An Intraoral Camera for Supporting Assistive Devices

Abstract

Thousands of patients around the globe are affected by paralysis which hinders the fulfilment of their basic needs such as mobility and speech. Several research topics have been dedicated to improve the livelihood of paralytic patients and a small subset of the topics has focused on capturing inputs from the tongue. The tongue is a muscular organ directly connected to the brain through a cranial nerve known as the hypoglossal nerve which is responsible for the motor functions of the tongue. Hence, tongue movements are not affected during spinal cord injuries, which is one of the major causes of paralysis. Given the importance of capturing inputs from the tongue, this research proposes a novel method of using an intraoral camera for this purpose. It discusses the methods used for capturing the images with the help of an Endoscope camera. It explains how the features were extracted in real-time using image processing techniques on each captured frame and how the orientation and position of the tongue was then accurately classified into one of the 11 possible categories to produce specific outputs which could be used by paralytic patients as inputs to any external system. After testing the system with a data entry application, an average of 19.34 correct entries per minute was calculated from 5 different experiments, and an average error rate of 3.96% was obtained, which outperforms systems such as the Resistopalatography and the MouthPad in terms of accuracy.