Dr. Yifan Peng Receives Prestigious National Science Foundation Award

Dr. Yifan Peng, assistant professor of population health sciences at Weill Cornell Medicine, has received a National Science Foundation (NSF) award to support his project, “Knowledge-enhanced and interpretable radiology report generation.” Part of the Faculty Early Career Development Program, this award recognizes faculty dedicated to building a foundation for leadership in integrating education and research.

Photo of Yifan Peng

Dr. Peng is primarily interested in developing and applying computational approaches to biomedical text data and medical images. His research lab is motivated by integrating clinical-inspired approaches to machine learning and, reciprocally, using these approaches to better understand decision-making in clinical systems.

The NSF-funded project aligns closely with Dr. Peng’s current work in applying information extracted through natural language processing (NLP) and image analysis. Radiology reports are the primary means of communication between radiologists and referring physicians while also serving as a legal document. To date, many studies have demonstrated the feasibility of using deep learning to automatically generate radiology reports from chest x-rays. However, existing approaches utilize only current chest x-ray images and do not consider historical images, associated electronic health records (EHRs), and domain-specific prior knowledge. As a result, the current computer-generated reports are far from accurate and complete and may adversely affect patient care. “There is a critical need to study new report generation techniques to handle large-scale, real-world healthcare data,” said Dr. Peng. “This project will develop and validate a framework to automatically generate radiology reports using longitudinal, multimodal EHR data and domain knowledge.”

From the biomedical informatics perspective, Dr. Peng’s proposed research will address the knowledge gaps in understanding the role of natural language, image analysis, and deep learning in report generation by leveraging the wealth of information in the EHR. From the clinical perspective, its contribution will profoundly mitigate radiologist burnout, improve clinical accuracy and efficiency, and enhance decision-making.

“To the best of our knowledge, this project will represent a new step towards building automatic systems with a higher-level understanding of radiology knowledge and decision-making,” said Dr. Peng. “It is expected to open new research horizons to employ techniques and theories from data science to support next-generation medical diagnostic reasoning from EHRs. It would also be the first attempt to return control of the digital workspace to the radiologists and encourage user acceptance. We sincerely look forward to working with NSF to take it to the next level.”

Population Health Sciences 402 E. 67th St. New York, NY 10065 Phone: (646) 962-8001