In the present medical era, the major cause of the rise in death rate worldwide is atherosclerosis disease and this diagnosis is complicated because initial signs are unattended. To reduce the costs of treatment and prevent serious events, it is necessary to improve the prediction accuracy of cardiovascular diseases during plaque formation. This proposal is intended to create a support system for the biosensor-assisted deep learning concepts for detecting atherosclerosis disease. With the clinical data, this mathematical model can predict heart disease based on deep learning-assisted k-means geometric distribution artificial neuron model. The atherosclerotic plaque formation mathematical model explains the early atherosclerotic lesion development in a more accurate manner. Further, the creation of the atherosclerotic plate, the test performs numerical simulations with idealized two-dimensional carotid artery bifurcation geometry. The proposed system has been analyzed using a variety of similarity tests such as the coefficient Matthews’s correlation (CMC). Furthermore, the results have reached 95.66% accuracy and 0.93 CMC, which are significantly higher than published conventional research.

在当前的医学时代，世界范围内死亡率上升的主要原因是动脉粥样硬化疾病，并且这种诊断是复杂的，因为初始症状是无人照看的。为了减少治疗费用并防止发生严重事件，有必要提高斑块形成过程中心血管疾病的预测准确性。该提议旨在为生物传感器辅助的深度学习概念创建一个支持系统，以检测动脉粥样硬化疾病。借助临床数据，该数学模型可以基于深度学习辅助k预测心脏病-表示几何分布人工神经元模型。动脉粥样硬化斑块形成数学模型以更准确的方式解释了早期动脉粥样硬化病变的发展。此外，通过创建动脉粥样硬化板，该测试可以对理想的二维颈动脉分叉几何形状进行数值模拟。已使用各种相似性测试（例如系数Matthews的相关性（CMC））对提出的系统进行了分析。此外，结果达到了95.66％的准确度和0.93 CMC，大大高于已发表的常规研究。