Continual Learning in Medical Image Analysis: Appropriate Use and Lifecycle Governance.
Abstract
AI models for medical imaging are typically trained once and deployed unchanged; however, as scanners, protocols, and patient populations continue to evolve, model performance may degrade over time. Continual learning offers a way to incorporate newly acquired data while limiting the loss of previously learned diagnostic knowledge; therefore, it is particularly suited to clinical environments in which historical images cannot be freely retained or reused. This review outlines the continual learning scenarios and methodological families most relevant to radiology, examines when continual updating is preferable to periodic retraining or simple recalibration, and summarizes the applications in classification, segmentation, detection, and multi-joint radiographic grading. It then addresses the clinical constraints that determine feasibility: privacy and regulatory limits on data retention, delayed and noisy labels, class imbalance, and evaluation requirements beyond the average accuracy. Therefore, continual learning should be implemented as a governed, auditable maintenance process rather than as a single algorithmic choice.
Related
- A picture is worth 1000 pathological learning points: Creating visual pathology teaching materials appropriate for preclerkship medical student learning
- Preventive governance for patient‐centred medical AI
- Introducing Infrastructure as Code for deployment and governance of medical AI: A case study
- Shared Governance for Effective Clinical Education in Medical Radiation Science: A Narrative Review
- Regulation of medical artificial intelligence in public health: governance lessons and implications for China
BibTeX
@article{yoo2026continual,
title = {{Continual Learning in Medical Image Analysis: Appropriate Use and Lifecycle Governance.}},
author = {Min Kyoon Yoo and Yu Rang Park},
journal = {Journal of the Korean Society of Radiology},
year = {2026},
month = sep,
doi = {10.3348/jksr.2026.0135},
url = {https://europepmc.org/article/MED/42903021},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Continual Learning in Medical Image Analysis: Appropriate Use and Lifecycle Governance.
AU - Min Kyoon Yoo
AU - Yu Rang Park
PY - 2026
DA - 2026/09/29
JO - Journal of the Korean Society of Radiology
DO - 10.3348/jksr.2026.0135
UR - https://europepmc.org/article/MED/42903021
AB - AI models for medical imaging are typically trained once and deployed unchanged; however, as scanners, protocols, and patient populations continue to evolve, model performance may degrade over time. Continual learning offers a way to incorporate newly acquired data while limiting the loss of previously learned diagnostic knowledge; therefore, it is particularly suited to clinical environments in which historical images cannot be freely retained or reused. This review outlines the continual learning scenarios and methodological families most relevant to radiology, examines when continual updating is preferable to periodic retraining or simple recalibration, and summarizes the applications in classification, segmentation, detection, and multi-joint radiographic grading. It then addresses the clinical constraints that determine feasibility: privacy and regulatory limits on data retention, delayed and noisy labels, class imbalance, and evaluation requirements beyond the average accuracy. Therefore, continual learning should be implemented as a governed, auditable maintenance process rather than as a single algorithmic choice.
SN - 2951-0805
ER -