A Sensorised Glove to Detect Scratching for Patients with Atopic Dermatitis

Area of Science:

• Biomedical Engineering
• Dermatology
• Machine Learning

Background:

• Atopic dermatitis (AD) management relies on subjective assessments of itch and scratching.
• Objective quantification of scratching is needed for accurate severity classification and treatment optimization.

Purpose of the Study:

• To develop and validate a wearable device for objective measurement of scratching in AD patients.
• To provide clinicians with quantifiable data on scratching frequency and duration.

Main Methods:

• Development of the SensorIsed Glove for Monitoring Atopic Dermatitis (SIGMA), a lightweight glove with microtubular stretchable sensors and an inertial measurement unit (IMU).
• Integration of sensor data with a machine learning model for scratch detection.
• Validation of the SIGMA device through algorithm accuracy testing and a pilot study in children with AD.

Main Results:

• The scratch prediction algorithm achieved 83% accuracy in validation.
• A 30-minute controlled trial demonstrated 99% accuracy.
• In a pilot study with 6 children, SIGMA detected 94.4% of scratching events.

Conclusions:

• SIGMA offers a quantifiable and objective method for assessing scratching in AD.
• The device has the potential to improve objective assessment of itch severity and guide personalized treatment decisions.
• This technology can empower dermatologists with better tools for managing AD.