Novel algorithm for detection of cognitive dysfunction using neural networks

According to the World Health Organization (WHO), the number of cases of neurodegenerative diseases will double by 2030 and triple by 2050, becoming an alarming problem for the health systems. Cognitive dysfunction is one of the first symptoms of neurodegenerative diseases, so cognitive dysfunction detection contributes to the early detection of these diseases. On the other hand, the electroencephalogram (EEG) is a non-invasive test that records the electrical activity of the brain and has a wide field of applications in the medical field, one of which is the detection of cognitive dysfunction. The combination of signal processing tools, feature extraction, and artificial intelligence applied to EEG information allows the creation of helpful tools for the automatic detection of cognitive dysfunction. This work aimed to implement an efficient and robust computational system focused on processing EEG signals through Wavelets and the automatic classification of cognitive dysfunction using neural networks. The proposed method achieved an efficiency of 97.5% and an F1 of 98.42% using 10 electrodes for EEG recording.