A pilot study used AI to scan videos of patients suspected of Parkinson’s Disease.

Researchers at the Skolkovo Institute of Science and Technology in Russia built a system that used computer-based, machine-learning algorithms to analyze video recordings of people, thought to have Parkinson’s Disease (PD), performing specific movement tasks. The system could identify between healthy and suspected PD patients. (Parkinson’s News Today)

How the AI system helps to diagnose PD

Diagnosing PD has historically been difficult to diagnose because of the non-availability of medical tests that can authoritatively flag the disease.

Physicians, therefore, rely upon symptoms such as shaking or tremor, slowness in movement, body rigidity, and difficulty in maintaining balance. These are the hallmarks of the disease.

Unfortunately, these symptoms sometimes prove to be red herrings because other conditions, such as essential tremors, may also exhibit similar systems.

However, researchers found that AI-enabled video analysis was a promising, non-invasive method. It is useful as a second opinion to confirm or rule out PD.

The study used videos of 83 persons performing a set of 15 exercises. This test population had a mix of people with or without neurodegenerative diseases.

“The exercises were designed under the supervision of neurologists and came from several different sources, including scales that are used for monitoring Parkinson’s disease and previous research done in this area,” Ekaterina Kovalenko, a study co-author. Previous research using sensors on patients had helped researchers to identify, and rank exercises according to their contribution to a reliable diagnosis of PD.

A software then processed the videos into a form that AI could analyze.

The results showed an accuracy score of 0.90 when comparing healthy participants to those with suspected PD. (On a scale in which 1 represented a perfect prediction).

In another reading, the system distinguished between PD and essential tremor with an accuracy of 0.84.

A dependable second opinion

“Our preliminary results show potential in improving diagnosis with the help of video analysis. Our goal is to provide a second opinion for doctors and clinicians, not to replace them,” the scientists wrote. “A video-based method perhaps is the most convenient for patients, as it is the most versatile and non-invasive when compared to various sensors and testing.”

Related Story:  AI Algorithm Diagnoses Alzheimer’s Disease Better Than Experts

Image Credit: PainSA (Creative Commons Licence)