Ģ��ֱ��

Ģ��ֱ�� dissertation studied deep learning methods for detecting and predicting epileptic seizures. Accurate detection of seizures could improve the quality of life for people with epilepsy. Study could help in developing portable devices that could effectively predict seizures in the future.

In his dissertation M.Sc. (Tech) Xiaoshuang Wang studied effective digital seizure detection and prediction methods by combining electroencephalogram (EEG) signals and deep learning techniques. Research on epileptic seizure detection and prediction can help computer-aided epilepsy diagnosis.

EEG, as a significant tool, has been widely utilized in the diagnosis of epilepsy. However, the use of EEG signals to detect and predict seizures is still challenging.

“The accurate detection and prediction of seizures will greatly reduce the suffering and improve the quality of life for people with epilepsy, because it can provide a time frame for people to take interventions to suppress the onset of seizures”, Wang says.

In Wang’s mind it is important to seek effective seizure detection and prediction methods based on EEG signals.

“Recently, deep learning and AI techniques have shown remarkable performances in image recognition and computer vision. Now we are using deep learning techniques also in our work.”, Wang explains.

Study might help in development of portable EEG equipment in seizure prediction

In his thesis Wang successfully applied deep learning techniques in EEG signals for seizure detection and prediction.

Considering the one-dimensional characteristics of EEG signals (time series), one-dimensional convolutional neural network (1D-CNN) is mainly applied for the analysis of seizure detection and prediction in this dissertation.

The methods of channel selection combined with deep learning related techniques showed remarkable performances in seizure prediction.

“In the future this may contribute to the development of portable EEG equipment with a reduced number of channels to effectively predict seizures in the future.”, Wang says.

M.Sc. (Tech) Xiaoshuang Wang defends his doctoral dissertation in Software and Communications Engineering "EEG-Based Detection and Prediction of Epileptic Seizures Using One-Dimensional Convolutional Neural Networks" 16 December. The audience can follow the dissertation online. Link to the online event:

Opponent Professor Zhiguo Zhang (Harbin Institute of Technology, China) and Custos Professor Tommi Kärkkäinen (Ģ��ֱ��). The doctoral dissertation is held in English.

Link to publication: