Analysis of Clinical featuRes and Echocardiographic Characteristics for Diagnosis of Infiltrative cardiomyopaThy (ACREDIT): Retrospective Multi-center Observational Study

• Conditions: Infiltrative Cardiomyopathy

• Registration Number: NCT05108168
• Lead Sponsor: Yonsei University

Brief Summary

This study sought to develop an algorithm by collecting echocardiographic image information and related clinical information capable of quantitatively evaluating changes of the myocardium through machine learning. Moreover, the researchers investigate the usefulness of an algorithm for early diagnosis and differential diagnosis of infiltrative cardiomyopathy.

Detailed Description

1. Study Design: Multicenter Retrospective Observational Study

2. Study method: If the above selection criteria are met, the index visit echocardiographic images which were performed immediately before or closest to the time of hospitalization for final diagnosis, echocardiographic images of the pre-visit and post-visit from the final diagnosis, and clinical information will be obtained. Chest X-ray, electrocardiogram, and echocardiography images are extracted in raw DICOM format and then analyzed in the core lab (Severance hospital). The characteristics of patients with infiltrative cardiomyopathy are identified through the collection of relevant clinical information, and a method for non-invasive early diagnosis and differential diagnosis of infiltrative cardiomyopathy is developed.

3. Quantative analysis of echocardiographic images using Radiomics

* Radiomics is a method of extracting a large number of quantitative image features (300-500 features such as shape, entropy, volume, etc.) from non-invasive medical images (CT, MRI, etc.) and statistically analyzing the features. Its value has been demonstrated through the studies for prediction of breast cancer recurrence and lesion classification.

* Using the open source platform PyRadiomics19, we extract the radiomic characteristics for brightness (Energy, Entropy, Mean, Median, etc.) and texture (Gray Level Co-occurrence Matrix Contrast, Difference Variance, Maximal Correlation Coefficient, etc.) from the set region of interest.

* The differences between infiltrative cardiomyopathy and normal control are identified using clinical information and radiomics features extracted from echocardiography at the time of the diagnosis visit. The algorithms to distinguish the disease will be developed using machine learning methods such as support vector machine classifier.

Study Timeline

• Study Start: 2021-01-04
• Status Verified: 2021-10
• Study First Submitted: 2021-10-24
• Study First Submitted QC: 2021-10-24
• Last Update Submitted: 2021-10-24
• Study First Posted: 2021-11-04
• Last Update Posted: 2021-11-04
• Primary Completion: 2021-12-16
• Study Completion: 2021-12-31

Recruitment & Eligibility

• Status: UNKNOWN
• Sex: All
• Target Recruitment: 500

Inclusion Criteria

Not provided

Exclusion Criteria

Not provided

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Area under the curve of the receiver operation characteristics	until June 30, 2022
Sensitivity	until June 30, 2022	Sensitivity (True Positive Rate) refers to the proportion of those who have the infiltrative cardiomyopathy that received a positive result on the diagnostic algorythm by machine learning.
Specificity	until June 30, 2022	Specificity refers to the proportion of those who do not have the infiltrative cardiomyopathy that received a negative result on the test.

Trial Locations

Locations (1)

Yonsei University Health System, Severance Hospital, Division of Cardiology; 🇰🇷 Seoul, Korea, Republic of