Neural collaborative filtering for unsupervised mitral valve segmentation in echocardiography

TL;DR

This paper introduces an unsupervised neural collaborative filtering approach for mitral valve segmentation in echocardiography videos, effectively handling low-quality and noisy data to support clinical tasks.

Contribution

It presents a novel unsupervised method using neural network collaborative filtering for mitral valve segmentation, reducing the need for extensive annotations and improving performance on challenging videos.

Findings

Outperforms existing unsupervised methods on low-quality videos

Achieves better results than supervised methods with sparse annotations

Effective across diverse mitral valve disease cases

Abstract

The segmentation of the mitral valve annulus and leaflets specifies a crucial first step to establish a machine learning pipeline that can support physicians in performing multiple tasks, e.g.\ diagnosis of mitral valve diseases, surgical planning, and intraoperative procedures. Current methods for mitral valve segmentation on 2D echocardiography videos require extensive interaction with annotators and perform poorly on low-quality and noisy videos. We propose an automated and unsupervised method for the mitral valve segmentation based on a low dimensional embedding of the echocardiography videos using neural network collaborative filtering. The method is evaluated in a collection of echocardiography videos of patients with a variety of mitral valve diseases, and additionally on an independent test cohort. It outperforms state-of-the-art \emph{unsupervised} and \emph{supervised} methods on low-quality videos or in the case of sparse annotation.