Accurate prognosis and prediction of a patient's current disease state is critical in an ICU. The use of vast amounts of digital medical information can help in predicting the best course of action for the diagnosis and treatment of patients. The proposed technique investigates the strength of using a combination of latent variable models (latent dirichlet allocation) and structured data to transform the information streams into potentially actionable knowledge. In this project, I use Apache Spark to predict mortality among ICU patients so that it can be used as an acuity surrogate to help physicians identify the patients in need of immediate care.
These instructions give you guidelines for preparing papers for IFAC technical meetings. Please use this document as a template to prepare your manuscript. For submission guidelines, follow instructions on paper submission system as well as the event website.
This Project posits elementary analogies of existing Probabilistic and Machine Learning models that have been used to find solutions to the problem of the Structural Segmentation of Musical audio. I have tried to use the idea that the chord of a given beat or frame of a song is an analogous representation of the states generated by trained Hidden Markov Models in generating feature vectors for the aforementioned problem; and that the knowledge of the temporal boundaries within which, a group of frames lie, can be used as constraints in creating the feature vectors that are eventually clustered to identify the pattern in which the various segments of a song repeat.