Optimization of Amyloidosis Detection Method for Right Ventricle and Its Deployment Using Machine Learning (ML) Approach

Abstract

When an aberrant protein called amyloid accumulates in our organs and tissues, it causes amyloidosis. When it happens, both their form and functionality are impacted and amyloidosis is a critical medical condition that may result in organ failure that is serious and may be lethal. In this present study, the author discussed about the amyloidosis in the right ventricle of the heart and its detection using the Machine Learning (ML). The methodology used for this research includes an infrastructure of amyloidosis detection, created by using supervised and unsupervised learning technique. The results show the precise diagnosis is crucial since the details of ailment will have a significant impact on the kind of therapy we get. The infrastructure gathers user input initially, which is utilised to prepare the data afterwards. Unsupervised data from the right ventricle is collected and supplied to the machine learning cycle after an effective data translation and model development. The study concludes that typical supervised machine learning datasets were used to train the algorithms. These methods may eventually enable longitudinal, automated, and objective monitoring of patients' ventricular function. Machine learning for vast and small datasets of Cardiac Amyloidosis (CA) and cardiac straining and functional parameters offers appealing diagnostics prediction accuracy for CA diagnosis when compared to state-of-the-art methods. The future potential of this research is the infrastructure can be used in an effective manner for the diagnosis of major diseases.