Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 18, 2026

Using Looming Visual Stimuli to Evaluate Mouse Vision
05:07

Using Looming Visual Stimuli to Evaluate Mouse Vision

Published on: June 13, 2019

MedQM: Medical Blind Image Quality Assessment for MLLM Vision.

Renwei Yang, Tianyi Chen, Dapeng Oliver Wu

    IEEE Journal of Biomedical and Health Informatics
    |June 16, 2026
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    You might also read

    Related Articles

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

    Sort by
    Same author

    Electrostatic Potential Tuning by Low-Volatility Halogenated Additive: Boosting PTQ10-Based Binary OPV to Near 20% Efficiency with High Scalability.

    Advanced materials (Deerfield Beach, Fla.)·2026
    Same author

    A Wearable Microneedle-Based Electrochemical Aptamer Sensor: Enabling Real-Time Dynamic NT-proBNP Monitoring for Enhanced Heart Failure Management.

    ACS sensors·2026
    Same author

    Pore size engineering in covalent organic frameworks for high-performance anion exchange membranes.

    Nanoscale·2026
    Same author

    Deep learning-driven decoding of ubiquitination: from regulatory mechanisms to targeted protein degradation.

    Biology direct·2026
    Same author

    H<sub>2</sub>O<sub>2</sub>-GA<sub>3</sub>-Na<sub>2</sub>WO<sub>4</sub> Synergistically Promotes Germination of Immature Winter Wheat Grains for Speed Breeding.

    Plants (Basel, Switzerland)·2026
    Same author

    <i>Lactiplantibacillus plantarum</i> CQPC03 Reduces Carrageenan-Induced Thrombosis in Mice via Controlling the Flora in the Gut.

    Food science & nutrition·2026

    We developed MedQM, the first blind image quality assessment (BIQA) framework for medical multi-modal large language models (MLLMs). MedQM improves diagnostic accuracy by evaluating image quality from the MLLM

    Area of Science:

    • Medical Imaging Analysis
    • Artificial Intelligence in Healthcare
    • Deep Learning for Diagnostics

    Background:

    • Multi-modal large language models (MLLMs) are increasingly used in medical applications.
    • Image quality is crucial for diagnostic performance, yet traditional BIQA methods based on human perception are inadequate for MLLMs.
    • Suboptimal image quality assessment by MLLMs can lead to diagnostic errors.

    Purpose of the Study:

    • To introduce MedQM, the first blind image quality assessment (BIQA) framework specifically designed for medical MLLMs.
    • To address the discrepancy between human visual perception and MLLM image understanding in medical contexts.
    • To enhance the reliability and accuracy of MLLM-based medical image diagnostics.

    Main Methods:

    • Developed MedQM-I, a novel BIQA model utilizing Medical Textual Priors and Implicit Feature Queries.

    More Related Videos

    Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients
    07:06

    Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients

    Published on: March 29, 2022

    Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
    07:45

    Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition

    Published on: July 21, 2020

    Related Experiment Videos

    Last Updated: Jun 18, 2026

    Using Looming Visual Stimuli to Evaluate Mouse Vision
    05:07

    Using Looming Visual Stimuli to Evaluate Mouse Vision

    Published on: June 13, 2019

    Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients
    07:06

    Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients

    Published on: March 29, 2022

    Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
    07:45

    Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition

    Published on: July 21, 2020

  • Employed a gated Mixture-of-Experts architecture within MedQM-I for adaptive and robust image quality scoring.
  • Introduced MedQM-D, an automatic approach to compute MLLM vision-oriented image diagnostic loss.
  • Implemented a sigmoid-based quality mapping to convert diagnostic loss into interpretable quality scores.

    • Main Results:

    • MedQM significantly outperforms traditional BIQA methods in assessing medical image quality for MLLMs.
    • The proposed MedQM-I model effectively guides attention to diagnostically relevant image regions and features.
    • MedQM-D provides an efficient and accurate method for labeling image quality scores, improving training stability and interpretability.
    • Experimental validation confirms the superior effectiveness of the MedQM framework.

    Conclusions:

    • MedQM represents a significant advancement in BIQA for medical MLLMs.
    • The framework enhances MLLM diagnostic utility by providing accurate and relevant image quality assessments.
    • MedQM has the potential to reduce misdiagnosis rates and improve patient outcomes in AI-assisted medical diagnostics.