The predictive value of deep learning-based cardiac ultrasound flow imaging for hypertrophic cardiomyopathy...

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Hypertrophic cardiomyopathy (HCM) is a complex condition often associated with arrhythmias.
• Accurate prediction of arrhythmias in HCM patients is crucial for timely intervention and improved outcomes.
• Current diagnostic methods may have limitations in predicting arrhythmia risk.

Purpose of the Study:

• To evaluate the predictive capability of deep learning-based cardiac ultrasound flow imaging for arrhythmias in patients with HCM.
• To identify specific imaging parameters that serve as risk factors for arrhythmias in HCM.

Main Methods:

• Retrospective analysis of clinical data from 158 hypertrophic cardiomyopathy patients (July 2019 - December 2021).
• Development and application of a deep learning model for cardiac ultrasound flow imaging.
• Extraction of various imaging parameters including LVEF, LAVI, E/e', vortex area change rate, circulation intensity change rate, mean blood flow velocity, and mean EL value.

Main Results:

• No statistically significant differences in general patient data (age, gender, BMI, blood pressure, HR) were observed between training, validation, and test groups.
• Similarly, no significant differences in general data were found between concurrent and non-concurrent arrhythmia groups within the training set.
• The deep learning model demonstrated high accuracy in identifying cardiac ultrasound images relevant to arrhythmia prediction.

Conclusions:

• Deep learning-based cardiac ultrasound flow imaging offers a powerful tool for predicting arrhythmias in HCM.
• Vortex area change rate, circulation intensity change rate, mean flow velocity, mean EL, LAVI, and E/e' are identified as significant risk factors for arrhythmias in HCM patients.