Beyond Knowledge Silos: Task Fingerprinting for Democratization of Medical Imaging AI

作     者：Godau, Patrick Srivastava, Akriti Adler, Tim Maier-Hein, Lena 

作者机构： NCT Heidelberg DKFZ University Hospital Heidelberg Germany  Division of Intelligent Medical Systems Heidelberg Germany Faculty of Mathematics and Computer Science Heidelberg University Germany HIDSS4Health - Helmholtz Information and Data Science School for Health Heidelberg Karlsruhe Germany Medical Faculty Heidelberg University Germany 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2024年

核心收录：

主　　题：Multi task learning 

摘      要：The field of medical imaging AI is currently undergoing rapid transformations, with methodical research increasingly translated into clinical practice. Despite these successes, research suffers from knowledge silos, hindering collaboration and progress: Existing knowledge is scattered across publications and many details remain unpublished, while privacy regulations restrict data sharing. In the spirit of democratizing of AI, we propose a framework for secure knowledge transfer in the field of medical image analysis. The key to our approach is datasetfingerprints, structured representations of feature distributions, that enable quantification of task similarity. We tested our approach across 71 distinct tasks and 12 medical imaging modalities by transferring neural architectures, pretraining, augmentation policies, and multi-task learning. According to comprehensive analyses, our method outperforms traditional methods for identifying relevant knowledge and facilitates collaborative model training. Our framework fosters the democratization of AI in medical imaging and could become a valuable tool for promoting faster scientific advancement. © 2024, CC BY.