Abstract

Abstract Background In the care of patient with atherosclerotic disease, some arterial segments were more prone to have lesion, while the whole arterial system was exposed with the same risk factors (hypertension, diabetes or hyperlipidemia, etc). Our team used the principles and practices of fluid mechanics to investigate the mechanism of atherosclerotic disease. In short, laminar flow protects the pipes or arteries while the turbulent flow damages the intima and starts the atherosclerotic process. In this study, our question is: Why many patients with peripheral arterial disease (PAD) had diffuse lesions in the femoral arteries while the iliac arteries showed only minimal or no lesion? Could a new dynamic angiographic review and Deep Learning analysis answer the above question? Method All patients who underwent diagnostic coronary angiogram or interventions were selected. After the procedure, an iliac and femoral angiogram were performed for vessel closure device deployment. At first, the iliac and femoral artery was filled completely with contrast, then the blood (in white color) moved down and replaced the black contrast. All images were recorded at 15 frames per second. The movement of blood flow, its shape (laminar or turbulent), direction (antegrade or retrograde) were identified and tabulated. At the same time, the Deep Learning models were trained based on the convolutional neural network. The U-net combined Densenet121 was used to train model number one to recognize the coronary and detect the stenosis area. Next, the CNN model was used to classify the stagnant flow. The patients with lesions in the iliac or femoral arteries were selected. The characteristics of the flow proximal and at the lesion site were selected and compared. Results 50 cases (70% male) met our criteria with 80% lesions at femoral arteries (figure 1) while only 20 % had lesions in the iliac arteries. The turbulent flow was seen in both arteries however it was persistent in the femoral arteries while in the iliac arteries, turbulent flow was intermittent with 40% of the time in laminar flow. The analysis by the Deep learning model had sensitivity in the detection of turbulent flow 0.91 and the sensitivity was 0.96. The explanation was that the persistent turbulent flow in the femoral arteries injured the intima However the laminar (non-turbulent flow) in the iliac arteries carried the LDL particle further distally and deposited them at the femoral level. This is why there were more atherosclerotic plaques in the femoral level than in the iliac arteries. Conclusion Persistent turbulence causes more damage to the intima in the femoral arteries while intermittent turbulence did less damage to the iliac arteries. LDL cholesterol while in suspension at the iliac level were carried and deposited in the femoral arteries. This sequence of event explained the predilection of femoral arteries diffuse lesions. Larger studies with more patients are needed to confirm the above results.Diffuse lesion in femoral artery

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call