Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 16, 2026Cardiovascular UltrasoundOpen Access

Quantification of mitral regurgitation: from traditional methods to artificial intelligence

View Full Paper
Ask AI
Bookmark
Share

Key result

AI-driven 3D Auto CFQ automates MR quantification to overcome geometric and temporal limitations of PISA.

Why the study?

Standard echocardiographic methods for mitral regurgitation rely on geometric assumptions and single-frame analysis that are inaccurate in eccentric, multiple, or non-holosystolic jets.

Comparison

Automated machine learning and AI quantification vs standard echocardiographic methods

Design

Review

Authors

KCKenneth ChoElectrophysiologyJSJimmy SuPhilips (United States)OBOdile BonnefousPhilips (France)

Discussion

Loading...

Member takes

Implication

May improve MR quantification accuracy in complex jets; leaves open prospective validation before clinical adoption.

Key Points

  • This research aims to compare traditional methods of quantifying mitral regurgitation with new AI techniques to improve assessment accuracy.
  • Utilized echocardiography to assess mitral regurgitation severity.
  • Implemented machine learning algorithms for automated quantification and analysis of regurgitant flow.
  • Analyzed AI performance against traditional methods and cardiac magnetic resonance imaging.
  • AI methods demonstrated favorable agreement with cardiac magnetic resonance imaging quantifications.
  • Enhanced accuracy in identifying and grading mitral regurgitation from echocardiographic clips.
  • AI techniques offer potential for effective screening in low-resource environments.

Structured PICO

I
Intervention
Artificial intelligence and machine learning frameworks (e.g., 3D Auto CFQ) for automated quantification of mitral regurgitation
C
Comparator
Traditional echocardiographic methods (e.g., 2D PISA)

AI-driven automated 3D quantification of mitral regurgitation overcomes the geometric and temporal limitations of traditional 2D PISA, offering a more accurate and reproducible approach for clinical practice.

Limitations

  • Current 3D MR flow quantification software yields smaller regurgitant volumes compared to 2D PISA, requiring future studies to establish optimal cutoff values for severe MR.
  • Some deep learning applications have only been validated in retrospective cohorts and are not yet widely or commercially available.
  • Current algorithms tend to yield smaller regurgitant volumes than conventional PISA-based approaches
  • Further studies are required to validate these measurements

Cite This Study

Cho et al. (2026) conducted a review in Mitral regurgitation. Artificial intelligence and 3D Auto Color Flow Quantification vs. Traditional methods (2D PISA) was evaluated. Artificial intelligence techniques, particularly 3D Auto CFQ, enable automated quantification of mitral regurgitation, overcoming the geometric and temporal limitations of traditional PISA methods.

synapsesocial.com/papers/6a31570faf7cf7f8256b3d02https://doi.org/10.1186/s12947-026-00374-6

Topics

Mitral valve interventionEchocardiographyArtificial intelligence in cardiology
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Transcatheter Mitral-Valve Repair in Patients with Heart Failure2018 · 2,999 citations
  2. 2High-Throughput Deep Learning Detection of Mitral Regurgitation2024 · 64 citations
  3. 3Self-supervised learning assisted diagnosis for mitral regurgitation severity classification based on color Doppler echocardiography2021 · 17 citations
  4. 4EasyPISA: Automatic Integrated PISA Measurements of Mitral Regurgitation From 2-D Color-Doppler Using Deep Learning2024 · 9 citations
  5. 5Quantification of Chronic Functional Mitral Regurgitation by Automated 3-Dimensional Peak and Integrated Proximal Isovelocity Surface Area and Stroke Volume Techniques Using Real-Time 3-Dimensional Volume Color Doppler Echocardiography2012 · 137 citations