Machine Learning and Big Data to reduce Chronic Disease Spending
Diabetes and heart diseases are the two most chronical conditions in the US. About $30 billion are spent in the US when it comes to hospitalizations. The majority of those costs are due to chronic diseases (30% heart diseases, 20% diabetes). Iaonnis Paschalidis, Professor of Engineering and Director of the Center for Information and Systems Engineering at Boston University, and his team promise to develop machine learning algorithms that will help to identify patients at higher risks of heart diseases or diabetes, and to prevent such hospitalization events. With machine learning and big data approaches, the team aims to deliver personalized predictions and recommendations to patients with the purpose of improving outcomes and reduce healthcare costs.
Despite this new method can benefit about 86 million Americans, these machine learning tools must be trustworthy and accurate with reliable data in order to work. An estimate of 33% of the US population already uses either fitness apps or smartwatches to track their wellness with. A lot of information can be retrieved from health trackers, using real-time health conditions, allowing Paschalidis and team to facilitate the development of such technology to make predictions.