Quantifying the robustness of [ 18 F]FDG-PET/CT radiomic features with respect to tumor delineation in head and neck and pancreatic cancer patients

Artificial intelligence for tumor [18F]FDG-PET imaging: Advancement and future trends-part I

The advent of sophisticated image analysis techniques has facilitated the extraction of increasingly complex data, such as radiomic features, from various imaging modalities, including [18F]FDG PET/CT, a well-established cornerstone of oncological imaging. Furthermore, the use of artificial intelligence (AI) algorithms has shown considerable promise in enhancing the interpretation of these quantitative parameters. Additionally, AI-driven models enable the integration of parameters from multiple imaging modalities along with clinical data, facilitating the development of comprehensive models with significant clinical impact. However, challenges remain regarding standardization and validation of the AI-powered models, as well as their implementation in real-world clinical practice. The variability in imaging acquisition protocols, segmentation methods, and feature extraction approaches across different institutions necessitates robust harmonization efforts to ensure reproducibility and clinical utility. Moreover, the successful translation of AI models into clinical practice requires prospective validation in large cohorts, as well as seamless integration into existing workflows to assess their ability to enhance clinicians' performance. This review aims to provide an overview of the literature and highlight three key applications: diagnostic impact, prediction of treatment response, and long-term patient prognostication. In the first part, we will focus on head and neck, lung, breast, gastroesophageal, colorectal, and gynecological malignancies.

Temporal Validation of an FDG-PET-Radiomic Model for Distant-Relapse-Free-Survival After Radio-Chemotherapy for Pancreatic Adenocarcinoma

Background/Objectives: Pancreatic cancer is a very aggressive disease with a poor prognosis, even when diagnosed at an early stage. This study aimed to validate and refine a radiomic-based [18F]FDG-PET model to predict distant relapse-free survival (DRFS) in patients with unresectable locally advanced pancreatic cancer (LAPC). Methods: A Cox regression model incorporating two radiomic features (RFs) and cancer stage (III vs. IV) was temporally validated using a larger cohort (215 patients treated between 2005-2022). Patients received concurrent chemoradiotherapy with capecitabine and hypo-fractionated Intensity Modulated Radiotherapy (IMRT). Data were split into training (145 patients, 2005-2017) and validation (70 patients, 2017-2022) groups. Seventy-eight RFs were extracted, harmonized, and analyzed using machine learning to develop refined models. Results: The model incorporating Statistical-Percentile10, Morphological-ComShift, and stage demonstrated moderate predictive accuracy (training: C-index = 0.632; validation: C-index = 0.590). When simplified to include only Statistical-Percentile10, performance improved slightly in the validation group (C-index = 0.601). Adding GLSZM3D-grayLevelVariance to Statistical-Percentile10, while excluding Morphological-ComShift, further enhanced accuracy (training: C-index = 0.654; validation: C-index = 0.623). Despite these refinements, all versions showed similar moderate ability to stratify patients into risk classes. Conclusions: [18F]FDG-PET radiomic features are robust predictors of DRFS after chemoradiotherapy in LAPC. Despite moderate performance, these models hold promise for patient risk stratification. Further validation with external cohorts is ongoing.

Radiomics for clinical decision support in radiation oncology

Radiomics is a promising tool for the development of quantitative biomarkers to support clinical decision-making. It has been shown to improve the prediction of response to treatment and outcome in different settings, particularly in the field of radiation oncology by optimising the dose delivery solutions and reducing the rate of radiation-induced side effects, leading to a fully personalised approach. Despite the promising results offered by radiomics at each of these stages, standardised methodologies, reproducibility and interpretability of results are still lacking, limiting the potential clinical impact of these tools. In this review, we briefly describe the principles of radiomics and the most relevant applications of radiomics at each stage of cancer management in the framework of radiation oncology. Furthermore, the integration of radiomics into clinical decision support systems is analysed, defining the challenges and offering possible solutions for translating radiomics into a clinically applicable tool.

The effect of harmonization on the variability of PET radiomic features extracted using various segmentation methods

PURPOSE

This study aimed to examine the robustness of positron emission tomography (PET) radiomic features extracted via different segmentation methods before and after ComBat harmonization in patients with non-small cell lung cancer (NSCLC).

METHODS

We included 120 patients (positive recurrence = 46 and negative recurrence = 74) referred for PET scanning as a routine part of their care. All patients had a biopsy-proven NSCLC. Nine segmentation methods were applied to each image, including manual delineation, K-means (KM), watershed, fuzzy-C-mean, region-growing, local active contour (LAC), and iterative thresholding (IT) with 40, 45, and 50% thresholds. Diverse image discretizations, both without a filter and with different wavelet decompositions, were applied to PET images. Overall, 6741 radiomic features were extracted from each image (749 radiomic features from each segmented area). Non-parametric empirical Bayes (NPEB) ComBat harmonization was used to harmonize the features. Linear Support Vector Classifier (LinearSVC) with L1 regularization For feature selection and Support Vector Machine classifier (SVM) with fivefold nested cross-validation was performed using StratifiedKFold with 'n_splits' set to 5 to predict recurrence in NSCLC patients and assess the impact of ComBat harmonization on the outcome.

RESULTS

From 749 extracted radiomic features, 206 (27%) and 389 (51%) features showed excellent reliability (ICC ≥ 0.90) against segmentation method variation before and after NPEB ComBat harmonization, respectively. Among all, 39 features demonstrated poor reliability, which declined to 10 after ComBat harmonization. The 64 fixed bin widths (without any filter) and wavelets (LLL)-based radiomic features set achieved the best performance in terms of robustness against diverse segmentation techniques before and after ComBat harmonization. The first-order and GLRLM and also first-order and NGTDM feature families showed the largest number of robust features before and after ComBat harmonization, respectively. In terms of predicting recurrence in NSCLC, our findings indicate that using ComBat harmonization can significantly enhance machine learning outcomes, particularly improving the accuracy of watershed segmentation, which initially had fewer reliable features than manual contouring. Following the application of ComBat harmonization, the majority of cases saw substantial increase in sensitivity and specificity.

CONCLUSION

Radiomic features are vulnerable to different segmentation methods. ComBat harmonization might be considered a solution to overcome the poor reliability of radiomic features.

Preoperative MRI radiomic analysis for predicting local tumor progression in colorectal liver metastases before microwave ablation

PURPOSE

Radiomics may aid in predicting prognosis in patients with colorectal liver metastases (CLM). Consistent data is available on CT, yet limited data is available on MRI. This study assesses the capability of MRI-derived radiomic features (RFs) to predict local tumor progression-free survival (LTPFS) in patients with CLMs treated with microwave ablation (MWA).

METHODS

All CLM patients with pre-operative Gadoxetic acid-MRI treated with MWA in a single institution between September 2015 and February 2022 were evaluated. Pre-procedural information was retrieved retrospectively. Two observers manually segmented CLMs on T2 and T1-Hepatobiliary phase (T1-HBP) scans. After inter-observer variability testing, 148/182 RFs showed robustness on T1-HBP, and 141/182 on T2 (ICC > 0.7).Cox multivariate analysis was run to establish clinical (CLIN-mod), radiomic (RAD-T1, RAD-T2), and combined (COMB-T1, COMB-T2) models for LTPFS prediction.

RESULTS

Seventy-six CLMs (43 patients) were assessed. Median follow-up was 14 months. LTP occurred in 19 lesions (25%).CLIN-mod was composed of minimal ablation margins (MAMs), intra-segment progression and primary tumor grade and exhibited moderately high discriminatory power in predicting LTPFS (AUC = 0.89, p = 0.0001). Both RAD-T1 and RAD-T2 were able to predict LTPFS: (RAD-T1: AUC = 0.83, p = 0.0003; RAD-T2: AUC = 0.79, p = 0.001). Combined models yielded the strongest performance (COMB-T1: AUC = 0.98, p = 0.0001; COMB-T2: AUC = 0.95, p = 0.0003). Both combined models included MAMs and tumor regression grade; COMB-T1 also featured 10th percentile of signal intensity, while tumor flatness was present in COMB-T2.

CONCLUSION

MRI-based radiomic evaluation of CLMs is feasible and potentially useful for LTP prediction. Combined models outperformed clinical or radiomic models alone for LTPFS prediction.

Comparison of semi-automatic and manual segmentation methods for tumor delineation on head and neck squamous cell carcinoma (HNSCC) positron emission tomography (PET) images

Objective. Accurate and reproducible tumor delineation on positron emission tomography (PET) images is required to validate predictive and prognostic models based on PET radiomic features. Manual segmentation of tumors is time-consuming whereas semi-automatic methods are easily implementable and inexpensive. This study assessed the reliability of semi-automatic segmentation methods over manual segmentation for tumor delineation in head and neck squamous cell carcinoma (HNSCC) PET images.Approach. We employed manual and six semi-automatic segmentation methods (just enough interaction (JEI), watershed, grow from seeds (GfS), flood filling (FF), 30% SUVmax and 40%SUVmax threshold) using 3D slicer software to extract 128 radiomic features from FDG-PET images of 100 HNSCC patients independently by three operators. We assessed the distributional properties of all features and considered 92 log-transformed features for subsequent analysis. For each paired comparison of a feature, we fitted a separate linear mixed effect model using the method (two levels; manual versus one semi-automatic method) as a fixed effect and the subject and the operator as the random effects. We estimated different statistics-the intraclass correlation coefficient agreement (aICC), limits of agreement (LoA), total deviation index (TDI), coverage probability (CP) and coefficient of individual agreement (CIA)-to evaluate the agreement between the manual and semi-automatic methods.Main results. Accounting for all statistics across 92 features, the JEI method consistently demonstrated acceptable agreement with the manual method, with median values of aICC = 0.86, TDI = 0.94, CP = 0.66, and CIA = 0.91.Significance. This study demonstrated that JEI method is a reliable semi-automatic method for tumor delineation on HNSCC PET images.

Semi-automatic fine delineation scheme for pancreatic cancer

BACKGROUND

Pancreatic cancer fine delineation in medical images by physicians is a major challenge due to the vast volume of medical images and the variability of patients.

PURPOSE

A semi-automatic fine delineation scheme was designed to assist doctors in accurately and quickly delineating the cancer target region to improve the delineation accuracy of pancreatic cancer in computed tomography (CT) images and effectively reduce the workload of doctors.

METHODS

A target delineation scheme in image blocks was also designed to provide more information for the deep learning delineation model. The start and end slices of the image block were manually delineated by physicians, and the cancer in the middle slices were accurately segmented using a three-dimensional Res U-Net model. Specifically, the input of the network is the CT image of the image block and the delineation of the cancer in the start and end slices, while the output of the network is the cancer area in the middle slices of the image block. Meanwhile, the model performance of pancreatic cancer delineation and the workload of doctors in different image block sizes were studied.

RESULTS

We used 37 3D CT volumes for training, 11 volumes for validating and 11 volumes for testing. The influence of different image block sizes on doctors' workload was compared quantitatively. Experimental results showed that the physician's workload was minimal when the image block size was 5, and all cancer could be accurately delineated. The Dice similarity coefficient was 0.894 ± 0.029, the 95% Hausdorff distance was 3.465 ± 0.710 mm, the normalized surface Dice was 0.969 ± 0.019. By completing the accurate delineation of all the CT images, the speed of the new method is 2.16 times faster than that of manual sketching.

CONCLUSION

Our proposed 3D semi-automatic delineative method based on the idea of block prediction could accurately delineate CT images of pancreatic cancer and effectively deal with the challenges of class imbalance, background distractions, and non-rigid geometrical features. This study had a significant advantage in reducing doctors' workload, and was expected to help doctors improve their work efficiency in clinical application.

Multicentric development and evaluation of [18F]FDG PET/CT and CT radiomic models to predict regional and/or distant recurrence in early-stage non-small cell lung cancer treated by stereotactic body radiation therapy

PURPOSE

To develop machine learning models to predict regional and/or distant recurrence in patients with early-stage non-small cell lung cancer (ES-NSCLC) after stereotactic body radiation therapy (SBRT) using [18F]FDG PET/CT and CT radiomics combined with clinical and dosimetric parameters.

METHODS

We retrospectively collected 464 patients (60% for training and 40% for testing) from University Hospital of Liège and 63 patients from University Hospital of Brest (external testing set) with ES-NSCLC treated with SBRT between 2010 and 2020 and who had undergone pretreatment [18F]FDG PET/CT and planning CT. Radiomic features were extracted using the PyRadiomics toolbox®. The ComBat harmonization method was applied to reduce the batch effect between centers. Clinical, radiomic, and combined models were trained and tested using a neural network approach to predict regional and/or distant recurrence.

RESULTS

In the training (n = 273) and testing sets (n = 191 and n = 63), the clinical model achieved moderate performances to predict regional and/or distant recurrence with C-statistics from 0.53 to 0.59 (95% CI, 0.41, 0.67). The radiomic (original_firstorder_Entropy, original_gldm_LowGrayLevelEmphasis and original_glcm_DifferenceAverage) model achieved higher predictive ability in the training set and kept the same performance in the testing sets, with C-statistics from 0.70 to 0.78 (95% CI, 0.63, 0.88) while the combined model performs moderately well with C-statistics from 0.50 to 0.62 (95% CI, 0.37, 0.69).

CONCLUSION

Radiomic features extracted from pre-SBRT analog and digital [18F]FDG PET/CT outperform clinical parameters in the prediction of regional and/or distant recurrence and to discuss an adjuvant systemic treatment in ES-NSCLC. Prospective validation of our models should now be carried out.

Interobserver agreement on definition of the target volume in stereotactic radiotherapy for pancreatic adenocarcinoma using different imaging modalities

PURPOSE

The aim of this study was to evaluate interobserver agreement (IOA) on target volume definition for pancreatic cancer (PACA) within the Radiosurgery and Stereotactic Radiotherapy Working Group of the German Society of Radiation Oncology (DEGRO) and to identify the influence of imaging modalities on the definition of the target volumes.

METHODS

Two cases of locally advanced PACA and one local recurrence were selected from a large SBRT database. Delineation was based on either a planning 4D CT with or without (w/wo) IV contrast, w/wo PET/CT, and w/wo diagnostic MRI. Novel compared to other studies, a combination of four metrics was used to integrate several aspects of target volume segmentation: the Dice coefficient (DSC), the Hausdorff distance (HD), the probabilistic distance (PBD), and the volumetric similarity (VS).

RESULTS

For all three GTVs, the median DSC was 0.75 (range 0.17-0.95), the median HD 15 (range 3.22-67.11) mm, the median PBD 0.33 (range 0.06-4.86), and the median VS was 0.88 (range 0.31-1). For ITVs and PTVs the results were similar. When comparing the imaging modalities for delineation, the best agreement for the GTV was achieved using PET/CT, and for the ITV and PTV using 4D PET/CT, in treatment position with abdominal compression.

CONCLUSION

Overall, there was good GTV agreement (DSC). Combined metrics appeared to allow a more valid detection of interobserver variation. For SBRT, either 4D PET/CT or 3D PET/CT in treatment position with abdominal compression leads to better agreement and should be considered as a very useful imaging modality for the definition of treatment volumes in pancreatic SBRT. Contouring does not appear to be the weakest link in the treatment planning chain of SBRT for PACA.

Stacking Ensemble Learning-Based [18F]FDG PET Radiomics for Outcome Prediction in Diffuse Large B-Cell Lymphoma

This study aimed to develop an analytic approach based on [18F]FDG PET radiomics using stacking ensemble learning to improve the outcome prediction in diffuse large B-cell lymphoma (DLBCL). Methods: In total, 240 DLBCL patients from 2 medical centers were divided into the training set (n = 141), internal testing set (n = 61), and external testing set (n = 38). Radiomics features were extracted from pretreatment [18F]FDG PET scans at the patient level using 4 semiautomatic segmentation methods (SUV threshold of 2.5, SUV threshold of 4.0 [SUV4.0], 41% of SUVmax, and SUV threshold of mean liver uptake [PERCIST]). All extracted features were harmonized with the ComBat method. The intraclass correlation coefficient was used to evaluate the reliability of radiomics features extracted by different segmentation methods. Features from the most reliable segmentation method were selected by Pearson correlation coefficient analysis and the LASSO (least absolute shrinkage and selection operator) algorithm. A stacking ensemble learning approach was applied to build radiomics-only and combined clinical-radiomics models for prediction of 2-y progression-free survival and overall survival based on 4 machine learning classifiers (support vector machine, random forests, gradient boosting decision tree, and adaptive boosting). Confusion matrix, receiver-operating-characteristic curve analysis, and survival analysis were used to evaluate the model performance. Results: Among 4 semiautomatic segmentation methods, SUV4.0 segmentation yielded the highest interobserver reliability, with 830 (66.7%) selected radiomics features. The combined model constructed by the stacking method achieved the best discrimination performance. For progression-free survival prediction in the external testing set, the areas under the receiver-operating-characteristic curve and accuracy of the stacking-based combined model were 0.771 and 0.789, respectively. For overall survival prediction, the stacking-based combined model achieved an area under the curve of 0.725 and an accuracy of 0.763 in the external testing set. The combined model also demonstrated a more distinct risk stratification than the International Prognostic Index in all sets (log-rank test, all P < 0.05). Conclusion: The combined model that incorporates [18F]FDG PET radiomics and clinical characteristics based on stacking ensemble learning could enable improved risk stratification in DLBCL.

Sensitivity of standardised radiomics algorithms to mask generation across different software platforms

The field of radiomics continues to converge on a standardised approach to image processing and feature extraction. Conventional radiomics requires a segmentation. Certain features can be sensitive to small contour variations. The industry standard for medical image communication stores contours as coordinate points that must be converted to a binary mask before image processing can take place. This study investigates the impact that the process of converting contours to mask can have on radiomic features calculation. To this end we used a popular open dataset for radiomics standardisation and we compared the impact of masks generated by importing the dataset into 4 medical imaging software. We interfaced our previously standardised radiomics platform with these software using their published application programming interface to access image volume, masks and other data needed to calculate features. Additionally, we used super-sampling strategies to systematically evaluate the impact of contour data pre processing methods on radiomic features calculation. Finally, we evaluated the effect that using different mask generation approaches could have on patient clustering in a multi-center radiomics study. The study shows that even when working on the same dataset, mask and feature discrepancy occurs depending on the contour to mask conversion technique implemented in various medical imaging software. We show that this also affects patient clustering and potentially radiomic-based modelling in multi-centre studies where a mix of mask generation software is used. We provide recommendations to negate this issue and facilitate reproducible and reliable radiomics.

Multicentric development and evaluation of 18F-FDG PET/CT and MRI radiomics models to predict para-aortic lymph node involvement in locally advanced cervical cancer

PURPOSE

To develop machine learning models to predict para-aortic lymph node (PALN) involvement in patients with locally advanced cervical cancer (LACC) before chemoradiotherapy (CRT) using 18F-FDG PET/CT and MRI radiomics combined with clinical parameters.

METHODS

We retrospectively collected 178 patients (60% for training and 40% for testing) in 2 centers and 61 patients corresponding to 2 further external testing cohorts with LACC between 2010 to 2022 and who had undergone pretreatment analog or digital 18F-FDG PET/CT, pelvic MRI and surgical PALN staging. Only primary tumor volumes were delineated. Radiomics features were extracted using the Radiomics toolbox®. The ComBat harmonization method was applied to reduce the batch effect between centers. Different prediction models were trained using a neural network approach with either clinical, radiomics or combined models. They were then evaluated on the testing and external validation sets and compared.

RESULTS

In the training set (n = 102), the clinical model achieved a good prediction of the risk of PALN involvement with a C-statistic of 0.80 (95% CI 0.71, 0.87). However, it performed in the testing (n = 76) and external testing sets (n = 30 and n = 31) with C-statistics of only 0.57 to 0.67 (95% CI 0.36, 0.83). The ComBat-radiomic (GLDZM_HISDE_PET_FBN64 and Shape_maxDiameter2D3_PET_FBW0.25) and ComBat-combined (FIGO 2018 and same radiomics features) models achieved very high predictive ability in the training set and both models kept the same performance in the testing sets, with C-statistics from 0.88 to 0.96 (95% CI 0.76, 1.00) and 0.85 to 0.92 (95% CI 0.75, 0.99), respectively.

CONCLUSIONS

Radiomic features extracted from pre-CRT analog and digital 18F-FDG PET/CT outperform clinical parameters in the decision to perform a para-aortic node staging or an extended field irradiation to PALN. Prospective validation of our models should now be carried out.

Role of PET/CT in Oropharyngeal Cancers

Oropharyngeal squamous cell carcinoma (OPSCC) comprises cancers of the tonsils, tongue base, soft palate, and uvula. The staging of oropharyngeal cancers varies depending upon the presence or absence of human papillomavirus (HPV)-directed pathogenesis. The incidence of HPV-associated oropharyngeal cancer (HPV + OPSCC) is expected to continue to rise over the coming decades. PET/CT is a useful modality for the diagnosis, staging, and follow up of patients with oropharyngeal cancers undergoing treatment and surveillance.

External validation of an 18F-FDG-PET radiomic model predicting survival after radiotherapy for oropharyngeal cancer

PURPOSE/OBJECTIVE

The purpose of the study is to externally validate published ¹⁸F-FDG-PET radiomic models for outcome prediction in patients with oropharyngeal cancer treated with chemoradiotherapy.

MATERIAL/METHODS

Outcome data and pre-radiotherapy PET images of 100 oropharyngeal cancer patients (stage IV:78) treated with concomitant chemotherapy to 66-69 Gy/30 fr were available. Tumors were segmented using a previously validated semi-automatic method; 450 radiomic features (RF) were extracted according to IBSI (Image Biomarker Standardization Initiative) guidelines. Only one model for cancer-specific survival (CSS) prediction was suitable to be independently tested, according to our criteria. This model, in addition to HPV status, SUVmean and SUVmax, included two independent meta-factors (F_i), resulting from combining selected RF clusters. In a subgroup of 66 patients with complete HPV information, the global risk score R was computed considering the original coefficients and was tested by Cox regression as predictive of CSS. Independently, only the radiomic risk score R_F derived from F_i was tested on the same subgroup to learn about the radiomics contribution to the model. The metabolic tumor volume (MTV) was also tested as a single predictor and its prediction performances were compared to the global and radiomic models. Finally, the validation of MTV and the radiomic score R_F were also tested on the entire dataset.

RESULTS

Regarding the analysis of the subgroup with HPV information, with a median follow-up of 41.6 months, seven patients died due to cancer. R was confirmed to be associated to CSS (p value = 0.05) with a C-index equal 0.75 (95% CI=0.62-0.85). The best cut-off value (equal to 0.15) showed high ability in patient stratification (p=0.01, HR=7.4, 95% CI=1.6-11.4). The 5-year CSS for R were 97% (95% CI: 93-100%) vs 74% (56-92%) for low- and high-risk groups, respectively. R_F and MTV alone were also significantly associated to CSS for the subgroup with an almost identical C-index. According to best cut-off value (R_F>0.12 and MTV>15.5cc), the 5-year CSS were 96% (95% CI: 89-100%) vs 65% (36-94%) and 97% (95% CI: 88-100%) vs 77% (58-93%) for RF and MTV, respectively. Results regarding RF and MTV were confirmed in the overall group.

CONCLUSION

A previously published PET radiomic model for CSS prediction was independently validated. Performances of the model were similar to the ones of using only the MTV, without improvement of prediction accuracy.

Prognostic analysis of curatively resected pancreatic cancer using harmonized positron emission tomography radiomic features

BACKGROUND

Texture features reflecting tumour heterogeneity enable us to investigate prognostic factors. The R package ComBat can harmonize the quantitative texture features among several positron emission tomography (PET) scanners. We aimed to identify prognostic factors among harmonized PET radiomic features and clinical information from pancreatic cancer patients who underwent curative surgery.

METHODS

Fifty-eight patients underwent preoperative enhanced dynamic computed tomography (CT) scanning and fluorodeoxyglucose PET/CT using four PET scanners. Using LIFEx software, we measured PET radiomic parameters including texture features with higher order and harmonized these PET parameters. For progression-free survival (PFS) and overall survival (OS), we evaluated clinical information, including age, TNM stage, and neural invasion, and the harmonized PET radiomic features based on univariate Cox proportional hazard regression. Next, we analysed the prognostic indices by multivariate Cox proportional hazard regression (1) by using either significant (p < 0.05) or borderline significant (p = 0.05-0.10) indices in the univariate analysis (first multivariate analysis) or (2) by using the selected features with random forest algorithms (second multivariate analysis). Finally, we checked these multivariate results by log-rank test.

RESULTS

Regarding the first multivariate analysis for PFS after univariate analysis, age was the significant prognostic factor (p = 0.020), and MTV and GLCM contrast were borderline significant (p = 0.051 and 0.075, respectively). Regarding the first multivariate analysis of OS, neural invasion, Shape sphericity and GLZLM LZLGE were significant (p = 0.019, 0.042 and 0.0076). In the second multivariate analysis, only MTV was significant (p = 0.046) for PFS, whereas GLZLM LZLGE was significant (p = 0.047), and Shape sphericity was borderline significant (p = 0.088) for OS. In the log-rank test, age, MTV and GLCM contrast were borderline significant for PFS (p = 0.08, 0.06 and 0.07, respectively), whereas neural invasion and Shape sphericity were significant (p = 0.03 and 0.04, respectively), and GLZLM LZLGE was borderline significant for OS (p = 0.08).

CONCLUSIONS

Other than the clinical factors, MTV and GLCM contrast for PFS and Shape sphericity and GLZLM LZLGE for OS may be prognostic PET parameters. A prospective multicentre study with a larger sample size may be warranted.

Does radiomics play a role in the diagnosis, staging and re-staging of gastroesophageal junction adenocarcinoma?

Radiomics is an emerging field of investigation in medicine consisting in the extraction of quantitative features from conventional medical images and exploring their potentials in improving diagnosis, prognosis and outcome prediction after therapy. Clinical applications are still limited, mostly due to reproducibility and repeatability issues as well as to limited interpretability of predictive radiomic-based features/signatures. In the specific case of gastroesophageal junction (GEJ) adenocarcinoma, the expectancies are particularly high, mainly due to its increasing incidence and to the limited performance of conventional imaging techniques in assessing correct diagnosis and accurate pre-surgical tumor characterization. Accordingly, current literature was reviewed, emphasizing the methodological quality. In addition, papers were scored according to the Radiomic Quality Score (RQS), weighting more the clinical applicability and generalizability of the resulting models. According to the criteria of the search, only two papers were retained: the resulting technical quality was relatively high for both, while the corresponding RQS were 15 and 19 (on a scale of 31). Although the potentials of radiomics in the setting of GEJ adenocarcinoma are relevant, they remain largely unexplored, warranting an urgent need of high-quality, possibly prospective, multicenter studies.

CT radiomic predictors of local relapse after SBRT for lung oligometastases from colorectal cancer: a single institute pilot study

OBJECTIVES

To assess the potential of radiomic features (RFs) extracted from simulation computed tomography (CT) images in discriminating local progression (LP) after stereotactic body radiotherapy (SBRT) in the management of lung oligometastases (LOM) from colorectal cancer (CRC).

MATERIALS AND METHODS

Thirty-eight patients with 70 LOM treated with SBRT were analyzed. The largest LOM was considered as most representative for each patient and was manually delineated by two blinded radiation oncologists. In all, 141 RFs were extracted from both contours according to IBSI (International Biomarker Standardization Initiative) recommendations. Based on the agreement between the two observers, 134/141 RFs were found to be robust against delineation (intraclass correlation coefficient [ICC] > 0.80); independent RFs were then assessed by Spearman correlation coefficients. The association between RFs and LP was assessed with Mann-Whitney test and univariate logistic regression (ULR): the discriminative power of the most informative RF was quantified by receiver-operating characteristics (ROC) analysis through area under curve (AUC).

RESULTS

In all, 15/38 patients presented LP. Median time to progression was 14.6 months (range 2.4-66 months); 5/141 RFs were significantly associated to LP at ULR analysis (p < 0.05); among them, 4 RFs were selected as robust and independent: Statistical_Variance (AUC = 0.75, p = 0.002), Statistical_Range (AUC = 0.72, p = 0.013), Grey Level Size Zone Matrix (GLSZM) _zoneSizeNonUniformity (AUC = 0.70, p = 0.022), Grey Level Dependence Zone Matrix (GLDZM) _zoneDistanceEntropy (AUC = 0.70, p = 0.026). Importantly, the RF with the best performance (Statisical_Variance) is simply representative of density heterogeneity within LOM.

CONCLUSION

Four RFs extracted from planning CT were significantly associated with LP of LOM from CRC treated with SBRT. Results encourage further research on a larger population aiming to define a usable radiomic score combining the most predictive RFs and, possibly, additional clinical features.

Radiomics: a primer on high-throughput image phenotyping

Radiomics is a high-throughput approach to image phenotyping. It uses computer algorithms to extract and analyze a large number of quantitative features from radiological images. These radiomic features collectively describe unique patterns that can serve as digital fingerprints of disease. They may also capture imaging characteristics that are difficult or impossible to characterize by the human eye. The rapid development of this field is motivated by systems biology, facilitated by data analytics, and powered by artificial intelligence. Here, as part of Abdominal Radiology's special issue on Quantitative Imaging, we provide an introduction to the field of radiomics. The technique is formally introduced as an advanced application of data analytics, with illustrating examples in abdominal radiology. Artificial intelligence is then presented as the main driving force of radiomics, and common techniques are defined and briefly compared. The complete step-by-step process of radiomic phenotyping is then broken down into five key phases. Potential pitfalls of each phase are highlighted, and recommendations are provided to reduce sources of variation, non-reproducibility, and error associated with radiomics.

Artificial Intelligence Applied to Pancreatic Imaging: A Narrative Review

The diagnosis, evaluation, and treatment planning of pancreatic pathologies usually require the combined use of different imaging modalities, mainly, computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). Artificial intelligence (AI) has the potential to transform the clinical practice of medical imaging and has been applied to various radiological techniques for different purposes, such as segmentation, lesion detection, characterization, risk stratification, or prediction of response to treatments. The aim of the present narrative review is to assess the available literature on the role of AI applied to pancreatic imaging. Up to now, the use of computer-aided diagnosis (CAD) and radiomics in pancreatic imaging has proven to be useful for both non-oncological and oncological purposes and represents a promising tool for personalized approaches to patients. Although great developments have occurred in recent years, it is important to address the obstacles that still need to be overcome before these technologies can be implemented into our clinical routine, mainly considering the heterogeneity among studies.

Impact of image quality on radiomics applications

Radiomics features extracted from medical images have been widely reported to be useful in the patient specific outcome modeling for variety of assessment and prediction purposes. Successful application of radiomics features as imaging biomarkers, however, is dependent on the robustness of the approach to the variation in each step of the modeling workflow. Variation in the input image quality is one of the main sources that impacts the reproducibility of radiomics analysis when a model is applied to broader range of medical imaging data. The quality of medical image is generally affected by both the scanner related factors such as image acquisition/reconstruction settings and the patient related factors such as patient motion. This article aimed to review the published literatures in this field that reported the impact of various imaging factors on the radiomics features through the change in image quality. The literatures were categorized by different imaging modalities and also tabulated based on the imaging parameters and the class of radiomics features included in the study. Strategies for image quality standardization were discussed based on the relevant literatures and recommendations for reducing the impact of image quality variation on the radiomics in multi-institutional clinical trial were summarized at the end of this article.

The role of volumetric and textural analysis of pretreatment 18F-fluorodeoxyglucose PET/computerized tomography images in predicting complete response to transarterial radioembolization in hepatocellular cancer

OBJECTIVE

This study evaluates the role of pretreatment 18F-FDG PET/CT in predicting the response to treatment in patients with hepatocellular cancer (HCC) who applied transarterial radioembolization (TARE) via the volumetric and texture features extracted from 18F-FDG PET/CT images.

METHODS

Thirty-three patients with HCC who had applied TARE [lobar (LT) or superselective (ST)] after 18F-FDG PET/CT were included in the study. Response to the treatment was evaluated from posttherapy magnetic resonance (MR). Patients were divided into two groups: the responder group (RG) (complete responders) and non-RG (NRG) (including partial response, stabile, and progressive). Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) and texture features were extracted from PET/CT images. The differences among MTV, TLG, and texture features between response groups were analyzed with the Mann-Whitney U test. ROC analysis was performed for features with P < 0.05. Spearman correlation analysis was used, and features with correlation coefficient < 0.8 were evaluated with the logistic regression analysis.

RESULTS

Significant differences were detected in TLG, MTV, SHAPE_compacity, GLCM_correlation, GLRLM_GLNU, GLRLM_RLNU, NGLDM_coarseness, NGLDM_busyness, GLZLM_LZHGE, GLZLM_GLNU, and GLZLM_ZLNU between RG and NRG. Multivariate analysis demonstrated that MTV was the only meaningful parameter with an AUC of 0.827 (P = 0.002; 95% CI, 0.688-0.966). The best cutoff value was determined as 74.11 ml with 78.9% sensitivity and 78.6% specificity in discriminating nonresponders.

CONCLUSION

In predicting the curative effect of TARE, multivariate analysis results demonstrated that MTV was the only independent predictor, and MTV higher than 74.11 ml were determined the best predictor of nonresponders.

Analysis of Cross-Combinations of Feature Selection and Machine-Learning Classification Methods Based on [18F]F-FDG PET/CT Radiomic Features for Metabolic Response Prediction of Metastatic Breast Cancer Lesions

BACKGROUND

This study aimed to identify optimal combinations between feature selection methods and machine-learning classifiers for predicting the metabolic response of individual metastatic breast cancer lesions, based on clinical variables and radiomic features extracted from pretreatment [18F]F-FDG PET/CT images.

METHODS

A total of 48 patients with confirmed metastatic breast cancer, who received different treatments, were included. All patients had an [18F]F-FDG PET/CT scan before and after the treatment. From 228 metastatic lesions identified, 127 were categorized as responders (complete or partial metabolic response) and 101 as non-responders (stable or progressive metabolic response), by using the percentage changes in SULpeak (peak standardized uptake values normalized for body lean body mass). The lesion pool was divided into training (n = 182) and testing cohorts (n = 46); for each lesion, 101 image features from both PET and CT were extracted (202 features per lesion). These features, along with clinical and pathological information, allowed the prediction model's construction by using seven popular feature selection methods in cross-combination with another seven machine-learning (ML) classifiers. The performance of the different models was investigated with the receiver-operating characteristic curve (ROC) analysis, using the area under the curve (AUC) and accuracy (ACC) metrics.

RESULTS

The combinations, least absolute shrinkage and selection operator (Lasso) + support vector machines (SVM), or random forest (RF) had the highest AUC in the cross-validation, with 0.93 ± 0.06 and 0.92 ± 0.03, respectively, whereas Lasso + neural network (NN) or SVM, and mutual information (MI) + RF, had the higher AUC and ACC in the validation cohort, with 0.90/0.72, 0.86/0.76, and 87/85, respectively. On average, the models with Lasso and models with SVM had the best mean performance for both AUC and ACC in both training and validation cohorts.

CONCLUSIONS

Image features obtained from a pretreatment [18F]F-FDG PET/CT along with clinical vaiables could predict the metabolic response of metastatic breast cancer lesions, by their incorporation into predictive models, whose performance depends on the selected combination between feature selection and ML classifier methods.

A systematic review and quality of reporting checklist for repeatability and reproducibility of radiomic features

•

Main factors impacting feature stability: Image acquisition, reconstruction, tumor segmentation, and interpolation.

•

Textural features are less robust than morphological or statistical features.

•

A checklist is provided including items that should be reported in a radiomic study.

Purpose

Although quantitative image biomarkers (radiomics) show promising value for cancer diagnosis, prognosis, and treatment assessment, these biomarkers still lack reproducibility. In this systematic review, we aimed to assess the progress in radiomics reproducibility and repeatability in the recent years.

Methods and materials

Four hundred fifty-one abstracts were retrieved according to the original PubMed search pattern with the publication dates ranging from 2017/05/01 to 2020/12/01. Each abstract including the keywords was independently screened by four observers. Forty-two full-text articles were selected for further analysis. Patient population data, radiomic feature classes, feature extraction software, image preprocessing, and reproducibility results were extracted from each article. To support the community with a standardized reporting strategy, we propose a specific reporting checklist to evaluate the feasibility to reproduce each study.

Results

Many studies continue to under-report essential reproducibility information: all but one clinical and all but two phantom studies missed to report at least one important item reporting image acquisition. The studies included in this review indicate that all radiomic features are sensitive to image acquisition, reconstruction, tumor segmentation, and interpolation. However, the amount of sensitivity is feature dependent, for instance, textural features were, in general, less robust than statistical features.

Conclusions

Radiomics repeatability, reproducibility, and reporting quality can substantially be improved regarding feature extraction software and settings, image preprocessing and acquisition, cutoff values for stable feature selection. Our proposed radiomics reporting checklist can serve to simplify and improve the reporting and, eventually, guarantee the possibility to fully replicate and validate radiomic studies.

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.

Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study

Histopathological growth patterns (HGPs) are independent prognosticators for colorectal liver metastases (CRLM). Currently, HGPs are determined postoperatively. In this study, we evaluated radiomics for preoperative prediction of HGPs on computed tomography (CT), and its robustness to segmentation and acquisition variations. Patients with pure HGPs [i.e. 100% desmoplastic (dHGP) or 100% replacement (rHGP)] and a CT-scan who were surgically treated at the Erasmus MC between 2003-2015 were included retrospectively. Each lesion was segmented by three clinicians and a convolutional neural network (CNN). A prediction model was created using 564 radiomics features and a combination of machine learning approaches by training on the clinician's and testing on the unseen CNN segmentations. The intra-class correlation coefficient (ICC) was used to select features robust to segmentation variations; ComBat was used to harmonize for acquisition variations. Evaluation was performed through a 100 × random-split cross-validation. The study included 93 CRLM in 76 patients (48% dHGP; 52% rHGP). Despite substantial differences between the segmentations of the three clinicians and the CNN, the radiomics model had a mean area under the curve of 0.69. ICC-based feature selection or ComBat yielded no improvement. Concluding, the combination of a CNN for segmentation and radiomics for classification has potential for automatically distinguishing dHGPs from rHGP, and is robust to segmentation and acquisition variations. Pending further optimization, including extension to mixed HGPs, our model may serve as a preoperative addition to postoperative HGP assessment, enabling further exploitation of HGPs as a biomarker.

The exploration of quantitative intra-tumoral metabolic heterogeneity in dual-time 18F-FDG PET/CT of pancreatic cancer

PURPOSE

We aimed to analyze the change of quantitative intra-tumoral metabolic heterogeneity consisting of texture features and conventional metabolic parameters of pancreatic cancer (PC) in dual-time 2-deoxy-2(18F) fluoro-D-glucose (18F-FDG) positron emission tomography-computed tomography (PET/CT).

METHODS

A retrospective analysis was conducted considering the texture features and conventional metabolic parameters in dual-time 18F-FDG PET/CT scans of PC patients. Features were extracted based on spatial distribution of 18F-FDG uptake in image. Firstly, the texture features and the conventional metabolic parameters of the delayed scan were both compared with that of the early scan. Statistically different data was defined among them. Secondly, the study evaluated the correlations between retention index (RI) of the texture features and the conventional metabolic parameters. Finally, the variation of texture features in dual-time PET/CT of resectable PC patients and unresectable PC patients was calculated separately.

RESULTS

In total, 183 PC patients were analyzed retrospectively in this research. The conventional metabolic parameters were all statistically different between the early and delayed scans except for metabolic tumor volume (MTV). In the radiomics, there were 59 textural features. Nineteen of 59 texture features were statistically different between the early and delayed scans. Features that were more than 10% different during two scans were observed in a substantial percentage of patients. Weak correlations were only found between MTV, TLG (Total lesion glycolysis), SUVpeak and the RI of some texture features in early or delayed scans. There were obviously fewer features with significant difference in resectable PC group than in unresectable PC group. Most features showing the difference in unresectable group while no significant difference in resectable group.

CONCLUSIONS

This study investigated the change and inner correlations of quantitative tumoral metabolic heterogeneity in the dual-time 18F-FDG-PET/CT scan of PC patients. Some features displayed the difference between dual-time scans. Conventional metabolic parameters were weakly related to the change of texture feature. The change of texture feature in resectable PC group was different from that in unresectable PC group. This result is potential to provide more information for the image evaluation of PC.

A different overview of staging PET/CT images in patients with esophageal cancer: the role of textural analysis with machine learning methods

OBJECTIVE

This study evaluates the ability of several machine learning (ML) algorithms, developed using volumetric and texture data extracted from baseline 18F-FDG PET/CT studies performed initial staging of patient with esophageal cancer (EC), to predict survival and histopathology.

METHODS

The initial staging 18F-FDG PET/CT images obtained on newly diagnosed EC patients between January 2008 and June 2019 were evaluated using LIFEx software. A region of interest (ROI) of the primary tumor was created and volumetric and textural features were obtained. A significant relationship between these features and pathological subtypes, 1-year, and 5-year survival was investigated. Due to the nonhomogeneity of the data, nonparametric test (The Mann-Whitney U test) was used for each feature, in pairwise comparisons of independent variables. A p value of < 0.05 was considered significant. Receiver operating curve (ROC) analysis was performed for features with p < 0.05. Correlation between the significant features was evaluated with Spearman correlation test; features with correlation coefficient < 0.8 were evaluated with several ML algorithms.

RESULTS

In predicting survival in a 1-year follow-up J48 was obtained as the most successful algorithm (AUC: 0.581, PRC: 0.565, MCC: 0.258, acc: 64.29%). 5-year survival results were more promising than 1-year survival results with (AUC: 0.820, PRC: 0.860, MCC: 271, acc: 81.36%) by logistic regression. It is revealed that the most successful algorithm was naive bayes (AUC: 0.680 PRC: 0.776, MCC: 0.298, acc: 82.66%) in the histopathological discrimination.

CONCLUSION

Texture analysis with ML algorithms could be predictive of overall survival and discriminating histopathological subtypes of EC.

Making Radiomics More Reproducible across Scanner and Imaging Protocol Variations: A Review of Harmonization Methods

Radiomics converts medical images into mineable data via a high-throughput extraction of quantitative features used for clinical decision support. However, these radiomic features are susceptible to variation across scanners, acquisition protocols, and reconstruction settings. Various investigations have assessed the reproducibility and validation of radiomic features across these discrepancies. In this narrative review, we combine systematic keyword searches with prior domain knowledge to discuss various harmonization solutions to make the radiomic features more reproducible across various scanners and protocol settings. Different harmonization solutions are discussed and divided into two main categories: image domain and feature domain. The image domain category comprises methods such as the standardization of image acquisition, post-processing of raw sensor-level image data, data augmentation techniques, and style transfer. The feature domain category consists of methods such as the identification of reproducible features and normalization techniques such as statistical normalization, intensity harmonization, ComBat and its derivatives, and normalization using deep learning. We also reflect upon the importance of deep learning solutions for addressing variability across multi-centric radiomic studies especially using generative adversarial networks (GANs), neural style transfer (NST) techniques, or a combination of both. We cover a broader range of methods especially GANs and NST methods in more detail than previous reviews.

Effects of interobserver and interdisciplinary segmentation variabilities on CT-based radiomics for pancreatic cancer

Radiomics is a method to mine large numbers of quantitative imaging features and develop predictive models. It has shown exciting promise for improved cancer decision support from early detection to personalized precision treatment, and therefore offers a desirable new direction for pancreatic cancer where the mortality remains high despite the current care and intense research. For radiomics, interobserver segmentation variability and its effect on radiomic feature stability is a crucial consideration. While investigations have been reported for high-contrast cancer sites such as lung cancer, no studies to date have investigated it on CT-based radiomics for pancreatic cancer. With three radiation oncology observers and three radiology observers independently contouring on the contrast CT of 21 pancreatic cancer patients, we conducted the first interobserver segmentation variability study on CT-based radiomics for pancreatic cancer. Moreover, our novel investigation assessed whether there exists an interdisciplinary difference between the two disciplines. For each patient, a consensus tumor volume was generated using the simultaneous truth and performance level expectation algorithm, using the dice similarity coefficient (DSC) to assess each observer's delineation against the consensus volume. Radiation oncology observers showed a higher average DSC of 0.81 ± 0.06 than the radiology observers at 0.69 ± 0.16 (p = 0.002). On a panel of 1277 radiomic features, the intraclass correlation coefficients (ICC) was calculated for all observers and those of each discipline. Large variations of ICCs were observed for different radiomic features, but ICCs were generally higher for the radiation oncology group than for the radiology group. Applying a threshold of ICC > 0.75 for considering a feature as stable, 448 features (35%) were found stable for the radiation oncology group and 214 features (16%) were stable from the radiology group. Among them, 205 features were found stable for both groups. Our results provide information for interobserver segmentation variability and its effect on CT-based radiomics for pancreatic cancer. An interesting interdisciplinary variability found in this study also introduces new considerations for the deployment of radiomics models.

Quantitative radiomics features in diffuse large B-cell lymphoma: does segmentation method matter?

INTRODUCTION

Radiomics features may predict outcome in diffuse large B-cell lymphoma (DLBCL). Currently, multiple segmentation methods are used to calculate metabolic tumor volume (MTV). We assessed the influence of segmentation method on the discriminative power of radiomics features in DLBCL for patient level and for the largest lesion. Methods: 50 baseline ¹⁸F-fluorodeoxyglucose positron emission tomography computed tomography (PET/CT) scans of DLBCL patients who progressed or relapsed within 2 years after diagnosis were matched on uptake time and reconstruction method with 50 baseline PET/CT scans of DLBCL patients without progression. Scans were analysed using 6 semi-automatic segmentation methods (standardized uptake value (SUV)4.0, SUV2.5, 41% of the maximum SUV, 50% of the SUVpeak, majority vote (MV)2 and MV3, respectively). Based on these segmentations, 490 radiomics features were extracted at patient level and 486 features for the largest lesion. To quantify the agreement between features extracted from different segmentation methods, the intra-class correlation (ICC) agreement was calculated for each method compared to SUV4.0. The feature space was reduced by deleting features that had high Pearson correlations (≥0.7) with the previously established predictors MTV and/or SUVpeak. Model performance was assessed using stratified repeated cross-validation with 5 folds and 2000 repeats yielding the mean receiver-operating characteristics curve integral (CV-AUC) for all segmentation methods using logistic regression with backward feature selection. Results: The percentage of features yielding an ICC ≥0.75 compared to the SUV4.0 segmentation was lowest for A50P both at patient level and for the largest lesion, with 77.3% and 66.7% of the features yielding an ICC ≥0.75, respectively. Features were not highly correlated with MTV, with at least 435 features at patient level and 409 features for the largest lesion for all segmentation methods with a correlation coefficient <0.7. Features were highly correlated with SUVpeak (at least 190 and 134 were uncorrelated, respectively). CV-AUCs ranged between 0.69±0.11 and 0.84±0.09 for patient level, and between 0.69±0.11 and 0.73±0.10 for lesion level. Conclusion: Even though there are differences in the actual radiomics feature values derived and selected features between segmentation methods, there is no substantial difference in the discriminative power of radiomics features between segmentation methods.

Evaluation of Hybrid PET/MRI for Gross Tumor Volume (GTV) Delineation in Colorectal Cancer Liver Metastases Radiotherapy

Purpose

Hybrid PET/MRI has been increasingly incorporated into the practice of radiation oncologists since it contains both anatomical and biological data and may bring about personalized radiation plans for each patient. The objective of this study was to evaluate the feasibility of GTV delineation from hybrid PET/MRI compared with that from current-practice MRI during radiotherapy planning in patients with colorectal liver metastases.

Patients and Methods

Twenty-four patients (thirty lesions) with colorectal liver metastases were prospectively enrolled in this study. Three physicians delineated the target volume with the most popular delineating methods-the visual method. First of all, differences among the three observers were assessed. The difference and correlation of GTV values obtained by MRI, PET, and hybrid PET/MRI were subjected to statistical analysis afterwards. Finally, the dice similarity coefficient (DSC) was calculated to assess the spatial overlap. Based on the value of DSC, we also evaluate the correlation between DSC and tumor size. GTV-MRI was set as a reference.

Results

There was no significant difference among observers in GTV-MRI (F=0.118, p=0.889), GTV-PET (F=0.070, p=0.933) and GTV-PET/MRI (F=0.40, p=0.961). 83.33% of GTV-PET/MRI and 63.33% of GTV-PET were larger than the reference GTV-MRI. Statistical analysis revealed that GTV-PET/MRI (p<0.001) and GTV-PET (p<0.05) diverged statistically significantly from GTV-MRI. GTV-PET (r=0.992, p<0.001) and GTV-PET/MRI (r=0.997, p<0.001) were significantly related to GTV-MRI. The average DSC value between GTV-MRI and GTV-PET was 0.51 (range 0-0.90) and that between GTV-MRI and GTV-PET/MRI was 0.72 (range 0.42-0.90). There was a positive correlation between the DSC and GTV-MRI (r=0.851, p<0.05).

Conclusion

With the database used, there is good agreement among observers. Hybrid PET/MRI in colorectal liver metastases radiotherapy may affect the GTV delineation. Moreover, the overlap degree between GTV-MRI and GTV-PET/MRI is higher and increases with volume.

Influence of inter-observer delineation variability on radiomic features of the parotid gland

PURPOSE

This study aimed to quantify the variability in the values of radiomic features extracted from a right parotid gland (RPG) delineated by a series of independent observers.

METHODS

This was a secondary analysis of anonymous data from a delineation workshop. Inter-observer variability of the RPG from 40 participants was quantified using DICE similarity coefficient (DSC) and Hausdorff distance (HD). An additional contour was generated using Varian SmartSegmentation. Radiomic features extracted include four shape features, six histogram features, and 32 texture features. The absolute mean paired percentage difference (PPD) in feature values from the expert and participants were ranked . Feature robustness was classified using pre- determined thresholds.

RESULTS

63% of participants achieved a DSC > 0.7, the auto- segmentation DSC was 0.76. The average HD for the participants was 16.16 mm ± 0.66 mm, and 15.16 mm for the auto-segmentation. 48% (n = 20) and 33% (n = 14) of features were deemed to be robust with a mean absolute PPD < 5%, for the auto-segmentation and manual delineations respectively; the majority of which were from the grey-run length matrix family. 7% (n = 3) of features from the auto- segmentation and 10% (n = 4) from the manual contours were deemed to be unstable with a mean absolute PPD > 50%. The value of the most robust feature was not related to DSC and HD.

CONCLUSION

Inter-observer delineation variability affects the value of the radiomic features extracted from the RPG. This study identifies the radiomic features least sensitive to these uncertainties. Further investigation of the clinical relevance of these features in prediction of xerostomia is warranted.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

Background: Although many works have supported the utility of PET radiomics, several authors have raised concerns over the robustness and replicability of the results. This study aimed to perform a systematic review on the topic of PET radiomics and the used methodologies. Methods: PubMed was searched up to 15 October 2020. Original research articles based on human data specifying at least one tumor type and PET image were included, excluding those that apply only first-order statistics and those including fewer than 20 patients. Each publication, cancer type, objective and several methodological parameters (number of patients and features, validation approach, among other things) were extracted. Results: A total of 290 studies were included. Lung (28%) and head and neck (24%) were the most studied cancers. The most common objective was prognosis/treatment response (46%), followed by diagnosis/staging (21%), tumor characterization (18%) and technical evaluations (15%). The average number of patients included was 114 (median = 71; range 20–1419), and the average number of high-order features calculated per study was 31 (median = 26, range 1–286). Conclusions: PET radiomics is a promising field, but the number of patients in most publications is insufficient, and very few papers perform in-depth validations. The role of standardization initiatives will be crucial in the upcoming years.

CT-derived radiomic features to discriminate histologic characteristics of pancreatic neuroendocrine tumors

PURPOSE

To assess the ability of radiomic features (RF) extracted from contrast-enhanced CT images (ceCT) and non-contrast-enhanced (non-ceCT) in discriminating histopathologic characteristics of pancreatic neuroendocrine tumors (panNET).

METHODS

panNET contours were delineated on pre-surgical ceCT and non-ceCT. First- second- and higher-order RF (adjusted to eliminate redundancy) were extracted and correlated with histological panNET grade (G1 vs G2/G3), metastasis, lymph node invasion, microscopic vascular infiltration. Mann-Whitney with Bonferroni corrected p values assessed differences. Discriminative power of significant RF was calculated for each of the end-points. The performance of conventional-imaged-based-parameters was also compared to RF.

RESULTS

Thirty-nine patients were included (mean age 55-years-old; 24 male). Mean diameters of the lesions were 24 × 27 mm. Sixty-nine RF were considered. Sphericity could discriminate high grade tumors (AUC = 0.79, p = 0.002). Tumor volume (AUC = 0.79, p = 0.003) and several non-ceCT and ceCT RF were able to identify microscopic vascular infiltration: voxel-alignment, neighborhood intensity-difference and intensity-size-zone families (AUC ≥ 0.75, p < 0.001); voxel-alignment, intensity-size-zone and co-occurrence families (AUC ≥ 0.78, p ≤ 0.002), respectively). Non-ceCT neighborhood-intensity-difference (AUC = 0.75, p = 0.009) and ceCT intensity-size-zone (AUC = 0.73, p = 0.014) identified lymph nodal invasion; several non-ceCT and ceCT voxel-alignment family features were discriminative for metastasis (p < 0.01, AUC = 0.80-0.85). Conventional CT 'necrosis' could discriminate for microscopic vascular invasion (AUC = 0.76, p = 0.004) and 'arterial vascular invasion' for microscopic metastasis (AUC = 0.86, p = 0.001). No conventional-imaged-based-parameter was significantly associated with grade and lymph node invasion.

CONCLUSIONS

Radiomic features can discriminate histopathology of panNET, suggesting a role of radiomics as a non-invasive tool for tumor characterization.

TRIAL REGISTRATION NUMBER

NCT03967951, 30/05/2019.

Evaluation of the Prognostic Value of FDG PET/CT Parameters for Patients With Surgically Treated Head and Neck Cancer: A Systematic Review

Importance

Head and neck squamous cell cancer (HNSCC) represents the seventh most frequent cancer worldwide. More than half of the patients diagnosed with HNSCC are treated with primary surgery.

Objective

To report the available evidence on the value of quantitative parameters of fluorodeoxyglucose F 18-labeled positron emission tomography and computed tomography (FDG-PET/CT) performed before surgical treatment of HNSCC to estimate overall survival (OS), disease-free survival (DFS), and distant metastasis (DM) and to discuss their limitations.

Evidence Review

A systematic review of the English-language literature in PubMed/MEDLINE and ScienceDirect published between January 2003 and February 15, 2019, was performed between March 1 and July 27, 2019, to identify articles addressing the association between preoperative FDG-PET/CT parameters and oncological outcomes among patients with HNSCC. Articles included those that addressed the following: (1) cancer of the oral cavity, oropharynx, hypopharynx, or larynx; (2) surgically treated (primary or for salvage); (3) pretreatment FDG-PET/CT; (4) quantitative or semiquantitative evaluation of the FDG-PET/CT parameters; and (5) the association between the value of FDG-PET/CT parameters and clinical outcomes. Quality assessment was performed using the Oxford Centre for Evidence-Based Medicine level of evidence.

Findings

A total of 128 studies were retrieved from the databases, and 36 studies met the inclusion criteria; these studies comprised 3585 unique patients with a median follow-up of 30.6 months (range, 16-53 months). Of these 36 studies, 32 showed an association between at least 1 FDG-PET/CT parameter and oncological outcomes (OS, DFS, and DM). The FDG-PET/CT volumetric parameters (metabolic tumor volume [MTV] and total lesion glycolysis [TLG]) were independent prognostic factors in most of the data, with a higher prognostic value than the maximum standard uptake value (SUVmax). For example, in univariate analysis of OS, the SUVmax was correlated with OS in 5 of 11 studies, MTV in 11 of 12 studies, and TLG in 6 of 9 studies. The spatial distribution of metabolism via textural indices seemed promising, although that factor is currently poorly evaluated: only 3 studies analyzed data from radiomics indices.

Conclusions and Relevance

The findings of this study suggest that the prognostic effectiveness of FDG-PET/CT parameters as biomarkers of OS, DFS, and DM among patients with HNSCC treated with surgery may be valuable. The volumetric parameters (MTV and TLG) seemed relevant for identifying patients with a higher risk of postsurgical disease progression who could receive early therapeutic intervention to improve their prognosis. However, further large-scale studies including exclusively surgery-treated patients stratified according to localization and further analysis of the textural indices are required to define a reliable FDG-PET/CT-based prognostic model of mortality and recurrence risk for these patients.

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

OBJECTIVE

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

MATERIAL AND METHODS

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

RESULTS

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

CONCLUSION

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

KEY POINTS

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

CT and MRI of pancreatic tumors: an update in the era of radiomics

Radiomics is a relatively new approach for image analysis. As a part of radiomics, texture analysis, which consists in extracting a great amount of quantitative data from original images, can be used to identify specific features that can help determining the actual nature of a pancreatic lesion and providing other information such as resectability, tumor grade, tumor response to neoadjuvant therapy or survival after surgery. In this review, the basic of radiomics, recent developments and the results of texture analysis using computed tomography and magnetic resonance imaging in the field of pancreatic tumors are presented. Future applications of radiomics, such as artificial intelligence, are discussed.

MRI-based radiomics in breast cancer: feature robustness with respect to inter-observer segmentation variability

Radiomics is an emerging field using the extraction of quantitative features from medical images for tissue characterization. While MRI-based radiomics is still at an early stage, it showed some promising results in studies focusing on breast cancer patients in improving diagnoses and therapy response assessment. Nevertheless, the use of radiomics raises a number of issues regarding feature quantification and robustness. Therefore, our study aim was to determine the robustness of radiomics features extracted by two commonly used radiomics software with respect to variability in manual breast tumor segmentation on MRI. A total of 129 histologically confirmed breast tumors were segmented manually in three dimensions on the first post-contrast T1-weighted MR exam by four observers: a dedicated breast radiologist, a resident, a Ph.D. candidate, and a medical student. Robust features were assessed using the intraclass correlation coefficient (ICC > 0.9). The inter-observer variability was evaluated by the volumetric Dice Similarity Coefficient (DSC). The mean DSC for all tumors was 0.81 (range 0.19–0.96), indicating a good spatial overlap of the segmentations based on observers of varying expertise. In total, 41.6% (552/1328) and 32.8% (273/833) of all RadiomiX and Pyradiomics features, respectively, were identified as robust and were independent of inter-observer manual segmentation variability.

Training and validation of a robust PET radiomic-based index to predict distant-relapse-free-survival after radio-chemotherapy for locally advanced pancreatic cancer

PURPOSE

To assess the value of 18F-Fluorodeoxyglucose (18F-FDG) PET Radiomic Features (RF) in predicting Distant Relapse Free Survival (DRFS) in patients with Locally AdvancedPancreaticCancer (LAPC) treated with radio-chemotherapy.

MATERIALS & METHODS

One-hundred-ninety-eight RFs were extracted using IBSI (Image Biomarker Standardization Initiative) consistent software from pre-radiotherapy images of 176 LAPC patients treated with moderate hypo-fractionation (44.25 Gy, 2.95 Gy/fr). Tumors were segmented by applying a previously validated semi-automatic method. One-hundred-twenty-six RFs were excluded due to poor reproducibility and/or repeatability and/or inter-scanner variability. The original cohort was randomly split into a training (n = 116) and a validation (n = 60) group. Multi-variable Cox regression was applied to the training group, including only independent RFs in the model. The resulting radiomic index was tested in the validation cohort. The impact of selected clinical variables was also investigated.

RESULTS

The resulting Cox model included two first order RFs: Center of Mass Shift (COMshift) and 10th Intensity percentile (P₁₀) (p = 0.0005, HR = 2.72, 95%CI = 1.54-4.80), showing worse outcomes for patients with lower COMshift and higher P₁₀. Once stratified by quartile values (<lowest quartile vs >highest quartile vs the remaining), the index properly stratified patients according to their DRFS (p = 0.0024, log-rank test). Performances were confirmed in the validation cohort (p = 0.03, HR = 2.53, 95%CI = 0.96-6.65). The addition of clinical factors did not significantly improve the models' performance.

CONCLUSIONS

A radiomic-based index including only two robust PET-RFs predicted DRFS of LAPC patients after radio-chemotherapy. The current results could find relevant applications in the treatment personalization of LAPC. A multi-institution independent validation has been planned.

Development of a new fully three-dimensional methodology for tumours delineation in functional images

Delineation of tumours in Positron Emission Tomography (PET) plays a crucial role in accurate diagnosis and radiotherapy treatment planning. In this context, it is of outmost importance to devise efficient and operator-independent segmentation algorithms capable of reconstructing the tumour three-dimensional (3D) shape. In previous work, we proposed a system for 3D tumour delineation on PET data (expressed in terms of Standardized Uptake Value - SUV), based on a two-step approach. Step 1 identified the slice enclosing the maximum SUV and generated a rough contour surrounding it. Such contour was then used to initialize step 2, where the 3D shape of the tumour was obtained by separately segmenting 2D PET slices, leveraging the slice-by-slice marching approach. Additionally, we combined active contours and machine learning components to improve performance. Despite its success, the slice marching approach poses unnecessary limitations that are naturally removed by performing the segmentation directly in 3D. In this paper, we migrate our system into 3D. In particular, the segmentation in step 2 is now performed by evolving an active surface directly in the 3D space. The key points of such an advancement are that it performs the shape reconstruction on the whole stack of slices simultaneously, naturally leveraging cross-slice information that could not be exploited before. Additionally, it does not require any specific stopping condition, as the active surface naturally reaches a stable topology once convergence is achieved. Performance of this fully 3D approach is evaluated on the same dataset discussed in our previous work, which comprises fifty PET scans of lung, head and neck, and brain tumours. The results have confirmed that a benefit is indeed achieved in practice for all investigated anatomical districts, both quantitatively, through a set of commonly used quality indicators (dice similarity coefficient >87.66%, Hausdorff distance < 1.48 voxel and Mahalanobis distance < 0.82 voxel), and qualitatively in terms of Likert score (>3 in 54% of the tumours).

Imaging phenotype using 18F-fluorodeoxyglucose positron emission tomography-based radiomics and genetic alterations of pancreatic ductal adenocarcinoma

PURPOSE

This study aimed to determine if major gene mutations including in KRAS, SMAD4, TP53, and CDKN2A were related to imaging phenotype using ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET)-based radiomics in patients with pancreatic ductal adenocarcinoma (PDAC).

METHODS

Data on 48 PDAC patients with pretreatment FDG PET/CT who underwent genomic analysis of their tumor tissue were retrospectively analyzed. A total of 35 unique quantitative radiomic features were extracted from PET images, including imaging phenotypes such as pixel intensity, shape, and textural features. Targeted exome sequencing using a customized cancer panel was used for genomic analysis. To assess the predictive performance of genetic alteration using PET-based radiomics, areas under the receiver operating characteristic curve (AUC) were used.

RESULTS

Mutation frequencies were KRAS 87.5%, TP53 70.8%, SMAD4 25.0%, and CDKN2A 18.8%. KRAS gene mutations were significantly associated with low-intensity textural features, including long-run emphasis (AUC = 0.806), zone emphasis (AUC = 0.794), and large-zone emphasis (AUC = 0.829). SMAD4 gene mutations showed significant relationships with standardized uptake value skewness (AUC = 0.727), long-run emphasis (AUC = 0.692), and high-intensity textural features such as run emphasis (AUC = 0.775), short-run emphasis (AUC = 0.736), zone emphasis (AUC = 0.750), and short-zone emphasis (AUC = 0.725). No significant associations were seen between the imaging phenotypes and genetic alterations in TP53 and CDKN2A.

CONCLUSION

Genetic alterations of KRAS and SMAD4 had significant associations with FDG PET-based radiomic features in PDAC. PET-based radiomics may help clinicians predict genetic alteration status in a noninvasive way.

The effects of volume of interest delineation on MRI-based radiomics analysis: evaluation with two disease groups

BACKGROUND

Manual delineation of volume of interest (VOI) is widely used in current radiomics analysis, suffering from high variability. The tolerance of delineation differences and possible influence on each step of radiomics analysis are not clear, requiring quantitative assessment. The purpose of our study was to investigate the effects of delineation of VOIs on radiomics analysis for the preoperative prediction of metastasis in nasopharyngeal carcinoma (NPC) and sentinel lymph node (SLN) metastasis in breast cancer.

METHODS

This study retrospectively enrolled two datasets (NPC group: 238 cases; SLN group: 146 cases). Three operations, namely, erosion, smoothing, and dilation, were implemented on the VOIs accurately delineated by radiologists to generate diverse VOI variations. Then, we extracted 2068 radiomics features and evaluated the effects of VOI differences on feature values by the intra-class correlation coefficient (ICC). Feature selection was conducted by Maximum Relevance Minimum Redundancy combined with 0.632+ bootstrap algorithms. The prediction performance of radiomics models with random forest classifier were tested on an independent validation cohort by the area under the receive operating characteristic curve (AUC).

RESULTS

The larger the VOIs changed, the fewer features with high ICCs. Under any variation, SLN group showed fewer features with ICC ≥ 0.9 compared with NPC group. Not more than 15% top-predictive features identical to the accurate VOIs were observed across feature selection. The differences of AUCs of models derived from VOIs across smoothing or dilation with 3 pixels were not statistically significant compared with the accurate VOIs (p > 0.05) except for T2-weighted fat suppression images (smoothing: 0.845 vs. 0.725, p = 0.001; dilation: 0.800 vs. 0.725, p = 0.042). Dilation with 5 and 7 pixels contributed to remarkable AUCs in SLN group but the opposite in NPC group. The radiomics models did not perform well when tested by data from other delineations.

CONCLUSIONS

Differences in delineation of VOIs affected radiomics analysis, related to specific disease and MRI sequences. Differences from smooth delineation or expansion with 3 pixels width around the tumors or lesions were acceptable. The delineation for radiomics analysis should follow a predefined and unified standard.

Real-time control of respiratory motion: Beyond radiation therapy

Motion management in radiation oncology is an important aspect of modern treatment planning and delivery. Special attention has been paid to control respiratory motion in recent years. However, other medical procedures related to both diagnosis and treatment are likely to benefit from the explicit control of breathing motion. Quantitative imaging - including increasingly important tools in radiology and nuclear medicine - is among the fields where a rapid development of motion control is most likely, due to the need for quantification accuracy. Emerging treatment modalities like focussed-ultrasound tumor ablation are also likely to benefit from a significant evolution of motion control in the near future. In the present article an overview of available respiratory motion systems along with ongoing research in this area is provided. Furthermore, an attempt is made to envision some of the most expected developments in this field in the near future.

Early variation of 18-fluorine-labelled fluorodeoxyglucose PET-derived parameters after chemoradiotherapy as predictors of survival in locally advanced pancreatic carcinoma patients

OBJECTIVE

To investigate if early variation of PET-derived parameters after concomitant chemoradiotherapy (CRT) predicts overall survival (OS), local relapse free survival (LRFS), distant relapse free survival (DRFS) and progression free survival (PFS) in locally advanced pancreatic cancer (LAPC) patients.

METHODS

Fifty-two LAPC patients (median age: 61 years; range: 35-85) with available FDG PET/CT before and after RT (2-6 months, median: 2) were enrolled from May 2005 to June 2015. The predictive value of the percentage variation of mean/maximum standard uptake value (ΔSUVmean/max), metabolic tumour volume (ΔMTV) and total lesion glycolysis (ΔTLG), estimated considering different uptake thresholds (40-50-60%), was investigated between pre- and post-RT PET. The percentage difference between gastrointestinal cancer-associated antigen (ΔGICA) levels measured at the time of PET was also considered. Log-rank test and Cox regression analysis were performed to assess the prognostic value of considered PET-derived parameters on survival outcomes.

RESULTS

The median follow-up was 13 months (range: 4-130). At univariate analysis, ΔTLG50 showed borderline significance in predicting OS (P = 0.05) and was the most significant parameter correlated to LRFS and PFS (P = 0.001). Median LRFS was 4 and 33 months if ΔTLG50 was below or above 35% respectively (P = 0.0003); similarly, median PFS was 3 vs 6 months (P = 0.0009). No significant correlation was found between PET-derived parameters and DRFS, while the ΔGICA was the only borderline significant prognostic value for this endpoint (P = 0.05).

CONCLUSION

PET-derived parameters predict survival in LAPC patients; in particular, ΔTLG50 is the strongest predictor. The combination of these biochemical and imaging biomarkers is promising in identifying patients at higher risk of earlier relapse.

Radiomics in nuclear medicine: robustness, reproducibility, standardization, and how to avoid data analysis traps and replication crisis

Radiomics in nuclear medicine is rapidly expanding. Reproducibility of radiomics studies in multicentre settings is an important criterion for clinical translation. We therefore performed a meta-analysis to investigate reproducibility of radiomics biomarkers in PET imaging and to obtain quantitative information regarding their sensitivity to variations in various imaging and radiomics-related factors as well as their inherent sensitivity. Additionally, we identify and describe data analysis pitfalls that affect the reproducibility and generalizability of radiomics studies. After a systematic literature search, 42 studies were included in the qualitative synthesis, and data from 21 were used for the quantitative meta-analysis. Data concerning measurement agreement and reliability were collected for 21 of 38 different factors associated with image acquisition, reconstruction, segmentation and radiomics-specific processing steps. Variations in voxel size, segmentation and several reconstruction parameters strongly affected reproducibility, but the level of evidence remained weak. Based on the meta-analysis, we also assessed inherent sensitivity to variations of 110 PET image biomarkers. SUV_mean and SUV_max were found to be reliable, whereas image biomarkers based on the neighbourhood grey tone difference matrix and most biomarkers based on the size zone matrix were found to be highly sensitive to variations, and should be used with care in multicentre settings. Lastly, we identify 11 data analysis pitfalls. These pitfalls concern model validation and information leakage during model development, but also relate to reporting and the software used for data analysis. Avoiding such pitfalls is essential for minimizing bias in the results and to enable reproduction and validation of radiomics studies.

PET/MRI-guided GTV delineation during radiotherapy planning in patients with squamous cell carcinoma of the tongue

Purpose

The aim of the study was to evaluate the usefulness and accuracy of 18-fluorine-labeled fluorodeoxyglucose (PET) and magnetic resonance imaging (MRI) hybrid in gross tumor volume (GTV) delineation during radiotherapy planning in patients with carcinoma of the tongue.

Methods

Ten patients with squamous cell carcinoma (SCC) of the tongue underwent computed tomography (CT) and PET/MRI examination. The GTV for primary tumor and lymph nodes (nGTV) were defined on CT (GTV-CT) and compared to GTVs obtained from PET (GTV-PET) and MRI (GTV-MRI) images. Two methods of GTV determination were used: visual interpretation of CT, PET (GTV-PET_vis) and MRI images and quantitative automatic method (Syngovia, Siemens) based on a chosen threshold value (20%, 30%, 40%, 50%) of standardized uptake values (SUV_max) from PET examination (GTV-PET_20%, GTV-PET_30%, etc.). Statistical analysis of differences in GTV values obtained from CT, PET and MRI studies was performed. GTV-CT was used as a reference.

Results

In all, 80% of GTV-MRI and 40% of GTV-PET_vis were larger than GTV-CT. Respectively, 20% of GTV-MRI and 60% of GTV-PET_vis were smaller than GTV-CT. Taking into account all threshold measurements, 70% of volumes were smaller than GTV-CT. GTV-PET_30% were the most closely related volumes to GTV-CT from all threshold methods in 50% of patients. GTV-PET_vis generated the most similar volumes in relation to GTV-CT from all PET measurements. Statistical analysis confirmed those results. Compared to nGTV-CT, 70% of nGTV-MRI and 20% of nGTV-PET_vis were larger. The remaining nGTV-MRI and nGTV-PET_vis measurements were smaller than nGTV-CT. Measurements of all thresholds nGTVs were smaller than nGTV-CTV in 52.5% of cases. nGTV-PET_20% were the most closely related volumes to nGTV-CT in 40% of the cases. Statistical analysis showed that nGTV-PET_20% (p = 0.0468), nGTV-PET_vis (p = 0.0166), and nGTV-PET_50% (p = 0.0166) diverge significantly from nGTV-CT results. nGTV-MRI (p = 0.1141), nGTV-PET_30% (p = 0.2845), and nGTV-PET_40% (p = 0.5076) were significantly related with nGTV-CT.

Conclusion

Combination of PET/MRI provides more information during target tumor mass delineation in radiotherapy planning of patients with SCC of the tongue than other standard imaging methods. The most frequently matching threshold value was 30% of SUV_max for primary tumor delineation and 30–40% of SUV_max for nGTV determination.

Electronic supplementary material

The online version of this article (10.1007/s00066-019-01480-3) contains supplementary material, which is available to authorized users.

Impact of acquisition count statistics reduction and SUV discretization on PET radiomic features in pediatric 18F-FDG-PET/MRI examinations

PURPOSE

The evaluation of features robustness with respect to acquisition and post-processing parameter changes is fundamental for the reliability of radiomics studies. The aim of this study was to investigate the sensitivity of PET radiomic features to acquisition statistics reduction and standardized-uptake-volume (SUV) discretization in PET/MRI pediatric examinations.

METHODS

Twenty-seven lesions were detected from the analysis of twenty-one 18F-FDG-PET/MRI pediatric examinations. By decreasing the count-statistics of the original list-mode data (3 MBq/kg), injected activity reduction was simulated. Two SUV discretization approaches were applied: 1) resampling lesion SUV range into fixed bins numbers (FBN); 2) rounding lesion SUV into fixed bin size (FBS). One hundred and six radiomic features were extracted. Intraclass Correlation Coefficient (ICC), Spearman correlation coefficient and coefficient-of-variation (COV) were calculated to assess feature reproducibility between low tracer activities and full tracer activity feature values.

RESULTS

More than 70% of Shape and first order features, and around 70% and 40% of textural features, when using FBS and FBN methods respectively, resulted robust till 1.2 MBk/kg. Differences in median features reproducibility (ICC) between FBS and FBN datasets were statistically significant for every activity level independently from bin number/size, with higher values for FBS. Differences in median Spearman coefficient (i.e. patient ranking according to feature values) were not statistically significant, varying the intensity resolution (i.e. bin number/size) for either FBS and FBN methods.

CONCLUSIONS

For each simulated count-statistic level, robust PET radiomic features were determined for pediatric PET/MRI examinations. A larger number of robust features were detected when using FBS methods.

Repeatability of 18 F-FDG PET radiomic features: A phantom study to explore sensitivity to image reconstruction settings, noise, and delineation method

BACKGROUND

18 F-fluoro-2-deoxy-D-Glucose positron emission tomography (18 F-FDG PET) radiomics has the potential to guide the clinical decision making in cancer patients, but validation is required before radiomics can be implemented in the clinical setting. The aim of this study was to explore how feature space reduction and repeatability of 18 F-FDG PET radiomic features are affected by various sources of variation such as underlying data (e.g., object size and uptake), image reconstruction methods and settings, noise, discretization method, and delineation method.

METHODS

The NEMA image quality phantom was scanned with various sphere-to-background ratios (SBR), simulating different activity uptakes, including spheres with low uptake, that is, SBR smaller than 1. Furthermore, images of a phantom containing 3D printed inserts reflecting realistic heterogeneity uptake patterns were acquired. Data were reconstructed using various matrix sizes, reconstruction algorithms, and scan durations (noise). For every specific reconstruction and noise level, ten statistically equal replicates were generated. The phantom inserts were delineated using CT and PET-based segmentation methods. A total of 246 radiomic features was extracted from each image dataset. Images were discretized with a fixed number of 64 bins (FBN) and a fixed bin width (FBW) of 0.25 for the high and a FBW of 0.05 for the low uptake data. In terms of feature reduction, we determined the impact of these factors on the composition of feature clusters, which were defined on the basis of Spearman's correlation matrices. To assess feature repeatability, the intraclass correlation coefficient was calculated over the ten replicates.

RESULTS

In general, larger spheres with high uptake resulted in better repeatability compared to smaller low uptake spheres. In terms of repeatability, features extracted from heterogeneous phantom inserts were comparable to features extracted from bigger high uptake spheres. For example, for an EARL-compliant reconstruction, larger and smaller high uptake spheres yielded good repeatability for 32% and 30% of the features, while the heterogeneous inserts resulted in 34% repeatable features. For the low uptake spheres, this was the case for 22% and 20% of the features for bigger and smaller spheres, respectively. Images reconstructed with point-spread-function (PSF) resulted in the highest repeatability when compared with OSEM or time-of-flight, for example, 53%, 30%, and 32% of repeatable features, respectively (for unsmoothed data, discretized with FBN, 300 s scan duration). Reducing image noise (increasing scan duration and smoothing) and using CT-based segmentation for the low uptake spheres yielded improved repeatability. FBW discretization resulted in higher repeatability than FBN discretization, for example, 89% and 35% of the features, respectively (for the EARL-compliant reconstruction and larger high uptake spheres).

CONCLUSION

Feature space reduction and repeatability of 18 F-FDG PET radiomic features depended on all studied factors. The high sensitivity of PET radiomic features to image quality suggests that a high level of image acquisition and preprocessing standardization is required to be used as clinical imaging biomarker.

Ct radiomic features of pancreatic neuroendocrine neoplasms (panNEN) are robust against delineation uncertainty

PURPOSE

The aim of this study was to quantify the impact of CT delineation uncertainty of pancreatic neuroendocrine neoplasms (panNEN) on Radiomic features (RF).

METHODS

Thirty-one previously operated patients were considered. Three expert radiologists contoured panNEN lesions on pre-surgical high-resolution contrast-enhanced CT images and contours were transferred onto pre-contrast CT. Volume agreement was quantified by the DICE index. After images resampling and re-binning, 69 RF were extracted and the impact of inter-observer variability was assessed by Intra-Class Correlation (ICC): ICC > 0.80 was considered as a threshold for "very high" inter-observer agreement.

RESULTS

The median volume was 1.3 cc (range: 0.2-110 cc); a satisfactory inter-observer volume agreement was found (mean DICE = 0.78). Only 4 RF showed ICC < 0.80 (0.48-0.73), including asphericity and three RFs (of five) of the neighborhood intensity difference matrix (NID).

CONCLUSIONS

The impact of inter-observer variability in delineating panNEN on RF was minimum, with the exception of the NID family and asphericity, showing a moderate agreement. These results support the feasibility of studies aiming to assess CT radiomic biomarkers for panNEN.