Pre-processing whole slide images (WSIs) can impact classification performance. Our study shows that using fixed hyper-parameters for pre-processing out-of-domain WSIs can significantly degrade performance. Therefore, it is critical to search domain-specific hyper-parameters during inference. However, searching for an optimal parameter set is time-consuming. To overcome this, we propose BAHOP, a novel Similarity-based Basin Hopping optimization for fast parameter tuning to enhance inference performance on out-of-domain data. The proposed BAHOP achieves 5\% to 30\% improvement in accuracy with ×5 times faster on average.
@article{arxiv.2404.11161,
title = {BAHOP: Similarity-based Basin Hopping for A fast hyper-parameter search in WSI classification},
author = {Jun Wang and Yu Mao and Yufei Cui and Nan Guan and Chun Jason Xue},
journal= {arXiv preprint arXiv:2404.11161},
year = {2024}
}