ABSTRACT
The current and future challenge for medical big data analysis is to develop methods and systems to remove noise, extract meaningful features, and learn from the dataset. Generally, there are three main steps to develop such a system and to make biosignals useful in real-world settings. These include real-time data collection, data processing (e.g., feature extraction and classification) by computer, and biofeedback to apply the desired action. The requirements of a practical system include methods for signal processing, machine learning, and data analysis in large datasets collected from user populations in real-time and in combination with their health records. Learning applications of big data in the form of real-time acquisition with the background of the electronic healthcare record provide for the generation of new knowledge that will aid in identifying outcome and, hence, prognosis. This situation calls for developing multimodal analysis methods for big data such as machine learning, deep learning. In this chapter, we review these concepts with biomedicine applications.
