Wearable Movement Exploration Device with Machine Learning Algorithm for Screening and Tracking Diabetic Neuropathy-A...

Area of Science:

• Biomedical Engineering
• Medical Technology
• Machine Learning in Healthcare

Background:

• Diabetic neuropathy is a common diabetes complication.
• Early detection and tracking are crucial for management.
• Novel devices are needed to improve accessibility and patient experience.

Purpose of the Study:

• To evaluate the Moveo device for detecting and tracking diabetic neuropathy.
• To assess the accuracy and usability of a machine learning-based system.
• To compare the Moveo device with traditional electrodiagnostic examinations.

Main Methods:

• A pilot study involving 23 participants (11 with diabetes, 11 with suspected radiculopathy, 4 healthy).
• The Moveo device uses sensors and a mobile app to collect movement data, processed by a cloud-based machine learning algorithm.
• Participants underwent both Moveo and electrodiagnostic examinations; user experience questionnaires were administered.

Main Results:

• The Moveo device achieved 82.1% overall accuracy, with 78% sensitivity and 87% specificity.
• A high linear correlation (up to 0.722) was found between Moveo and electrodiagnostic examination (EDx) features.
• Most participants preferred the Moveo device due to its less painful nature.

Conclusions:

• The Moveo device is an accurate and user-friendly tool for detecting diabetic neuropathy.
• Its suitability for home environments shows significant potential for future clinical application.
• Moveo offers a promising, accessible alternative for managing diabetic neuropathy.