Deep Learning Algorithms Detects Physiological Changes Linked to Cardiac Disease


Machine learning can capture intrinsic frequency algorithm, which can be used to calculate physically relevant variables related to cardiac functions of patients

New method developed by researchers at USC Viterbi School of Engineering coupled machine learning model with a patient’s pulse data to measure a key risk factor for cardiovascular diseases and arterial stiffness, using just a smart phone. Arterial stiffening is characterized by loss in rigidity of arteries resulting into increased blood and pulse pressure. Furthermore, it is associated with diseases such as diabetes and renal failure. Their study was published in Nature Scientific Reports in January, 2017.

According to Centers for Disease Control and Prevention (CDC), heart disease are the leading cause of death worldwide. “If the aorta is stiff, then when it transfers the pulse energy all the way to the peripheral vasculature – to small vessels – it can cause end organ damage. So, if the kidneys are sitting at the end, the kidneys get hurt; if the brain is sitting at the end, the brain gets hurt,” said Niema Pahlevan, assistant professor of aerospace and mechanical engineering and medicine.

The novel method developed by Pahlevan, Marianne Razavi and Peyman Tavallali uses a single, uncalibrated carotid pressure wave that can be captured with a smart phone camera. This method captures the shape of a patient’s pulse wave for the mathematical model, called intrinsic frequency, to calculate key variables related to the phases of the patient’s heartbeat. These variables are then used in a machine learning model that determines pulse wave velocity (PWV) and arterial stiffness.

As a part of study, 4,798 patients were analyzed for PWV measurement. The measurements of these patients were found to be associated with the onset of cardiovascular diseases over a ten-year follow up period. Furthermore, team is working on expanding intrinsic frequency algorithm for number of other applications, such as detecting silent heart attacks.


About Author

Comments are closed.