Carpal tunnel syndrome is a condition in which the nerves in the wrist are compressed, causing numbness in the hand and difficulty moving the fingers. Roughly 2 to 4 percent of middle aged women and older are said to be affected by the condition, and early diagnosis and treatment is recommended because patients can lose the ability to carry out fine movements, such as grasping items with their thumbs, if the condition is allowed to progress. However, there has long been a need for quick and easy diagnosis of the condition, because it can currently only be diagnosed accurately using a nerve conduction study (NCS), to test nerve conduction speed, which requires expensive equipment and expert knowledge.
Associate Professor Yuta Sugiura and his colleagues at the Faculty of Science and Technology, Keio University , have developed a method for detecting signs of the condition by having patients play a game app for smartphones for 30 seconds to a minute that focuses on thumb movement. The team first had individuals without symptoms of the condition play the game to generate thumb tracking data. They then developed a program to detect potential signs of the condition by using that data for machine learning. Finally, they examined the accuracy of the predictions by comparing the thumb tracking data of a different group without symptoms versus a group diagnosed with carpel tunnel syndrome. They found that the accuracy of the approach was equivalent to or greater than a physical examination carried out by a specialist doctor.
Moving forward, the research team hopes to develop a system using the smartphone app to detect signs of the condition in those without significant symptoms or when a specialist doctor is not immediately available to encourage further diagnosis by a specialist. This methodology is also expected to help in screening patients for other rare conditions.
Players move the animals with their thumbs to pick vegetables, which appear one after another on screen from 12 directions. The players’ other fingers are held down in a harness so that only thumb movements are tracked.
Takafumi Koyama, Shusuke Sato, Madoka Torium, Takuro Watanabe, Akimoto Nimura, Atsushi Okawa Yuta Sugiura and Koji Fujita, “A Screening Method Using Anomaly Detection on a Smartphone for Patients With Carpal Tunnel Syndrome: Diagnostic Case-Control Study”, JMIR mHealth and uHealth, Published online March 14, 2021, doi: 10.2196/26320