Personalized Federated Segmentation with Shared Feature Aggregation and Boundary-Focused Calibration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tashdeed, Ishmam, Rahman, Md. Atiqur, Islam, Sabrina, Hossain, Md. Azam
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914168677859328
author Tashdeed, Ishmam
Rahman, Md. Atiqur
Islam, Sabrina
Hossain, Md. Azam
author_facet Tashdeed, Ishmam
Rahman, Md. Atiqur
Islam, Sabrina
Hossain, Md. Azam
contents Personalized federated learning (PFL) possesses the unique capability of preserving data confidentiality among clients while tackling the data heterogeneity problem of non-independent and identically distributed (Non-IID) data. Its advantages have led to widespread adoption in domains such as medical image segmentation. However, the existing approaches mostly overlook the potential benefits of leveraging shared features across clients, where each client contains segmentation data of different organs. In this work, we introduce a novel personalized federated approach for organ agnostic tumor segmentation (FedOAP), that utilizes cross-attention to model long-range dependencies among the shared features of different clients and a boundary-aware loss to improve segmentation consistency. FedOAP employs a decoupled cross-attention (DCA), which enables each client to retain local queries while attending to globally shared key-value pairs aggregated from all clients, thereby capturing long-range inter-organ feature dependencies. Additionally, we introduce perturbed boundary loss (PBL) which focuses on the inconsistencies of the predicted mask's boundary for each client, forcing the model to localize the margins more precisely. We evaluate FedOAP on diverse tumor segmentation tasks spanning different organs. Extensive experiments demonstrate that FedOAP consistently outperforms existing state-of-the-art federated and personalized segmentation methods.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18847
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Personalized Federated Segmentation with Shared Feature Aggregation and Boundary-Focused Calibration
Tashdeed, Ishmam
Rahman, Md. Atiqur
Islam, Sabrina
Hossain, Md. Azam
Computer Vision and Pattern Recognition
Artificial Intelligence
Personalized federated learning (PFL) possesses the unique capability of preserving data confidentiality among clients while tackling the data heterogeneity problem of non-independent and identically distributed (Non-IID) data. Its advantages have led to widespread adoption in domains such as medical image segmentation. However, the existing approaches mostly overlook the potential benefits of leveraging shared features across clients, where each client contains segmentation data of different organs. In this work, we introduce a novel personalized federated approach for organ agnostic tumor segmentation (FedOAP), that utilizes cross-attention to model long-range dependencies among the shared features of different clients and a boundary-aware loss to improve segmentation consistency. FedOAP employs a decoupled cross-attention (DCA), which enables each client to retain local queries while attending to globally shared key-value pairs aggregated from all clients, thereby capturing long-range inter-organ feature dependencies. Additionally, we introduce perturbed boundary loss (PBL) which focuses on the inconsistencies of the predicted mask's boundary for each client, forcing the model to localize the margins more precisely. We evaluate FedOAP on diverse tumor segmentation tasks spanning different organs. Extensive experiments demonstrate that FedOAP consistently outperforms existing state-of-the-art federated and personalized segmentation methods.
title Personalized Federated Segmentation with Shared Feature Aggregation and Boundary-Focused Calibration
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2511.18847