loading page

Developing an Echocardiography-Based, Automatic Deep Learning Framework for the Differentiation of Increased Left Ventricular Wall Thickness Etiologies
  • +24
  • Imon Banerjee,
  • James Li,
  • Chieh-Ju Chao, ,
  • Jiwoong Jason Jeong,
  • Amith Seri R,
  • Timothy Barry,
  • Hana Neuman,
  • Megan Campany,
  • Merna Abdou,
  • Michael O’Shea,
  • Sean Smith,
  • Bishoy Abraham,
  • Seyedeh Maryam Hosseini,
  • Yuxiang Wang,
  • Steven Lester,
  • Said Alsidawi,
  • susan Wilansky,
  • Eric Steidley,
  • Julie Rosenthal,
  • Chadi Ayoub,
  • Christopher Appleton,
  • Win-Kuang Shen,
  • Martha Grogan,
  • Garvan Kane,
  • Jae Oh,
  • Bhavik N. Patel,
  • Reza Arsanjani
Imon Banerjee
Mayo Clinic Research in Arizona

Corresponding Author:banerjee.imon@mayo.edu

Author Profile
James Li
Mayo Clinic Research in Arizona
Author Profile
Chieh-Ju Chao,
Mayo Clinic Minnesota
Author Profile
Jiwoong Jason Jeong
Mayo Clinic Research in Arizona
Author Profile
Amith Seri R
Mayo Clinic Research in Arizona
Author Profile
Timothy Barry
Mayo Clinic Research in Arizona
Author Profile
Hana Neuman
Mayo Clinic Research in Arizona
Author Profile
Megan Campany
Mayo Clinic Research in Arizona
Author Profile
Merna Abdou
Mayo Clinic Research in Arizona
Author Profile
Michael O’Shea
Mayo Clinic Research in Arizona
Author Profile
Sean Smith
Mayo Clinic Research in Arizona
Author Profile
Bishoy Abraham
Mayo Clinic Research in Arizona
Author Profile
Seyedeh Maryam Hosseini
Mayo Clinic Research in Arizona
Author Profile
Yuxiang Wang
Mayo Clinic Research in Arizona
Author Profile
Steven Lester
Mayo Clinic Research in Arizona
Author Profile
Said Alsidawi
Mayo Clinic Research in Arizona
Author Profile
susan Wilansky
Mayo Clinic Research in Arizona
Author Profile
Eric Steidley
Mayo Clinic Research in Arizona
Author Profile
Julie Rosenthal
Mayo Clinic Research in Arizona
Author Profile
Chadi Ayoub
Mayo Clinic Research in Arizona
Author Profile
Christopher Appleton
Mayo Clinic Research in Arizona
Author Profile
Win-Kuang Shen
Mayo Clinic Research in Arizona
Author Profile
Martha Grogan
Mayo Clinic Minnesota
Author Profile
Garvan Kane
Mayo Clinic Minnesota
Author Profile
Jae Oh
Mayo Clinic Minnesota
Author Profile
Bhavik N. Patel
Mayo Clinic Research in Arizona
Author Profile
Reza Arsanjani
Mayo Clinic Research in Arizona
Author Profile

Abstract

Aims Increased LV wall thickness is frequently encountered in transthoracic echocardiography (TTE). While accurate and early diagnosis is clinically important, given the differences in available therapeutic options and prognosis, an extensive workup is often required to establish the diagnosis. We propose the first echo-based, automated deep learning model with a fusion architecture to facilitate the evaluation and diagnosis of increased left ventricular (LV) wall thickness. Methods and Results Patients with an established diagnosis of increased LV wall thickness (hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), and hypertensive heart disease (HTN)/others) between 1/2015 to 11/2019 at Mayo Clinic Arizona were identified. The cohort was divided into 80%/10%/10% for training, validation, and testing sets, respectively. Six baseline TTE views were used to optimize a pre-trained InceptionResnetV2 model. Each model output was used to train a meta-learner under a fusion architecture. Model performance was assessed by multiclass area under the receiver operating characteristic curve (AUROC). A total of 586 patients were used for the final analysis (194 HCM, 201 CA, and 191 HTN/others). The mean age was 55.0 years, and 57.8% were male. Among the individual view-dependent models, the apical 4 chamber model had the best performance (AUROC: HCM: 0.94, CA: 0.73, and HTN/other: 0.87). The final fusion model outperformed all the view-dependent models (AUROC: CA: 0.90, HCM: 0.93, and HTN/other: 0.92). Conclusion The echo-based InceptionResnetV2 fusion model can accurately classify the main etiologies of increased LV wall thickness and can facilitate the process of diagnosis and workup.