Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

ReclAIm: A Multiagent Framework for Monitoring and Correcting Performance Decline in Medical Imaging AI.

Eleftherios Tzanis^1,2, Michail E Klontzas^1,3,2

¹Artificial Intelligence and Translational Imaging (ATI) Lab, Department of Radiology, School of Medicine, University of Crete, Voutes, 71003 Heraklion, Crete, Greece.

Radiology. Artificial Intelligence

|June 3, 2026

Summary

Related Concept Videos

Issues And Trends In Healthcare Delivery System

Issues And Trends In Healthcare Delivery System

The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...

Radiological Investigation II: MRI and Ventilation Perfusion Scan

Radiological Investigation II: MRI and Ventilation Perfusion Scan

Description
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Phalangeal Bone Mineral Density Mapping Using Quantitative CT: Implications for Hand Surgery Fixation Planning.

Diagnostics (Basel, Switzerland)·2026

Same author

OpenRad: a curated repository of open-access AI models for radiology.

European radiology·2026

Same author

Artificial intelligence-driven radiomics: developing valuable radiomics signatures with the use of artificial intelligence.

BJR artificial intelligence·2026

Same author

Reporting checklist for foundation and large language models in medical research (REFINE): an international consensus guideline.

Diagnostic and interventional radiology (Ankara, Turkey)·2026

Same author

Strategic Integration of Artificial Intelligence in Musculoskeletal Interventions: Scope, Success, and Pursuit of Personalized Precision Care Models.

Seminars in musculoskeletal radiology·2026

Same author

Rethinking Privacy in Medical Imaging AI: From Metadata and Pixel-Level Identification Risks to Federated Learning and Synthetic Data Challenges.

Radiology. Artificial intelligence·2025

This summary is machine-generated.

A new multiagent framework, ReclAIm, automates medical image classification model monitoring and correction. It detects performance declines and uses fine-tuning to restore accuracy, improving model reliability.

Area of Science:

Artificial Intelligence in Medical Imaging
Machine Learning Model Monitoring
Deep Learning for Healthcare

Background:

Medical image classification models require continuous monitoring to ensure sustained performance.
Performance degradation in AI models can impact diagnostic accuracy and clinical utility.
Automated systems are needed to manage and maintain these models effectively.

Purpose of the Study:

To develop and evaluate ReclAIm, a multiagent framework for automated monitoring, detection, and correction of performance decline in medical image classification models.
To leverage natural language interaction for system operation and accessibility.
To address catastrophic forgetting during model fine-tuning.

Main Methods:

ReclAIm, a large language model-based multiagent system, was developed.

Keywords:

AI Agents CT Continual Learning Conventional Radiography Convolutional Neural Network Diagnosis Fine-tuning Large Language Model MR Imaging Medical Image Classification Monitoring Performance Postmarket Surveillance

Related Experiment Videos

A master agent coordinated task-specific agents for performance evaluation and fine-tuning triggers.

Fine-tuning incorporated data augmentation, class imbalance handling, and parameter-anchoring regularization.

The system was tested on brain MRI, chest CT, and chest radiography datasets.

Main Results:

ReclAIm successfully managed training, evaluation, and monitoring across all tested datasets.
Performance discrepancies were detected in 8 of 18 models, triggering fine-tuning.
Fine-tuning restored performance metrics to within ± 2% of baseline values, even after declines up to 40.6%.

Conclusions:

ReclAIm offers a prototype for automated monitoring and targeted fine-tuning of medical image classification models.
The natural language interface enhances accessibility for research and clinical applications.
This framework supports the maintenance and reliability of AI in medical imaging.