Detecting Disease Early Through AI

Abhinav Raj, Writer
@uxconnections

Scientists at TU Dresden have successfully demonstrated the use of biocompatible chips that detect ‘pathological changes’ without medical supervision.

Artificial intelligence continues to change the paradigm of healthcare and medicine—in big (and minuscule) ways. 

Scientists at the Chair of Optoelectronics at the Dresden University of Technology have successfully developed a biocompatible and implantable AI platform on a polymer network that is able to detect malign patterns in a patient’s biological signals. 

The platform, built on biocompatible AI chip and polymer-based fibre networks can classify a patient’s biosignals between healthy or diseased with a never-before-seen accuracy rate and alert doctors and patients via IoT devices in cases of anomalous behaviour—while consuming less energy than the average pacemaker. 

In the study published in the journal Science Advances, the Chair of Optoelectronics Prof. Karl Leo and his research team propose pattern recognition for the detection of malign patterns through active biosignal monitoring. 

To demonstrate the utility of the AI-enabled biosignal analysis platform, a research trial was conducted wherein the AI was required to classify four classes of arrhythmic heartbeats. According to the results, the AI was able to distinguish healthy heartbeats from arrhythmias with an accuracy rate of 88 per cent. 

The biosignal analysis platform can detect irregular heartbeats or arrhythmias. (Image: Wikimedia Commons)

“The vision of combining modern electronics with biology has come a long way in recent years with the development of so-called organic mixed conductors,” explicated the first author of the paper, Matteo Cucchi. 

“By harnessing the power of neuromorphic computing, such as reservoir computing used here, we have succeeded in not only solving complex classification tasks in real time but we will also potentially be able to do this within the human body.”

Biosignal analysis may go a long way. By implanting the body with discrete, biocompatible AI monitors, faster diagnosis and swift action can be achieved and medical attention can be made available to those who require it. 

In the future, AI-enabled bio-implants might just save a million lives by sending a timely notification on your iPhone or smartwatch.