DSP and Applications

Signal processing is a broad engineering discipline that is concerned with extracting, manipulating, and storing information embedded in complex signals. Signal processing applications consist of audio, speech, image, and video processing; graphics; biological & biomedical signals; computer vision; synthetic signals; astronomical signals. 

At the faculty, we have currently a research group doing an automatic sleep stage classification based on EEG signal. Sleep stage classification is a basic discipline of the sleep stage research area that is concerned to label a signal segment of 30 seconds such as EEG signal, ECG signal,… as one of 6 different sleep stages. This research is useful in supporting decision making in diagnostic studies of sleep disorders in experimental medicine. Currently, the methods are typically based on the analysis of visual signals by specialists. This work takes more time, effort and further classification results are not uniform depending on the professional knowledge of analyzer. Therefore, a signal processing system that can classify and annotate the sleep period will be extremely useful and help to analyze the process easier. 

Initially, the group had a result of features to perform simple single-channel EEG data and the Mahalanobis metric learning from training data for classification of sleep problems. In the future, the group will continue to develop applications for multi-channel EEG signals and to build different classifiers in order to increase accuracy of classification.

Group members: Phan Quốc Huy, Đỗ Đức Minh Quân, Đỗ Thế Luân, Cao Văn Hưng.