Atrial fibrillation risk assessment method based on machine learning

The invention relates to an atrial fibrillation risk assessment method based on machine learning. Firstly, multi-source data including electrocardiogram (ECG), echocardiography, lifestyle and clinical medical records are collected and preprocessed, and the preprocessing step includes data cleaning, denoising, normalization and missing value processing; then carrying out feature engineering, namely extracting the shape and time features of QRS complex waves, P waves and T waves from the electrocardiogram, extracting the size and function indexes of the left atrium from the echocardiogram, and selecting features with high prediction values by using a feature selection algorithm; training a single-modal machine learning model for each data type, and dynamically adjusting the weight of each modal data according to the data characteristics and quality of different patients by adopting an adaptive multi-modal fusion algorithm so as to optimize the overall prediction performance of the model; and finally, adjusting