Linear Combination Of Mitral Valve And Machine Learning To Predict Mitral Annuloplasty Band Size in Endoscopic Mitral Valve Surgery

MO058 

Abstract Submission 

Rafik Margaryan (1), Nicola Martini (2), Massimiliano Mariani (3), Marco Solinas (4)

(1) Ospedale Del Cuore, Massa, Tuscany, (2) Ospedale Del Cuore Fondazione Toscana G Monasterio, Massa, NA, (3) Ospedale Del Cuore Fondazione G Monasterio, Massa, NA, (4) Fondazione Toscana Gabriele Monasterio, Massa

Rafik Margaryan    -  Contact Me
Ospedale Del Cuore

Nicola Martini    -  Contact Me
Ospedale Del Cuore Fondazione Toscana G Monasterio

Massimiliano Mariani    -  Contact Me
Ospedale Del Cuore Fondazione G Monasterio

Marco Solinas    -  Contact Me
Fondazione Toscana Gabriele Monasterio

Rafik Margaryan    -  Contact Me
Ospedale Del Cuore

Background: Endoscopic mitral valve surgery is becoming standard of care. However it has steep learning curve and annular sizing is difficult with conventional sizers.

Methods: A retrospective analysis was performed on 104 patients who underwent mitral repair at one institution from July 2015 to September 2022. 93 of them had CT scans suitable for mitral valvular apparatus measurements (intertrigonal distance, anterior leaflet max length, etc). A machine learning pipeline was created in order to predict annuloplasty band size (from 26 to 38 mm). Mitral measurements were done in MPR modality.

Results: A final model (best model among the trained) was able to predict the annular size with 72% accuracy. A linear combination (PCA fo two components) of intertrigonal distance and anterior leaflet lengths was the main linear combination in the model

Conclusions: Linear combination of mitral valve measurements can predict annular size in clinical setting. Its value is highly valuable for young and inexperienced groups.

Minimally Invasive & Robotic Mitral Valve Repair