Turning your smartphone in a medical diagnosic device

Carotid arterial waveform captured using an unmodified iPhone 5S camera by placing the iPhone camera over the carotid pulse. The accuracy matches the one obtained with a professional medical echocardiography system. Credit: Niema M. Pahlevan et al./Critical Card Medicine

The general availability of smartphones pushed researchers to leverage on them as diagnostic tools. In the past few years a number of “adds on” have been created to let smartphones morph into virus/bacteria detector, measurer of a variety of substances, like glucose and so on. Basically, researchers take advantage of the processing power of the smartphone to analyse a variety of data that can be retrieved using specific sensors. Of course they also leverage on the communications capability to connect to a medical centre / doctor and on the display to provide information to a para-medic or lay person to take action.

The progress in signal processing is now making possible to use the smartphone itself as a diagnostic medical instrument leveraging on its camera.

I have been using for a few years now the VitalSigns app (by Philips) to check my pulse and there are several more for a variety of situations. However, these apps are a bit “borderline” if you look at them as real, reliable, diagnostic tools, au pair with professional ones. They can surely help but you are better off with the real thing if something is wrong.

This is why the news coming from Caltech is so interesting and may signal a new era in diagnostics. Researchers have developed a signal processing software running on an iPhone (but it can run on any smartphone in principle) that is able to analyse the images captured by the phone camera and assess with very good precision the amount of blood pumped by the heart as it beats. This quantity, that is called in medical lingo LVEF (Left Ventricular Ejection Fraction), is quite important in determining a number of hearth pathologies. In practice it measures the amount of blood that is ejected from the left ventricle into the aorta vs the amount of blood remaining in the ventricle. The normal ratio should be over 50%, in general it is between 50 and 70%. The higher the number the more effective the pulse and the less fatigue for the heart. A lower number may indicate a weakening heart, a problem with the valve…

It is obvious that this value should be accurate. The best way to determine it is through MRI, but this is seldom done because of the scarcity of these equipment, the cost and complexity involved. Medical doctors make do with the cheaper echocardiograph that provides a value accurate to +/- 20%. It does not seem very accurate but for a diagnoses is pretty good and if the value may be critical a further MRI exam can be run.

The software developed is able to provide a value with an accuracy within +/- 19%, au pair (actually slightly better) with the one performed with an echocardiograph and it does that in just 2 minutes, versus the 30 minutes required by the echocardiograph.

This result shows how signal processing progress can change the rules of the game. Our smartphone can become n the next decade an accepted medical diagnostic tool. Couple this to advances in Artificial Intelligence and you start to see some changes in health care practice…

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the New Initiative Committee and co-chairs the Digital Reality Initiative. He is a member of the IEEE in 2050 Ad Hoc Committee. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.