Domain-Specific Pre-training Improves Confidence in Whole Slide Image Classification

TL;DR

This work tackles the challenge of classifying whole slide histopathology images (WSIs) with weak labels by examining domain-specific pre-training. It evaluates two modern MIL models, CLAM and TransMIL, across three backbones (ResNet50, DenseNet121, KimiaNet) using the DBTA brain tumor dataset to assess accuracy, AUC, and a defined confidence metric. The results show that domain-specific pre-training with KimiaNet improves model confidence by about 0.3%–1.3% and can achieve state-of-the-art performance for glioma classification when paired with TransMIL, underscoring the importance of feature extractor architecture. The findings suggest domain-specific pre-training can enhance the clinical reliability of WSI-based diagnosis and motivate further exploration of advanced backbone architectures.

Abstract

Whole Slide Images (WSIs) or histopathology images are used in digital pathology. WSIs pose great challenges to deep learning models for clinical diagnosis, owing to their size and lack of pixel-level annotations. With the recent advancements in computational pathology, newer multiple-instance learning-based models have been proposed. Multiple-instance learning for WSIs necessitates creating patches and uses the encoding of these patches for diagnosis. These models use generic pre-trained models (ResNet-50 pre-trained on ImageNet) for patch encoding. The recently proposed KimiaNet, a DenseNet121 model pre-trained on TCGA slides, is a domain-specific pre-trained model. This paper shows the effect of domain-specific pre-training on WSI classification. To investigate the effect of domain-specific pre-training, we considered the current state-of-the-art multiple-instance learning models, 1) CLAM, an attention-based model, and 2) TransMIL, a self-attention-based model, and evaluated the models' confidence and predictive performance in detecting primary brain tumors - gliomas. Domain-specific pre-training improves the confidence of the models and also achieves a new state-of-the-art performance of WSI-based glioma subtype classification, showing a high clinical applicability in assisting glioma diagnosis. We will publicly share our code and experimental results at https://github.com/soham-chitnis10/WSI-domain-specific.

Figures (1)

• Figure 1: The WSI pre-processing pipeline.