Jayashree Kalpathy-Cramer
贾亚斯里·卡尔帕西-克拉默
PhD
Professor of Radiology and Biomedical Imaging, UCSF; Director of Artificial Intelligence, Department of Radiology and Biomedical Imaging, UCSFUCSF放射学与生物医学成像学教授；UCSF放射学与生物医学成像科人工智能主任
👥Biography 个人简介
Jayashree Kalpathy-Cramer, PhD is Professor of Radiology and Biomedical Imaging and Director of Artificial Intelligence at the UCSF Department of Radiology and Biomedical Imaging. Previously, she was a faculty member at Massachusetts General Hospital/Harvard Medical School where she led the Quantitative Tumor Imaging laboratory and the QTIM (Quantitative Tumor Imaging at MGH) group. Dr. Kalpathy-Cramer is one of the most prominent researchers in AI for medical imaging, with a career spanning early development of image analysis methods for brain tumors, to leading the development of deep learning systems for retinal disease, to large-scale federated learning platforms for cross-institutional medical AI training. Her work on AI for diabetic retinopathy and retinopathy of prematurity (ROP) contributed to the regulatory approval of AI-based retinal screening systems, making her one of only a small number of academic researchers whose work has directly contributed to FDA-cleared AI medical devices. In neuro-oncology AI, her laboratory developed MRI-based deep learning systems for glioma grading, IDH mutation prediction, molecular subtype classification, and treatment response assessment in glioblastoma, providing tools for non-invasive tumor characterization that have been validated in multi-institutional imaging datasets. She is a founder of the FeTS (Federated Tumor Segmentation) initiative and a leader in privacy-preserving federated learning for multi-site clinical AI development.
🧪Research Fields 研究领域
🎓Key Contributions 主要贡献
Brain Tumor AI — Glioma Segmentation, Grading, and Molecular Prediction from MRI
Led development of deep learning algorithms for automated brain tumor segmentation, glioma grading, IDH mutation status prediction, MGMT methylation classification, and treatment response assessment from multi-parametric MRI (T1, T2, FLAIR, T1-post-contrast sequences). These algorithms achieved expert-level performance in multi-center BraTS (Brain Tumor Segmentation) benchmark challenges and were validated in retrospective clinical cohorts, providing accurate, reproducible automated tumor characterization that reduces radiology workload and enables high-throughput neuro-oncology research.
Retinopathy AI — Contributing to FDA-Cleared Algorithms for Retinal Disease Screening
Developed and validated deep learning algorithms for automated diabetic retinopathy (DR) grading and retinopathy of prematurity (ROP) screening from fundus photography and RetCam images. Contributed to clinical validation datasets and performance characterization studies that supported FDA De Novo authorization and 510(k) clearances for AI-based retinal screening devices, making Dr. Kalpathy-Cramer one of the rare academic investigators whose research has directly translated to cleared AI medical devices deployed in clinical screening programs.
Federated Learning — FeTS Initiative for Privacy-Preserving Multi-Site AI
Co-founded and led the FeTS (Federated Tumor Segmentation) initiative, one of the largest federated learning collaborations in medical AI, involving over 70 institutions sharing model updates (rather than patient data) to train globally generalizable glioblastoma segmentation AI. Published methods and results in Nature Communications demonstrating that federated learning can achieve performance matching or exceeding centralized training while preserving data privacy and complying with HIPAA and GDPR, establishing federated learning as a viable paradigm for clinical AI at global scale.
Quantitative Imaging Biomarkers — Radiomics and AI for Treatment Response Assessment
Developed quantitative imaging biomarker frameworks for assessing treatment response in glioblastoma, lung cancer, and breast cancer using radiomic feature extraction and deep learning from multi-parametric MRI and CT, addressing the critical challenge of pseudoprogression versus true progression in GBM patients on bevacizumab and immunotherapy. Published validated models for distinguishing true radiographic progression from treatment-related changes, providing AI decision support for neuro-oncology management decisions at imaging follow-up.
Representative Works 代表性著作
Federated Learning Enables Big Data for Rare Cancer Boundary Detection
Nature Communications (2022)
FeTS initiative federated learning across 71 sites for brain tumor segmentation, demonstrating federated AI achieves performance comparable to centralized training while enabling global multi-institutional collaboration without data sharing.
International Evaluation of an AI System for Breast Cancer Screening
Nature (2020)
Landmark multi-national evaluation of AI for mammography screening demonstrating superior performance to radiologists in breast cancer detection, including major contributions to validation methodology.
Automated Retinopathy of Prematurity Case Detection with Deep Neural Networks
JAMA Ophthalmology (2018)
Deep learning system for ROP severity grading from RetCam images with expert-level performance, contributing to AI screening tool development for premature infant ophthalmology.
Predicting IDH Mutation Status in Diffuse Gliomas Using Multiparametric MRI and Deep Learning
Neuro-Oncology (2021)
Deep learning model predicting IDH mutation status from pre-operative multi-parametric MRI in diffuse gliomas with AUC >0.89, enabling non-invasive molecular subtype classification.
🏆Awards & Recognition 奖项与荣誉
📄Data Sources 数据来源
Last updated: 2026-04-06 | All information from publicly available academic sources
Related Experts 相关专家
Faisal Mahmood
Brigham and Women's Hospital; Harvard Medical School
Kun-Hsing Yu
Harvard Medical School; Brigham and Women's Hospital
Ziad Obermeyer
University of California, Berkeley; UC San Francisco
Andrew H. Beck
PathAI; Harvard Medical School (formerly)
关注 贾亚斯里·卡尔帕西-克拉默 的研究动态
Follow Jayashree Kalpathy-Cramer's research updates
留下邮箱，当我们发布与 Jayashree Kalpathy-Cramer（University of California, San Francisco）相关的新研究或访谈时，我们会通知你。
Explore More Experts
Discover the researchers shaping the future of cancer treatment