Healthcare

Fingernail Sensors and AI Can Help Clinicians to Monitor Health and Disease Progression

Share this post:

Grip strength is a useful metric in a surprisingly broad set of health issues. It has been associated with the effectiveness of medication in individuals with Parkinson’s disease, the degree of cognitive function in schizophrenics, the state of an individual’s cardiovascular health, and all-cause mortality in geriatrics. At IBM Research, one of our ongoing challenges is to obtain a better understanding of the effects of diseases on an individual’s overall health, as well as how AI can help clinicians to monitor individuals in their natural environments, and potentially point to indicators and clues into the progression of a patient’s conditions. In new research published in Scientific Reports today, our team details a first-of-a-kind “fingernail sensor” prototype to help monitor human health. The wearable, wireless device continuously measures how a person’s fingernail bends and moves, which is a key indicator of grip strength.

Fingernail sensor that can monitor your activities/health. IBM Lab, Yorktown Heights, NY. (Feature Photo Service)

The project began as an attempt to capture the medication state of people with Parkinson’s disease. Getting a new therapy approved requires quantifying how people on the therapy are doing in relation to controls. The majority of people with Parkinson’s are older, an age group with increasingly brittle, friable skin.

Comprised of skin, nails and hair, the integumentary system covers most of our bodies. Its primary purpose is to protect our internal components from pathogens, toxins, ultraviolet radiation, dehydration, and changes in temperature. It also provides a structure for sensory receptors of the somatosensory system of neurons across our bodies.

One method to measure a disease’s progression is to attach skin-based sensors to capture things like motion, the health of muscles and nerve cells, or changes in sweat gland activity, which can reflect the intensity of a person’s emotional state. But with older patients, such skin-based sensors can often cause problems, including infection.

This is where the potential of a fingernail sensor comes into play. We interact with objects throughout the day using our hands, such as the tactile sensing of pressure, temperature, surface textures and more. Our team realized it might be possible to derive interesting signals from how the fingernail bends throughout the course of a day, as we use our fingers to interact with our environment, and tap into the power of AI and machine learning to analyze and derive valuable insights from that data.

One of the functions of human fingernails is to focus the finger-tip pulp on the object being manipulated. It turns out that our fingernails deform – bend and move — in stereotypic ways when we use them for gripping, grasping, and even flexing and extending our fingers. This deformation is usually on the order of single digit microns and not visible to the naked eye. However, it can easily detected with strain gauge sensors. For context, a typical human hair is between 50 and 100 microns across and a red blood cell is usually less than 10 microns across.

Since nails are so tough, we decided to glue a sensor system to a fingernail without worrying about any of the issues associated with attaching to skin. Our dynamometer experiments demonstrated we could extract a consistent enough signal from the nail to give good grip force prediction in a variety of grip types.

We also found it is possible to deconvolve subtle finger movements from nail deformation. We were able to differentiate typical daily activities which all involve pronation and supination such as turning a key, opening a doorknob or using a screwdriver. An even more subtle activity is finger writing, and we trained a neural network to produce a very good accuracy (.94)  at detecting digits written by a finger wearing the sensor.

Our system consists of strain gauges attached to the fingernail and a small computer that samples strain values, collects accelerometer data and communicates with a smart watch. The watch also runs machine learning models to rate bradykinesia, tremor, and dyskinesia which are all symptoms of Parkinson’s disease.

By pushing computation to the end of our fingers we’ve found a new use for our nails by detecting and characterizing their subtle movements. With the sensor, we can derive health state insights and enable a new type of user interface. This work has also served as the inspiration for a new device modeled on the structure of the fingertip that could one day help quadriplegics communicate.


Katsuyuki Sakuma, Avner Abrami, Gaddi Blumrosen, Stanislav Lukashov, Rajeev Narayanan, Joseph W. Ligman, Vittorio Caggiano & Stephen J. Heisig. Wearable Nail Deformation Sensing for Behavioral and Biomechanical Monitoring and Human-Computer Interaction. Scientific Reports 8, Article number:18031 (2018).

More Healthcare stories

Dark Matter Matters: AI Makes DNA Dark Matter Useful

What is the minimal description that captures a space? Asking a mathematician’s basic question of a  biological dataset reveals interesting answers about biology itself. This summarizes our underlying approach to subtyping hematological cancer. Disease subtyping is a central tenet of precision medicine, and is the challenging task of identifying and classifying patients with similar presentations […]

Continue reading

Helping to Untangle Cancer Drug Resistance with Data

Why do targeted cancer therapies often fail? We have acquired so much more understanding about cancer in the last fifty years than in the last five thousand years. Approaches to patient treatments have dramatically changed, and statistics show significant improvement in patient response and outcomes to therapy in the last half a century [1]. Yet […]

Continue reading

Novel AI tools to accelerate cancer research

At the 18th European Conference on Computational Biology and the 27th Conference on Intelligent Systems for Molecular Biology, IBM will present significant, novel research that led to the implementation of three machine learning solutions aimed at accelerating and guiding cancer research.

Continue reading