Domain-incremental White Blood Cell Classification with Privacy-aware Continual Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kumari, Pratibha, Bozorgpour, Afshin, Reisenbüchler, Daniel, Jost, Edgar, Crysandt, Martina, Matek, Christian, Merhof, Dorit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915213213696000
author Kumari, Pratibha
Bozorgpour, Afshin
Reisenbüchler, Daniel
Jost, Edgar
Crysandt, Martina
Matek, Christian
Merhof, Dorit
author_facet Kumari, Pratibha
Bozorgpour, Afshin
Reisenbüchler, Daniel
Jost, Edgar
Crysandt, Martina
Matek, Christian
Merhof, Dorit
contents White blood cell (WBC) classification plays a vital role in hematology for diagnosing various medical conditions. However, it faces significant challenges due to domain shifts caused by variations in sample sources (e.g., blood or bone marrow) and differing imaging conditions across hospitals. Traditional deep learning models often suffer from catastrophic forgetting in such dynamic environments, while foundation models, though generally robust, experience performance degradation when the distribution of inference data differs from that of the training data. To address these challenges, we propose a generative replay-based Continual Learning (CL) strategy designed to prevent forgetting in foundation models for WBC classification. Our method employs lightweight generators to mimic past data with a synthetic latent representation to enable privacy-preserving replay. To showcase the effectiveness, we carry out extensive experiments with a total of four datasets with different task ordering and four backbone models including ResNet50, RetCCL, CTransPath, and UNI. Experimental results demonstrate that conventional fine-tuning methods degrade performance on previously learned tasks and struggle with domain shifts. In contrast, our continual learning strategy effectively mitigates catastrophic forgetting, preserving model performance across varying domains. This work presents a practical solution for maintaining reliable WBC classification in real-world clinical settings, where data distributions frequently evolve.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19819
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Domain-incremental White Blood Cell Classification with Privacy-aware Continual Learning
Kumari, Pratibha
Bozorgpour, Afshin
Reisenbüchler, Daniel
Jost, Edgar
Crysandt, Martina
Matek, Christian
Merhof, Dorit
Machine Learning
Computer Vision and Pattern Recognition
White blood cell (WBC) classification plays a vital role in hematology for diagnosing various medical conditions. However, it faces significant challenges due to domain shifts caused by variations in sample sources (e.g., blood or bone marrow) and differing imaging conditions across hospitals. Traditional deep learning models often suffer from catastrophic forgetting in such dynamic environments, while foundation models, though generally robust, experience performance degradation when the distribution of inference data differs from that of the training data. To address these challenges, we propose a generative replay-based Continual Learning (CL) strategy designed to prevent forgetting in foundation models for WBC classification. Our method employs lightweight generators to mimic past data with a synthetic latent representation to enable privacy-preserving replay. To showcase the effectiveness, we carry out extensive experiments with a total of four datasets with different task ordering and four backbone models including ResNet50, RetCCL, CTransPath, and UNI. Experimental results demonstrate that conventional fine-tuning methods degrade performance on previously learned tasks and struggle with domain shifts. In contrast, our continual learning strategy effectively mitigates catastrophic forgetting, preserving model performance across varying domains. This work presents a practical solution for maintaining reliable WBC classification in real-world clinical settings, where data distributions frequently evolve.
title Domain-incremental White Blood Cell Classification with Privacy-aware Continual Learning
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.19819