Project Grant 2348440

This Project Grant award of $174,990.00 from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) supports the development of novel deep learning models and methods for automated detection of cardiomegaly (enlargement of the heart) in animals. The key goals are to: 1) Develop deep learning models that can yield understandable predictions aligned with traditional clinical metrics to improve trustworthiness of automated cardiomegaly detection; 2) Create an end-to-end deep detection system to predict biomarker positions and accurately calculate the Vertebral Heart Scale for cardiomegaly diagnosis; and 3) Design a unified human-computer interaction interface for cardiomegaly labeling, detection, and report generation. This award will support research and educational activities at Yeshiva University, including the engagement of PhD, master's, and undergraduate students, as well as the development of a graduate-level capstone course. The project aims to ultimately provide a high-precision, open-source software tool to help clinicians and the general public improve the accuracy, reduce the cost, and lower the emotional stress of cardiomegaly diagnosis, particularly for pet owners.