Clinical Scorecard: Impact of Sequential Institution Integration on Automated Lesion Detection Performance in Federated Learning Systems
At a Glance
Category
Detail
Condition
Automated Lesion Detection
Key Mechanisms
Federated Learning, Machine Learning Algorithms
Target Population
Patients undergoing MRI for cerebral aneurysms and brain metastases
Care Setting
Clinical practice utilizing computer-aided detection software
Key Highlights
Federated learning allows institutions to train models while retaining local data.
Sequential addition of institutions may enhance CAD software performance.
Two CAD software types were evaluated: for cerebral aneurysms and brain metastases.
Different training approaches were assessed: from-scratch, fine-tuning, and full fine-tuning.
Centralized models used pooled training data from participating clients.
Guideline-Based Recommendations
Diagnosis
Incorporate data from multiple institutions to improve model robustness.
Management
Utilize federated learning frameworks for training CAD systems.
Monitoring & Follow-up
Evaluate performance metrics as new institutions are added.
Risks
Performance may deteriorate if training data do not match clinical data.
Patient & Prescribing Data
Patients with cerebral aneurysms and brain metastases undergoing MRI.
CAD software can assist in the detection of lesions in MRI images.
Clinical Best Practices
Use federated learning to enhance the generalizability of CAD systems.
Regularly update models with new data from additional institutions.
Related Resources & Content
Suraj Rajendran et al., ASCO Publications, 2021 -- Cloud-Based Federated Learning Implementation Across Medical Centers
Collaborative 3D segmentation of multiple organs using both labeled and unlabeled datasets, Springer, 2024
Is Federated Learning the Answer to Patient Privacy in AI?, ASCO AI in Oncology, 2026
Transcript for CDRH Webinar Final Guidance: Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions, FDA, 2024
Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI), ScienceDirect, 2024
Federated Learning with Contrastive Prototypes for Detecting Fetal Standard Planes Amid Noisy Labels
Transcript for CDRH Webinar Final Guidance: Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions
Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI): a randomised, controlled, parallel-group, non-inferiority, single-blinded, screening accuracy study - ScienceDirect
Advancements and challenges of federated learning in medical imaging: a systematic literature review | Artificial Intelligence Review | Springer Nature Link
Optical analysis of resection-margin tissue identified tumor infiltration as the strongest contributor to predictions of where glioblastoma later recurred.
