Divyam Goel

BACKGROUND: Static cold storage of cardiac allografts at 4–8 °C or 10 °C has yielded promising heart transplant outcomes compared with ice storage. However, direct comparisons between these 2 preservation temperatures are lacking. This dual-center study is the first to compare adult heart transplant outcomes using allografts preserved at 10 °C versus 4–8 °C static cold storage. METHODS: All single-organ, donation-after-brain-death adult heart transplants performed at 2 high-volume centers between January 2020 and April 2025 were retrospectively analyzed. Multiorgan, adult congenital, and donation-after-circulatory-death cases were excluded. A 3:1 nearest-neighbor propensity score matching was applied using a standardized mean difference P =0.004), greater donor-recipient sex mismatch, and more frequent donor undersizing by predicted heart mass ratio. Recipients in this cohort were older and had more re-sternotomies and higher serum creatinine levels at transplant. Patients in the 4–8 °C group were more often listed as Status 2. After matching, 10 °C preservation was associated with less primary graft dysfunction (2 [2.9%] versus 19 [14.6%]; P =0.008), less left and right ventricular dysfunction, reduced new intraaortic balloon pump use, and improved 1-year survival (95.7% versus 86.2%; P =0.02). Other outcomes, including cardiac indices and posttransplant length of stay, were not significantly different between groups. CONCLUSIONS: Preservation of cardiac allografts at 10 °C may yield superior early graft function compared with 4–8 °C static cold storage. However, further prospective studies are required to delineate the optimal donor heart preservation temperature.

Introduction: Cardiovascular disease (CVD) remains the leading cause of mortality worldwide, with coronary artery disease (CAD), also known as ischemic heart disease (IHD), responsible for approximately 13% of global deaths in 2021. Studies applying machine learning (ML) and deep learning (DL) to heart disease classification have demonstrated promising results in risk prediction and feature extraction. Background/Objectives: In this study, we develop an AI/ML framework to predict and classify ischemic heart disease risk using publicly available datasets, the Framingham Heart Study and the Cleveland subset of the UCI Heart Disease dataset, along with explanations for how predictions were made by a process called SHAP (SHapley Additive exPlanations). Methods: We implemented a leakage-controlled machine learning pipeline that included data cleaning, stratified 80/20 train-test splitting, training-fold-only feature scaling and class balancing, 5-fold hyperparameter tuning, SHAP interpretability, and Brier score-based calibration assessment. Logistic regression, random forest, K-nearest neighbors, XGBoost, and a feedforward neural network were evaluated on the Framingham dataset and the Cleveland subset of the UCI Heart Disease dataset. Performance was assessed using accuracy, precision, recall, F1-score, Matthews correlation coefficient, AUROC, and Brier score. Results: After leakage-controlled evaluation, Framingham performance was more modest than in the preliminary analysis. Logistic regression achieved the highest AUROC on the Framingham dataset (0.7234), while random forest achieved the lowest Brier score (0.1750), and the feedforward neural network achieved the highest accuracy (0.7719). On the Cleveland subset, logistic regression achieved the strongest threshold-based performance (accuracy 0.8667, precision 0.8571, recall 0.8571, F1-score 0.8571, MCC 0.7321), whereas K-nearest neighbors achieved the highest AUROC (0.9531) and lowest Brier score (0.0942). SHAP highlighted systolic blood pressure, smoking status, and hypertension as influential predictors (Framingham) and number of major vessels, chest pain type, thallium stress-test result (thal; normal, fixed defect, or reversible defect), and age (Cleveland) as top predictors. Conclusions: Optimal model performance is dataset-dependent, and SHAP enhances clinical interpretability. Broader access to high-quality, de-identified medical data could accelerate reproducible ML research in cardiology.

HeartMate III (HM3) pump failure due to intrapump obstruction from biodebris or thrombosis is an uncommon but life-threatening condition. Here, we describe a case of HM3 failure from biodebris occluding the inflow cannula and pump interior 11 months after initial implant.

By Gliceida M Galarza Fortuna, Nishita Tripathi & 16 more. Abstract from 2023 Precision Oncology Conferences

Background: Only 74% of infectious diseases (ID) training positions were filled in the 2022 fellowship match, indicating a need to find increasingly novel and creative solutions for ID recruitment and outreach. Methods: The websites of 2321 universities and 181 medical schools across the United States were manually searched for the presence of undergraduate clubs and interest groups, respectively, for multiple medical specialties and subspecialties, including ID. Geographic data were used to compare the proximity of ID fellowships to undergraduate institutions. Results: ID student groups were extremely rare among the categories studied throughout undergraduate institutions (6 out of 2048, or 0.29%). Only 58 of 163 (35.6%) medical schools nationwide included an ID student group. Geographic comparison found that every adult ID fellowship is in the same county and/or city as at least 1 undergraduate institution and 28.5% of adult ID fellowships are in the same zip code as at least 1 undergraduate institution. Conclusions: The relative paucity of ID student interest groups presents an opportunity for the ID community to begin outreach and recruitment at the undergraduate and medical student levels, specifically through student groups.