AI Empowered Cerebro-Cardiovascular Health Engineering

This research topic focuses on the diagnosis and treatment of cardiovascular and cerebrovascular diseases. Several innovative methods have been proposed to improve the accuracy and reliability of diagnosis, including DR-LCT-UNet, personalized aortic flow waveform, machine learning-aided scheme for heart failure detection, complex-valued phase-amplitude coupling, improved mel-frequency cepstrum coefficient features, preoperative CTA and CFD simulation, real-time three-dimensional echocardiographic transilluminated imaging, decision tree ID3 algorithm, directed transfer function with surrogate analysis, network pharmacology and molecular docking technology, and support vector machine technique.

These methods have been found to be effective in improving the accuracy and reliability of diagnosis, with accuracies ranging from 83% to 94.97%. Additionally, these methods have been used to improve the accuracy of evaluation and treatment of CHD, predict the outcome of Y-shaped extracardiac conduits Fontan, localize epileptogenic tissue, estimate the innervation zone of a muscle, and predict the prognosis of patients with severe acute myocardial infarction.

The proposed methods have also been used to improve the accuracy of wave reflection indices, detect intracranial pressure changes, and improve hand-eye coordination of patients during a game-like experience. Furthermore, these methods have been used to predict the practical components and potential mechanisms of Artemisia annua L. inhibiting the occurrence and development of abdominal aortic aneurysm.