Data-driven Healthcare using Affordable Sensing: Screening, Diagnosis and Therapy

Abstract

Today's Healthcare systems are built around an "illness"-driven model where all the stakeholders benefit when people become "ill". There needs to be a paradigm shift to convert this into a "wellness"-driven model. However, in order to create such systems, one needs to build affordable, easily-usable and mass-deployable solutions. We look at three use cases and try to provide solutions -- a) It is seen that a vast majority of the population is affected by "silent-killer" lifestyle diseases like cardiac artery disease (CAD), chronic obstructive pulmonary disease (COPD) and diabetes -- early detection and screening for such diseases are really useful. In this paper, we present how to detect early onset of these using mobile phones and low-cost attachments to mobile phones followed by signal processing and machine learning based analytics b) In rural / semi-urban areas of developing countries, a big problem is early diagnosis of hypertension among pregnant mothers -- it is seen a large number of complications during birth can be avoided if hypertension of pregnant mothers are controlled. In this paper we present how to measure cuff-less blood pressure affordably using just a smart phone and its camera for this. c) Finally, there is a huge number of stroke patients who need rehabilitation therapy -- such treatment typically requires sophisticated rehab-labs in hospitals which is not only costly, but also has accessibility issues. We discuss how a Kinect-sensor based system at home can be used followed by sensor data analytics to help in diagnosis, guidance and compliance to the therapy.