A new method of heart target region recognition based on dense pixel features

In order to more effectively improve the efficiency and accuracy of doctors' diagnosis of diseases, and improve the effect of intelligent algorithm on doctors' auxiliary diagnosis, the advantages of deep learning technology in the field of image processing are integrated with medical detection technology. Echocardiography is an important basis for the evaluation of cardiac function, which reflects the diversity and complexity of information. Therefore, a method of cardiac function analysis based on echocardiography and semantic segmentation is proposed in this paper. By building a dynamic detection model based on dense pixel features, the pixel level semantic segmentation of left ventricular region is performed on apical four chamber echocardiography images with end diastolic and end systolic features respectively. According to the segmentation results of the model, various parameters of the left ventricle in different periods can be calculated, and then the dynamic function of the heart can be analyzed. Experimental results show that the average recognition accuracy of the proposed method is 98.60%.