DEEP Learning Approach Using Modified Auto-Encoder for PPG Signal Compression

PPG (Photoplethysmography) signal measures the heart's pumping capacity which uses a light-emitting diode to track the blood flow which controls the actions of the heart rate. PPG is widely used in all healthcare applications, which range from monitoring a person's oxygen saturation level within the ICU (intensive care units) for measuring the pulse rate in preferably devices that are wrist-worn. In this research, we formulate an algorithm called as bioelectrical signals with encoder and decoder and feature engineering). This combined method incorporates a method by combining FE and encoder-decoder. In the end, our algorithm differentiates between relevant tasks and gets all the features required for classification. Our proposed method accomplishes the feature engineering on higher-level learning features; this also ensures encoder and decoder for constructing the generative units. The experimental outcome of our proposed model sets a benchmark to state that our model works better in comparison with existing methods.