Wearable devices

Smart insoles and artificial intelligence: steps become diagnostics

From diabetic foot to Parkinson's disease: new self-powered system reveals health status by walking

by Federico Mereta

23 April 2025

3' min read

Key points

How the system works
For whom it might be useful

3' min read

On the one hand, intelligent detectors on footwear that can monitor how people walk. On the other, Artificial Intelligence (Ai) that analyses the state of movement step by step (it has to be said). In between, a series of technologies that make it possible to constantly analyse walking. All this, in order to analyse the movements of the feet and arrive at a personalised strategy to improve posture, helping to combat back pain, as well as a targeted approach to specific problems such as plantar fasciitis or the diabetic foot, or even an early diagnosis (or monitoring of the picture) of Parkinson's disease.

In short: 'tell me how you walk and I will tell you how you are'. Launching this hypothesis, presenting an innovative smart insole system that constantly and non-stop studies the way people walk, is research published in Science Advances coordinated by Qi Wang and Jinghua Li, of Ohio State University.

How the system works

The device is based on 22 small pressure sensors and is powered by small solar panels placed on the top of the shoes. In this way, it collects unique data related to the way a person walks, which is a unique biomechanical process. The collected data then travels via Bluetooth to a smartphone for a quick and detailed initial analysis.

At this point, Ai comes into play. Thanks to an advanced machine learning model, the wearable device is able to recognise eight different states of movement, from static ones such as sitting and standing to more dynamic movements such as running and squatting. In this way, different subjective movements, pressure and load points of the foot can be analysed, resulting in a kind of 'footprint' unique to each person.

All this, without any particular problems with energy supply. In fact, the device is based on slabs made of flexible and safe materials and can run for a long time, thanks to the action of solar cells that convert light into energy, which in turn is stored in tiny lithium batteries. In short: in addition to reliability, the studied prototype lasts a long time, is efficient and precise in data collection and, above all, should have no power supply problems.

Li himself confirms this in a press note from the American university: 'our device is innovative in terms of high resolution, spatial sensing, self-powering capability and the ability to integrate with machine learning algorithms'. It is mainly in terms of durability, however, that the difference could be made. Indeed, according to the research, the smart insoles showed no significant deterioration in performance after 180,000 compression and decompression cycles, demonstrating their long-term durability.

For whom it might be useful

According to Li, monitoring the well-being of the foot and especially changes in load and pressure points could be of great help to health. "Our body contains a lot of useful information that we are not even aware of," is his comment. "These states also change over time, so our goal is to use electronics to extract and decode these signals and encourage better self-diagnosis.

In particular, by analysing the feedback from the various sensors placed along the entire architecture of the foot, it was seen how these can change and how they are different when walking or running. When walking, pressure is applied sequentially from the heel to the toes, whereas during running, almost all sensors are pressurised simultaneously. Moreover, during walking, the time of pressure application accounts for about half of the total time, while during running only about a quarter. The foot, therefore, can talk. And offer important information. Thanks to smart insoles, gait could be analysed and abnormalities associated with pressure-related foot disorders (such as diabetic foot ulcers), musculoskeletal disorders (such as plantar fasciitis) and neurological disorders (such as Parkinson's disease) could be identified at an early stage.

All this, knowing that the eventual physical rehabilitation could also be helped by the device, capable of correcting any postural problems. Researchers predict that the technology will probably be commercially available within the next three to five years. But targeted tests are still needed to make the insoles capable of supporting (and bearing) any foot, given the variations in gait between individuals.