<ul><li>Self-supervised pre-training of medical foundation models on large-scale datasets is a common approach for achieving good performance.</li><li>Current methods for increasing pre-training data volume do not necessarily improve model performance.</li><li>The introduction of V-information in self-supervised pre-training provides a theoretical foundation for sample selection.</li><li>OptiDEL, an optimized data-effective learning method, outperforms existing approaches on multiple datasets by using 20x less training data.</li></ul>

A Self-Supervised Paradigm for Data-Efficient Medical Foundation Model Pre-training: V-information Optimization Framework

Discover more