Non-parametric regularization for class imbalance federated medical image classification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911958195765248 |
|---|---|
| author | Wicaksana, Jeffry Yan, Zengqiang Cheng, Kwang-Ting |
| author_facet | Wicaksana, Jeffry Yan, Zengqiang Cheng, Kwang-Ting |
| contents | Limited training data and severe class imbalance pose significant challenges to developing clinically robust deep learning models. Federated learning (FL) addresses the former by enabling different medical clients to collaboratively train a deep model without sharing privacy-sensitive data. However, class imbalance worsens due to variation in inter-client class distribution. We propose federated learning with non-parametric regularization (FedNPR and FedNPR-Per, a personalized version of FedNPR) to regularize the feature extractor and enhance useful and discriminative signal in the feature space. Our extensive experiments show that FedNPR outperform the existing state-of-the art FL approaches in class imbalance skin lesion classification and intracranial hemorrhage identification. Additionally, the non-parametric regularization module consistently improves the performance of existing state-of-the-art FL approaches. We believe that NPR is a valuable tool in FL under clinical settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12446 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Non-parametric regularization for class imbalance federated medical image classification Wicaksana, Jeffry Yan, Zengqiang Cheng, Kwang-Ting Computer Vision and Pattern Recognition Limited training data and severe class imbalance pose significant challenges to developing clinically robust deep learning models. Federated learning (FL) addresses the former by enabling different medical clients to collaboratively train a deep model without sharing privacy-sensitive data. However, class imbalance worsens due to variation in inter-client class distribution. We propose federated learning with non-parametric regularization (FedNPR and FedNPR-Per, a personalized version of FedNPR) to regularize the feature extractor and enhance useful and discriminative signal in the feature space. Our extensive experiments show that FedNPR outperform the existing state-of-the art FL approaches in class imbalance skin lesion classification and intracranial hemorrhage identification. Additionally, the non-parametric regularization module consistently improves the performance of existing state-of-the-art FL approaches. We believe that NPR is a valuable tool in FL under clinical settings. |
| title | Non-parametric regularization for class imbalance federated medical image classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.12446 |