Radiomic Features Predict Esophagitis Risk in Non-Small Cell Lung Cancer Patients Treated with Radiation

MRI-based delta-radiomic features for prediction of local control in liver lesions treated with stereotactic body radiation therapy

Real-time magnetic resonance image guided stereotactic ablative radiotherapy (MRgSBRT) is used to treat abdominal tumors. Longitudinal data is generated from daily setup images. Our study aimed to identify delta radiomic texture features extracted from these images to predict for local control in patients with liver tumors treated with MRgSBRT. Retrospective analysis of an IRB-approved database identified patients treated with MRgSBRT for primary liver and secondary metastasis histologies. Daily low field strength (0.35 T) images were retrieved, and the gross tumor volume was identified on each image. Next, images' gray levels were equalized, and 39 s-order texture features were extracted. Delta-radiomics were calculated as the difference between feature values on the initial scan and after delivered biological effective doses (BED, α/β = 10) of 20 Gy and 40 Gy. Then, features were ranked by the Gini Index during training of a random forest model. Finally, the area under the receiver operating characteristic curve (AUC) was estimated using a bootstrapped logistic regression with the top two features. We identified 22 patients for analysis. The median dose delivered was 50 Gy in 5 fractions. The top two features identified after delivery of BED 20 Gy were gray level co-occurrence matrix features energy and gray level size zone matrix based large zone emphasis. The model generated an AUC = 0.9011 (0.752-1.0) during bootstrapped logistic regression. The same two features were selected after delivery of a BED 40 Gy, with an AUC = 0.716 (0.600-0.786). Delta-radiomic features after a single fraction of SBRT predicted local control in this exploratory cohort. If confirmed in larger studies, these features may identify patients with radioresistant disease and provide an opportunity for physicians to alter management much sooner than standard restaging after 3 months. Expansion of the patient database is warranted for further analysis of delta-radiomic features.

Lung Subregion Partitioning by Incremental Dose Intervals Improves Omics-Based Prediction for Acute Radiation Pneumonitis in Non-Small-Cell Lung Cancer Patients

Purpose: To evaluate the effectiveness of features obtained from our proposed incremental-dose-interval-based lung subregion segmentation (IDLSS) for predicting grade ≥ 2 acute radiation pneumonitis (ARP) in lung cancer patients upon intensity-modulated radiotherapy (IMRT). (1) Materials and Methods: A total of 126 non-small-cell lung cancer patients treated with IMRT were retrospectively analyzed. Five lung subregions (SRs) were generated by the intersection of the whole lung (WL) and five sub-regions receiving incremental dose intervals. A total of 4610 radiomics features (RF) from pre-treatment planning computed tomographic (CT) and 213 dosiomics features (DF) were extracted. Six feature groups, including WL-RF, WL-DF, SR-RF, SR-DF, and the combined feature sets of WL-RDF and SR-RDF, were generated. Features were selected by using a variance threshold, followed by a Student t-test. Pearson’s correlation test was applied to remove redundant features. Subsequently, Ridge regression was adopted to develop six models for ARP using the six feature groups. Thirty iterations of resampling were implemented to assess overall model performance by using the area under the Receiver-Operating-Characteristic curve (AUC), accuracy, precision, recall, and F1-score. (2) Results: The SR-RDF model achieved the best classification performance and provided significantly better predictability than the WL-RDF model in training cohort (Average AUC: 0.98 ± 0.01 vs. 0.90 ± 0.02, p < 0.001) and testing cohort (Average AUC: 0.88 ± 0.05 vs. 0.80 ± 0.04, p < 0.001). Similarly, predictability of the SR-DF model was significantly stronger than that of the WL-DF model in training cohort (Average AUC: 0.88 ± 0.03 vs. 0.70 ± 0.030, p < 0.001) and in testing cohort (Average AUC: 0.74 ± 0.08 vs. 0.65 ± 0.06, p < 0.001). By contrast, the SR-RF model significantly outperformed the WL-RF model only in the training set (Average AUC: 0.93 ± 0.02 vs. 0.85 ± 0.03, p < 0.001), but not in the testing set (Average AUC: 0.79 ± 0.05 vs. 0.77 ± 0.07, p = 0.13). (3) Conclusions: Our results demonstrated that the IDLSS method improved model performance for classifying ARP with grade ≥ 2 when using dosiomics or combined radiomics-dosiomics features.

Impact of contouring variability on oncological PET radiomics features in the lung

Radiomics features extracted from oncological PET images are currently under intense scrutiny within the context of risk stratification for a variety of cancers. However, the lack of robustness assessment poses problems for their application across institutions and for broader patient populations. The objective of the current study was to examine the extent to which radiomics parameters from oncological PET vary in response to manual contouring variability in lung cancer. Imaging data employed in the study consisted of 26 PET scans with lesions in the lung being created through the use of an anthropomorphic phantom in conjunction with Monte Carlo simulations. From each of the simulated lesions, 25 radiomics features related to the gray-level co-occurrence matrices (GLCOM), gray-level size zone matrices (GLSZM), and gray-level neighborhood difference matrices (GLNDM) were extracted from ground truth contour and from manual contours provided by 10 raters in regard to four intensity discretization schemes with number of gray levels of 32, 64, 128, and 256, respectively. The impact of interrater variability in tumor delineation upon the agreement between raters on radiomics features was examined via interclass correlation and leave-p-out assessment. Only weak and moderate correlations were found between segmentation accuracy as measured by the Dice coefficient and percent feature error from ground truth for the vast majority of the features being examined. GLNDM-based texture parameters emerged as the top performing category of radiomcs features in terms of robustness against contouring variability for discretization schemes engaging number of gray levels of 32, 64, and 128 while GLCOM-based parameters stood out for discretization scheme engaging 256 gray levels. How and to what extent interrater reliability of radiomics features vary in response to the number of raters were largely feature-dependent. It was concluded that impact of contouring variability on PET-based radiomics features is present to varying degrees and could be experienced as a barrier to convey PET-based radiomics research to clinical relevance.

Rectal wall MRI radiomics in prostate cancer patients: prediction of and correlation with early rectal toxicity

PURPOSE

To investigate MRI radiomic analysis to assess IMRT associated rectal wall changes and also for predicting radiotherapy induced rectal toxicity.

MATERIAL AND METHODS

At first, a machine learning radiomic analysis was applied on T2-weighted (T2W) and apparent diffusion coefficient (ADC) rectal wall MR images of prostate cancer patients' pre- and post-IMRT to predict rectal toxicity. Next, Wilcoxon singed ranked test was performed to find radiomic features with significant changes pre- and post-IMRT. A logistic regression classifier was used to find correlation between features with significant changes and radiation toxicity. Area under the curve (AUC) of receiver operating characteristic (ROC) curve was used in two levels of study for finding performances.

RESULTS

AUC_mean, 0.68 ± 0.086 and 0.61 ± 0.065 were obtained for pre- and post-IMRT T2 radiomic models, respectively. For ADC radiomic models, AUC_mean was 0.58 ± 0.034 for pre-IMRT and was 0.56 ± 0.038 for post-IMRT. Wilcoxon-signed rank test revealed that 9 T2 radiomic features vary significantly post-IMRT. The AUC of logistic-regression was in the range of 0.46-0.58 for single significant features and was 0.81 when all significant features were combined.

CONCLUSIONS

Pre-IMRT MR image radiomic features could predict rectal toxicity in prostate cancer patients. Radiotherapy associated complications may be assessed by studying the changes in the MR radiomic features.