Intelligent Brushing Monitoring Using a Smart Toothbrush with Recurrent Probabilistic Neural Network.

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Sensor Technology

Background:

• Smart toothbrushes utilize inertial sensors for personalized oral health monitoring.
• Real-time processing of nine-axis inertial sensing for toothbrush posture recognition demands significant computational resources.
• Existing deep learning models like CNN and LSTM face challenges with high computational requirements.

Purpose of the Study:

• To propose a Recurrent Probabilistic Neural Network (RPNN) for efficient and accurate toothbrush posture recognition.
• To develop a system requiring low computational resources for smart toothbrush applications.
• To enable real-time monitoring of brushing technique correctness and integrity, specifically the Bass Brushing Technique.

Main Methods:

• Implementation of a Recurrent Probabilistic Neural Network (RPNN) model.
• Training the RPNN for toothbrush posture recognition and brushing position identification.
• Evaluating the RPNN's performance against Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) models.

Main Results:

• The RPNN achieved a recognition accuracy of 99.08% in experimental evaluations.
• RPNN demonstrated superior accuracy compared to CNN (+16.2%) and LSTM (+21.21%).
• The proposed model significantly reduces the computational power required for processing inertial sensor data.

Conclusions:

• The RPNN is a highly accurate and efficient model for smart toothbrush posture recognition.
• The low computational resource requirement of RPNN allows for direct implementation on smartphones.
• This advancement facilitates accessible, personalized oral health care through smart toothbrush technology.