Objectives. Radiomic features extracted from diverse MRI modalities have been investigated regarding their predictive and/or prognostic value in a variety of cancers. With the aid of a 3D realistic digital MRI phantom of the brain, the aim of this study was to examine the impact of pulse sequence parameter selection on MRI-based textural parameters of the brain. Methods. MR images of the employed digital phantom were realized with SimuBloch, a simulation package made for fast generation of image sequences based on the Bloch equations. Pulse sequences being investigated consisted of spin echo (SE), gradient echo (GRE), spoiled gradient echo (SP-GRE), inversion recovery spin echo (IR-SE), and inversion recovery gradient echo (IR-GRE). Twenty-nine radiomic textural features related, respectively, to gray-level intensity histograms (GLIH), cooccurrence matrices (GLCOM), zone size matrices (GLZSM), and neighborhood difference matrices (GLNDM) were evaluated for the obtained MR realizations, and differences were identified. Results. It was found that radiomic features vary considerably among images generated by the five different T1-weighted pulse sequences, and the deviations from those measured on the T1 map vary among features, from a few percent to over 100%. Radiomic features extracted from T1-weighted spin-echo images with TR varying from 360 ms to 620 ms and TE = 3.4 ms showed coefficients of variation (CV) up to 45%, while up to 70%, for T2-weighted spin-echo images with TE varying over the range 60–120 ms and TR = 6400 ms. Conclusion. Variability of radiologic textural appearance on MR realizations with respect to the choice of pulse sequence and imaging parameters is feature-dependent and can be substantial. It calls for caution in employing MRI-derived radiomic features especially when pooling imaging data from multiple institutions with intention of correlating with clinical endpoints.

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.

Purpose: Though failure mode and effects analysis (FMEA) is becoming more widely adopted for risk assessment in radiation therapy, to our knowledge, its output has never been validated against data on errors that actually occur. The objective of this study was to perform FMEA of a stereotactic body radiation therapy (SBRT) treatment planning process and validate the results against data recorded within an incident learning system. Methods: FMEA on the SBRT treatment planning process was carried out by a multidisciplinary group including radiation oncologists, medical physicists, dosimetrists, and IT technologists. Potential failure modes were identified through a systematic review of the process map. Failure modes were rated for severity, occurrence, and detectability on a scale of one to ten and risk priority number (RPN) was computed. Failure modes were then compared with historical reports identified as relevant to SBRT planning within a departmental incident learning system that has been active for two and a half years. Differences between FMEA anticipated failure modes and existing incidents were identified. Results: FMEA identified 63 failure modes. RPN values for the top 25% of failure modes ranged from 60 to 336. Analysis of the incident learning database identified 33 reported near‐miss events related to SBRT planning. Combining both methods yielded a total of 76 possible process failures, of which 13 (17%) were missed by FMEA while 43 (57%) identified by FMEA only. When scored for RPN, the 13 events missed by FMEA ranked within the lower half of all failure modes and exhibited significantly lower severity relative to those identified by FMEA (p = 0.02). Conclusions: FMEA, though valuable, is subject to certain limitations. In this study, FMEA failed to identify 17% of actual failure modes, though these were of lower risk. Similarly, an incident learning system alone fails to identify a large number of potentially high‐severity process errors. Using FMEA in combination with incident learning may render an improved overview of risks within a process.

(1) Background and purpose: clinical trials have unsuccessfully tried to de-escalate treatment in locally advanced human papillomavirus positive (HPV+) oropharyngeal squamous cell carcinoma (OPSCC) with the goal of reducing treatment toxicity. The aim of this study was to explore the role of radiomics for risk stratification in this patient population to guide treatment. (2) Methods: the study population consisted of 225 patients with locally advanced HPV+ OPSCC treated with curative-intent radiation or chemoradiation therapy. Appearance of distant metastasis was used as the endpoint event. Radiomics data were extracted from the gross tumor volumes (GTVs) identified on the planning CT, with gray level being discretized using three different bin widths (8, 16, and 32). The data extracted for the groups with and without distant metastasis were subsequently balanced using three different algorithms including synthetic minority over-sampling technique (SMOTE), adaptive synthetic sampling (ADASYN), and borderline SMOTE. From these different combinations, a total of nine radiomics datasets were derived. Top features that minimized redundancy while maximizing relevance to the endpoint were selected individually and collectively for the nine radiomics datasets to build support vector machine (SVM) based predictive classifiers. Performance of the developed classifiers was evaluated by receiver operating characteristic (ROC) curve analysis. (3) Results: of the 225 locally advanced HPV+ OPSCC patients being studied, 9.3% had developed distant metastases at last follow-up. SVM classifiers built for the nine radiomics dataset using either their own respective top features or the top consensus ones were all able to differentiate the two cohorts at a level of excellence or beyond, with ROC area under curve (AUC) ranging from 0.84 to 0.95 (median = 0.90). ROC comparisons further revealed that the majority of the built classifiers did not distinguish the two cohorts significantly better than each other. (4) Conclusions: radiomics demonstrated discriminative ability in distinguishing patients with locally advanced HPV+ OPSCC who went on to develop distant metastasis after completion of definitive chemoradiation or radiation alone and may serve to risk stratify this patient population with the purpose of guiding the appropriate therapy.

Metabolic tumor volume (MTV) is a robust prognostic biomarker in diffuse large B-cell lymphoma (DLBCL). The available semiautomatic software for calculating MTV requires manual input limiting its routine application in clinical research. Our objective was to develop a fully automated method (AM) for calculating MTV and to validate the method by comparing its results with those from two nuclear medicine (NM) readers. The automated method designed for this study employed a deep convolutional neural network to segment normal physiologic structures from the computed tomography (CT) scans that demonstrate intense avidity on positron emission tomography (PET) scans. The study cohort consisted of 100 patients with newly diagnosed DLBCL who were randomly selected from the Alliance/CALGB 50,303 (NCT00118209) trial. We observed high concordance in MTV calculations between the AM and readers with Pearson’s correlation coefficients and interclass correlations comparing reader 1 to AM of 0.9814 (p < 0.0001) and 0.98 (p < 0.001; 95%CI = 0.96 to 0.99), respectively; and comparing reader 2 to AM of 0.9818 (p < 0.0001) and 0.98 (p < 0.0001; 95%CI = 0.96 to 0.99), respectively. The Bland–Altman plots showed only relatively small systematic errors between the proposed method and readers for both MTV and maximum standardized uptake value (SUVmax). This approach may possess the potential to integrate PET-based biomarkers in clinical trials.