A Multi-center Study on Artificial Intelligence-Based Quantitative Evaluation of Echocardiography

Recruiting

• Conditions: Artificial Intelligence (AI); Artificial Intelligence (AI) in Diagnosis; Cardiovascular Diseases (CVD); Echocardiography

• Registration Number: NCT07133516
• Lead Sponsor: First Hospital of China Medical University

Brief Summary

This project aims to collaborate with multiple medical institutions to verify the accuracy, stability, and clinical application value of AI algorithms in echocardiographic quantitative measurement through multi-center clinical research. Specific objectives include:

1. Compare the automatic measurement results of AI with the manual measurement data from physicians of different levels, and analyze the measurement deviation and consistency of AI in key parameters such as intracardiac diameter, volume, and function.

2. Investigate whether AI-assisted measurement can significantly reduce echocardiogram analysis time and optimize clinical workflows. Through multi-center data validation, establish a standardized reference system for AI ultrasound measurement, promote the promotion and application of AI technology in medical institutions at all levels, and reduce diagnostic differences between different hospitals and physicians.

3. Exploring the application of AI in special cases: Assessing the measurement stability of AI algorithms in complex cases (such as cardiomyopathy, valvular disease, coronary heart disease, etc.), and optimizing AI models to meet broader clinical needs.

Detailed Description

Cardiovascular disease is a major threat to the health of Chinese residents, and echocardiography, as its core diagnostic tool, directly affects clinical decision-making in terms of measurement accuracy and efficiency. However, traditional ultrasound evaluation heavily relies on physician experience, with pain points such as strong subjectivity, time-consuming measurements, and uneven levels of primary diagnosis. There is an urgent need for technological innovation to improve diagnostic standardization. In recent years, artificial intelligence (AI) technology has shown great potential in the field of medical image analysis, which can achieve automated quantitative measurement of cardiac chamber structure and function. However, existing AI models generally have problems such as insufficient multi center validation and limited adaptability to complex cases, which restrict their clinical translation and application.

To overcome these bottlenecks, this project collaborates with multiple medical institutions to conduct clinical research, systematically evaluating the measurement differences between AI algorithms and physicians of different levels, and assessing the accuracy and stability of AI algorithms. The research will focus on verifying the value of AI technology in improving diagnostic consistency, optimizing workflows, and exploring its potential applications in complex cardiovascular diseases. By establishing a standardized evaluation system, this project aims to promote the standardized application of AI ultrasound technology, ultimately achieving the goal of improving diagnosis and treatment efficiency, promoting the sinking of high-quality medical resources, and helping to improve the overall level of cardiovascular disease prevention and treatment.

Study Timeline

• Study Start: 2025-07-22
• Study First Submitted: 2025-07-30
• Status Verified: 2025-08
• Study First Submitted QC: 2025-08-18
• Last Update Submitted: 2025-08-18
• Study First Posted: 2025-08-21
• Last Update Posted: 2025-08-21
• Primary Completion: 2026-06-30
• Study Completion: 2026-07-31

Recruitment & Eligibility

• Status: RECRUITING
• Sex: All
• Target Recruitment: 1600

Inclusion Criteria

1. Age ≥18 - 80 years;

2. Types of diseases (8 in total, 200 cases each):

   1. Normal heart
   2. Coronary heart disease (with segmental thinning and abnormal movement)
   3. Valve disease (valve stenosis or reflux)
   4. Hypertensive heart disease
   5. Atrial fibrillation
   6. Heart failure
   7. Dilated cardiomyopathy
   8. Hypertrophic cardiomyopathy

Exclusion Criteria

1. Patients with congenital heart disease
2. Patients with poor image quality

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
The indicators of left ventricular size and function are measured by AI, senior physicians at the PI unit, and intermediate physicians at each sub center respectively (all parameters are measured by Mindray ultrasonic machines on the machine)	Artificial intelligence and intermediate doctors measurement results of each sub center will be completed within one day after data collection. Senior physicians measurement results of PI unit will be completed within one month after data collection.	Left ventricular end diastolic volume (EDV, ml), end systolic volume (ESV, ml), and ejection fraction (EF, %) (calculated by (EDV-ESV)/EDV\*100% ) in the four chamber view and two chamber view, as well as the Simpson biplane left ventricular volume (ml) and ejection fraction (%) results.
The indicators of right ventricular size are measured by AI, senior physicians at the PI unit, and intermediate physicians at each sub center respectively (all parameters are measured by Mindray ultrasonic machines on the machine)	Artificial intelligence and intermediate doctors measurement results of each sub center will be completed within one day after data collection. Senior physicians measurement results of PI unit will be completed within one month after data collection.	Right ventricular end diastolic area (EDA) and end systolic area (ESA).
The indicator of right ventricular function is measured by AI, senior physicians at the PI unit, and intermediate physicians at each sub center respectively (all parameters are measured by Mindray ultrasonic machines on the machine)	Artificial intelligence and intermediate doctors measurement results of each sub center will be completed within one day after data collection. Senior physicians measurement results of PI unit will be completed within one month after data collection.	Right ventricular area change rate calculated by (EDA-ESA)/EDA\*100%.

Secondary Outcome Measures

Name	Time	Method
Doppler ultrasound measurement indicators by are measured by AI, senior physicians at the PI unit, and intermediate physicians at each sub center respectively (all parameters are measured by Mindray ultrasonic machines on the machine)	Artificial intelligence and intermediate doctors measurement results of each sub center will be completed within one day after data collection. Senior physicians measurement results of PI unit will be completed within one month after data collection.	Forward blood flow velocity of mitral and tricuspid valve (m/s).
Mitral valve annulus and tricuspid valve annulus displacement are measured by AI, senior physicians at the PI unit, and intermediate physicians at each sub center respectively (all parameters are measured by Mindray ultrasonic machines on the machine)	Artificial intelligence and intermediate doctors measurement results of each sub center will be completed within one day after data collection. Senior physicians measurement results of PI unit will be completed within one month after data collection.	Mitral valve annulus and tricuspid valve annulus displacement (mm)

Trial Locations

Locations (37)

Chaoyang Central Hospital; 🇨🇳 Chaoyang, Liaoning, China

Fushun Central Hospital; 🇨🇳 Fushun, Liaoning, China

Shengjing Hospital of China Medical University; 🇨🇳 Shenyang, Liaoning, China

The First Hospital of China Medical University; 🇨🇳 Shenyang, Liaoning, China

Qinghai Provincial Hospital of Cardiovascular and Cerebrovascular Diseases; 🇨🇳 Xining, Qinghai, China

The Central Hospital Affiliated to Shandong First Medical University; 🇨🇳 Tai'an, Shandong, China

Linfen People's Hospital; 🇨🇳 Linfen, Shanxi, China

The Second Affiliated Hospital of Xi'an Jiaotong University; 🇨🇳 Xi'an, Shanxi, China

Jiangyou 903 Hospital; 🇨🇳 Jiangyou, Sichuan, China

Bishan Hospital Affiliated to Chongqing Medical University; 🇨🇳 Chongqing, China

Scroll for more (27 remaining)