Literature DB >> 34897594

Automatic morphological classification of mitral valve diseases in echocardiographic images based on explainable deep learning methods.

Majid Vafaeezadeh1, Hamid Behnam2, Ali Hosseinsabet3, Parisa Gifani4.   

Abstract

PURPOSE: Carpentier's functional classification is a guide to explain the types of mitral valve regurgitation based on morphological features. There are four types of pathological morphologies, regardless of the presence or absence of mitral regurgitation: Type I, normal; Type II, mitral valve prolapse; Type IIIa, mitral valve stenosis; and Type IIIb, restricted mitral leaflet motion. The aim of this study was to automatically classify mitral valves using echocardiographic images.
METHODS: In our procedure, after the classification of apical 4-chamber (A4C) and parasternal long-axis (PLA) views, we extracted the systolic/diastolic phase of the cardiac cycle by calculating the left ventricular area. Six typical pre-trained models were fine-tuned with a 4-class model for the PLA and a 3-class model for the A4C views. As an additional contribution, to provide explainability, we applied the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm to visualize areas of echocardiographic images where the different models generated a prediction.
RESULTS: This approach conferred a proper understanding of where various networks "look" into echocardiographic images to predict the four types of pathological mitral valve morphologies. Considering the accuracy metric and Grad-CAM maps and by applying the Inception-ResNet-v2 architecture to classify Type II in the PLA view and ResNeXt50 architecture to classify the other three classes in the A4C view, we achieved an 80% rate of model accuracy in the test data set.
CONCLUSIONS: We suggest an explainable, fully automated, and rule-based procedure to classify the four types of mitral valve morphologies based on Carpentier's functional classification using deep learning on transthoracic echocardiographic images. Our study results infer the feasibility of the use of deep learning models to prepare quick and precise assessments of mitral valve morphologies in echocardiograms. According to our knowledge, our study is the first one that provides a public data set regarding the Carpentier classification of MV pathologies.
© 2021. CARS.

Entities:  

Keywords:  Carpentier classification; DCNN; Echocardiography; Mitral valve morphology

Mesh:

Year:  2021        PMID: 34897594     DOI: 10.1007/s11548-021-02542-7

Source DB:  PubMed          Journal:  Int J Comput Assist Radiol Surg        ISSN: 1861-6410            Impact factor:   2.924


  7 in total

1.  Deep learning in medical image analysis: A third eye for doctors.

Authors:  A Fourcade; R H Khonsari
Journal:  J Stomatol Oral Maxillofac Surg       Date:  2019-06-26       Impact factor: 1.569

2.  Real-Time Standard View Classification in Transthoracic Echocardiography Using Convolutional Neural Networks.

Authors:  Andreas Østvik; Erik Smistad; Svein Arne Aase; Bjørn Olav Haugen; Lasse Lovstakken
Journal:  Ultrasound Med Biol       Date:  2018-11-20       Impact factor: 2.998

3.  Automatic biplane left ventricular ejection fraction estimation with mobile point-of-care ultrasound using multi-task learning and adversarial training.

Authors:  Mohammad H Jafari; Hany Girgis; Nathan Van Woudenberg; Zhibin Liao; Robert Rohling; Ken Gin; Purang Abolmaesumi; Terasa Tsang
Journal:  Int J Comput Assist Radiol Surg       Date:  2019-04-02       Impact factor: 2.924

4.  Cardiac valve surgery--the "French correction".

Authors:  A Carpentier
Journal:  J Thorac Cardiovasc Surg       Date:  1983-09       Impact factor: 5.209

5.  2020 Focused Update of the 2017 ACC Expert Consensus Decision Pathway on the Management of Mitral Regurgitation: A Report of the American College of Cardiology Solution Set Oversight Committee.

Authors:  Robert O Bonow; Patrick T O'Gara; David H Adams; Vinay Badhwar; Joseph E Bavaria; Sammy Elmariah; Judy W Hung; JoAnn Lindenfeld; Alanna A Morris; Ruby Satpathy; Brian Whisenant; Y Joseph Woo
Journal:  J Am Coll Cardiol       Date:  2020-02-14       Impact factor: 24.094

6.  Real-Time Automatic Ejection Fraction and Foreshortening Detection Using Deep Learning.

Authors:  Erik Smistad; Andreas Ostvik; Ivar Mjaland Salte; Daniela Melichova; Thuy Mi Nguyen; Kristina Haugaa; Harald Brunvand; Thor Edvardsen; Sarah Leclerc; Olivier Bernard; Bjornar Grenne; Lasse Lovstakken
Journal:  IEEE Trans Ultrason Ferroelectr Freq Control       Date:  2020-11-24       Impact factor: 2.725

7.  Measuring Left Ventricular Volumes in Two-Dimensional Echocardiography Image Sequence Using Level-set Method for Automatic Detection of End-Diastole and End-systole Frames.

Authors:  Saeed Darvishi; Hamid Behnam; Majid Pouladian; Niloufar Samiei
Journal:  Res Cardiovasc Med       Date:  2013-02-24
  7 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.