Transfer Learning Strategies for Pathological Foundation Models: A Systematic Evaluation in Brain Tumor Classification
This work addresses the problem of efficient AI-assisted diagnosis in clinical pathology, offering practical implications for reducing data requirements, though it is incremental in refining existing transfer learning approaches.
The study tackled brain tumor classification by systematically evaluating transfer learning strategies for pathological foundation models, finding that foundation models achieved robust performance with only 10 patches per case and that linear probing was sufficient while fine-tuning often degraded results.
Foundation models pretrained on large-scale pathology datasets have shown promising results across various diagnostic tasks. Here, we present a systematic evaluation of transfer learning strategies for brain tumor classification using these models. We analyzed 254 cases comprising five major tumor types: glioblastoma, astrocytoma, oligodendroglioma, primary central nervous system lymphoma, and metastatic tumors. Comparing state-of-the-art foundation models with conventional approaches, we found that foundation models demonstrated robust classification performance with as few as 10 patches per case, despite the traditional assumption that extensive per-case image sampling is necessary. Furthermore, our evaluation revealed that simple transfer learning strategies like linear probing were sufficient, while fine-tuning often degraded model performance. These findings suggest a paradigm shift from "training encoders on extensive pathological data" to "querying pre-trained encoders with labeled datasets", providing practical implications for implementing AI-assisted diagnosis in clinical pathology.