The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation · 2026
An integrated clinical-histopathologic prediction model for cardiac allograft rejection: Translating machine learning into clinical risk frameworks.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Kim Dale D, Madabhushi Anant, Margulies Kenneth B, Peyster Eliot G
Аннотация
BACKGROUND: Cardiac allograft rejection (CAR) remains the leading cause of early graft failure after heart transplantation (HT). Current diagnostics, including histologic grading of endomyocardial biopsy (EMB) and blood-based assays, lack accurate predictive power for future CAR risk. We developed a predictive model integrating routine clinical data with EMB-derived quantitative morphologic features to demonstrate the precision medicine potential of mining existing data sources in post-HT care.
METHODS: In a retrospective cohort of 484 HT recipients with 1,188 EMB encounters within 6 months post-transplant, we extracted 370 quantitative pathology features from digitized H&E-stained slides and 268 longitudinal clinical features from routine lab testing. Trained using the XGBoost algorithm, we compared model performance across time (cross-sectional vs longitudinal) and across data domains (clinical vs morphologic). The top predictors of our best-in-class model informed the derivation of a simplified Integrated Rejection Risk Index (IRRI) for patient risk stratification. Model performance was evaluated by AUROC, AUPRC, and time-to-event hazard ratios.
RESULTS: The fully integrated longitudinal model, inclusive of clinical and morphologic domains, achieved superior predictive accuracy (AUROC 0.86, AUPRC 0.74) compared to cross-sectional or single-domain models. IRRI stratified patients into risk categories with distinct future CAR hazards: high-risk patients showed a markedly increased CAR risk (HR = 6.15, 95% CI: 4.17-9.09), while low-risk patients had significantly reduced risk (HR = 0.52, 95% CI: 0.33-0.84).
CONCLUSIONS: By integrating longitudinal clinical and biopsy morphologic features, IRRI provides a scalable, interpretable tool for proactive CAR risk assessment. This precision-based approach offers a promising pathway toward risk-adaptive surveillance and immunosuppressive strategies.
METHODS: In a retrospective cohort of 484 HT recipients with 1,188 EMB encounters within 6 months post-transplant, we extracted 370 quantitative pathology features from digitized H&E-stained slides and 268 longitudinal clinical features from routine lab testing. Trained using the XGBoost algorithm, we compared model performance across time (cross-sectional vs longitudinal) and across data domains (clinical vs morphologic). The top predictors of our best-in-class model informed the derivation of a simplified Integrated Rejection Risk Index (IRRI) for patient risk stratification. Model performance was evaluated by AUROC, AUPRC, and time-to-event hazard ratios.
RESULTS: The fully integrated longitudinal model, inclusive of clinical and morphologic domains, achieved superior predictive accuracy (AUROC 0.86, AUPRC 0.74) compared to cross-sectional or single-domain models. IRRI stratified patients into risk categories with distinct future CAR hazards: high-risk patients showed a markedly increased CAR risk (HR = 6.15, 95% CI: 4.17-9.09), while low-risk patients had significantly reduced risk (HR = 0.52, 95% CI: 0.33-0.84).
CONCLUSIONS: By integrating longitudinal clinical and biopsy morphologic features, IRRI provides a scalable, interpretable tool for proactive CAR risk assessment. This precision-based approach offers a promising pathway toward risk-adaptive surveillance and immunosuppressive strategies.