Related Experiment Video
Updated: Feb 28, 2026
08:02
A Mouse Ear Model for Allergic Contact Dermatitis Evaluation
Published on: March 24, 2023
5.1K
Detection of Nocturnal Scratching Movements in Patients with Atopic Dermatitis Using Accelerometers and Recurrent
IEEE Journal of Biomedical and Health Informatics
|June 15, 2017
Summary
This study developed an algorithm using wrist actigraphy to accurately detect nocturnal scratching in patients with atopic dermatitis. The method distinguishes scratching from other movements, aiding in objective assessment of pruritic conditions.
Area of Science:
- Biomedical Engineering
- Dermatology
- Computational Neuroscience
Background:
- Atopic dermatitis is a chronic inflammatory skin condition causing significant pruritus (itching) in both children and adults.
- Objective quantification of nocturnal scratching is crucial for developing effective therapies for atopic dermatitis and other pruritic disorders.
- High-resolution wrist actigraphy offers a noninvasive method for recording nocturnal movement patterns.
Purpose of the Study:
- To develop and validate an algorithm for accurately detecting nocturnal scratching events using wrist actigraphy data.
- To differentiate scratching movements from other nocturnal motions.
- To compare the algorithm's performance against a reference method and manual scoring.
Main Methods:
- A twofold algorithm was developed: data segmentation into "no motion," "single handed motion," and "both handed motion," followed by classification using a bidirectional recurrent neural network.
- The algorithm was trained and tested using high-resolution wrist actigraphy data (20 Hz, 3D accelerometer).
- Performance was validated against manually scored infrared video data from 24 subjects (6 healthy, 18 atopic dermatitis patients).
Main Results:
- The developed algorithm achieved a score of 0.68 and a rank correlation of 0.945 when compared to manual video scoring.
- The algorithm significantly outperformed a previously published wrist actigraphy-based method, which obtained a score of 0.09 and a rank correlation of 0.466.
- Scratching movements were accurately discriminated from other nocturnal movements.
Conclusions:
- The proposed algorithm effectively detects nocturnal scratching events using wrist actigraphy.
- This objective method provides a valuable tool for assessing pruritus in atopic dermatitis and related conditions.
- The findings support the use of advanced actigraphy analysis for objective measurement in dermatological research.
