World's First AI-aided Wearable Fingernail Sensor to help with Cardio, Parkinson’s treatment

The use of AI in various sectors across the globe has seen a sudden increase in the last year. Artificial Intelligence alone, in some cases, is the sole reason why growth has skyrocketed for businesses.

The medical and healthcare industry is one area where Artificial Intelligence can be effectively used to detect as well as cure life-threatening diseases. Grip strength has proven to be useful to link the effectiveness of a particular medication in people with Parkinson's disease, cognitive function in schizophrenics, an individual's cardiovascular strength or the overall mortality in geriatrics.

According to a recent study published in the Scientific Reports, IBM's team has developed a prototype of the world's first "fingernail sensor" which will be used to monitor a person's health. It will continuously keep a tab on how the fingernail bends or moves, which in turn indicates grip strength.

It has been found that our fingernails deform, bend or move in various ways when we extend our fingers, flex, or grasp anything. The deformation is so minuscule that it spans in the order of single digit microns and is easily detectable by strain gauge sensors. 

That being said, there are some alternative ways to measure how much a particular disease has progressed by monitoring certain things like motion, nerve cells, deviation in sweat gland activity and so on. These sensors are usually attached to the skin but can cause problems with older patients and might also lead to infections.

Using the data from the sensor, the researchers were also able to distinguish the typical activities such as turning a key or opening a doorknob. With the help of a neural network, the sensor was even able to predict digits written by the user's finger with an accuracy of 94 percent.

It was primarily developed for monitoring the effects of medication of people with Parkinson's disease. The system consists of strain gauges which are attached to a fingernail which feeds the incoming data to a small computer responsible for sampling strain values, collecting accelerometer data and is even connected to a smartwatch.

However, this is not possible with just AI. The watch is also responsible for running machine learning models to rate symptoms related to Parkinson's such as bradykinesia, tremor, and dyskinesia.

The use of fingernail sensors is not limited to just Parkinson's but has opened up new ways in which they can be used to keep a constant check on the person's health thus enabling a new type of user interface. This technology can hopefully, someday, be used to help quadriplegics communicate by making micro-movements with just fingertips.

PC: Pablo, Unsplash
VC: IBM Research