Robust Pre-Clinical Software System for Real Time NIRS and EEG Monitoring

This master's thesis presents the design and implementation of a real-time software system to support a bimodal NIRS and EEG brain imaging device. Real-time information on brain activity is an important factor in early detection and diagnosis at the top level of the cortex of various brain disorders. Current software systems provide limited real-time parameter adjustment and automated features for quick and easy analysis. The project presented in this master's thesis is part of the multidisciplinary IMAGINC research group, with the objective of developing a wireless, non-invasive and portable brain imaging system that allows imaging of the whole cortex in real time. The hardware system is capable of recording data from 128 NIRS and 32 EEG channels, as well as additional accelerometer and analog channels through the optodes and electrodes mounted onto the helmet. The software system acquires the real-time data from the hardware module using a wireless connection and displays the hemodynamic variations on the user interface. The change in hemodynamic activity is displayed on a 2D map of the brain, with selection of different views. Remote monitoring is also possible since the data can be transferred wirelessly to another computer. Through the user-friendly and intuitive user interface, the user can control and adjust various test parameters throughout the acquisition without any interruption. In order to achieve maximum illumination setting for individual subjects there is an automatic calibration function that quickly adjusts the illumination intensity for each of the emitters in just a few seconds. Previously defined NIRS and EEG configuration files (bipolar and referential montage) can be uploaded for easy testing. An automated analysis feature quickly analyzes and reports the status of all NIRS channels during the test to ensure good connection and valid results. The designed system can successfully record and process data for a continuous period of up to 24 hours. The results have been validated using similar NIRS data analysis software during figure tapping tasks and the hemodynamic variations were as expected.