A Neural Network-Based Lower Extremity Joint Angle Estimation from Insole Data

Alemayoh, Tsige Tadesse; Lee, Jae Hoon; Okamoto, Shogo

doi:10.1109/ur57808.2023.10202438

Cited by 1 publication

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, capacitive- and resistive-based sensors can be disadvantageous for certain biomedical applications as they require high power usage [ 2 , 3 ], are usually heavily wired, and are thus more sensitive to electromagnetic interference and temperature and humidity fluctuation [ 7 , 8 , 9 ]. Conversely, the accuracy of piezoelectric sensors in distinguishing shear forces from other compressive forces is still inconsistent despite recent improvements in these technologies [ 10 , 11 ]. Compared to capacitive and resistive-based sensors, optical-based sensors would have certain advantages for medical applications because they are unaffected by electromagnetic fields and are more stable in relation to environmental factors like temperature and humidity [ 12 , 13 ].…”

Section: Introductionmentioning

confidence: 99%

An Optical Sensor for Measuring Displacement between Parallel Surfaces

Ahamed,

McGeehan,

Ong

2024

Sensors

View full text Add to dashboard Cite

An optoelectronic sensor was developed to measure the in-plane displacement between two parallel surfaces. This sensor used a photodetector, which was placed on one of the parallel surfaces, to measure the intensity of the red (R), green (G), blue (B), and white/clear (C) light spectra of a broad-spectrum light that was reflected off a color grid on the opposing surface. The in-plane displacement between these two surfaces caused a change in the reflected RGB and C light intensity, allowing the prediction of the displacement direction and magnitude by using a polynomial regression prediction algorithm to convert the RGB and C light intensity to in-plane displacement. Results from benchtop experiments showed that the sensor can achieve accurate displacement predictions with a coefficient of determination R2 > 0.97, a root mean squared error (RMSE) < 0.3 mm, and a mean absolute error (MAE) < 0.36 mm. By measuring the in-plane displacement between two surfaces, this sensor can be applied to measure the shear of a flexible layer, such as a shoe’s insole or the lining of a limb prosthesis. This sensor would allow slippage detection in wearable devices such as orthotics, prostheses, and footwear to quantify the overfitting or underfitting of these devices.

show abstract