Tongue-Trackpad: An Augmentative and an Assistive Device Aimed at Enhancing the Digital Life

A successful implementation of assistive devices targeted for individuals with upper limb impairment, that enables interaction with smart digital interfaces, requires functionality which is as close as possible to a traditional mouse or trackpad. This thesis is focused on development of the Tongue-Trackpad, which is a wearable device that is readily recognizable by computer as a Bluetooth mouse. This device is controlled by swiping of the tongue on the surface of the upper palate. The device is composed of a matrix of sensors embedded inside a wearable oral retainer. This device features a Cypress PSoC 4 BLE chip which allows for fast development of capacitive sensing applications inside the humid environment of the oral cavity. The mouth environment, characterized by the presence of saliva, requires the employment of liquid tolerance sensing algorithm which allow high performance even when the saliva accumulation increases. Our trackpad works through self-capacitance detection, meaning a voltage is applied during scanning to a specific column/row and a shield signal is applied to all the other columns and rows. Hence allowing for the additive capacitance created by the presence of the saliva to be nullified. We have implemented this algorithm to a matrix of 7 rows by 5 columns of copper sensors, and also to a matrix of 5 rows and 5 columns. Hence, enabling a precise detection of the contact of the tip of the tongue with the wearable device. The Tongue-Trackpad is programmed to be recognized as a Human Interface Device (HID) mouse, which allows for its recognition by computers, tablets and smartphones (iOS and Android) without requiring any firmware modifications. The Tongue-Trackpad has been tested on an expert subject. The Tongue-Trackpad was evaluated while the user typed phrases using an onscreen keyboard, hence allowing for the assessment of the device as a pointer control and also allowing for evaluation of the typing speed. Preliminary results suggested that it was possible to achieve a maximum typing speed of 5.4 Correct Words Per Minute (CWPM), this speed was increased to 6.9 CWPM when Predictive Text algorithm of the onscreen keyboard was employed. The obtained results, are in general higher than others intraoral and assistive devices utilized by other research groups. In conclusion in this thesis we demonstrate the preliminary feasibility of using the Tongue-Trackpad as an assistive device to gain control over digital interactions with either smartphones, tablets and computers. The Tongue-Trackpad achieved a continuous control of the pointer and enabled performing a click gesture. It is important to note that the testing evaluation of the device on more subjects was not possible due the Covid-19 Pandemic.