Uncertainty-Aware Multi-Expert Knowledge Distillation for Imbalanced Disease Grading

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tong, Shuo, Gao, Shangde, Liu, Ke, Huang, Zihang, Xu, Hongxia, Ying, Haochao, Wu, Jian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909598328291328
author Tong, Shuo
Gao, Shangde
Liu, Ke
Huang, Zihang
Xu, Hongxia
Ying, Haochao
Wu, Jian
author_facet Tong, Shuo
Gao, Shangde
Liu, Ke
Huang, Zihang
Xu, Hongxia
Ying, Haochao
Wu, Jian
contents Automatic disease image grading is a significant application of artificial intelligence for healthcare, enabling faster and more accurate patient assessments. However, domain shifts, which are exacerbated by data imbalance, introduce bias into the model, posing deployment difficulties in clinical applications. To address the problem, we propose a novel \textbf{U}ncertainty-aware \textbf{M}ulti-experts \textbf{K}nowledge \textbf{D}istillation (UMKD) framework to transfer knowledge from multiple expert models to a single student model. Specifically, to extract discriminative features, UMKD decouples task-agnostic and task-specific features with shallow and compact feature alignment in the feature space. At the output space, an uncertainty-aware decoupled distillation (UDD) mechanism dynamically adjusts knowledge transfer weights based on expert model uncertainties, ensuring robust and reliable distillation. Additionally, UMKD also tackles the problems of model architecture heterogeneity and distribution discrepancies between source and target domains, which are inadequately tackled by previous KD approaches. Extensive experiments on histology prostate grading (\textit{SICAPv2}) and fundus image grading (\textit{APTOS}) demonstrate that UMKD achieves a new state-of-the-art in both source-imbalanced and target-imbalanced scenarios, offering a robust and practical solution for real-world disease image grading.
format Preprint
id arxiv_https___arxiv_org_abs_2505_00592
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Uncertainty-Aware Multi-Expert Knowledge Distillation for Imbalanced Disease Grading
Tong, Shuo
Gao, Shangde
Liu, Ke
Huang, Zihang
Xu, Hongxia
Ying, Haochao
Wu, Jian
Computer Vision and Pattern Recognition
Machine Learning
Automatic disease image grading is a significant application of artificial intelligence for healthcare, enabling faster and more accurate patient assessments. However, domain shifts, which are exacerbated by data imbalance, introduce bias into the model, posing deployment difficulties in clinical applications. To address the problem, we propose a novel \textbf{U}ncertainty-aware \textbf{M}ulti-experts \textbf{K}nowledge \textbf{D}istillation (UMKD) framework to transfer knowledge from multiple expert models to a single student model. Specifically, to extract discriminative features, UMKD decouples task-agnostic and task-specific features with shallow and compact feature alignment in the feature space. At the output space, an uncertainty-aware decoupled distillation (UDD) mechanism dynamically adjusts knowledge transfer weights based on expert model uncertainties, ensuring robust and reliable distillation. Additionally, UMKD also tackles the problems of model architecture heterogeneity and distribution discrepancies between source and target domains, which are inadequately tackled by previous KD approaches. Extensive experiments on histology prostate grading (\textit{SICAPv2}) and fundus image grading (\textit{APTOS}) demonstrate that UMKD achieves a new state-of-the-art in both source-imbalanced and target-imbalanced scenarios, offering a robust and practical solution for real-world disease image grading.
title Uncertainty-Aware Multi-Expert Knowledge Distillation for Imbalanced Disease Grading
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2505.00592