Dataset for toothbrushing activity using brush-attached and wearable sensors

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Health Informatics

Background:

• Maintaining oral hygiene is crucial for overall health.
• Poor toothbrushing techniques contribute significantly to dental issues like tooth decay and gum disease.
• Current technological solutions for assessing toothbrushing quality are limited.

Purpose of the Study:

• To address the challenge of assessing toothbrushing quality by framing it as a human activity recognition (HAR) problem.
• To extend a previous dataset by incorporating more experiments and a new sensor for toothbrushing activity analysis.
• To investigate and analyze the collection of a comprehensive dataset for toothbrushing activity recognition.

Main Methods:

• Utilized an Inertial Measurement Unit (IMU) sensor to gather time-series data during toothbrushing.
• Recruited 22 healthy participants to collect data in diverse settings and using both electric and manual toothbrushes.
• Recorded a total of 120 toothbrushing sessions, employing both brush-attached and wearable sensors.

Main Results:

• Collected an expanded dataset of toothbrushing activities using novel sensor configurations.
• Analyzed the feasibility of using HAR techniques for evaluating toothbrushing adherence.
• Established a foundation for developing technology to assess and improve toothbrushing quality.

Conclusions:

• The study successfully collected and analyzed a comprehensive dataset for toothbrushing activity recognition.
• Framing toothbrushing quality assessment as an HAR problem shows promise for technological intervention.
• The findings contribute to the development of tools for promoting better oral hygiene through improved toothbrushing practices.