Search

Purpose To develop and evaluate a publicly available deep learning model for segmenting and classifying cardiac implantable electronic devices (CIEDs) on Digital Imaging and Communications in Medicine (DICOM) and smartphone-based chest radiographs. Materials and Methods This institutional review board-approved retrospective study included patients with implantable pacemakers, cardioverter defibrillators, cardiac resynchronization therapy devices, and cardiac monitors who underwent chest radiography between January 2012 and January 2022. A U-Net model with a ResNet-50 backbone was created to classify CIEDs on DICOM and smartphone images. Using 2321 chest radiographs in 897 patients (median age, 76 years [range, 18-96 years]; 625 male, 272 female), CIEDs were categorized into four manufacturers, 27 models, and one "other" category. Five smartphones were used to acquire 11 072 images. Performance was reported using the Dice coefficient on the validation set for segmentation or balanced accuracy on the test set for manufacturer and model classification, respectively. Results The segmentation tool achieved a mean Dice coefficient of 0.936 (IQR: 0.890-0.958). The model had an accuracy of 94.36% (95% CI: 90.93%, 96.84%; 251 of 266) for CIED manufacturer classification and 84.21% (95% CI: 79.31%, 88.30%; 224 of 266) for CIED model classification. Conclusion The proposed deep learning model, trained on both traditional DICOM and smartphone images, showed high accuracy for segmentation and classification of CIEDs on chest radiographs. Keywords: Conventional Radiography, Segmentation Supplemental material is available for this article . © RSNA, 2024 See also the commentary by Júdice de Mattos Farina and Celi in this issue.

Background: Postoperative recurrence risk for pediatric low-grade gliomas (pLGGs) is challenging to predict by conventional clinical, radiographic, and genomic factors. We investigated if deep learning of MRI tumor features could improve postoperative pLGG risk stratification., Methods: We used pre-trained deep learning (DL) tool designed for pLGG segmentation to extract pLGG imaging features from preoperative T2-weighted MRI from patients who underwent surgery (DL-MRI features). Patients were pooled from two institutions: Dana Farber/Boston Children's Hospital (DF/BCH) and the Children's Brain Tumor Network (CBTN). We trained three DL logistic hazard models to predict postoperative event-free survival (EFS) probabilities with 1) clinical features, 2) DL-MRI features, and 3) multimodal (clinical and DL-MRI features). We evaluated the models with a time-dependent Concordance Index (Ctd) and risk group stratification with Kaplan Meier plots and log-rank tests. We developed an automated pipeline integrating pLGG segmentation and EFS prediction with the best model., Results: Of the 396 patients analyzed (median follow-up: 85 months, range: 1.5-329 months), 214 (54%) underwent gross total resection and 110 (28%) recurred. The multimodal model improved EFS prediction compared to the DL-MRI and clinical models (Ctd: 0.85 (95% CI: 0.81-0.93), 0.79 (95% CI: 0.70-0.88), and 0.72 (95% CI: 0.57-0.77), respectively). The multimodal model improved risk-group stratification (3-year EFS for predicted high-risk: 31% versus low-risk: 92%, p<0.0001)., Conclusions: DL extracts imaging features that can inform postoperative recurrence prediction for pLGG. Multimodal DL improves postoperative risk stratification for pLGG and may guide postoperative decision-making. Larger, multicenter training data may be needed to improve model generalizability., (© The Author(s) 2024. Published by Oxford University Press on behalf of the Society for Neuro-Oncology.)

Artificial intelligence (AI) algorithms hold the potential to revolutionize radiology. However, a significant portion of the published literature lacks transparency and reproducibility, which hampers sustained progress toward clinical translation. Although several reporting guidelines have been proposed, identifying practical means to address these issues remains challenging. Here, we show the potential of cloud-based infrastructure for implementing and sharing transparent and reproducible AI-based radiology pipelines. We demonstrate end-to-end reproducibility from retrieving cloud-hosted data, through data pre-processing, deep learning inference, and post-processing, to the analysis and reporting of the final results. We successfully implement two distinct use cases, starting from recent literature on AI-based biomarkers for cancer imaging. Using cloud-hosted data and computing, we confirm the findings of these studies and extend the validation to previously unseen data for one of the use cases. Furthermore, we provide the community with transparent and easy-to-extend examples of pipelines impactful for the broader oncology field. Our approach demonstrates the potential of cloud resources for implementing, sharing, and using reproducible and transparent AI pipelines, which can accelerate the translation into clinical solutions., (© 2024. The Author(s).)

Background and Purpose: Lung cancer is a leading cause of cancer-related mortality, and stereotactic body radiotherapy (SBRT) has become a standard treatment for early-stage lung cancer. However, the heterogeneous response to radiation at the tumor level poses challenges. Currently, standardized dosage regimens lack adaptation based on individual patient or tumor characteristics. Thus, we explore the potential of delta radiomics from on-treatment magnetic resonance (MR) imaging to track radiation dose response, inform personalized radiotherapy dosing, and predict outcomes., Materials and Methods: A retrospective study of 47 MR-guided lung SBRT treatments for 39 patients was conducted. Radiomic features were extracted using Pyradiomics, and stability was evaluated temporally and spatially. Delta radiomics were correlated with radiation dose delivery and assessed for associations with tumor control and survival with Cox regressions., Results: Among 107 features, 49 demonstrated temporal stability, and 57 showed spatial stability. Fifteen stable and non-collinear features were analyzed. Median Skewness and surface to volume ratio decreased with radiation dose fraction delivery, while coarseness and 90th percentile values increased. Skewness had the largest relative median absolute changes (22 %-45 %) per fraction from baseline and was associated with locoregional failure (p = 0.012) by analysis of covariance. Skewness, Elongation, and Flatness were significantly associated with local recurrence-free survival, while tumor diameter and volume were not., Conclusions: Our study establishes the feasibility and stability of delta radiomics analysis for MR-guided lung SBRT. Findings suggest that MR delta radiomics can capture short-term radiographic manifestations of the intra-tumoral radiation effect., Competing Interests: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Benjamin Kann, Raymond Mak, Shu-Hui Hsu, and Jonathan Leeman receive research support from ViewRay, Inc, not directly related to the work reported in this paper., (© 2024 The Author(s).)

Purpose To develop, externally test, and evaluate clinical acceptability of a deep learning pediatric brain tumor segmentation model using stepwise transfer learning. Materials and Methods In this retrospective study, the authors leveraged two T2-weighted MRI datasets (May 2001 through December 2015) from a national brain tumor consortium ( n = 184; median age, 7 years [range, 1-23 years]; 94 male patients) and a pediatric cancer center ( n = 100; median age, 8 years [range, 1-19 years]; 47 male patients) to develop and evaluate deep learning neural networks for pediatric low-grade glioma segmentation using a stepwise transfer learning approach to maximize performance in a limited data scenario. The best model was externally tested on an independent test set and subjected to randomized blinded evaluation by three clinicians, wherein they assessed clinical acceptability of expert- and artificial intelligence (AI)-generated segmentations via 10-point Likert scales and Turing tests. Results The best AI model used in-domain stepwise transfer learning (median Dice score coefficient, 0.88 [IQR, 0.72-0.91] vs 0.812 [IQR, 0.56-0.89] for baseline model; P = .049). With external testing, the AI model yielded excellent accuracy using reference standards from three clinical experts (median Dice similarity coefficients: expert 1, 0.83 [IQR, 0.75-0.90]; expert 2, 0.81 [IQR, 0.70-0.89]; expert 3, 0.81 [IQR, 0.68-0.88]; mean accuracy, 0.82). For clinical benchmarking ( n = 100 scans), experts rated AI-based segmentations higher on average compared with other experts (median Likert score, 9 [IQR, 7-9] vs 7 [IQR 7-9]) and rated more AI segmentations as clinically acceptable (80.2% vs 65.4%). Experts correctly predicted the origin of AI segmentations in an average of 26.0% of cases. Conclusion Stepwise transfer learning enabled expert-level automated pediatric brain tumor autosegmentation and volumetric measurement with a high level of clinical acceptability. Keywords: Stepwise Transfer Learning, Pediatric Brain Tumors, MRI Segmentation, Deep Learning Supplemental material is available for this article . © RSNA, 2024.

Competing Interests: DSB reports being an Associate Editor of Radiation Oncology at HemOnc.org (no financial compensation, unrelated to this work, and recieving funding from American Association for Cancer Research, unrelated to this work. HJWLA reports advising and consulting for Onc.AI, Love Health, Sphera, Editas, AstraZeneca, and Bristol Myers Squibb, unrelated to this work. RHM reports being on an Advisory Board for ViewRay and AstraZeneca; Consulting for Varian Medical Systems and Sio Capital Management; and honorarium from Novartis and Springer Nature. JL reports research funding from Viewray, NH Theraguix, and Varian. ML reports advisory and consulting for Pfizer, Gilead, Novartis, and AstraZeneca, unrelated to this work. BHK reports research funding from Botha-Chan Low Grade Glioma Consortium (National institutes of Health [NIH]-USA K08DE030216-01). All other authors declare no competing interests. The authors acknowledge financial support from the Woods Foundation (DSB, RHM, BHK, and HJWLA) NIH (NIH-USA U54CA274516-01A1 (SC, MG, BHK, HJWLA, GKS, and DSB), NIH-USA U24CA194354 (HJWLA), NIH-USA U01CA190234 (HJWLA), NIH-USA U01CA209414 (HJWLA), and NIH-USA R35CA22052 (HJWLA), NIH-NIDA R01DA051464 (MA), R01GM114355 (GKS), NIH-USA R01LM012973 (TM and MA), NIH-USA R01MH126977 (TM), NIH-USA U54 TW012043-01 (JG and LAC), NIH-USA OT2OD032701 (JG and LAC), NIH-USA R01EB017205 (LAC), and the EU European Research Council (HJWLA 866504), all outside of the submitted work. All data collected and generated in this study, after de-identification, are available at https://github.com/AIM-Harvard/OncQA. SC: conceptualisation, data curation, formal analysis, investigation, methodology, visualisation, and writing (original draft, review, and editing). MG: conceptualisation, data curation, and formal analysis. SM, FH EH, BHK, FEC, JL: data curation, investigation, and methodology. RHM: data curation, investigation, methodology, and writing (review and editing). HJWLA: investigation, methodology, resources, and writing (review and editing). JG: formal analysis, investigation, methodology, visualisation, and writing (review and editing). TM and GKS: formal analysis, investigation, methodology, and writing (review and editing). ML data curation, formal analysis, investigation, and methodology. LAC formal analysis, investigation, supervision, and writing (review and editing). MA: conceptualisation, data curation, formal analysis, investigation, methodology, supervision, and writing (review and editing). DSB: conceptualisation, data curation, formal analysis, investigation, methodology, supervision, visualisation, resources, and writing (original draft, review, and editing). SC and DSB directly accessed and verified the underlying data reported in the manuscript. All authors have full access to all the data in the study and accept responsibility to submit for publication.

Importance: The association between body composition (BC) and cancer outcomes is complex and incompletely understood. Previous research in non-small-cell lung cancer (NSCLC) has been limited to small, single-institution studies and yielded promising, albeit heterogeneous, results., Objectives: To evaluate the association of BC with oncologic outcomes in patients receiving immunotherapy for advanced or metastatic NSCLC., Design, Setting, and Participants: This comprehensive multicohort analysis included clinical data from cohorts receiving treatment at the Dana-Farber Brigham Cancer Center (DFBCC) who received immunotherapy given alone or in combination with chemotherapy and prospectively collected data from the phase 1/2 Study 1108 and the chemotherapy arm of the phase 3 MYSTIC trial. Baseline and follow-up computed tomography (CT) scans were collected and analyzed using deep neural networks for automatic L3 slice selection and body compartment segmentation (skeletal muscle [SM], subcutaneous adipose tissue [SAT], and visceral adipose tissue). Outcomes were compared based on baseline BC measures or their change at the first follow-up scan. The data were analyzed between July 2022 and April 2023., Main Outcomes and Measures: Hazard ratios (HRs) for the association of BC measurements with overall survival (OS) and progression-free survival (PFS)., Results: A total of 1791 patients (878 women [49%]) with NSCLC were analyzed, of whom 487 (27.2%) received chemoimmunotherapy at DFBCC (DFBCC-CIO), 825 (46.1%) received ICI monotherapy at DFBCC (DFBCC-IO), 222 (12.4%) were treated with durvalumab monotherapy on Study 1108, and 257 (14.3%) were treated with chemotherapy on MYSTIC; median (IQR) ages were 65 (58-74), 66 (57-71), 65 (26-87), and 63 (30-84) years, respectively. A loss in SM mass, as indicated by a change in the L3 SM area, was associated with worse oncologic outcome across patient groups (HR, 0.59 [95% CI, 0.43-0.81] and 0.61 [95% CI, 0.47-0.79] for OS and PFS, respectively, in DFBCC-CIO; HR, 0.74 [95% CI, 0.60-0.91] for OS in DFBCC-IO; HR, 0.46 [95% CI, 0.33-0.64] and 0.47 [95% CI, 0.34-0.64] for OS and PFS, respectively, in Study 1108; HR, 0.76 [95% CI, 0.61-0.96] for PFS in the MYSTIC trial). This association was most prominent among male patients, with a nonsignificant association among female patients in the MYSTIC trial and DFBCC-CIO cohorts on Kaplan-Meier analysis. An increase of more than 5% in SAT density, as quantified by the average CT attenuation in Hounsfield units of the SAT compartment, was associated with poorer OS in 3 patient cohorts (HR, 0.61 [95% CI, 0.43-0.86] for DFBCC-CIO; HR, 0.62 [95% CI, 0.49-0.79] for DFBCC-IO; and HR, 0.56 [95% CI, 0.40-0.77] for Study 1108). The change in SAT density was also associated with PFS for DFBCC-CIO (HR, 0.73; 95% CI, 0.54-0.97). This was primarily observed in female patients on Kaplan-Meier analysis., Conclusions and Relevance: The results of this multicohort study suggest that loss in SM mass during systemic therapy for NSCLC is a marker of poor outcomes, especially in male patients. SAT density changes are also associated with prognosis, particularly in female patients. Automated CT-derived BC measurements should be considered in determining NSCLC prognosis.

Purpose To develop and externally test a scan-to-prediction deep learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pediatric low-grade glioma. Materials and Methods This retrospective study included two pediatric low-grade glioma datasets with linked genomic and diagnostic T2-weighted MRI data of patients: Dana-Farber/Boston Children's Hospital (development dataset, n = 214 [113 (52.8%) male; 104 (48.6%) BRAF wild type, 60 (28.0%) BRAF fusion, and 50 (23.4%) BRAF V600E]) and the Children's Brain Tumor Network (external testing, n = 112 [55 (49.1%) male; 35 (31.2%) BRAF wild type, 60 (53.6%) BRAF fusion, and 17 (15.2%) BRAF V600E]). A deep learning pipeline was developed to classify BRAF mutational status ( BRAF wild type vs BRAF fusion vs BRAF V600E) via a two-stage process: (a) three-dimensional tumor segmentation and extraction of axial tumor images and (b) section-wise, deep learning-based classification of mutational status. Knowledge-transfer and self-supervised approaches were investigated to prevent model overfitting, with a primary end point of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, a novel metric, center of mass distance, was developed to quantify the model attention around the tumor. Results A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest classification performance with an AUC of 0.82 (95% CI: 0.72, 0.91), 0.87 (95% CI: 0.61, 0.97), and 0.85 (95% CI: 0.66, 0.95) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively, on internal testing. On external testing, the pipeline yielded an AUC of 0.72 (95% CI: 0.64, 0.86), 0.78 (95% CI: 0.61, 0.89), and 0.72 (95% CI: 0.64, 0.88) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively. Conclusion Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pediatric low-grade glioma mutational status prediction in a limited data scenario. Keywords: Pediatrics, MRI, CNS, Brain/Brain Stem, Oncology, Feature Detection, Diagnosis, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2024.

Objective: Large language models (LLMs) have shown impressive ability in biomedical question-answering, but have not been adequately investigated for more specific biomedical applications. This study investigates ChatGPT family of models (GPT-3.5, GPT-4) in biomedical tasks beyond question-answering., Materials and Methods: We evaluated model performance with 11 122 samples for two fundamental tasks in the biomedical domain-classification (n = 8676) and reasoning (n = 2446). The first task involves classifying health advice in scientific literature, while the second task is detecting causal relations in biomedical literature. We used 20% of the dataset for prompt development, including zero- and few-shot settings with and without chain-of-thought (CoT). We then evaluated the best prompts from each setting on the remaining dataset, comparing them to models using simple features (BoW with logistic regression) and fine-tuned BioBERT models., Results: Fine-tuning BioBERT produced the best classification (F1: 0.800-0.902) and reasoning (F1: 0.851) results. Among LLM approaches, few-shot CoT achieved the best classification (F1: 0.671-0.770) and reasoning (F1: 0.682) results, comparable to the BoW model (F1: 0.602-0.753 and 0.675 for classification and reasoning, respectively). It took 78 h to obtain the best LLM results, compared to 0.078 and 0.008 h for the top-performing BioBERT and BoW models, respectively., Discussion: The simple BoW model performed similarly to the most complex LLM prompting. Prompt engineering required significant investment., Conclusion: Despite the excitement around viral ChatGPT, fine-tuning for two fundamental biomedical natural language processing tasks remained the best strategy., (© The Author(s) 2024. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Background: Guidelines for primary prevention of atherosclerotic cardiovascular disease (ASCVD) recommend a risk calculator (ASCVD risk score) to estimate 10-year risk for major adverse cardiovascular events (MACE). Because the necessary inputs are often missing, complementary approaches for opportunistic risk assessment are desirable., Objective: To develop and test a deep-learning model (CXR CVD-Risk) that estimates 10-year risk for MACE from a routine chest radiograph (CXR) and compare its performance with that of the traditional ASCVD risk score for implications for statin eligibility., Design: Risk prediction study., Setting: Outpatients potentially eligible for primary cardiovascular prevention., Participants: The CXR CVD-Risk model was developed using data from a cancer screening trial. It was externally validated in 8869 outpatients with unknown ASCVD risk because of missing inputs to calculate the ASCVD risk score and in 2132 outpatients with known risk whose ASCVD risk score could be calculated., Measurements: 10-year MACE predicted by CXR CVD-Risk versus the ASCVD risk score., Results: Among 8869 outpatients with unknown ASCVD risk, those with a risk of 7.5% or higher as predicted by CXR CVD-Risk had higher 10-year risk for MACE after adjustment for risk factors (adjusted hazard ratio [HR], 1.73 [95% CI, 1.47 to 2.03]). In the additional 2132 outpatients with known ASCVD risk, CXR CVD-Risk predicted MACE beyond the traditional ASCVD risk score (adjusted HR, 1.88 [CI, 1.24 to 2.85])., Limitation: Retrospective study design using electronic medical records., Conclusion: On the basis of a single CXR, CXR CVD-Risk predicts 10-year MACE beyond the clinical standard and may help identify individuals at high risk whose ASCVD risk score cannot be calculated because of missing data., Primary Funding Source: None., Competing Interests: Disclosures: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M23-1898.

Background: Heavy smokers are at increased risk for cardiovascular disease and may benefit from individualized risk quantification using routine lung cancer screening chest computed tomography. We investigated the prognostic value of deep learning-based automated epicardial adipose tissue quantification and compared it to established cardiovascular risk factors and coronary artery calcium., Methods: We investigated the prognostic value of automated epicardial adipose tissue quantification in heavy smokers enrolled in the National Lung Screening Trial and followed for 12.3 (11.9-12.8) years. The epicardial adipose tissue was segmented and quantified on non-ECG-synchronized, non-contrast low-dose chest computed tomography scans using a validated deep-learning algorithm. Multivariable survival regression analyses were then utilized to determine the associations of epicardial adipose tissue volume and density with all-cause and cardiovascular mortality (myocardial infarction and stroke)., Results: Here we show in 24,090 adult heavy smokers (59% men; 61 ± 5 years) that epicardial adipose tissue volume and density are independently associated with all-cause (adjusted hazard ratios: 1.10 and 1.38; P < 0.001) and cardiovascular mortality (adjusted hazard ratios: 1.14 and 1.78; P < 0.001) beyond demographics, clinical risk factors, body habitus, level of education, and coronary artery calcium score., Conclusions: Our findings suggest that automated assessment of epicardial adipose tissue from low-dose lung cancer screening images offers prognostic value in heavy smokers, with potential implications for cardiovascular risk stratification in this high-risk population., (© 2024. The Author(s).)

Manual segmentation of tumors and organs-at-risk (OAR) in 3D imaging for radiation-therapy planning is time-consuming and subject to variation between different observers. Artificial intelligence (AI) can assist with segmentation, but challenges exist in ensuring high-quality segmentation, especially for small, variable structures, such as the esophagus. We investigated the effect of variation in segmentation quality and style of physicians for training deep-learning models for esophagus segmentation and proposed a new metric, edge roughness, for evaluating/quantifying slice-to-slice inconsistency. This study includes a real-world cohort of 394 patients who each received radiation therapy (mainly for lung cancer). Segmentation of the esophagus was performed by 8 physicians as part of routine clinical care. We evaluated manual segmentation by comparing the length and edge roughness of segmentations among physicians to analyze inconsistencies. We trained eight multiple- and individual-physician segmentation models in total, based on U-Net architectures and residual backbones. We used the volumetric Dice coefficient to measure the performance for each model. We proposed a metric, edge roughness, to quantify the shift of segmentation among adjacent slices by calculating the curvature of edges of the 2D sagittal- and coronal-view projections. The auto-segmentation model trained on multiple physicians (MD1-7) achieved the highest mean Dice of 73.7 ± 14.8%. The individual-physician model (MD7) with the highest edge roughness (mean ± SD: 0.106 ± 0.016) demonstrated significantly lower volumetric Dice for test cases compared with other individual models (MD7: 58.5 ± 15.8%, MD6: 67.1 ± 16.8%, p < 0.001). A multiple-physician model trained after removing the MD7 data resulted in fewer outliers (e.g., Dice ≤ 40%: 4 cases for MD1-6, 7 cases for MD1-7, N total  = 394). While we initially detected this pattern in a single clinician, we validated the edge roughness metric across the entire dataset. The model trained with the lowest-quantile edge roughness (MD ER -Q1, N train  = 62) achieved significantly higher Dice (N test  = 270) than the model trained with the highest-quantile ones (MD ER -Q4, N train  = 62) (MD ER -Q1: 67.8 ± 14.8%, MD ER -Q4: 62.8 ± 15.7%, p < 0.001). This study demonstrates that there is significant variation in style and quality in manual segmentations in clinical care, and that training AI auto-segmentation algorithms from real-world, clinical datasets may result in unexpectedly under-performing algorithms with the inclusion of outliers. Importantly, this study provides a novel evaluation metric, edge roughness, to quantify physician variation in segmentation which will allow developers to filter clinical training data to optimize model performance., (© 2024. The Author(s).)

Artificial intelligence (AI) techniques are increasingly applied across various domains, favoured by the growing acquisition and public availability of large, complex datasets. Despite this trend, AI publications often suffer from lack of reproducibility and poor generalisation of findings, undermining scientific value and contributing to global research waste. To address these issues and focusing on the learning aspect of the AI field, we present RENOIR (REpeated random sampliNg fOr machIne leaRning), a modular open-source platform for robust and reproducible machine learning (ML) analysis. RENOIR adopts standardised pipelines for model training and testing, introducing elements of novelty, such as the dependence of the performance of the algorithm on the sample size. Additionally, RENOIR offers automated generation of transparent and usable reports, aiming to enhance the quality and reproducibility of AI studies. To demonstrate the versatility of our tool, we applied it to benchmark datasets from health, computer science, and STEM (Science, Technology, Engineering, and Mathematics) domains. Furthermore, we showcase RENOIR's successful application in recently published studies, where it identified classifiers for SET2D and TP53 mutation status in cancer. Finally, we present a use case where RENOIR was employed to address a significant pharmacological challenge-predicting drug efficacy. RENOIR is freely available at https://github.com/alebarberis/renoir ., (© 2024. The Author(s).)

Social determinants of health (SDoH) play a critical role in patient outcomes, yet their documentation is often missing or incomplete in the structured data of electronic health records (EHRs). Large language models (LLMs) could enable high-throughput extraction of SDoH from the EHR to support research and clinical care. However, class imbalance and data limitations present challenges for this sparsely documented yet critical information. Here, we investigated the optimal methods for using LLMs to extract six SDoH categories from narrative text in the EHR: employment, housing, transportation, parental status, relationship, and social support. The best-performing models were fine-tuned Flan-T5 XL for any SDoH mentions (macro-F1 0.71), and Flan-T5 XXL for adverse SDoH mentions (macro-F1 0.70). Adding LLM-generated synthetic data to training varied across models and architecture, but improved the performance of smaller Flan-T5 models (delta F1 + 0.12 to +0.23). Our best-fine-tuned models outperformed zero- and few-shot performance of ChatGPT-family models in the zero- and few-shot setting, except GPT4 with 10-shot prompting for adverse SDoH. Fine-tuned models were less likely than ChatGPT to change their prediction when race/ethnicity and gender descriptors were added to the text, suggesting less algorithmic bias (p < 0.05). Our models identified 93.8% of patients with adverse SDoH, while ICD-10 codes captured 2.0%. These results demonstrate the potential of LLMs in improving real-world evidence on SDoH and assisting in identifying patients who could benefit from resource support., (© 2024. The Author(s).)

Introduction: Artificial intelligence (AI)-based technologies embody countless solutions in radiation oncology, yet translation of AI-assisted software tools to actual clinical environments remains unrealized. We present the Deep Learning On-Demand Assistant (DL-ODA), a fully automated, end-to-end clinical platform that enables AI interventions for any disease site featuring an automated model-training pipeline, auto-segmentations, and QA reporting., Materials and Methods: We developed, tested, and prospectively deployed the DL-ODA system at a large university affiliated hospital center. Medical professionals activate the DL-ODA via two pathways (1): On-Demand, used for immediate AI decision support for a patient-specific treatment plan, and (2) Ambient, in which QA is provided for all daily radiotherapy (RT) plans by comparing DL segmentations with manual delineations and calculating the dosimetric impact. To demonstrate the implementation of a new anatomy segmentation, we used the model-training pipeline to generate a breast segmentation model based on a large clinical dataset. Additionally, the contour QA functionality of existing models was assessed using a retrospective cohort of 3,399 lung and 885 spine RT cases. Ambient QA was performed for various disease sites including spine RT and heart for dosimetric sparing., Results: Successful training of the breast model was completed in less than a day and resulted in clinically viable whole breast contours. For the retrospective analysis, we evaluated manual-versus-AI similarity for the ten most common structures. The DL-ODA detected high similarities in heart, lung, liver, and kidney delineations but lower for esophagus, trachea, stomach, and small bowel due largely to incomplete manual contouring. The deployed Ambient QAs for heart and spine sites have prospectively processed over 2,500 cases and 230 cases over 9 months and 5 months, respectively, automatically alerting the RT personnel., Discussion: The DL-ODA capabilities in providing universal AI interventions were demonstrated for On-Demand contour QA, DL segmentations, and automated model training, and confirmed successful integration of the system into a large academic radiotherapy department. The novelty of deploying the DL-ODA as a multi-modal, fully automated end-to-end AI clinical implementation solution marks a significant step towards a generalizable framework that leverages AI to improve the efficiency and reliability of RT systems., Competing Interests: RM: Advisory Board ViewRay, AstraZeneca, Consulting Varian Medical Systems, Sio Capital Management, Honorarium Novartis, Springer Nature, King Faisal Specialist Hospital, American Association of Physicists in Medicine, Research Funding National Institute of Health, ViewRay. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Kehayias, Yan, Bontempi, Quirk, Bitterman, Bredfeldt, Aerts, Mak and Guthier.)

Public imaging datasets are critical for the development and evaluation of automated tools in cancer imaging. Unfortunately, many do not include annotations or image-derived features, complicating downstream analysis. Artificial intelligence-based annotation tools have been shown to achieve acceptable performance and can be used to automatically annotate large datasets. As part of the effort to enrich public data available within NCI Imaging Data Commons (IDC), here we introduce AI-generated annotations for two collections containing computed tomography images of the chest, NSCLC-Radiomics, and a subset of the National Lung Screening Trial. Using publicly available AI algorithms, we derived volumetric annotations of thoracic organs-at-risk, their corresponding radiomics features, and slice-level annotations of anatomical landmarks and regions. The resulting annotations are publicly available within IDC, where the DICOM format is used to harmonize the data and achieve FAIR (Findable, Accessible, Interoperable, Reusable) data principles. The annotations are accompanied by cloud-enabled notebooks demonstrating their use. This study reinforces the need for large, publicly accessible curated datasets and demonstrates how AI can aid in cancer imaging., (© 2024. The Author(s).)

Foundation models in deep learning are characterized by a single large-scale model trained on vast amounts of data serving as the foundation for various downstream tasks. Foundation models are generally trained using self-supervised learning and excel in reducing the demand for training samples in downstream applications. This is especially important in medicine, where large labelled datasets are often scarce. Here, we developed a foundation model for cancer imaging biomarker discovery by training a convolutional encoder through self-supervised learning using a comprehensive dataset of 11,467 radiographic lesions. The foundation model was evaluated in distinct and clinically relevant applications of cancer imaging-based biomarkers. We found that it facilitated better and more efficient learning of imaging biomarkers and yielded task-specific models that significantly outperformed conventional supervised and other state-of-the-art pretrained implementations on downstream tasks, especially when training dataset sizes were very limited. Furthermore, the foundation model was more stable to input variations and showed strong associations with underlying biology. Our results demonstrate the tremendous potential of foundation models in discovering new imaging biomarkers that may extend to other clinical use cases and can accelerate the widespread translation of imaging biomarkers into clinical settings., Competing Interests: Competing interestsThe authors declare no competing interests., (© The Author(s) 2024.)

The remarkable advances of artificial intelligence (AI) technology are revolutionizing established approaches to the acquisition, interpretation, and analysis of biomedical imaging data. Development, validation, and continuous refinement of AI tools requires easy access to large high-quality annotated datasets, which are both representative and diverse. The National Cancer Institute (NCI) Imaging Data Commons (IDC) hosts large and diverse publicly available cancer image data collections. By harmonizing all data based on industry standards and colocalizing it with analysis and exploration resources, the IDC aims to facilitate the development, validation, and clinical translation of AI tools and address the well-documented challenges of establishing reproducible and transparent AI processing pipelines. Balanced use of established commercial products with open-source solutions, interconnected by standard interfaces, provides value and performance, while preserving sufficient agility to address the evolving needs of the research community. Emphasis on the development of tools, use cases to demonstrate the utility of uniform data representation, and cloud-based analysis aim to ease adoption and help define best practices. Integration with other data in the broader NCI Cancer Research Data Commons infrastructure opens opportunities for multiomics studies incorporating imaging data to further empower the research community to accelerate breakthroughs in cancer detection, diagnosis, and treatment. Published under a CC BY 4.0 license.

Lean muscle mass (LMM) is an important aspect of human health. Temporalis muscle thickness is a promising LMM marker but has had limited utility due to its unknown normal growth trajectory and reference ranges and lack of standardized measurement. Here, we develop an automated deep learning pipeline to accurately measure temporalis muscle thickness (iTMT) from routine brain magnetic resonance imaging (MRI). We apply iTMT to 23,876 MRIs of healthy subjects, ages 4 through 35, and generate sex-specific iTMT normal growth charts with percentiles. We find that iTMT was associated with specific physiologic traits, including caloric intake, physical activity, sex hormone levels, and presence of malignancy. We validate iTMT across multiple demographic groups and in children with brain tumors and demonstrate feasibility for individualized longitudinal monitoring. The iTMT pipeline provides unprecedented insights into temporalis muscle growth during human development and enables the use of LMM tracking to inform clinical decision-making., (© 2023. The Author(s).)

In the past decade, there has been a sharp increase in publications describing applications of convolutional neural networks (CNNs) in medical image analysis. However, recent reviews have warned of the lack of reproducibility of most such studies, which has impeded closer examination of the models and, in turn, their implementation in healthcare. On the other hand, the performance of these models is highly dependent on decisions on architecture and image pre-processing. In this work, we assess the reproducibility of three studies that use CNNs for head and neck cancer outcome prediction by attempting to reproduce the published results. In addition, we propose a new network structure and assess the impact of image pre-processing and model selection criteria on performance. We used two publicly available datasets: one with 298 patients for training and validation and another with 137 patients from a different institute for testing. All three studies failed to report elements required to reproduce their results thoroughly, mainly the image pre-processing steps and the random seed. Our model either outperforms or achieves similar performance to the existing models with considerably fewer parameters. We also observed that the pre-processing efforts significantly impact the model's performance and that some model selection criteria may lead to suboptimal models. Although there have been improvements in the reproducibility of deep learning models, our work suggests that wider implementation of reporting standards is required to avoid a reproducibility crisis., (© 2023. Springer Nature Limited.)

Importance: Sarcopenia is an established prognostic factor in patients with head and neck squamous cell carcinoma (HNSCC); the quantification of sarcopenia assessed by imaging is typically achieved through the skeletal muscle index (SMI), which can be derived from cervical skeletal muscle segmentation and cross-sectional area. However, manual muscle segmentation is labor intensive, prone to interobserver variability, and impractical for large-scale clinical use., Objective: To develop and externally validate a fully automated image-based deep learning platform for cervical vertebral muscle segmentation and SMI calculation and evaluate associations with survival and treatment toxicity outcomes., Design, Setting, and Participants: For this prognostic study, a model development data set was curated from publicly available and deidentified data from patients with HNSCC treated at MD Anderson Cancer Center between January 1, 2003, and December 31, 2013. A total of 899 patients undergoing primary radiation for HNSCC with abdominal computed tomography scans and complete clinical information were selected. An external validation data set was retrospectively collected from patients undergoing primary radiation therapy between January 1, 1996, and December 31, 2013, at Brigham and Women's Hospital. The data analysis was performed between May 1, 2022, and March 31, 2023., Exposure: C3 vertebral skeletal muscle segmentation during radiation therapy for HNSCC., Main Outcomes and Measures: Overall survival and treatment toxicity outcomes of HNSCC., Results: The total patient cohort comprised 899 patients with HNSCC (median [range] age, 58 [24-90] years; 140 female [15.6%] and 755 male [84.0%]). Dice similarity coefficients for the validation set (n = 96) and internal test set (n = 48) were 0.90 (95% CI, 0.90-0.91) and 0.90 (95% CI, 0.89-0.91), respectively, with a mean 96.2% acceptable rate between 2 reviewers on external clinical testing (n = 377). Estimated cross-sectional area and SMI values were associated with manually annotated values (Pearson r = 0.99; P < .001) across data sets. On multivariable Cox proportional hazards regression, SMI-derived sarcopenia was associated with worse overall survival (hazard ratio, 2.05; 95% CI, 1.04-4.04; P = .04) and longer feeding tube duration (median [range], 162 [6-1477] vs 134 [15-1255] days; hazard ratio, 0.66; 95% CI, 0.48-0.89; P = .006) than no sarcopenia., Conclusions and Relevance: This prognostic study's findings show external validation of a fully automated deep learning pipeline to accurately measure sarcopenia in HNSCC and an association with important disease outcomes. The pipeline could enable the integration of sarcopenia assessment into clinical decision making for individuals with HNSCC.

Purpose: Radiotherapy (RT) toxicities can impair survival and quality of life, yet remain understudied. Real-world evidence holds potential to improve our understanding of toxicities, but toxicity information is often only in clinical notes. We developed natural language processing (NLP) models to identify the presence and severity of esophagitis from notes of patients treated with thoracic RT., Methods: Our corpus consisted of a gold-labeled data set of 1,524 clinical notes from 124 patients with lung cancer treated with RT, manually annotated for Common Terminology Criteria for Adverse Events (CTCAE) v5.0 esophagitis grade, and a silver-labeled data set of 2,420 notes from 1,832 patients from whom toxicity grades had been collected as structured data during clinical care. We fine-tuned statistical and pretrained Bidirectional Encoder Representations from Transformers-based models for three esophagitis classification tasks: task 1, no esophagitis versus grade 1-3; task 2, grade ≤1 versus >1; and task 3, no esophagitis versus grade 1 versus grade 2-3. Transferability was tested on 345 notes from patients with esophageal cancer undergoing RT., Results: Fine-tuning of PubMedBERT yielded the best performance. The best macro-F1 was 0.92, 0.82, and 0.74 for tasks 1, 2, and 3, respectively. Selecting the most informative note sections during fine-tuning improved macro-F1 by ≥2% for all tasks. Silver-labeled data improved the macro-F1 by ≥3% across all tasks. For the esophageal cancer notes, the best macro-F1 was 0.73, 0.74, and 0.65 for tasks 1, 2, and 3, respectively, without additional fine-tuning., Conclusion: To our knowledge, this is the first effort to automatically extract esophagitis toxicity severity according to CTCAE guidelines from clinical notes. This provides proof of concept for NLP-based automated detailed toxicity monitoring in expanded domains.

Artificial intelligence (AI) and machine learning (ML) are becoming critical in developing and deploying personalized medicine and targeted clinical trials. Recent advances in ML have enabled the integration of wider ranges of data including both medical records and imaging (radiomics). However, the development of prognostic models is complex as no modeling strategy is universally superior to others and validation of developed models requires large and diverse datasets to demonstrate that prognostic models developed (regardless of method) from one dataset are applicable to other datasets both internally and externally. Using a retrospective dataset of 2,552 patients from a single institution and a strict evaluation framework that included external validation on three external patient cohorts (873 patients), we crowdsourced the development of ML models to predict overall survival in head and neck cancer (HNC) using electronic medical records (EMR) and pretreatment radiological images. To assess the relative contributions of radiomics in predicting HNC prognosis, we compared 12 different models using imaging and/or EMR data. The model with the highest accuracy used multitask learning on clinical data and tumor volume, achieving high prognostic accuracy for 2-year and lifetime survival prediction, outperforming models relying on clinical data only, engineered radiomics, or complex deep neural network architecture. However, when we attempted to extend the best performing models from this large training dataset to other institutions, we observed significant reductions in the performance of the model in those datasets, highlighting the importance of detailed population-based reporting for AI/ML model utility and stronger validation frameworks. We have developed highly prognostic models for overall survival in HNC using EMRs and pretreatment radiological images based on a large, retrospective dataset of 2,552 patients from our institution.Diverse ML approaches were used by independent investigators. The model with the highest accuracy used multitask learning on clinical data and tumor volume.External validation of the top three performing models on three datasets (873 patients) with significant differences in the distributions of clinical and demographic variables demonstrated significant decreases in model performance., Significance: ML combined with simple prognostic factors outperformed multiple advanced CT radiomics and deep learning methods. ML models provided diverse solutions for prognosis of patients with HNC but their prognostic value is affected by differences in patient populations and require extensive validation., (© 2023 The Authors; Published by the American Association for Cancer Research.)

Background: Pretreatment identification of pathological extranodal extension (ENE) would guide therapy de-escalation strategies for in human papillomavirus (HPV)-associated oropharyngeal carcinoma but is diagnostically challenging. ECOG-ACRIN Cancer Research Group E3311 was a multicentre trial wherein patients with HPV-associated oropharyngeal carcinoma were treated surgically and assigned to a pathological risk-based adjuvant strategy of observation, radiation, or concurrent chemoradiation. Despite protocol exclusion of patients with overt radiographic ENE, more than 30% had pathological ENE and required postoperative chemoradiation. We aimed to evaluate a CT-based deep learning algorithm for prediction of ENE in E3311, a diagnostically challenging cohort wherein algorithm use would be impactful in guiding decision-making., Methods: For this retrospective evaluation of deep learning algorithm performance, we obtained pretreatment CTs and corresponding surgical pathology reports from the multicentre, randomised de-escalation trial E3311. All enrolled patients on E3311 required pretreatment and diagnostic head and neck imaging; patients with radiographically overt ENE were excluded per study protocol. The lymph node with largest short-axis diameter and up to two additional nodes were segmented on each scan and annotated for ENE per pathology reports. Deep learning algorithm performance for ENE prediction was compared with four board-certified head and neck radiologists. The primary endpoint was the area under the curve (AUC) of the receiver operating characteristic., Findings: From 178 collected scans, 313 nodes were annotated: 71 (23%) with ENE in general, 39 (13%) with ENE larger than 1 mm ENE. The deep learning algorithm AUC for ENE classification was 0·86 (95% CI 0·82-0·90), outperforming all readers (p<0·0001 for each). Among radiologists, there was high variability in specificity (43-86%) and sensitivity (45-96%) with poor inter-reader agreement (κ 0·32). Matching the algorithm specificity to that of the reader with highest AUC (R2, false positive rate 22%) yielded improved sensitivity to 75% (+ 13%). Setting the algorithm false positive rate to 30% yielded 90% sensitivity. The algorithm showed improved performance compared with radiologists for ENE larger than 1 mm (p<0·0001) and in nodes with short-axis diameter 1 cm or larger., Interpretation: The deep learning algorithm outperformed experts in predicting pathological ENE on a challenging cohort of patients with HPV-associated oropharyngeal carcinoma from a randomised clinical trial. Deep learning algorithms should be evaluated prospectively as a treatment selection tool., Funding: ECOG-ACRIN Cancer Research Group and the National Cancer Institute of the US National Institutes of Health., Competing Interests: Declaration of interests BHK reports research support from the US National Institutes of Health (NIH; grant number K08DE030216-01) and the Radiological Society of North America, honoraria from Dedham Group, and stock options in Monogram Orthopedics. JDS reports research support paid to their institution from Merck, Bristol Myers Squibb (BMS), Regeneron, Debiopharm, and Merck KGA; consulting, scientific advisory board, and travel fees from Castle Biosciences, Genentech, Immunitas, Debiopharm, BMS, Nanobiotix, Tilos, AstraZeneca, LEK Consulting, Catenion, ACI Clinical, Astellas, Stimit, and Merck KGA; expert witness fees; stock options in Immunitas; and equity in Doximity. RH reports payment for consulting from Celgene, Nanobiotix, ISA, Merck, Eisai, BMS, AstraZeneca, Pfizer, Loxo, Genentech, Immunomic Therapeutics, GlaxoSmithKline, Gilead Sciences, Vaccinex, Emmanuel Merck Darmstadt (EMD), Serono, BioNTech, Achilles Therapeutics, Bayer, Coherus Biosciences, Boehringer Ingelheim, and Mirati Therapeutics; and research funding from Boehringer Ingelheim, Merck, BMS, Celgene, AstraZeneca, Genentech, Pfizer, Kura Oncology, EMD, and Serono. SA reports research support from the Agency for Healthcare Research and Quality, National Cancer Institute, National Science Foundation, American Cancer Society, American Society of Clinical Oncology, Amazon, and Patterson Trust; fees from American Society of Clinical Oncology, HighCape Capital, Emmerson Collective; and stock options in Prophet Brand Strategy. HJWLA reports research support from NIH (grant numbers U24CA194354, U01CA190234, U01CA209414, and R35CA22052), and the European Union—European research Council (grant number 866504) and stock and other ownership interests in Onc.Ai, Love Health, Health-AI, and Bayerische Motoren Werke AG. BAB reports a consulting or advisory role with Merck, Debiopharm Group, CUE Biopharma, Maverick Therapeutics, Rakuten Medical, Nanobiotix, Macrogenics, ALX Oncology, IO Biotech, Ipsen, Genentech/Roche, Kura Oncology, Merck KGA, Pharmaceutical Product Development Global, and Exelixis; research funding to their institution from Merck, Aduro Biotech, Formation Biologics, BMS, and CUE Biopharma; and travel, accommodations, or expenses from Merck and Debiopharm Group. RLF reports stock and other ownership interests in Novasenta; a consulting or advisory role with Merck, Pfizer, EMD Serono, Numab, Macrogenics, Aduro Biotech, Novasenta, Sanofi, Zymeworks, and BMS; and research funding from Bristol Myers Squibb, AstraZeneca/MedImmune, Merck, Tesaro, and Novasenta. All other authors declare no competing interests., (Copyright © 2023 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.)

Prevention and management of chronic lung diseases (asthma, lung cancer, etc.) are of great importance. While tests are available for reliable diagnosis, accurate identification of those who will develop severe morbidity/mortality is currently limited. Here, we developed a deep learning model, CXR Lung-Risk, to predict the risk of lung disease mortality from a chest x-ray. The model was trained using 147,497 x-ray images of 40,643 individuals and tested in three independent cohorts comprising 15,976 individuals. We found that CXR Lung-Risk showed a graded association with lung disease mortality after adjustment for risk factors, including age, smoking, and radiologic findings (Hazard ratios up to 11.86 [8.64-16.27]; p < 0.001). Adding CXR Lung-Risk to a multivariable model improved estimates of lung disease mortality in all cohorts. Our results demonstrate that deep learning can identify individuals at risk of lung disease mortality on easily obtainable x-rays, which may improve personalized prevention and treatment strategies., (© 2023. The Author(s).)

Cancer-associated cachexia (CAC) is a major contributor to morbidity and mortality in individuals with non-small cell lung cancer. Key features of CAC include alterations in body composition and body weight. Here, we explore the association between body composition and body weight with survival and delineate potential biological processes and mediators that contribute to the development of CAC. Computed tomography-based body composition analysis of 651 individuals in the TRACERx (TRAcking non-small cell lung Cancer Evolution through therapy (Rx)) study suggested that individuals in the bottom 20th percentile of the distribution of skeletal muscle or adipose tissue area at the time of lung cancer diagnosis, had significantly shorter lung cancer-specific survival and overall survival. This finding was validated in 420 individuals in the independent Boston Lung Cancer Study. Individuals classified as having developed CAC according to one or more features at relapse encompassing loss of adipose or muscle tissue, or body mass index-adjusted weight loss were found to have distinct tumor genomic and transcriptomic profiles compared with individuals who did not develop such features. Primary non-small cell lung cancers from individuals who developed CAC were characterized by enrichment of inflammatory signaling and epithelial-mesenchymal transitional pathways, and differentially expressed genes upregulated in these tumors included cancer-testis antigen MAGEA6 and matrix metalloproteinases, such as ADAMTS3. In an exploratory proteomic analysis of circulating putative mediators of cachexia performed in a subset of 110 individuals from TRACERx, a significant association between circulating GDF15 and loss of body weight, skeletal muscle and adipose tissue was identified at relapse, supporting the potential therapeutic relevance of targeting GDF15 in the management of CAC., (© 2023. The Author(s) under exclusive license to Springer Nature America, Inc.)

Circulating tumour DNA (ctDNA) can be used to detect and profile residual tumour cells persisting after curative intent therapy 1 . The study of large patient cohorts incorporating longitudinal plasma sampling and extended follow-up is required to determine the role of ctDNA as a phylogenetic biomarker of relapse in early-stage non-small-cell lung cancer (NSCLC). Here we developed ctDNA methods tracking a median of 200 mutations identified in resected NSCLC tissue across 1,069 plasma samples collected from 197 patients enrolled in the TRACERx study 2 . A lack of preoperative ctDNA detection distinguished biologically indolent lung adenocarcinoma with good clinical outcome. Postoperative plasma analyses were interpreted within the context of standard-of-care radiological surveillance and administration of cytotoxic adjuvant therapy. Landmark analyses of plasma samples collected within 120 days after surgery revealed ctDNA detection in 25% of patients, including 49% of all patients who experienced clinical relapse; 3 to 6 monthly ctDNA surveillance identified impending disease relapse in an additional 20% of landmark-negative patients. We developed a bioinformatic tool (ECLIPSE) for non-invasive tracking of subclonal architecture at low ctDNA levels. ECLIPSE identified patients with polyclonal metastatic dissemination, which was associated with a poor clinical outcome. By measuring subclone cancer cell fractions in preoperative plasma, we found that subclones seeding future metastases were significantly more expanded compared with non-metastatic subclones. Our findings will support (neo)adjuvant trial advances and provide insights into the process of metastatic dissemination using low-ctDNA-level liquid biopsy., (© 2023. The Author(s).)

Generative models, such as DALL-E 2 (OpenAI), could represent promising future tools for image generation, augmentation, and manipulation for artificial intelligence research in radiology, provided that these models have sufficient medical domain knowledge. Herein, we show that DALL-E 2 has learned relevant representations of x-ray images, with promising capabilities in terms of zero-shot text-to-image generation of new images, the continuation of an image beyond its original boundaries, and the removal of elements; however, its capabilities for the generation of images with pathological abnormalities (eg, tumors, fractures, and inflammation) or computed tomography, magnetic resonance imaging, or ultrasound images are still limited. The use of generative models for augmenting and generating radiological data thus seems feasible, even if the further fine-tuning and adaptation of these models to their respective domains are required first., (©Lisa C Adams, Felix Busch, Daniel Truhn, Marcus R Makowski, Hugo J W L Aerts, Keno K Bressem. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 16.03.2023.)

Introduction: Distant metastases (DMs) are the primary driver of mortality for patients with early stage NSCLC receiving stereotactic body radiation therapy (SBRT), yet patient-level risk is difficult to predict. We developed and validated a model to predict individualized risk of DM in this population., Methods: We used a multi-institutional database of 1280 patients with cT1-3N0M0 NSCLC treated with SBRT from 2006 to 2015 for model development and internal validation. A Fine and Gray (FG) regression model was built to predict 1-year DM risk and compared with a random survival forests model. The higher performing model was evaluated on an external data set of 130 patients from a separate institution. Discriminatory performance was evaluated using the time-dependent area under the curve (AUC). Calibration was assessed graphically and with Brier scores., Results: The FG model yielded an AUC of 0.71 (95% confidence interval [CI]: 0.57-0.86) compared with the AUC of random survival forest at 0.69 (95% CI: 0.63-0.85) in the internal test set and was selected for further testing. On external validation, the FG model yielded an AUC of 0.70 (95% CI: 0.57-0.83) with good calibration (Brier score: 0.08). The model identified a high-risk patient subgroup with greater 1-year DM rates in the internal test (20.0% [3 of 15] versus 2.9% [7 of 241], p = 0.001) and external validation (21.4% [3 of 15] versus 7.8% [9 of 116], p = 0.095). A model nomogram and online application was made available., Conclusions: We developed and externally validated a practical model that predicts DM risk in patients with NSCLC receiving SBRT which may help select patients for systemic therapy., (Copyright © 2022 International Association for the Study of Lung Cancer. Published by Elsevier Inc. All rights reserved.)

Background: Cancer trial accrual is a national priority, yet up to 20% of trials fail to accrue. Trial eligibility criteria growth may be associated with accrual failure. We sought to quantify eligibility criteria growth within National Cancer Institute (NCI)-affiliated trials and determine impact on accrual., Methods: Utilizing the Aggregated Analysis of ClinicalTrials.gov, we analyzed phase II/III interventional NCI-affiliated trials initiated between 2008 and 2018. Eligibility criteria growth was assessed via number of unique content words within combined inclusion and exclusion criteria. Association between unique word count and accrual failure was evaluated with multivariable logistic regression, adjusting for known predictors of failure. Medical terms associated with accrual failure were identified via natural language processing and categorized., Results: Of 1197 trials, 231 (19.3%) failed due to low accrual. Accrual failure rate increased with eligibility criteria growth, from 11.8% in the lowest decile (12-112 words) to 29.4% in the highest decile (445-750 words). Median eligibility criteria increased over time, from 214 (IQR [23, 282]) unique content words in 2008 to 417 (IQR [289, 514]) in 2018 (r 2  = 0.73, P < 0.001). Eligibility criteria growth was independently associated with accrual failure (OR: 1.09 per decile, 95% CI [1.03-1.15], p = 0.004). Eighteen exclusion criteria categories were significantly associated with accrual failure, including renal, pulmonary, and diabetic, among others (Bonferroni-corrected p < 0.001)., Conclusions: Eligibility criteria content growth is increasing dramatically among NCI-affiliated trials and is strongly associated with accrual failure. These findings support national initiatives to simplify eligibility criteria and suggest that further efforts are warranted to improve cancer trial accrual., (© 2022 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.)

Importance: Lung cancer screening with chest computed tomography (CT) prevents lung cancer death; however, fewer than 5% of eligible Americans are screened. CXR-LC, an open-source deep learning tool that estimates lung cancer risk from existing chest radiograph images and commonly available electronic medical record (EMR) data, may enable automated identification of high-risk patients as a step toward improving lung cancer screening participation., Objective: To validate CXR-LC using EMR data to identify individuals at high-risk for lung cancer to complement 2022 US Centers for Medicare & Medicaid Services (CMS) lung cancer screening eligibility guidelines., Design, Setting, and Participants: This prognostic study compared CXR-LC estimates with CMS screening guidelines using patient data from a large US hospital system. Included participants were persons who currently or formerly smoked cigarettes with an outpatient posterior-anterior chest radiograph between January 1, 2013, and December 31, 2014, with no history of lung cancer or screening CT. Data analysis was performed between May 2021 and June 2022., Exposures: CXR-LC lung cancer screening eligibility (previously defined as having a 3.297% or greater 12-year risk) based on inputs (chest radiograph image, age, sex, and whether currently smoking) extracted from the EMR., Main Outcomes and Measures: 6-year incident lung cancer., Results: A total of 14 737 persons were included in the study population (mean [SD] age, 62.6 [6.8] years; 7154 [48.5%] male; 204 [1.4%] Asian, 1051 [7.3%] Black, 432 [2.9%] Hispanic, 12 330 [85.2%] White) with a 2.4% rate of incident lung cancer over 6 years (361 patients with cancer). CMS eligibility could be determined in 6277 patients (42.6%) using smoking pack-year and quit-date from the EMR. Patients eligible by both CXR-LC and 2022 CMS criteria had a high rate of lung cancer (83 of 974 patients [8.5%]), higher than those eligible by 2022 CMS criteria alone (5 of 177 patients [2.8%]; P < .001). Patients eligible by CXR-LC but not 2022 CMS criteria also had a high 6-year incidence of lung cancer (121 of 3703 [3.3%]). In the 8460 cases (57.4%) where CMS eligibility was unknown, CXR-LC eligible patients had a 5-fold higher rate of lung cancer than ineligible (127 of 5177 [2.5%] vs 18 of 2283 [0.5%]; P < .001). Similar results were found in subgroups, including female patients and Black persons., Conclusions and Relevance: Using routine chest radiographs and other data automatically extracted from the EMR, CXR-LC identified high-risk individuals who may benefit from lung cancer screening CT.

Background: Artificial intelligence (AI) and deep learning have shown great potential in streamlining clinical tasks. However, most studies remain confined to in silico validation in small internal cohorts, without external validation or data on real-world clinical utility. We developed a strategy for the clinical validation of deep learning models for segmenting primary non-small-cell lung cancer (NSCLC) tumours and involved lymph nodes in CT images, which is a time-intensive step in radiation treatment planning, with large variability among experts., Methods: In this observational study, CT images and segmentations were collected from eight internal and external sources from the USA, the Netherlands, Canada, and China, with patients from the Maastro and Harvard-RT1 datasets used for model discovery (segmented by a single expert). Validation consisted of interobserver and intraobserver benchmarking, primary validation, functional validation, and end-user testing on the following datasets: multi-delineation, Harvard-RT1, Harvard-RT2, RTOG-0617, NSCLC-radiogenomics, Lung-PET-CT-Dx, RIDER, and thorax phantom. Primary validation consisted of stepwise testing on increasingly external datasets using measures of overlap including volumetric dice (VD) and surface dice (SD). Functional validation explored dosimetric effect, model failure modes, test-retest stability, and accuracy. End-user testing with eight experts assessed automated segmentations in a simulated clinical setting., Findings: We included 2208 patients imaged between 2001 and 2015, with 787 patients used for model discovery and 1421 for model validation, including 28 patients for end-user testing. Models showed an improvement over the interobserver benchmark (multi-delineation dataset; VD 0·91 [IQR 0·83-0·92], p=0·0062; SD 0·86 [0·71-0·91], p=0·0005), and were within the intraobserver benchmark. For primary validation, AI performance on internal Harvard-RT1 data (segmented by the same expert who segmented the discovery data) was VD 0·83 (IQR 0·76-0·88) and SD 0·79 (0·68-0·88), within the interobserver benchmark. Performance on internal Harvard-RT2 data segmented by other experts was VD 0·70 (0·56-0·80) and SD 0·50 (0·34-0·71). Performance on RTOG-0617 clinical trial data was VD 0·71 (0·60-0·81) and SD 0·47 (0·35-0·59), with similar results on diagnostic radiology datasets NSCLC-radiogenomics and Lung-PET-CT-Dx. Despite these geometric overlap results, models yielded target volumes with equivalent radiation dose coverage to those of experts. We also found non-significant differences between de novo expert and AI-assisted segmentations. AI assistance led to a 65% reduction in segmentation time (5·4 min; p<0·0001) and a 32% reduction in interobserver variability (SD; p=0·013)., Interpretation: We present a clinical validation strategy for AI models. We found that in silico geometric segmentation metrics might not correlate with clinical utility of the models. Experts' segmentation style and preference might affect model performance., Funding: US National Institutes of Health and EU European Research Council., Competing Interests: Declaration of interests AH reports consultancy fees for Altis Labs; and owning stock in Altis Labs. HJWLA reports being a scientific adviser and shareholder for Onc.Ai, Love Health, Health-AI, and Bristol Myers Squibb. RHM reports a research grant from ViewRay; honoraria from NewRT and ViewRay; is on an advisory board for AstraZeneca; and reports travel expenses from NewRT. DEK reports honoraria from RefleXion; and is a consultant and on an advisory board for Genentech. All other authors declare no competing interests., (Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.)

Identifying the presence of intravenous contrast material on CT scans is an important component of data curation for medical imaging-based artificial intelligence model development and deployment. Use of intravenous contrast material is often poorly documented in imaging metadata, necessitating impractical manual annotation by clinician experts. Authors developed a convolutional neural network (CNN)-based deep learning platform to identify intravenous contrast enhancement on CT scans. For model development and validation, authors used six independent datasets of head and neck (HN) and chest CT scans, totaling 133 480 axial two-dimensional sections from 1979 scans, which were manually annotated by clinical experts. Five CNN models were trained first on HN scans for contrast enhancement detection. Model performances were evaluated at the patient level on a holdout set and external test set. Models were then fine-tuned on chest CT data and externally validated. This study found that Digital Imaging and Communications in Medicine metadata tags for intravenous contrast material were missing or erroneous for 1496 scans (75.6%). An EfficientNetB4-based model showed the best performance, with areas under the curve (AUCs) of 0.996 and 1.0 in HN holdout ( n = 216) and external ( n = 595) sets, respectively, and AUCs of 1.0 and 0.980 in the chest holdout ( n = 53) and external ( n = 402) sets, respectively. This automated, scan-to-prediction platform is highly accurate at CT contrast enhancement detection and may be helpful for artificial intelligence model development and clinical application. Keywords: CT, Head and Neck, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN), Machine Learning Algorithms, Contrast Material Supplemental material is available for this article. © RSNA, 2022., Competing Interests: Disclosures of conflicts of interest: Z.Y. No relevant relationships. J.M.Q. No relevant relationships. A.H. Consultant for Altis Labs; shareholder in Altis Labs. R.Z. No relevant relationships. D.P. No relevant relationships. J.L. No relevant relationships. Z.Z. No relevant relationships. R.H.M. Contract/grant from ViewRay; consulting for ViewRay and AstraZeneca; payment for expert testimony from U.S. District Attorney's Office of New York. H.J.W.L.A. No relevant relationships. B.H.K. RSNA Research Scholar Award NIH K08DE030216., (© 2022 by the Radiological Society of North America, Inc.)

Background and Purpose: Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (LRC) in oropharyngeal squamous cell carcinoma (OPSCC) tumors., Materials and Methods: Radiomic features were extracted from postcontrast T1-weighted MRIs of 177 OPSCC primary tumors using six different manual delineation strategies. LRC prediction models based on recursive feature elimination combined with logistic regression were built. Models were trained and tested on data from each separate delineation. Additionally, the model derived from segmentations from the experienced reader was tested by each of the alternative delineations. Complementary, this was repeated with removal of size and shape features. Model performance was evaluated using area under the curve (AUC)., Results: Prediction performance of the experienced radiologist tumor delineation (AUC: 0.74) was superior compared to all other delineations when trained and tested (AUCs: 0.41-0.56) or trained on experienced delineations and tested (AUC: 0.56-0.67) on alternative segmentations. Removal of size and shape features considerably decreases prediction performance (AUC: 0.54). Applying the model based on expert delineations to spherical or single slice delineations makes prediction worthless since these models predict one class., Conclusion: Fast or readily available contours cannot substitute full expert tumor delineations in radiomics models predictive of LRC in OPSCC., (Copyright © 2022 Elsevier B.V. All rights reserved.)

Purpose: Coronary artery calcium (CAC) quantified on computed tomography (CT) scans is a robust predictor of atherosclerotic coronary disease; however, the feasibility and relevance of quantitating CAC from lung cancer radiotherapy planning CT scans is unknown. We used a previously validated deep learning (DL) model to assess whether CAC is a predictor of all-cause mortality and major adverse cardiac events (MACEs)., Methods: Retrospective analysis of non-contrast-enhanced radiotherapy planning CT scans from 428 patients with locally advanced lung cancer is performed. The DL-CAC algorithm was previously trained on 1,636 cardiac-gated CT scans and tested on four clinical trial cohorts. Plaques ≥ 1 cubic millimeter were measured to generate an Agatston-like DL-CAC score and grouped as DL-CAC = 0 (very low risk) and DL-CAC ≥ 1 (elevated risk). Cox and Fine and Gray regressions were adjusted for lung cancer and cardiovascular factors., Results: The median follow-up was 18.1 months. The majority (61.4%) had a DL-CAC ≥ 1. There was an increased risk of all-cause mortality with DL-CAC ≥ 1 versus DL-CAC = 0 (adjusted hazard ratio, 1.51; 95% CI, 1.01 to 2.26; P = .04), with 2-year estimates of 56.2% versus 45.4%, respectively. There was a trend toward increased risk of major adverse cardiac events with DL-CAC ≥ 1 versus DL-CAC = 0 (hazard ratio, 1.80; 95% CI, 0.87 to 3.74; P = .11), with 2-year estimates of 7.3% versus 1.2%, respectively., Conclusion: In this proof-of-concept study, CAC was effectively measured from routinely acquired radiotherapy planning CT scans using an automated model. Elevated CAC, as predicted by the DL model, was associated with an increased risk of mortality, suggesting a potential benefit for automated cardiac risk screening before cancer therapy begins., Competing Interests: Jakob WeissConsulting or Advisory Role: Onc.AI Danielle S. BittermanEmployment: Brigham and Women's Hospital Elizabeth HuynhResearch Funding: ViewRay David E. KozonoHonoraria: Reflexion Medical Anju NohriaConsulting or Advisory Role: Takeda, AstraZeneca, Boehringer IngelheimResearch Funding: Amgen (Inst)Uncompensated Relationships: Triple Gene Udo HoffmannConsulting or Advisory Role: ReCor Medical, Duke Clinical Research InstituteResearch Funding: AstraZeneca/MedImmune (Inst), Kowa Pharmaceutical (Inst), Heartflow (Inst) Hugo J. W. L. AertsLeadership: SpheraStock and Other Ownership Interests: Onc.AIConsulting or Advisory Role: Onc.AIResearch Funding: Varian Medical Systems Raymond H. MakConsulting or Advisory Role: AstraZeneca, ViewRayResearch Funding: ViewRay, AstraZeneca (Inst)No other potential conflicts of interest were reported.

Objectives: The goal of this study was to assess whether a deep learning estimate of age from a chest radiograph image (CXR-Age) can predict longevity beyond chronological age., Background: Chronological age is an imperfect measure of longevity. Biological age, a measure of overall health, may improve personalized care. This paper proposes a new way to estimate biological age using a convolutional neural network that takes as input a CXR image and outputs a chest x-ray age (in years) as a measure of long-term mortality risk., Methods: CXR-Age was developed using CXR from 116,035 individuals and validated in 2 held-out testing sets: 1) 75% of the CXR arm of PLCO (Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial) (N = 40,967); and 2) the CXR arm of NLST (National Lung Screening Trial) (N = 5,414). CXR-Age was compared to chronological age and a multivariable regression model of chronological age, risk factors, and radiograph findings to predict all-cause and cardiovascular mortality with a maximum 23 years and 13 years of follow-up, respectively. The primary outcome was observed mortality; results are provided for the testing datasets only., Results: In the PLCO testing dataset, a 5-year increase in CXR-Age carried a higher risk of all-cause mortality than a 5-year increase in chronological age (CXR-Age hazard ratio [HR]: 2.26 [95% confidence interval (CI): 2.24 to 2.29] vs. chronological age HR: 1.77 [95% CI: 1.75 to 1.78]; p < 0.001). A similar pattern was found for cardiovascular mortality (CXR-Age cause-specific HR: 2.45 per 5 years [95% CI: 2.34 to 2.56] vs. chronological age HR: 1.82 per 5 years [95% CI: 1.74 to 1.90]). Similar results were seen for both outcomes in the NLST external testing dataset. Adding CXR-Age to the multivariable model resulted in significant improvements for predicting both outcomes in both testing datasets (p < 0.001 for all comparisons)., Conclusions: Based on a CXR image, CXR-Age predicted long-term all-cause and cardiovascular mortality., Competing Interests: Funding Support and Author Disclosures This work was supported by the National Academy of Medicine Healthy Longevity Grand Challenge (2000011734). A graphics processing unit used for this research was donated to Dr Lu as an unrestricted gift through the Nvidia Corporation Academic Program. Original data collection for the ACRIN 6654 trial (NLST) was supported by National Cancer Institute Cancer Imaging Program grants. Dr Lu is supported by American Heart Association grants 18UNPG34030172 and 810966; has common stock in Nvidia and Advanced Micro Devices; reports research funding as a co-investigator to Massachusetts General Hospital from Kowa Company Limited and Medimmune/AstraZeneca; and received personal fees from PQBypass unrelated to this work. Dr Hoffmann is supported by an NIH grant (5K24HL113128); has received research support on behalf of his institution from Duke University (Abbott), HeartFlow, Kowa Company Limited, and MedImmune/AstraZeneca; and has received consulting fees from Duke University (NIH) and Recor Medical unrelated to this research. Dr Raghu is supported by NIH grant T32HL076136; and has common stock in Nvidia, Alphabet, and Apple. Dr Aerts has received personal fees from Sphera, Genospace, and Onc. AI outside the submitted work. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose., (Copyright © 2021 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)

Deep learning convolutional neural network (CNN) can predict mortality from chest radiographs, yet, it is unknown whether radiologists can perform the same task. Here, we investigate whether radiologists can visually assess image gestalt (defined as deviation from an unremarkable chest radiograph associated with the likelihood of 6-year mortality) of a chest radiograph to predict 6-year mortality. The assessment was validated in an independent testing dataset and compared to the performance of a CNN developed for mortality prediction. Results are reported for the testing dataset only (n = 100; age 62.5 ± 5.2; male 55%, event rate 50%). The probability of 6-year mortality based on image gestalt had high accuracy (AUC: 0.68 (95% CI 0.58-0.78), similar to that of the CNN (AUC: 0.67 (95% CI 0.57-0.77); p = 0.90). Patients with high/very high image gestalt ratings were significantly more likely to die when compared to those rated as very low (p ≤ 0.04). Assignment to risk categories was not explained by patient characteristics or traditional risk factors and imaging findings (p ≥ 0.2). In conclusion, assessing image gestalt on chest radiographs by radiologists renders high prognostic accuracy for the probability of mortality, similar to that of a specifically trained CNN. Further studies are warranted to confirm this concept and to determine potential clinical benefits., (© 2021. The Author(s).)

Purpose: Patients with locally advanced non-small cell lung cancer (LA-NSCLC) have a high prevalence of pre-existing coronary heart disease and face excess cardiac risk after thoracic radiation therapy. We sought to assess whether statin therapy is a predictor of overall survival (OS) after thoracic radiation therapy., Methods and Materials: We performed a retrospective analysis of 748 patients with LA-NSCLC treated with thoracic radiation therapy, using Kaplan-Meier OS estimates and Cox regression., Results: Statin use among high cardiac risk patients (Framingham risk ≥20% or pre-existing coronary heart disease; n = 496) was 51.2%. After adjustment for baseline cardiac risk and other prognostic factors, statin therapy was associated with a significantly increased risk of all-cause mortality (adjusted hazard ratio, 1.39; 95% confidence interval [CI], 1.00-1.91; P = .048) but not major adverse cardiac events (adjusted hazard ratio, 1.18; 95% CI, 0.52-2.68; P = .69). Among statin-naïve patients, mean heart dose ≥10 Gy versus <10 Gy was associated with a significantly increased risk of all-cause mortality (hazard ratio, 1.32; 95% CI, 1.04-1.68; P = .022), with 2-year OS estimates of 46.9% versus 60.0%, respectively. However, OS did not differ by heart dose among patients on statin therapy (hazard ratio, 1.00; 95% CI, 0.76-1.32; P = 1.00; P-interaction = .031), with 2-year OS estimates of 46.9% versus 50.3%, respectively., Conclusions: Among patients with LA-NSCLC, only half of statin-eligible high cardiac risk patients were on statin therapy, reflecting the highest cardiac risk level of our cohort. Statin use was an independent predictor of all-cause mortality but not major adverse cardiac events. Elevated mean heart dose (≥10 Gy) was associated with increased risk of all-cause mortality in statin-naïve patients but not among those on statin therapy, identifying a group of patients in which early intervention with statins may mitigate the deleterious effects of high heart radiation therapy dose. This warrants evaluation in prospective trials., (Copyright © 2021 The Authors. Published by Elsevier Inc. All rights reserved.)

The National Cancer Institute (NCI) Cancer Research Data Commons (CRDC) aims to establish a national cloud-based data science infrastructure. Imaging Data Commons (IDC) is a new component of CRDC supported by the Cancer Moonshot. The goal of IDC is to enable a broad spectrum of cancer researchers, with and without imaging expertise, to easily access and explore the value of deidentified imaging data and to support integrated analyses with nonimaging data. We achieve this goal by colocating versatile imaging collections with cloud-based computing resources and data exploration, visualization, and analysis tools. The IDC pilot was released in October 2020 and is being continuously populated with radiology and histopathology collections. IDC provides access to curated imaging collections, accompanied by documentation, a user forum, and a growing number of analysis use cases that aim to demonstrate the value of a data commons framework applied to cancer imaging research. SIGNIFICANCE: This study introduces NCI Imaging Data Commons, a new repository of the NCI Cancer Research Data Commons, which will support cancer imaging research on the cloud., (©2021 The Authors; Published by the American Association for Cancer Research.)

Objectives: The size of the heart may predict major cardiovascular events (MACE) in patients with stable chest pain. We aimed to evaluate the prognostic value of 3D whole heart volume (WHV) derived from non-contrast cardiac computed tomography (CT)., Methods: Among participants randomized to the CT arm of the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE), we used deep learning to extract WHV, defined as the volume of the pericardial sac. We compared the WHV across categories of cardiovascular risk factors and coronary artery disease (CAD) characteristics and determined the association of WHV with MACE (all-cause death, myocardial infarction, unstable angina; median follow-up: 26 months)., Results: In the 3798 included patients (60.5 ± 8.2 years; 51.5% women), the WHV was 351.9 ± 57.6 cm 3 /m 2 . We found smaller WHV in no- or non-obstructive CAD, women, people with diabetes, sedentary lifestyle, and metabolic syndrome. Larger WHV was found in obstructive CAD, men, and increased atherosclerosis cardiovascular disease (ASCVD) risk score (p < 0.05). In a time-to-event analysis, small WHV was associated with over 4.4-fold risk of MACE (HR (per one standard deviation) = 0.221; 95% CI: 0.068-0.721; p = 0.012) independent of ASCVD risk score and CT-derived CAD characteristics. In patients with non-obstructive CAD, but not in those with no- or obstructive CAD, WHV increased the discriminatory capacity of ASCVD and CT-derived CAD characteristics significantly., Conclusions: Small WHV may represent a novel imaging marker of MACE in stable chest pain. In particular, WHV may improve risk stratification in patients with non-obstructive CAD, a cohort with an unmet need for better risk stratification., Key Points: • Heart volume is easily assessable from non-contrast cardiac computed tomography. • Small heart volume may be an imaging marker of major adverse cardiac events independent and incremental to traditional cardiovascular risk factors and established CT measures of CAD. • Heart volume may improve cardiovascular risk stratification in patients with non-obstructive CAD., (© 2021. European Society of Radiology.)

Purpose: Mean heart dose (MHD) over 10 Gy and left anterior descending (LAD) coronary artery volume (V) receiving 15 Gy (V15Gy) greater than 10% can significantly increase the risk of major adverse cardiac events (MACE) in patients with non-small cell lung cancer (NSCLC). We sought to characterize the discordance between MHD and LAD dose and the association of this classification on the risk of MACE after radiation therapy., Methods and Materials: The coefficient of determination for MHD and LAD V15Gy was calculated in this retrospective analysis of 701 patients with locally advanced NSCLC treated with radiation therapy. Four groups were defined on the basis of high or low MHD (≥10 Gy vs <10 Gy) and LAD V15Gy (≥10% vs <10%). MACE (unstable angina, heart failure, myocardial infarction, coronary revascularization, and cardiac death) cumulative incidence was estimated, and Fine and Gray regressions were performed., Results: The proportion of variance in LAD V15Gy predictable from MHD was only 54.5% (R 2  = 0.545). There was discordance (where MHD was high [≥10 Gy] and LAD low [V15Gy < 10%], or vice versa) in 23.1% of patients (n = 162). Two-year MACE estimates were 4.2% (MHD high /LAD low ), 7.6% (MHD high /LAD high ), 1.8% (MHD low /LAD low ), and 13.0% (MHD low /LAD high ). Adjusting for pre-existing coronary heart disease and other prognostic factors, MHD high /LAD low (subdistribution hazard ratio [SHR], 0.34; 95% CI, 0.13-0.93; P = .036) and MHD low /LAD low (SHR, 0.24; 95% CI, 0.10-0.53; P < .001) were associated with a significantly reduced risk of MACE., Conclusions: MHD is insufficient to predict LAD V15Gy with confidence. When MHD and LAD V15Gy dose exposure is discordant, isolated low LAD V15Gy significantly reduces the risk of MACE in patients with locally advanced NSCLC after radiation therapy, suggesting that the validity of whole heart metrics for optimally predicting cardiac events should be reassessed., (Copyright © 2021 The Authors. Published by Elsevier Inc. All rights reserved.)