The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights.

Filters are commonly used to enhance specific structures and patterns in images, such as vessels or peritumoral regions, to enable clinical insights beyond the visible image using radiomics. However, their lack of standardization restricts reproducibility and clinical translation of radiomics decision support tools. In this special report, teams of researchers who developed radiomics software participated in a three-phase study (September 2020 to December 2022) to establish a standardized set of filters. The first two phases focused on finding reference filtered images and reference feature values for commonly used convolutional filters: mean, Laplacian of Gaussian, Laws and Gabor kernels, separable and nonseparable wavelets (including decomposed forms), and Riesz transformations. In the first phase, 15 teams used digital phantoms to establish 33 reference filtered images of 36 filter configurations. In phase 2, 11 teams used a chest CT image to derive reference values for 323 of 396 features computed from filtered images using 22 filter and image processing configurations. Reference filtered images and feature values for Riesz transformations were not established. Reproducibility of standardized convolutional filters was validated on a public data set of multimodal imaging (CT, fluorodeoxyglucose PET, and T1-weighted MRI) in 51 patients with soft-tissue sarcoma. At validation, reproducibility of 486 features computed from filtered images using nine configurations × three imaging modalities was assessed using the lower bounds of 95% CIs of intraclass correlation coefficients. Out of 486 features, 458 were found to be reproducible across nine teams with lower bounds of 95% CIs of intraclass correlation coefficients greater than 0.75. In conclusion, eight filter types were standardized with reference filtered images and reference feature values for verifying and calibrating radiomics software packages. A web-based tool is available for compliance checking.

Exploring published and novel pre-treatment CT and PET radiomics to stratify risk of progression among early-stage non-small cell lung cancer patients treated with stereotactic radiation.

PURPOSE: Disease progression after definitive stereotactic body radiation therapy (SBRT) for early-stage non-small cell lung cancer (NSCLC) occurs in 20-40% of patients. Here, we explored published and novel pre-treatment CT and PET radiomics features to identify patients at risk of progression. MATERIALS/METHODS: Published CT and PET features were identified and explored along with 15 other CT and PET features in 408 consecutively treated early-stage NSCLC patients having CT and PET < 3 months pre-SBRT (training/set-aside validation subsets: n = 286/122). Features were associated with progression-free survival (PFS) using bootstrapped Cox regression (Bonferroni-corrected univariate predictor: p ≤ 0.002) and only non-strongly correlated predictors were retained (|Rs|<0.70) in forward-stepwise multivariate analysis. RESULTS: Tumor diameter and SUVmax were the two most frequently reported features associated with progression/survival (in 6/20 and 10/20 identified studies). These two features and 12 of the 15 additional features (CT: 6; PET: 6) were candidate PFS predictors. A re-fitted model including diameter and SUVmax presented with the best performance (c-index: 0.78; log-rank p-value < 0.0001). A model built with the two best additional features (CTspiculation1 and SUVentropy) had a c-index of 0.75 (log-rank p-value < 0.0001). CONCLUSIONS: A re-fitted pre-treatment model using the two most frequently published features - tumor diameter and SUVmax - successfully stratified early-stage NSCLC patients by PFS after receiving SBRT.

Simulating the Potential of Model-Based Individualized Prescriptions for Ultracentral Lung Tumors.

Purpose: The use of stereotactic body radiation therapy for ultracentral lung tumors is limited by increased toxicity. We hypothesized that using published normal tissue complication probability (NTCP) and tumor control probability (TCP) models could improve the therapeutic ratio between tumor control and toxicity. A proposed model-based approach was applied to virtually replan early-stage non-small cell lung cancer (NSCLC) tumors. Methods and Materials: The analysis included 63 patients with ultracentral NSCLC tumors treated at our center between 2008 and 2017. Along with current clinical constraints, additional NTCP model-based criteria, including for grade 3+ radiation pneumonitis (RP3+) and grade 2+ esophagitis, were implemented using 4 different fractionation schemes. Scaled dose distributions resulting in the highest TCP without violating constraints were selected (optimal plan [Planopt]). Planopt predictions were compared with the observed local control and toxicities. Results: The observed 2-year local control rate was 72% (95% CI, 57%-88%) compared with 87% (range, 6%-93%) for Planopt TCP. Thirty-nine patients had Planopt with TCP > 80%, and 14 patients had Planopt TCP < 50%. The Planopt NTCPs for RP3+ were reduced by nearly half compared with patients' observed RP3+. The RP3+ NTCP was the most frequent reason for TCP of Planopt < 80% (14/24 patients), followed by grade 2+ esophagitis NTCP (5/24 patients) due to larger tumors (>40 cc vs ≤40 cc; P = .002) or a shorter tumor to esophagus distance (≥5 cm vs <5 cm; P < .001). Conclusions: We demonstrated the potential for model-based prescriptions to yield higher TCP while respecting NTCP for patients with ultracentral NSCLC. Individualizing treatments based on NTCP- and TCP-driven simulations halved the predicted relative to the observed rates of RP3+. Our simulations also identified patients whose TCP could not be improved without violating NTCP due to larger tumors or a near tumor to esophagus proximity.

ROE (Radiotherapy Outcomes Estimator): An open-source tool for optimizing radiotherapy prescriptions.

BACKGROUND AND OBJECTIVES: Radiotherapy prescriptions currently derive from population-wide guidelines established through large clinical trials. We provide an open-source software tool for patient-specific prescription determination using personalized dose-response curves. METHODS: We developed ROE, a plugin to the Computational Environment for Radiotherapy Research to visualize predicted tumor control and normal tissue complication simultaneously, as a function of prescription dose. ROE can be used natively with MATLAB and is additionally made accessible in GNU Octave and Python, eliminating the need for commercial licenses. It provides a curated library of published and validated predictive models and incorporates clinical restrictions on normal tissue outcomes. ROE additionally provides batch-mode tools to evaluate and select among different fractionation schemes and analyze radiotherapy outcomes across patient cohorts. CONCLUSION: ROE is an open-source, GPL-copyrighted tool for interactive exploration of the dose-response relationship to aid in radiotherapy planning. We demonstrate its potential clinical relevance in (1) improving patient awareness by quantifying the risks and benefits of a given treatment protocol (2) assessing the potential for dose escalation across patient cohorts and (3) estimating accrual rates of new protocols.

External validation of pulmonary radiotherapy toxicity models for ultracentral lung tumors.

Introduction: Pulmonary toxicity is dose-limiting in stereotactic body radiation therapy (SBRT) for tumors that abut the proximal bronchial tree (PBT), esophagus, or other mediastinal structures. In this work we explored published models of pulmonary toxicity following SBRT for such ultracentral tumors in an independent cohort of patients. Methods: The PubMed database was searched for pulmonary toxicity models. Identified models were tested in a cohort of patients with ultracentral lung tumors treated between 2008 and 2017 at one large center (N = 88). This cohort included 60 % primary and 40 % metastatic tumors treated to 45 Gy in 5 fractions (fx), 50 Gy in 5 fx, 60 Gy in 8 fx, or 60 Gy in 15 fx prescribed as 100 % dose to PTV. Results: Seven published NTCP models from two studies were identified. The NTCP models utilized PBT max point dose (Dmax), D0.2 cm3, V65, V100, and V130. Within the independent cohort, the ≥ grade 3 toxicity and grade 5 toxicity rates were 18 % and 7-10 %, respectively, and the Dmax models best described pulmonary toxicity. The Dmax to 0.1 cm3 model was better calibrated and had increased steepness compared to the Dmax model. A re-planning study minimizing PBT 0.1 cm3 to below 122 Gy in EQD23 (for a 10 % ≥grade 3 pulmonary toxicity) was demonstrated to be completely feasible in 4/6 patients, and dose to PBT 0.1 cm3 was considerably lowered in all six patients. Conclusions: Pulmonary toxicity models were identified from two studies and explored within an independent ultracentral lung tumor cohort. A modified Dmax to 0.1 cm3 PBT model displayed the best performance. This model could be utilized as a starting point for rationally constructed airways constraints in ultracentral patients treated with SBRT or hypofractionation.

Acute Lymphoblastic Leukemia presenting as a Pathologic Fibular Fracture.

The authors present a case of a 2-year-old girl with left ankle pain. On examination, there was tenderness but no sign of superficial swelling, erythema, or deformity. Imaging studies revealed a heterogeneous lytic lesion in the distal diaphysis of the left fibula, causing a pathologic fracture. The patient was treated with ankle splinting, analgesia, and referred to a pediatric orthopedic physician. Ultimately the diagnosis of acute lymphoblastic leukemia (ALL) was made. The authors present the significance of discerning skeletal abnormalities and orthopedic pain as the initial manifestation of leukemia.

Prospectively-validated deep learning model for segmenting swallowing and chewing structures in CT.

Objective.Delineating swallowing and chewing structures aids in radiotherapy (RT) treatment planning to limit dysphagia, trismus, and speech dysfunction. We aim to develop an accurate and efficient method to automate this process.Approach.CT scans of 242 head and neck (H&N) cancer patients acquired from 2004 to 2009 at our institution were used to develop auto-segmentation models for the masseters, medial pterygoids, larynx, and pharyngeal constrictor muscle using DeepLabV3+. A cascaded framework was used, wherein models were trained sequentially to spatially constrain each structure group based on prior segmentations. Additionally, an ensemble of models, combining contextual information from axial, coronal, and sagittal views was used to improve segmentation accuracy. Prospective evaluation was conducted by measuring the amount of manual editing required in 91 H&N CT scans acquired February-May 2021.Main results. Medians and inter-quartile ranges of Dice similarity coefficients (DSC) computed on the retrospective testing set (N = 24) were 0.87 (0.85-0.89) for the masseters, 0.80 (0.79-0.81) for the medial pterygoids, 0.81 (0.79-0.84) for the larynx, and 0.69 (0.67-0.71) for the constrictor. Auto-segmentations, when compared to two sets of manual segmentations in 10 randomly selected scans, showed better agreement (DSC) with each observer than inter-observer DSC. Prospective analysis showed most manual modifications needed for clinical use were minor, suggesting auto-contouring could increase clinical efficiency. Trained segmentation models are available for research use upon request viahttps://github.com/cerr/CERR/wiki/Auto-Segmentation-models.Significance.We developed deep learning-based auto-segmentation models for swallowing and chewing structures in CT and demonstrated its potential for use in treatment planning to limit complications post-RT. To the best of our knowledge, this is the only prospectively-validated deep learning-based model for segmenting chewing and swallowing structures in CT. Segmentation models have been made open-source to facilitate reproducibility and multi-institutional research.

Deep learning auto-segmentation and automated treatment planning for trismus risk reduction in head and neck cancer radiotherapy.

BACKGROUND AND PURPOSE: Reducing trismus in radiotherapy for head and neck cancer (HNC) is important. Automated deep learning (DL) segmentation and automated planning was used to introduce new and rarely segmented masticatory structures to study if trismus risk could be decreased. MATERIALS AND METHODS: Auto-segmentation was based on purpose-built DL, and automated planning used our in-house system, ECHO. Treatment plans for ten HNC patients, treated with 2 Gy × 35 fractions, were optimized (ECHO0). Six manually segmented OARs were replaced with DL auto-segmentations and the plans re-optimized (ECHO1). In a third set of plans, mean doses for auto-segmented ipsilateral masseter and medial pterygoid (MIMean, MPIMean), derived from a trismus risk model, were implemented as dose-volume objectives (ECHO2). Clinical dose-volume criteria were compared between the two scenarios (ECHO0 vs. ECHO1; ECHO1 vs. ECHO2; Wilcoxon signed-rank test; significance: p < 0.01). RESULTS: Small systematic differences were observed between the doses to the six auto-segmented OARs and their manual counterparts (median: ECHO1 = 6.2 (range: 0.4, 21) Gy vs. ECHO0 = 6.6 (range: 0.3, 22) Gy; p = 0.007), and the ECHO1 plans provided improved normal tissue sparing across a larger dose-volume range. Only in the ECHO2 plans, all patients fulfilled both MIMean and MPIMean criteria. The population median MIMean and MPIMean were considerably lower than those suggested by the trismus model (ECHO0: MIMean = 13 Gy vs. ≤42 Gy; MPIMean = 29 Gy vs. ≤68 Gy). CONCLUSIONS: Automated treatment planning can efficiently incorporate new structures from DL auto-segmentation, which results in trismus risk sparing without deteriorating treatment plan quality. Auto-planning and deep learning auto-segmentation together provide a powerful platform to further improve treatment planning.

Reproducibility of radiomic features using network analysis and its application in Wasserstein k-means clustering.

Purpose: The goal of this study is to develop innovative methods for identifying radiomic features that are reproducible over varying image acquisition settings. Approach: We propose a regularized partial correlation network to identify reliable and reproducible radiomic features. This approach was tested on two radiomic feature sets generated using two different reconstruction methods on computed tomography (CT) scans from a cohort of 47 lung cancer patients. The largest common network component between the two networks was tested on phantom data consisting of five cancer samples. To further investigate whether radiomic features found can identify phenotypes, we propose a k -means clustering algorithm coupled with the optimal mass transport theory. This approach following the regularized partial correlation network analysis was tested on CT scans from 77 head and neck squamous cell carcinoma (HNSCC) patients in the Cancer Imaging Archive (TCIA) and validated using an independent dataset. Results: A set of common radiomic features was found in relatively large network components between the resultant two partial correlation networks resulting from a cohort of lung cancer patients. The reliability and reproducibility of those radiomic features were further validated on phantom data using the Wasserstein distance. Further analysis using the network-based Wasserstein k -means algorithm on the TCIA HNSCC data showed that the resulting clusters separate tumor subsites as well as HPV status, and this was validated on an independent dataset. Conclusion: We showed that a network-based analysis enables identifying reproducible radiomic features and use of the selected set of features can enhance clustering results.

Using Auto-Segmentation to Reduce Contouring and Dose Inconsistency in Clinical Trials: The Simulated Impact on RTOG 0617.

PURPOSE: Contouring inconsistencies are known but understudied in clinical radiation therapy trials. We applied auto-contouring to the Radiation Therapy Oncology Group (RTOG) 0617 dose escalation trial data. We hypothesized that the trial heart doses were higher than reported due to inconsistent and insufficient heart segmentation. We tested our hypothesis by comparing doses between deep-learning (DL) segmented hearts and trial hearts. METHODS AND MATERIALS: The RTOG 0617 data were downloaded from The Cancer Imaging Archive; the 442 patients with trial hearts and dose distributions were included. All hearts were resegmented using our DL pipeline and quality assured to meet the requirements for clinical implementation. Dose (V5%, V30%, and mean heart dose) was compared between the 2 sets of hearts (Wilcoxon signed-rank test). Each dose metric was associated with overall survival (Cox proportional hazards). Lastly, 18 volume similarity metrics were assessed for the hearts and correlated with |DoseDL - DoseRTOG0617| (linear regression; significance: P ≤ .0028; corrected for 18 tests). RESULTS: Dose metrics were significantly higher for DL hearts compared with trial hearts (eg, mean heart dose: 15 Gy vs 12 Gy; P = 5.8E-16). All 3 DL heart dose metrics were stronger overall survival predictors than those of the trial hearts (median, P = 2.8E-5 vs 2.0E-4). Thirteen similarity metrics explained |DoseDL - DoseRTOG0617|; the axial distance between the 2 centers of mass was the strongest predictor (CENTAxial; median, R2 = 0.47; P = 6.1E-62). CENTAxial agreed with the qualitatively identified inconsistencies in the superior direction. The trial's qualitative heart contouring score was not correlated with |DoseDL - DoseRTOG0617| (median, R2 = 0.01; P = .02) or with any of the similarity metrics (median, Rs = 0.13 [range, -0.22 to 0.31]). CONCLUSIONS: Using a coherent heart definition, as enabled through our open-source DL algorithm, the trial heart doses in RTOG 0617 were found to be significantly higher than previously reported, which may have led to an even more rapid mortality accumulation. Auto-segmentation is likely to reduce contouring and dose inconsistencies and increase the quality of clinical RT trials.

Library of deep-learning image segmentation and outcomes model-implementations.

An open-source library of implementations for deep-learning-based image segmentation and outcomes models based on radiotherapy and radiomics is presented. As oncology treatment planning becomes increasingly driven by automation, such a library of model implementations is crucial to (i) validate existing models on datasets collected at different institutions, (ii) automate segmentation, (iii) create ensembles for improving performance and (iv) incorporate validated models in the clinical workflow. Inclusion of deep-learning-based image segmentation and outcomes models in the same library provides a fully automated and reproduceable pipeline to estimate prognosis. The library was developed with the Computational Environment for Radiological Research (CERR) software platform. Centralizing model implementations in CERR builds upon its rich set of radiotherapy and radiomics tools and caters to the world-wide user base. CERR provides well-validated feature extraction pipelines for radiotherapy dosimetry and radiomics with fine control over the calculation settings, allowing users to select appropriate parameters used in model derivation. Models for automatic image segmentation are distributed via containers, allowing them to be deployed with a variety of scientific computing architectures. The library includes implementations of popular DVH-based models outlined in the Quantitative Analysis of Normal Tissue Effects in the Clinic effort and recently published literature. Radiomics models include features from the Image Biomarker Standardization Initiative and application-specific features found to be relevant across multiple sites and image modalities. The library is distributed as a module within CERR at https://www.github.com/cerr/CERR under the GNU-GPL copyleft with additional restrictions on clinical and commercial use and provision to dual license in future.

A novel kernel Wasserstein distance on Gaussian measures: An application of identifying dental artifacts in head and neck computed tomography.

The Wasserstein distance is a powerful metric based on the theory of optimal mass transport. It gives a natural measure of the distance between two distributions with a wide range of applications. In contrast to a number of the common divergences on distributions such as Kullback-Leibler or Jensen-Shannon, it is (weakly) continuous, and thus ideal for analyzing corrupted and noisy data. Until recently, however, no kernel methods for dealing with nonlinear data have been proposed via the Wasserstein distance. In this work, we develop a novel method to compute the L2-Wasserstein distance in reproducing kernel Hilbert spaces (RKHS) called kernel L2-Wasserstein distance, which is implemented using the kernel trick. The latter is a general method in machine learning employed to handle data in a nonlinear manner. We evaluate the proposed approach in identifying computed tomography (CT) slices with dental artifacts in head and neck cancer, performing unsupervised hierarchical clustering on the resulting Wasserstein distance matrix that is computed on imaging texture features extracted from each CT slice. We further compare the performance of kernel Wasserstein distance with alternatives including kernel Kullback-Leibler divergence we previously developed. Our experiments show that the kernel approach outperforms classical non-kernel approaches in identifying CT slices with artifacts.

Histogram Analysis and Visual Heterogeneity of Diffusion-Weighted Imaging with Apparent Diffusion Coefficient Mapping in the Prediction of Molecular Subtypes of Invasive Breast Cancers.

Objective: To investigate if histogram analysis and visually assessed heterogeneity of diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) mapping can predict molecular subtypes of invasive breast cancers. Materials and Methods: In this retrospective study, 91 patients with invasive breast carcinoma who underwent preoperative magnetic resonance imaging (MRI) with DWI at our institution were included. Two radiologists delineated a 2-D region of interest (ROI) on ADC maps in consensus. Tumors were also independently classified into low and high heterogeneity based on visual assessment of DWI. First-order statistics extracted through histogram analysis within the ROI of the ADC maps (mean, 10th percentile, 50th percentile, 90th percentile, standard deviation, kurtosis, and skewness) and visually assessed heterogeneity were evaluated for associations with tumor receptor status (ER, PR, and HER2 status) as well as molecular subtype. Results: HER2-positive lesions demonstrated significantly higher mean (p=0.034), Perc50 (p=0.046), and Perc90 (p=0.040), with AUCs of 0.605, 0.592, and 0.652, respectively, than HER2-negative lesions. No significant differences were found in the histogram values for ER and PR statuses. Neither quantitative histogram analysis based on ADC maps nor qualitative visual heterogeneity assessment of DWI images was able to significantly differentiate between molecular subtypes, i.e., luminal A versus all other subtypes (luminal B, HER2-enriched, and triple negative) combined, luminal A and B combined versus HER2-enriched and triple negative combined, and triple negative versus all other types combined. Conclusion: Histogram analysis and visual heterogeneity assessment cannot be used to differentiate molecular subtypes of invasive breast cancer.

Preoperative MRI-radiomics features improve prediction of survival in glioblastoma patients over MGMT methylation status alone.

BACKGROUND: Glioblastoma (GBM) is the most common malignant central nervous system tumor, and MGMT promoter hypermethylation in this tumor has been shown to be associated with better prognosis. We evaluated the capacity of radiomics features to add complementary information to MGMT status, to improve the ability to predict prognosis. METHODS: 159 patients with untreated GBM were included in this study and divided into training and independent test sets. 286 radiomics features were extracted from the magnetic resonance images acquired prior to any treatments. A least absolute shrinkage selection operator (LASSO) selection followed by Kaplan-Meier analysis was used to determine the prognostic value of radiomics features to predict overall survival (OS). The combination of MGMT status with radiomics was also investigated and all results were validated on the independent test set. RESULTS: LASSO analysis identified 8 out of the 286 radiomic features to be relevant which were then used for determining association to OS. One feature (edge descriptor) remained significant on the external validation cohort after multiple testing (p=0.04) and the combination with MGMT identified a group of patients with the best prognosis with a survival probability of 0.61 after 43 months (p=0.0005). CONCLUSION: Our results suggest that combining radiomics with MGMT is more accurate in stratifying patients into groups of different survival risks when compared to with using these predictors in isolation. We identified two subgroups within patients who have methylated MGMT: one with a similar survival to unmethylated MGMT patients and the other with a significantly longer OS.