The way that a person grabs hold of an object or carries out an everyday movement can give an insight into the state of their muscles and, more generally, their health. In order to provide the healthcare industry with a tool that is more useful and more accurate than traditional skin sensors, IBM have developed a wireless device that can be stuck to the patient’s fingernail and linked with a connected watch or smartphone.  This technology could be used to monitor patients with Parkinson’s disease.

Several scientific studies have shown how useful evaluating the strength of a patient’s grasp can be when diagnosing and monitoring various illnesses, such as Parkinson’s disease, schizophrenia or even cardiovascular disease. This strength is generally measured using sensors placed on the skin. 

Unfortunately, in the case of illnesses that affect elderly patients, who also suffer from dermatological issues, this method is not particularly suitable. This is why IBM have developed a miniature sensor that is not placed on the skin, but rather on the patient’s fingernail, reports website Futura-Sciences.

A miniature wireless sensor linked up to a connected watch 

Stuck directly onto the fingernail – which is much less sensitive than the skin – the miniature sensor is linked up to a connected watch (which should be worn on the same arm as the sensor) or even a smartphone, via Bluetooth Low Energy. The way the fingernail moves and twists when the patient grasps an object is measured using strain detectors, while an accelerometer evaluates the movements’ amplitude and fluidity. 

The device, which IBM’s researchers claim is so robust that the wearer can throw a baseball without displacing it, still needs some improvement. It must be reduced in size down to micrometre scale, in order to reduce its electric consumption, and its electronic components need improving.  

The micro-sensor is so sensitive that it can detect deformations measuring just a few microns. This accuracy is extremely impressive when you think that a single hair measures between 50 and 100 microns, and a red blood cell measures less than 10 microns, highlights Futura-Sciences. It is enough to differentiate between a vast range of specific movements, like turning a key, cutting vegetables, opening a jar, etc. 

Symptoms of Parkinson’s disease can be identified using AI 

The data collected by the micro-sensor regarding the pressure, movement and quality of the patient’s grasp will then be  sent to an application, where it will be used to put together a reference database. This database, analysed by artificial intelligence, will be stored on the Cloud as opposed to on the watch or smartphone, for better reliability.

Any weakening or changes in motor control can therefore be identified. All of the symptoms of Parkinson’s disease – bradykinesia (slow movement), shaking and dyskinesia – will be assessed by artificial intelligence using learning models as a base. IBM plans to develop a second application in the future, whose role will be to detect movements that might present a danger to sufferers.

Allianz Partners