FedA3I: Annotation Quality-Aware Aggregation for Federated Medical Image Segmentation against Heterogeneous Annotation Noise

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wu, Nannan, Sun, Zhaobin, Yan, Zengqiang, Yu, Li
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929213930471424
author Wu, Nannan
Sun, Zhaobin
Yan, Zengqiang
Yu, Li
author_facet Wu, Nannan
Sun, Zhaobin
Yan, Zengqiang
Yu, Li
contents Federated learning (FL) has emerged as a promising paradigm for training segmentation models on decentralized medical data, owing to its privacy-preserving property. However, existing research overlooks the prevalent annotation noise encountered in real-world medical datasets, which limits the performance ceilings of FL. In this paper, we, for the first time, identify and tackle this problem. For problem formulation, we propose a contour evolution for modeling non-independent and identically distributed (Non-IID) noise across pixels within each client and then extend it to the case of multi-source data to form a heterogeneous noise model (i.e., Non-IID annotation noise across clients). For robust learning from annotations with such two-level Non-IID noise, we emphasize the importance of data quality in model aggregation, allowing high-quality clients to have a greater impact on FL. To achieve this, we propose Federated learning with Annotation quAlity-aware AggregatIon, named FedA3I, by introducing a quality factor based on client-wise noise estimation. Specifically, noise estimation at each client is accomplished through the Gaussian mixture model and then incorporated into model aggregation in a layer-wise manner to up-weight high-quality clients. Extensive experiments on two real-world medical image segmentation datasets demonstrate the superior performance of FedA$^3$I against the state-of-the-art approaches in dealing with cross-client annotation noise. The code is available at https://github.com/wnn2000/FedAAAI.
format Preprint
id arxiv_https___arxiv_org_abs_2312_12838
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle FedA3I: Annotation Quality-Aware Aggregation for Federated Medical Image Segmentation against Heterogeneous Annotation Noise
Wu, Nannan
Sun, Zhaobin
Yan, Zengqiang
Yu, Li
Machine Learning
Computer Vision and Pattern Recognition
Federated learning (FL) has emerged as a promising paradigm for training segmentation models on decentralized medical data, owing to its privacy-preserving property. However, existing research overlooks the prevalent annotation noise encountered in real-world medical datasets, which limits the performance ceilings of FL. In this paper, we, for the first time, identify and tackle this problem. For problem formulation, we propose a contour evolution for modeling non-independent and identically distributed (Non-IID) noise across pixels within each client and then extend it to the case of multi-source data to form a heterogeneous noise model (i.e., Non-IID annotation noise across clients). For robust learning from annotations with such two-level Non-IID noise, we emphasize the importance of data quality in model aggregation, allowing high-quality clients to have a greater impact on FL. To achieve this, we propose Federated learning with Annotation quAlity-aware AggregatIon, named FedA3I, by introducing a quality factor based on client-wise noise estimation. Specifically, noise estimation at each client is accomplished through the Gaussian mixture model and then incorporated into model aggregation in a layer-wise manner to up-weight high-quality clients. Extensive experiments on two real-world medical image segmentation datasets demonstrate the superior performance of FedA$^3$I against the state-of-the-art approaches in dealing with cross-client annotation noise. The code is available at https://github.com/wnn2000/FedAAAI.
title FedA3I: Annotation Quality-Aware Aggregation for Federated Medical Image Segmentation against Heterogeneous Annotation Noise
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.12838