SUMMARY
Yashbir (Yash) Singh, M.E., Ph.D., specializes in developing tools that extract meaningful data from medical images to enhance care. His research focuses on large language models, deep learning-based artificial intelligence (AI) algorithms and topological data analysis in medical imaging.
Dr. Singh’s main research objective is developing methods to improve the interpretability of AI models analyzing medical images. He specifically investigates imaging biomarkers in primary sclerosing cholangitis and cholangiocarcinoma. His work combines advanced imaging analysis techniques with clinical applications.
With his background as a biomedical engineer and cancer data scientist, Dr. Singh brings unique expertise to the field of AI-driven medical imaging analysis. He has demonstrated success in implementing and validating AI models for clinical use, effectively bridging the gap between technical innovation and practical medical needs.
Dr. Singh has established research collaborations with scientists globally, contributing to publications on machine learning, topological data analysis and persistent topology. His research philosophy emphasizes the value of geometric insights in advancing these fields.
Focus areas
Primary sclerosing cholangitis
- Dr. Singh developed a computational approach using algebraic topology combined with machine learning to predict hepatic decompensation.
- He developed novel imaging-based early detection features through collaborative research with Mayo Clinic, Norwegian liver transplant centers and the University of Toronto.
- Dr. Singh’s work focuses on computational approaches to improve patient outcomes and prediction of disease progression.
Cholangiocarcinoma
- Dr. Singh developed an innovative deep learning model for perihilar cholangiocarcinoma detection using medical imaging.
- He conducted a comparative analysis between the model’s performance and radiologist assessments.
- He also is working with leading cholangiocarcinoma expert Gregory J. Gores, M.D., to develop a multimodal approach to detect cholangiocarcinoma.
Advanced computational methods in medical imaging
- Dr. Singh is developing multimodal approaches for comprehensive cholangiocarcinoma analysis.
- He is implementing uncertainty quantification methods for liver cancer and coronary artery disease assessment.
- He is integrating diverse data sources to improve diagnostic accuracy and treatment planning.
Applied algebraic topology in medical research
- Dr. Singh is applying advanced topological methods to analyze complex medical imaging data.
- He is developing novel approaches to characterize rare diseases through topological data analysis.
- Working with Bradley J. Erickson, M.D., Ph.D., he is developing deep learning and semisupervised approaches combined with topological data analysis for image analysis.
- Dr. Singh also is focusing on translating mathematical insights into improved individualized patient care.
Coronary artery disease research
- Dr. Singh is using algebraic topology to identify novel patterns in cardiac imaging data.
- He is developing methods to extract clinically significant information from complex cardiovascular imaging.
- He also is working on innovative approaches to improve disease characterization and risk assessment.
Significance to patient care
Dr. Singh’s research advances in deep learning-based topological data analysis are transforming medical image interpretation across multiple modalities, including echocardiography, computerized tomography (CT) and magnetic resonance imaging (MRI).
His innovative approach combines sophisticated mathematical techniques with practical clinical applications, leading to improved diagnostic accuracy and treatment planning. The integration of deep learning with topological analysis has shown promise in early disease detection and progression monitoring, especially in complex conditions such as primary sclerosing cholangitis and cholangiocarcinoma.
Dr. Singh’s work demonstrates the potential for advanced computational methods to enhance individualized patient care through more-precise disease characterization and risk assessment. These developments are advancing current clinical practice and opening new avenues for clinical research, particularly in the study and treatment of rare liver diseases and complex cardiovascular conditions.
Professional highlights
- Editorial board, Discover Imaging, 2024-present.
- Editorial board, Oncotarget, 2024-present.
- Editorial board, Insights Into Imaging, 2024-present.
- Early career reviewer, Center for Scientific Review, National Institutes of Health (NIH), 2023-present.
- Member, Machine Learning Tools/Research Subcommittee, Society for Imaging Informatics in Medicine, 2022-present.
- Trainee Research Prize, for the paper “Cholangio-Net: A Deep Learning Approach for the Early Detection of Cholangiocarcinoma Using MRI,” Radiological Society of North America, 2024.
- Poster distinction award, American Association for the Study of Liver Diseases (AASLD), 2024.
- The AI Powered Science and Discovery Awards in Cancer, Mayo Clinic Comprehensive Cancer Center, 2024.
- Member, International Cholangiocarcinoma Research Network (ICRN), Cholangiocarcinoma Foundation, 2024.
- National Imaging Informatics Course, Society for Imaging Informatics in Medicine, 2024.
- Travel grant, The New York Academy of Sciences, 2023, 2024.
- NIH supplement travel grant, Stanford University, San Jose, California, 2023.
- Visiting faculty, Harvard Medical School, Boston, Massachusetts, 2023.
- Finalist, Arthur E. Weyman Young Investigator’s Award, American Society of Echocardiography, 2021.
- Short-term research fellowship, Deutscher Akademischer Austauschdienst (DAAD), 2018.