# Federated Transfer Learning for Medical Image Classification under Mixed Skew

DevFeed: [Federated Transfer Learning for Medical Image Classification under Mixed Skew](<https://devfeed.tech/articles/federated-transfer-learning-for-medical-image-classification-under-mixed-skew-117008.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.05035>)

Author: Alexandre G. Leclercq, S\\'ebastien Bougleux, Aur\\'elien Corroyer-Dulmont, Romain H\\'erault

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Federated Learning](<https://devfeed.tech/topics/federated-learning.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Medical imaging](<https://devfeed.tech/topics/medical-imaging.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [federated](<https://devfeed.tech/tags/federated.md>), [image-classification](<https://devfeed.tech/tags/image-classification.md>), [medical](<https://devfeed.tech/tags/medical.md>), [transfer-learning](<https://devfeed.tech/tags/transfer-learning.md>)

## AI overview

The Personalized Aligner Federated Pretrained (PAFP) framework aims to address mixed feature, label, concept, and quantity shifts in multicenter medical imaging, where limited data makes training personalized federated models from scratch unrealistic. It uses a pretrained backbone with client-specific Feature Alignment modules that adjust hidden feature distributions toward a server-aggregated target. On the FedISIC benchmark, the authors report the highest unweighted balanced accuracy and roughly six-fold lower inter-client performance variance than the strongest baseline. The supplied abstract does not give absolute scores or details of the comparison.

## Source excerpt

arXiv:2611.05035v1 Announce Type: new Abstract: Deep learning in medical imaging is constrained by data scarcity and by privacy regulations that make centralizing patient data across institutions impractical. Federated Learning (FL) addresses these constraints by keeping data local, but its distributed nature exposes training to strong statistical heterogeneity. Multicentric medical data is particularly affected, exhibiting a mixed skew that combines feature, label, concept, and quantity shifts across centers. Personalized FL mitigates heterogeneity by tailoring a model to each client, yet most methods assume training from scratch, which is unrealistic given the high dimensionality of medical features relative to the limited data available. We propose the Personalized Aligner Federated Pretrained (PAFP) framework, a personalized FL approach built on a pre-trained backbone. PAFP inserts client-specific Feature Alignment modules before the fine-tuned blocks to realign hidden feature distributions toward a server-aggregated target. On the FedISIC benchmark, PAFP attains the highest unweighted balanced accuracy while reducing inter-client performance variance roughly six-fold relative to the strongest baseline, yielding accuracy and cross-client equity jointly. Our code is publicly available at https://github.com/Alexandre-Leclercq/PAFP.