Medically reviewed by Susan Kerrigan, MD and Marianne Madsen
Artificial intelligence (AI) has helped us make giant leaps in healthcare. It has been used with great results in areas such as cancer, neurology, and cardiology. It analyzes a variety of diagnostic data, such as ECGs, EEGs, and x-rays, often as well (or better than!) humans.
Up until now, AI has only been able to help diagnose illnesses when it is used outside of the body. Implanting AI diagnostic tools into the body to detect real-time changes has been difficult. However, researchers recently developed a new type of implantable AI platform that helps healthcare professionals detect pathological patterns immediately.
The Technology
Researchers recently developed a bio-compatible, implantable form of AI that uses polymer-based fiber networks. These networks actually use the human brain as their model for how they function. They form a recurrent network from randomly arranged polymer fibers. This biocompatible computer platform can be the model for hardware-based artificial neural networks that consume very little power but are able to interact with tissues and body fluids.
What It Can Do
These small, implantable devices can provide healthcare professionals–and patients–with information in real time. For example, if someone has recently had heart surgery and monitoring heartbeat rhythms is necessary, this device could be implanted at the time of surgery to monitor the patient. If anything goes wrong, data could be sent by smart phone to healthcare providers and the patient to get help immediately. No waiting for the doctor, no waiting for tests, no waiting for results from tests, no trying to catch the arrhythmia when it happens–data is available and people are alerted right now.
In a study, these devices were able to detect three common types of arrhythmia and distinguish between those and a healthy heartbeat with 88% accuracy. It did all this using less energy than a traditional pacemaker.
Solving complex tasks in real-time with this type of electronic biological sensor has been out of reach until now. The only type of problem solving previously available was that of individual synapses or sensors which are simpler electronic components. Matteo Cucchi, one of the study’s authors, says, “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.” He continues, “This approach will make it possible to develop further intelligent systems in the future that can help save human lives.”
