About the Speaker
Professor Mathukumalli Vidyasagar FRS is a leading control theorist and a Fellow of Royal Society. Currently a Distinguished Professor in Electrical Engineering at IIT Hyderabad, previously he was the Cecil & Ida Green Chair of Systems Biology Science at the University of Texas at Dallas, Executive vice-president at Tata Consultancy Services (TCS) where he headed the Advanced Technology Centre, and the director of Centre for Artificial Intelligence and Robotics (CAIR), a DRDO defence lab in Bangalore. In 2017 University of Wisconsin named him as one of the 125 "People of Impact" from around the world during the 125th anniversary of the Department of Electrical Engineering
Title of the Talk
"Machine Learning Methods in Computational Cancer Biology".
Can a team of engineers who do not themselves carry out experiments or treat cancer patients contribute anything useful to the development of "personalised" medicine for cancer (or as it is now called, "precision therapy" for cancer)? In this talk he will show that the answer is definitely YES! At present there is sufficient amount of public domain data in cancer that interested researchers can develop their own algorithms on the data and cross-validate it on independent data. However, working with clinicians (who actually treat patients) would enhance the impact of the research. The biggest challenge in applying machine learning methods to cancer biology data is that the data is noisy, non-repeatable, and full of errors. Thus "off the shelf" algorithms developed for "clean" engineering data do not work on biology data. He will point out some common sources of error, and how algorithms can be developed that are specifically tailored for such scenarios.