A Smart Bluetooth Insole Lets Toe-Tapping Reveal Fall Risk in Parkinson's Disease Patients
Measuring toe-tapping tests with on-board accelerometers, these smart insoles could ease fall risk diagnoses.
Researchers from Texas A&M University have come up with an interesting approach to diagnosing fall risk in people with Parkinson's disease: a smart insole that monitors them as they tap their toes.
"[A] toe-tapping motion is one of the important tasks in [the Unified Parkinson's Disease Rating Scale]," explains Ya Wang in an interview with IEEE Spectrum, "but there is a very limited number of studies working on analyzing toe tapping with smart wearable devices in the field. We started to wonder if we could use toe-tapping motion, which has a simpler and more interesting setup compared to walking motion, to assess the risk of falling."
To test the idea, the team built a wearable designed to slip into a shoe: An insole equipped with on-board accelerometers, a microcontroller, and a Bluetooth Low Energy (BLE) radio to send captured data to a nearby computer for analysis.
"In total," the team notes, "87 time-axis and 1,152 acceleration-axis features are extracted and analyzed between the two groups," referring to test subjects with Parkinson's disease, the PD group, and aged-matched healthy control subjects in the HC group, all of whom were asked to tap their toes in either an alternating or synchronized pattern.
The results are promising, though show that more work is required for clinical use. The results of the toe-tapping data-gathering exercise showed an ability to distinguish between the PD group and the HC group, and a close-correlation between the toe-tapping results and a PD group's fall risk on the UPDR scale — though with a somewhat reduced accuracy compared to a traditional walking test.
Wang and colleagues are planning to perform additional experiments, potentially extending the smart insole to include electromyography (EMG) sensors to capture data from muscle movements.
The team's work has been published in the journal IEEE Sensors Letters under closed-access terms; more information is available on IEEE Spectrum.