A knowledge-based modeling for plantar pressure image reconstruction.

Area of Science:

• Biomedical Engineering
• Biomechanics
• Medical Device Technology

Background:

• Prolonged plantar pressure is a primary cause of foot ulcers.
• Current in-shoe electronic pressure monitoring systems have limited sensor counts (K < 10) due to cost and energy constraints.
• Reliability of these systems depends on sensor spatial resolution.

Purpose of the Study:

• To develop a knowledge-based regression model (SCPM) for reconstructing continuous plantar pressure images from limited sensor data.
• To improve the spatial resolution of in-shoe pressure monitoring systems.
• To bridge the gap between technological limitations and medical needs in foot ulcer prevention.

Main Methods:

• Utilized a knowledge-based regression model (SCPM).
• Trained a per-subject regression function using high-resolution clinical pressure data.
• Reconstructed spatially continuous plantar pressure images from a small number of sensors.

Main Results:

• SCPM outperformed other tested interpolation methods for K < 60 sensors.
• Achieved less than one-third of the error compared to other methods when using K = 10 sensors.
• Demonstrated effective reconstruction of plantar pressure from limited sensor data.

Conclusions:

• SCPM significantly enhances the spatial resolution of in-shoe pressure monitoring.
• The model improves the accuracy of plantar pressure mapping, aiding in foot ulcer prevention.
• SCPM facilitates wider adoption of sensing insoles for various medical applications.