Deep learning model for sepsis prediction using high-frequency data

Sepsis is a potentially life-threatening condition caused by the body's response to an infection. The body normally releases chemicals into the bloodstream to fight an infection. Sepsis occurs when the body's response to these chemicals is out of balance, triggering changes that can damage multiple organ systems. Our main goal here is to train a deep learning model in python using all of its symptoms for the prediction of early onset of sepsis. Depending upon the values fed into the application, a doctor should get a good idea whether a person is susceptible to sepsis and get an early alert which can be critical for diagnosis. The application should be able to make these predictions using only a minimal set of streaming physiological data in real-time. During the course of this project, new deep learning methods, using temporal convolutional neural networks or quasi RNN, a model will be developed to identify markers that predict the onset of sepsis in patients admitted to the intensive care unit. We shall develop this application using the eICU database.