Deep Neural Network method for Sleep Stages Classification using spectrogram of signal based on transfer learning with different domain data

Sleep stages Classification is a useful way to diagnose sleep problems. This based on the processing of bio-signals (ECG, EEG, EOG, PPG). The less complex this signal is, better detection and processing. Feature extraction methods using hand are tedious long lasting. Extraction features without intervention deep features, which usually extracted from images. Analysis time-frequency characteristics non-static very important has information. In study, image was ECG spectrogram were convolutional neural network. After extracting classified transfer learning method. Network training performed one testing with other channel.The results show that it possible detect acceptable accuracy different amplitudes signals. detected 98.92% 96.52% sensitivity.