BAHOP: Similarity-based Basin Hopping for A fast hyper-parameter search in WSI classification
This addresses the need for efficient hyper-parameter tuning in medical imaging to improve classification accuracy on out-of-domain whole slide images, representing an incremental advance in optimization methods.
The paper tackles the problem of performance degradation in whole slide image classification when using fixed pre-processing hyper-parameters on out-of-domain data, proposing BAHOP for fast domain-specific tuning, which achieves 5% to 30% accuracy improvement and is 5 times faster on average.
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 $\times5$ times faster on average.