IVCVMay 20, 2024

Comparing ImageNet Pre-training with Digital Pathology Foundation Models for Whole Slide Image-Based Survival Analysis

arXiv:2405.17446v43 citationsh-index: 2
Originality Synthesis-oriented
AI Analysis

This work addresses survival prediction in medical imaging, offering incremental improvements by integrating domain-specific foundation models into existing frameworks.

The study compared using ImageNet pre-trained models versus digital pathology foundation models for survival analysis from whole slide images, finding that ensembles of histopathology-specific models improved baseline accuracy but benefits decreased with more complex architectures.

The abundance of information present in Whole Slide Images (WSIs) renders them an essential tool for survival analysis. Several Multiple Instance Learning frameworks proposed for this task utilize a ResNet50 backbone pre-trained on natural images. By leveraging recenetly released histopathological foundation models such as UNI and Hibou, the predictive prowess of existing MIL networks can be enhanced. Furthermore, deploying an ensemble of digital pathology foundation models yields higher baseline accuracy, although the benefits appear to diminish with more complex MIL architectures. Our code will be made publicly available upon acceptance.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes