate whether radiomic features from (¹⁸F)-fluorodeoxyglucose positron emission tomography/computed tomography [(¹⁸F)-FDG PET/CT] can predict epidermal growth factor receptor (EGFR) mutation status and prognosis in patients with lung adenocarcinoma.

Methods

One hundred and seventy-four consecutive patients with lung adenocarcinoma underwent (¹⁸F)-FDG PET/CT and EGFR gene testing were retrospectively analyzed. Radiomic features combined with clinicopathological factors to construct a random forest (RF) model to identify EGFR mutation status. The mutant/wild-type model was trained on a training group (n=139) and validated in an independent validation group (n=35). The second RF classifier predicting the 19/21 mutation site was also built and evaluated in an EGFR mutation subset (training group, n=80; validation group, n=25). Radiomic score and 5 clinicopathological factors were integrated into a multivariate Cox proportional hazard (CPH) model for predicting overall survival (OS). AUC (the area under the receiver characteristic curve) and C-index were calculated to evaluate the model’s performance.

Results

Of 174 patients, 109 (62.6%) harbored EGFR mutations, 21L858R was the most common mutation type [55.9% (61/109)]. The mutant/wild-type model was identified in the training (AUC, 0.77) and validation (AUC, 0.71) groups. The 19/21 mutation site model had an AUC of 0.82 and 0.73 in the training and validation groups, respectively. The C-index of the CPH model was 0.757. The survival time between targeted therapy and chemotherapy for patients with EGFR mutations was significantly different (P=0.03).

Conclusions

Radiomic features based on (¹⁸F)-FDG PET/CT combined with clinicopathological factors could reflect genetic differences and predict EGFR mutation type and prognosis.

Do we need to see to believe?-radiomics for lung nodule classification and lung cancer risk stratification

Despite multiple recent advances, the diagnosis and management of lung cancer remain challenging and it continues to be the deadliest malignancy. In 2011, the National Lung Screening Trial (NLST) reported 20% reduction in lung cancer related mortality using annual low dose chest computed tomography (CT). These results led to the approval and nationwide establishment of lung cancer CT-based lung cancer screening programs. These findings have been further validated by the recently published Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON) and Multicentric Italian Lung Detection (MILD) trials, the latter showing benefit of screening even beyond the 5 years. However, the implementation of lung cancer screening has been impeded by several challenges, including the differentiation between benign and malignant nodules, the large number of false positive studies and the detection of indolent, potentially clinically insignificant lung cancers (overdiagnosis). Hence, the development of non-invasive strategies to accurately classify and risk stratify screen-detected pulmonary nodules in order to individualize clinical management remains a high priority area of research. Radiomics is a recently coined term which refers to the process of imaging feature extraction and quantitative analysis of clinical diagnostic images to characterize the nodule phenotype beyond what is possible with conventional radiologist assessment. Even though it is still in early phase, several studies have already demonstrated that radiomics approaches are potentially useful for lung nodule classification, risk stratification, individualized management and prediction of overall prognosis. The goal of this review is to summarize the current literature regarding the radiomics of screen-detected lung nodules, highlight potential challenges and discuss its clinical application along with future goals and challenges.

Computed tomography-derived radiomic signature of head and neck squamous cell carcinoma (peri)tumoral tissue for the prediction of locoregional recurrence and distant metastasis after concurrent chemo-radiotherapy

Introduction

In this study, we investigate the role of radiomics for prediction of overall survival (OS), locoregional recurrence (LRR) and distant metastases (DM) in stage III and IV HNSCC patients treated by chemoradiotherapy. We hypothesize that radiomic analysis of (peri-)tumoral tissue may detect invasion of surrounding tissues indicating a higher chance of locoregional recurrence and distant metastasis.

Methods

Two comprehensive data sources were used: the Dutch Cancer Society Database (Alp 7072, DESIGN) and “Big Data To Decide” (BD2Decide). The gross tumor volumes (GTV) were delineated on contrast-enhanced CT. Radiomic features were extracted using the RadiomiX Discovery Toolbox (OncoRadiomics, Liege, Belgium). Clinical patient features such as age, gender, performance status etc. were collected. Two machine learning methods were chosen for their ability to handle censored data: Cox proportional hazards regression and random survival forest (RSF). Multivariable clinical and radiomic Cox/ RSF models were generated based on significance in univariable cox regression/ RSF analyses on the held out data in the training dataset. Features were selected according to a decreasing hazard ratio for Cox and relative importance for RSF.

Results

A total of 444 patients with radiotherapy planning CT-scans were included in this study: 301 head and neck squamous cell carcinoma (HNSCC) patients in the training cohort (DESIGN) and 143 patients in the validation cohort (BD2DECIDE). We found that the highest performing model was a clinical model that was able to predict distant metastasis in oropharyngeal cancer cases with an external validation C-index of 0.74 and 0.65 with the RSF and Cox models respectively. Peritumoral radiomics based prediction models performed poorly in the external validation, with C-index values ranging from 0.32 to 0.61 utilizing both feature selection and model generation methods.

Conclusion

Our results suggest that radiomic features from the peritumoral regions are not useful for the prediction of time to OS, LR and DM.

Nuclear medicine radiomics in precision medicine: why we can't do without artificial intelligence

In recent years, radiomics, defined as the extraction of large amounts of quantitative features from medical images, has gained emerging interest. Radiomics consists of the extraction of handcrafted features combined with sophisticated statistical methods or machine learning algorithms for modelling, or deep learning algorithms that both learn features from raw data and perform modelling. These features have the potential to serve as non-invasive biomarkers for tumor characterization, prognostic stratification and response prediction, thereby contributing to precision medicine. However, especially in nuclear medicine, variable results are obtained when using radiomics for these purposes. Individual studies show promising results, but due to small numbers of patients per study and little standardization, it is difficult to compare and validate results on other datasets. This review describes the radiomic pipeline, its applications and the increasing role of artificial intelligence within the field. Furthermore, the challenges that need to be overcome to achieve clinical translation are discussed, so that, eventually, radiomics, combined with clinical data and other biomarkers, can contribute to precision medicine, by providing the right treatment to the right patient, with the right dose, at the right time.

Radiomics and deep learning in lung cancer

Lung malignancies have been extensively characterized through radiomics and deep learning. By providing a three-dimensional characterization of the lesion, models based on radiomic features from computed tomography (CT) and positron-emission tomography (PET) have been developed to detect nodules, distinguish malignant from benign lesions, characterize their histology, stage, and genotype. Deep learning models have been applied to automatically segment organs at risk in lung cancer radiotherapy, stratify patients according to the risk for local and distant recurrence, and identify patients candidate for molecular targeted therapy and immunotherapy. Moreover, radiomics has also been applied successfully to predict side effects such as radiation- and immunotherapy-induced pneumonitis and differentiate lung injury from recurrence. Radiomics could also untap the potential for further use of the cone beam CT acquired for treatment image guidance, four-dimensional CT, and dose-volume data from radiotherapy treatment plans. Radiomics is expected to increasingly affect the clinical practice of treatment of lung tumors, optimizing the end-to-end diagnosis-treatment-follow-up chain. The main goal of this article is to provide an update on the current status of lung cancer radiomics.

CT Radiomics Signature of Tumor and Peritumoral Lung Parenchyma to Predict Nonsmall Cell Lung Cancer Postsurgical Recurrence Risk

RATIONALE AND OBJECTIVES

To estimate recurrence risk after surgery in nonsmall cell lung cancer (NSCLC) patients by employing tumoral and peritumoral radiomics analysis.

MATERIALS AND METHODS

One-hundred twenty-four surgically treated stage IA-IIB NSCLC patients' data from 2008 to 2013 were retrospectively collected. Patient outcome was defined as local recurrence (LR), distant metastasis (DM), and (sum of LR and DM) total recurrence (TR) at follow-up. Volumetric region of interests (ROIs) were drawn for the tumor, peritumoral lung parenchyma (2 cm around the tumor) and involved lobe on CT images. Ninety-four (morphological, first-order, textural, fractal-based) radiomics features were extracted from the ROIs and datasets were created from single or combined ROIs. Predictive models were built with radiomics signature (RS) and clinicopathological data, and the area under the curve (AUC) was used to evaluate the performance. Radiomics score was calculated with the best models' feature coefficients, low- and high-risk groups of patients defined accordingly. Kaplan-Meier curves were built, and the log-rank test was used for comparison among low- and high-risk groups. Differences in recurrence risk among the two risk groups were calculated (chi-square test).

RESULTS

Fifty-six patients developed TR (25 LR, 31 DM). The tumor-node-metastasis (TNM) stage recurrence predictability (AUC_TR 0.680; AUC_DM 0.672; AUC_LR 0.580) was substantially improved when RS was added to the predictive model (AUC_TR 0.760; AUC_DM 0.759; AUC_LR 0.750). Seventy-five percent of high-risk patients developed TR. Recurrence risk of the high-risk group was 16-fold higher than that of the low-risk group (p < 0.001).

CONCLUSION

Combination of the tumoral and peritumoral RS with TNM staging system outperformed TNM staging alone in individualized recurrence risk estimation of patients with surgically treated NSCLC.

Clinical, Conventional CT and Radiomic Feature-Based Machine Learning Models for Predicting ALK Rearrangement Status in Lung Adenocarcinoma Patients

Objectives: To predict the anaplastic lymphoma kinase (ALK) mutations in lung adenocarcinoma patients non-invasively with machine learning models that combine clinical, conventional CT and radiomic features.

Methods: This retrospective study included 335 lung adenocarcinoma patients who were randomly divided into a primary cohort (268 patients; 90 ALK-rearranged; and 178 ALK wild-type) and a test cohort (67 patients; 22 ALK-rearranged; and 45 ALK wild-type). One thousand two hundred and eighteen quantitative radiomic features were extracted from the semi-automatically delineated volume of interest (VOI) of the entire tumor using both the original and the pre-processed non-enhanced CT images. Twelve conventional CT features and seven clinical features were also collected. Normalized features were selected using a sequential of the F-test-based method, the density-based spatial clustering of applications with noise (DBSCAN) method, and the recursive feature elimination (RFE) method. Selected features were then used to build three predictive models (radiomic, radiological, and integrated models) for the ALK-rearranged phenotype by a soft voting classifier. Models were evaluated in the test cohort using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity, and the performances of three models were compared using the DeLong test.

Results: Our results showed that the addition of clinical information and conventional CT features significantly enhanced the validation performance of the radiomic model in the primary cohort (AUC = 0.83–0.88, P = 0.01), but not in the test cohort (AUC = 0.80–0.88, P = 0.29). The majority of radiomic features associated with ALK mutations reflected information around and within the high-intensity voxels of lesions. The presence of the cavity and left lower lobe location were new imaging phenotypic patterns in association with ALK-rearranged tumors. Current smoking was strongly correlated with non-ALK-mutated lung adenocarcinoma.

Conclusions: Our study demonstrates that radiomics-derived machine learning models can potentially serve as a non-invasive tool to identify ALK mutation of lung adenocarcinoma.

Stable and discriminating radiomic predictor of recurrence in early stage non-small cell lung cancer: Multi-site study

OBJECTIVES

To evaluate whether combining stability and discriminability criteria in building radiomic classifiers will improve the prognosis of cancer recurrence in early stage non-small cell lung cancer on non-contrast computer tomography (CT).

MATERIALS AND METHODS

CT scans of 610 patients with early stage (IA, IB, IIA) NSCLC from four independent cohorts were evaluated. A total of 350 patients from Cleveland Clinic Foundation and University of Pennsylvania were divided into two equal sets for training (D₁) and validation set (D₂). 80 patients from The Cancer Genome Atlas Lung Adenocarcinoma and Squamous Cell Carcinoma and 195 patients from The Cancer Imaging Archive, were used as independent second (D₃) and third (D₄) validation sets. A linear discriminant analysis (LDA) classifier was built based on the most stable and discriminate features. In addition, a radiomic risk score (RRS) was generated by using least absolute shrinkage and selection operator, Cox regression model to predict time to progression (TTP) following surgery.

RESULTS

A feature selection strategy focusing on both feature discriminability and stability resulted in the classifier having a higher discriminability on validation datasets compared to the discriminability alone criteria in discriminating cancer recurrence (D₂, AUC of 0.75 vs. 0.65; D₃, 0.74 vs. 0.62; D₄, 0.76 vs. 0.63). The RRS generated by most stable-discriminating features was significantly associated with TTP compared to discriminating alone criteria (HR = 1.66, C-index of 0.72 vs. HR = 1.04, C-index of 0.62).

CONCLUSION

Accounting for both stability and discriminability yielded a more generalizable classifier for predicting cancer recurrence and TTP in early stage NSCLC.

A multi-scanner study of MRI radiomics in uterine cervical cancer: prediction of in-field tumor control after definitive radiotherapy based on a machine learning method including peritumoral regions

PURPOSE

This study aimed to identify the most appropriate volume of interest (VOI) setting in prognostic prediction using pretreatment magnetic resonance imaging (MRI) radiomic analysis for cervical cancer (CC) treated with definitive radiotherapy.

MATERIALS AND METHODS

The study participants were 87 patients who had undergone pretreatment MRI and definitive radiotherapy for CC. VOI_tumor was created with tumor alone and VOI_+4 mm-VOI_+20 mm mechanically expanded by 4-20 mm around each VOI_tumor in axial T2-weighted images (T2WI) and an apparent diffusion coefficient (ADC) map. A model was constructed to predict recurrence within the irradiation field within 2 years after treatment using imaging features from the VOI of each sequence. Sorting ability was evaluated by area under the receiver operator characteristic curve (AUC-ROC) analysis.

RESULTS

VOI expansion improved AUC-ROCs compared with the predictive models of VOI_tumor (0.59 and 0.67 in T2WI and ADC, respectively). The AUC-ROCs of the models with imaging features from expanded VOI_+4 mm in T2WI and VOI_+4 mm and VOI_+8 mm in ADC were 0.82, 0.82, and 0.86, respectively.

CONCLUSION

Recurrence could be predicted with high accuracy using expanded VOI for CC treated with definitive radiotherapy, suggesting that including the pathological characteristics of invasive margins in radiomics may improve predictive ability.

Stability and reproducibility of computed tomography radiomic features extracted from peritumoral regions of lung cancer lesions

PURPOSE

Recent efforts have demonstrated that radiomic features extracted from the peritumoral region, the area surrounding the tumor parenchyma, have clinical utility in various cancer types. However, as like any radiomic features, peritumoral features could also be unstable and/or nonreproducible. Hence, the purpose of this study was to assess the stability and reproducibility of computed tomography (CT) radiomic features extracted from the peritumoral regions of lung lesions where stability was defined as the consistency of a feature by different segmentations, and reproducibility was defined as the consistency of a feature to different image acquisitions.

METHODS

Stability was measured utilizing the "moist run" dataset and reproducibility was measured utilizing the Reference Image Database to Evaluate Therapy Response test-retest dataset. Peritumoral radiomic features were extracted from incremental distances of 3-12 mm outside the tumor segmentation. A total of 264 statistical, histogram, and texture radiomic features were assessed from the selected peritumoral region-of-interests (ROIs). All features (except wavelet texture features) were extracted using standardized algorithms defined by the Image Biomarker Standardisation Initiative. Stability and reproducibility of features were assessed using the concordance correlation coefficient. The clinical utility of stable and reproducible peritumoral features was tested in three previously published lung cancer datasets using overall survival as the endpoint.

RESULTS

Features found to be stable and reproducible, regardless of the peritumoral distances, included statistical, histogram, and a subset of texture features suggesting that these features are less affected by changes (e.g., size or shape) of the peritumoral region due to different segmentations and image acquisitions. The stability and reproducibility of Laws and wavelet texture features were inconsistent across all peritumoral distances. The analyses also revealed that a subset of features were consistently stable irrespective of the initial parameters (e.g., seed point) for a given segmentation algorithm. No significant differences were found in stability for features that were extracted from ROIs bounded by a lung parenchyma mask versus ROIs that were not bounded by a lung parenchyma mask (i.e., peritumoral regions that extended outside of lung parenchyma). After testing the clinical utility of peritumoral features, stable and reproducible features were shown to be more likely to create repeatable models than unstable and nonreproducible features.

CONCLUSIONS

This study identified a subset of stable and reproducible CT radiomic features extracted from the peritumoral region of lung lesions. The stable and reproducible features identified in this study could be applied to a feature selection pipeline for CT radiomic analyses. According to our findings, top performing features in survival models were more likely to be stable and reproducible hence, it may be best practice to utilize them to achieve repeatable studies and reduce the chance of overfitting.