PathBench-MIL: A Comprehensive AutoML and Benchmarking Framework for Multiple Instance Learning in Histopathology
This framework addresses the need for rapid experimentation and standardization in histopathology MIL, though it is incremental as it builds on existing MIL methods and tools.
The authors tackled the challenge of automating and standardizing multiple instance learning (MIL) pipelines in histopathology by introducing PathBench-MIL, an open-source AutoML and benchmarking framework that automates end-to-end pipeline construction and benchmarks dozens of models and feature extractors.
We introduce PathBench-MIL, an open-source AutoML and benchmarking framework for multiple instance learning (MIL) in histopathology. The system automates end-to-end MIL pipeline construction, including preprocessing, feature extraction, and MIL-aggregation, and provides reproducible benchmarking of dozens of MIL models and feature extractors. PathBench-MIL integrates visualization tooling, a unified configuration system, and modular extensibility, enabling rapid experimentation and standardization across datasets and tasks. PathBench-MIL is publicly available at https://github.com/Sbrussee/PathBench-MIL