Talk "Machine Learning, Data Science and Decisions for a Better Planet"
Mihaela van der Schaar, Oxford – Man Institute of Quantitative Finance (OMI)
Machine Learning and Data Science are making great leaps forward with the goal of changing the world for the better. This talk describes some of my work that contributes to this goal.
The main focus of the talk is on the development and application of new machine learning methods to timely intervention for hospitalized patients. More than 200,000 hospitalized patients have cardiac arrests each year in the US, and 75% of those die. Many of these patients could be saved by timely risk assessment and intervention: transfer to an intensive care unit (ICU) followed by appropriate treatment. To enable this, I develop a new, versatile probabilistic model – the Hidden Absorbing Semi-Markov Model (HASMM) – that uses the modern electronic health record (EHR) data to calculate a risk score for recommending timely decisions on transfer of patients from hospital wards to the ICU. Unlike existing models, HASMM accommodates irregularly sampled, temporally correlated, and informatively censored physiological data, and can describe non-stationary clinical state transitions. Learning the parameters of an HASMM from EHR data is achieved via a novel forward-filtering backward-sampling Monte-Carlo EM algorithm that exploits the knowledge of the end-point clinical outcomes (informative censoring) in the EHR data, and implements the E-step by sequentially sampling the patients’ clinical states in the reverse-time direction while conditioning on the future states. This method has been extensively tested and significantly outperforms all previous methods (e.g. the Rothman Index) and is currently under implementation at the UCLA Ronald Reagan Medical Center.
Other medical applications include breast cancer diagnosis and treatment; non-medical applications include personalized education, finance and smart cities.
Mihaela van der Schaar is the Man Professor of Quantitative Finance, Oxford-Man Institute, Department of Engineering Science, University of Oxford. Her research interests include 1) Machine learning for medicine; 2) Machine learning for a better Planet (including education etc.); 3) Quantitative Finance; 4) Data science and decisions; 5) Game Theory; 6) Social, economic and biological networks. Before coming to Oxford, she was Chancellor's Professor of Electrical Engineering at University of California, Los Angeles where she founded the UCLA Center for Engineering Economics, Learning, and Networks. She is an IEEE Fellow (2009) and has been a Distinguished Lecturer of the Communications Society, the Editor in Chief of IEEE Transactions on Multimedia, and member of the Senior Editorial Board member of IEEE Journal on Emerging and Selected Topics in Circuits and Systems (JETCAS) and IEEE Journal on Selected Topics in Signal Processing (JSTSP). She received an NSF CAREER Award (2004), the Best Paper Award from IEEE Transactions on Circuits and Systems for Video Technology (2005), the Okawa Foundation Award (2006), the IBM Faculty Award (2005, 2007, 2008), the Most Cited Paper Award from EURASIP: Image Communications Journal (2006), the Gamenets Conference Best Paper Award (2011) and the 2011 IEEE Circuits and Systems Society Darlington Award Best Paper Award. She received three ISO awards for her contributions to the MPEG video compression and streaming international standardization activities, and holds 33 granted US patents.