Presentation is loading. Please wait.

Presentation is loading. Please wait.

Machine Learning for Individualized Medicine

Similar presentations


Presentation on theme: "Machine Learning for Individualized Medicine"— Presentation transcript:

1 Machine Learning for Individualized Medicine
Bachelor-/Master thesis Contact: Prof. Dr.-Ing. G. Dartmann, Bachelor-/Master thesis Contact: Prof. Dr.-Ing. G. Dartmann Prof. Dr.-Ing. A. Schmeink Motivation Use Case Methodology This project is an innovative, IT-based decision support system for individualized risk stratification, monitoring and therapy management in intensive care medicine, which is developed and prototypically implemented in this R&D project and its usability tested in a user study. It links a patient's data with medical expertise and guideline knowledge as well as patient condition and treatment pathway patterns obtained from data by pattern recognition. The project uses advances in the digitalization of the health care system and new technologies for automated data analysis to achieve treatment paths that are better adapted to the patient and thus avoid undesired long-term consequences such as the need for care and long-term respiration. Due to its learning ability, the system allows to continuously learn new knowledge from the data and to store it in a research knowledge database separate from the productive knowledge database. We investigate a novel holistic approach for a machine learning in the intensive care unit. Methods Data Analytics & Machine Learning Principle of Machine Learning System identification with different models Prediction of future states, early warning Hybrid Petri Nets Neural Networks Unsupervised and supervised learning Decision Making P. Vieting, R. C. de Lamare, L. Martin, G. Dartmann and A. Schmeink, An Adaptive Learning Approach to Parameter Estimation for Hybrid Petri Nets in Systems Biology, IEEE Statistical Signal Processing Workshop (SSP), 2018 Tasks Requirements Research Questions and Tasks Students Natural language processing Hybrid models for modelling diseases and patients GUI programming for patient data management systems Data-Analytics and anomaly detection Information technology Computer Science Cooperation Partners University Hospital Aachen, OIM RWTH Aachen University, ICE & ISEK Trier University of Applied Sciences


Download ppt "Machine Learning for Individualized Medicine"

Similar presentations


Ads by Google