| abstract |
The invention discloses a method for analyzing stenotic coronary blood flow changes under multiple exercise intensities. Firstly, an image sample library is constructed for coronary stenosis CT images, and label category processing is performed; The wavelet correlation method is used for denoising and enhancement preprocessing; the convolutional neural network model CNN in the deep learning method is used for feature extraction, and the Otsu method is used for segmentation, and finally 3D reconstruction is performed; the 3D reconstructed image is limited Meta-analysis; finally, the patient's wall shear force, blood flow velocity, wall pressure and wall deformation will be tested under the condition of resting state and different exercise intensities. The invention measures the hemodynamic analysis of coronary atherosclerosis patients at rest and at three different levels of exercise intensity, realizes accurate judgment in controlling the disease, and avoids the harm caused by invasive surgery to the patient. |