Wearable Sensors Most Reliable for Diagnosing Parkinson’s Disease
Scientists have compared three techniques – handwriting, cameras, and wearable sensors – and found sensors to be the most reliable in diagnosing Parkinson’s disease, although they are uncomfortable for patients in the later stages of the disease.
“Not all people with Parkinson’s show each and every symptom of the disease, and we tried to figure out if identification could rely on a single diagnostic modality: a sensor, cameras, or handwriting,” Aleksei Shcherbak, a PhD student at Skoltech said in a press release. “Even though sensor data analysis showed the best results, it’s inconvenient for the patients at a later stage of PD [Parkinson’s disease]. But the patients diagnosed at early stages report that wearable sensors are reasonably comfortable to wear, meaning that in the near future, wearable technologies could collect and assess the data continuously to track the disease development and patients’ response to medical treatment.”
Their results were published in IEEE Transactions on Instrumentation and Measurement in the study “Comparative Study of Wearable Sensors, Video, and Handwriting to Detect Parkinson’s Disease“.
Although the exact cause of Parkinson’s disease is unknown, age, genetic, and environmental factors play a role in the risk of developing it.
The disease affects the dopamine-producing cells in the substantia nigra, a part of the brain. Over time, the death and dysfunction of these dopamine-producing cells affects coordination and movement, leading to tremor, abnormally slow movements, and muscle stiffness, the disease’s characteristic signs.
Parkinson’s can be challenging to diagnose early because it shares symptoms with other brain conditions such as dementia with Lewy bodies and multiple system atrophy. A late diagnosis often has a negative effect on therapeutic options and a patient’s quality of life, but an early diagnosis could significantly improve a patient’s prognosis and treatment options.
Scientists are now turning to artificial intelligence to improve the diagnosis and monitor the progression of difficult-to-detect conditions such as Parkinson’s. Researchers at Skoltech in Russia designed specific movement tasks to detect Parkinson’s in middle-aged adults. They collected data from 35 people without Parkinson’s and 85 with Parkinson’s (mean age, 58.5) and had them complete motor tasks in front of cameras while wearing sensors, using computer-generated algorithms to assess their performance. The tasks included folding a towel, filling a glass with water, tapping the index finger to the thumb, reading a complex sentence aloud, writing it down and tracing a spiral. The tasks generally take about 15 minutes to complete, the researchers noted.
The exercises were designed under the supervision of neurologists and came from several different sources, including scales used for monitoring Parkinson’s disease and previous research done in this area. Each exercise has a target symptom that it can reveal,” Ekaterina Kovalenko, a PhD student said.
To analyze the data obtained, researchers relied on machine and deep learning methods.
Although the sensors were the most reliable, the handwriting and video tests were the most comfortable and less time-consuming for patients as they could be done relatively quickly and without the restrictions of wearing a device, “offering a sensible trade-off for the patients at the later stages of [Parkinson’s],” the researchers wrote.
“…Doctors need objective tools to assess motor fluctuation in patients with [Parkinson’s disease]. These evaluated tools can provide a more personalized approach to therapy and help doctors make decisions on medical treatment and, if needed, neurosurgical interventions,” Ekaterina Bril, MD, PhD, a head of neurology research at A.I. Burnazyan Federal Medical and Biophysical Center, said.
About the Author
