Distinguished Lecture: The Knowledge Revolution in Healthcare
Healthcare is on the verge of a revolution, which will be driven by knowledge. This knowledge is being derived both by traditional clinical studies and also by mining vast quantities of patient data. Today’s medicine is “one size fits all”, in that treatment and diagnostic decisions are based on set of guidelines that are uniformly applied to the patient population. The use of knowledge by decision-support systems to personalize the delivery of care has the potential to transform the practice of medicine.
There are two fundamental challenges that must be overcome:
The use of knowledge: There is a great need for computerized decision-support systems that can automatically combine relevant knowledge with patient data to assist the physician at the point of care. Widespread use of knowledge-driven systems is hampered by the poor quality of patient data; most of the useful clinical information in patient records is recorded in unstructured form (images, free text, mass-spect arrays) and existing decision-support systems work primarily with structured data. In this talk, we discuss how machine learning & probabilistic inference methods can help deal with the problems of applying medical knowledge to unstructured patient records.
The discovery of new medical knowledge: The personalization of medicine is being further fueled by recent advances in gene sequencing and molecular imaging technology. Machine learning methods are being used to discover biomarkers – these are essentially, classifiers that extract information from patient data, and can be used to predict the risk of disease, to diagnose a patient, to provide patient prognosis conditioned on different treatments (and thus select therapy), and to monitor the impact of a therapy. In addition to being customized to the individual patient characteristics (in theory, down to the genome), decision-support systems must also be personalized with respect to what is known about each patient (i.e., not all patients come with the same tests, diagnoses, etc.). Here, we discuss some of the particular machine learning challenges involved with learning from medical data (some of these aspects were covered in KDD Cups 2006 & 2008).
We conclude by identifying some challenges for the data mining community.
R. Bharat Rao is the Head of the Knowledge Solutions group for the Image and Knowledge Management (IKM CKS) Division in Siemens Healthcare , Inc, a subsidiary of Siemens AG . The Knowledge Solutions group of Siemens Healthcare is based in Malvern, PA, USA, and focuses on developing products and services that (a) help improve patient outcomes by integrating medical knowledge with various parts of a patient record (free text, images, labs, pharmacy, genomics, etc.), and (b) supports the increasing drive to personalize medicine.
Dr. Rao received a B.Tech in Electronics Engineering from the Indian Institute of Technology, Madras in 1985, and an M.S. and Ph.D. focusing on machine learning from the Dept. of Electrical Engineering, University of Illinois, Urbana-Champaign, in 1993. He joined Siemens Corporate Research in 1993, and formed the Data Mining group there in 1996. In 2002, he moved to Siemens Healthcare to help found the “Computer-Aided Diagnosis” group (currently the “CAD & Knowledge Solutions” Group in IKM).
Dr. Rao's research interests include probabilistic inference, machine learning, natural language processing, classification, and graphical models, with a focus on developing decision-support systems that can help physicians improve the quality of patient care. He is particularly interested in the development of novel data mining methods to collectively mine the structured and unstructured parts of a patient record and the automatic integration of medical domain knowledge into the mining process. He has published over 100 papers in peer-reviewed scientific journals and conferences in machine learning and medicine and has filed over 50 patents. In 2005, Siemens honored him with its "Inventor of the Year" award for “outstanding contributions related to improving the technical expertise and the economic success of the company” for developing REMIND (Reliable Extraction and Meaningful Inference from Nonstructured Data). REMIND is a platform that supports both the integration of knowledge into medical decision-support, as well as the discovery of novel medical knowledge to support personalized medicine. He has since received the inaugural IEEE Data Mining Practice Prize for the best deployed industrial and government data mining application.
Dr. Rao's passions outside of the sphere of Science and Business include the sport of Cricket, Classic Rock, the history of Science, and the study of Philosophy and Religion. He is married and has two children.