Federated learning in cancer imaging: a comprehensive review and future outlook
- 1The Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, China
- 2The Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD, United States
Abstract
The rapid development of artificial intelligence has promoted the advancement of data-driven AI medical models, bringing transformative changes to the field of traditional cancer imaging by significantly improving diagnostic accuracy. However, despite these remarkable achievements, a fundamental challenge remains: data privacy. The sensitive nature of medical data and stringent privacy regulations often limit data sharing and collaboration between institutions, hindering the development of robust and generalizable AI models. Federated learning has emerged as a promising solution to address this challenge. By enabling multiple institutions to collaboratively train artificial intelligence models without sharing their local datasets, federated learning reduces the privacy leakage risks associated with conventional centralized AI training while utilizing the collective power of diverse medical data. From 2019 to 2026, extensive research has explored the applications of federated learning in cancer imaging, demonstrating its potential to advance the field while preserving patient confidentiality. In this review, we provide a comprehensive overview of the progress in applying federated learning to cancer imaging, highlighting key studies and their contributions. We also discuss the unique challenges facing federated learning, such as data security, data heterogeneity, and communication overhead. Finally, we explore some potential future research directions. This review aims to provide insights into the evolving role of federated learning in cancer imaging.
Related
- Artificial Intelligence in Oncologic Imaging: Deep Learning, Radiomics, and Clinical Integration for Precision Cancer Diagnosis
- Novel liquid biopsy based on comprehensive small RNA profiling for screening of cancer in dogs utilizing machine learning.
- DEEP REINFORCEMENT LEARNING IN ONCOLOGY: ADVANCES IN CANCER IMAGING, RADIOTHERAPY, AND PERSONALIZED TREATMENT
- CT imaging-based radiomics and deep learning models for predicting chemotherapy response in advanced pancreatic cancer
- Evaluation of accelerated whole-body diffusion weighted imaging with deep learning reconstruction in patients with metastatic prostate cancer: assessment of image quality and ADC estimates.
BibTeX
@article{shi2026federated,
title = {{Federated learning in cancer imaging: a comprehensive review and future outlook}},
author = {Yiming Shi and Wen Li and Jing Cai},
journal = {Frontiers in Radiology},
year = {2026},
month = oct,
doi = {10.3389/fradi.2026.1976540},
url = {https://doi.org/10.3389/fradi.2026.1976540},
note = {CC BY},
}
RIS
TY - JOUR
TI - Federated learning in cancer imaging: a comprehensive review and future outlook
AU - Yiming Shi
AU - Wen Li
AU - Jing Cai
PY - 2026
DA - 2026/10/28
JO - Frontiers in Radiology
DO - 10.3389/fradi.2026.1976540
UR - https://doi.org/10.3389/fradi.2026.1976540
AB - The rapid development of artificial intelligence has promoted the advancement of data-driven AI medical models, bringing transformative changes to the field of traditional cancer imaging by significantly improving diagnostic accuracy. However, despite these remarkable achievements, a fundamental challenge remains: data privacy. The sensitive nature of medical data and stringent privacy regulations often limit data sharing and collaboration between institutions, hindering the development of robust and generalizable AI models. Federated learning has emerged as a promising solution to address this challenge. By enabling multiple institutions to collaboratively train artificial intelligence models without sharing their local datasets, federated learning reduces the privacy leakage risks associated with conventional centralized AI training while utilizing the collective power of diverse medical data. From 2019 to 2026, extensive research has explored the applications of federated learning in cancer imaging, demonstrating its potential to advance the field while preserving patient confidentiality. In this review, we provide a comprehensive overview of the progress in applying federated learning to cancer imaging, highlighting key studies and their contributions. We also discuss the unique challenges facing federated learning, such as data security, data heterogeneity, and communication overhead. Finally, we explore some potential future research directions. This review aims to provide insights into the evolving role of federated learning in cancer imaging.
SN - 2673-8740
ER -