MetaHistoSeg: A Python Framework for Meta Learning in Histopathology Image Segmentation
This work addresses the problem of validating segmentation models across different cancer types for researchers in computational pathology, though it is incremental as it builds on existing meta-learning and transfer learning approaches.
The authors tackled the challenge of cross-domain generalization in histopathology image segmentation by introducing MetaHistoSeg, a Python framework for meta-learning and transfer learning, and a curated meta-dataset for benchmarking, finding that both methods deliver comparable results on average but vary in effectiveness across tasks.
Few-shot learning is a standard practice in most deep learning based histopathology image segmentation, given the relatively low number of digitized slides that are generally available. While many models have been developed for domain specific histopathology image segmentation, cross-domain generalization remains a key challenge for properly validating models. Here, tooling and datasets to benchmark model performance across histopathological domains are lacking. To address this limitation, we introduce MetaHistoSeg - a Python framework that implements unique scenarios in both meta learning and instance based transfer learning. Designed for easy extension to customized datasets and task sampling schemes, the framework empowers researchers with the ability of rapid model design and experimentation. We also curate a histopathology meta dataset - a benchmark dataset for training and validating models on out-of-distribution performance across a range of cancer types. In experiments we showcase the usage of MetaHistoSeg with the meta dataset and find that both meta-learning and instance based transfer learning deliver comparable results on average, but in some cases tasks can greatly benefit from one over the other.