Bayesian Methods and Deep Learning for Healthcare

Maksym Korablyov

Our November meeting will feature two talks. Prof. Gil Alterovitz, of Harvard Medical School, will talk about bayesian methods in healthcare and clinical genomics. Maksym Korablyov, software engineer at Boston Children’s hospital and visiting student at Prof. Alterovitz's lab, will talk about neural networks for deep learning of proteins and a Kaggle competition they are planning to launch in early 2017.

First Talk

Title:

Applying Bayesian methods for point of care and translational research clinical genomic apps.

Abstract:

This talk will discuss how new interoperability technologies are enabling clinical genomics to be used for clinical genomic point of care applications and translational research- and use of Bayesian methods to aid in clinical decision support.

Speaker Bio:

Prof. Gil Alterovitz is a faculty at Harvard Medical School in biomedical informatics. He is Director of the Biomedical Cybernetics Laboratory and core faculty member of the Computational Health Informatics Program. He is also affiliated with the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology. He is co-chair of the HL7 Clinical Genomics workgroup and on the Clinical Workgroup core executive group of the Global Alliance.

His research focuses on integrating genomics within healthcare settings as well as international collaborations involving genomics/clinical data integration. His work on integrative methods for “big data” in the biomedical informatics space has been published or presented in more than 30 peer-reviewed publications ranging from academic journals and international conferences to three books (including “Systems Bioinformatics: An Engineering Case-based Approach,” ranked #1 in new Amazon bioinformatics category). His lab site is at: http://bcl.med.harvard.edu

Second Talk:

Title:

Harvard affinity project, - deep convolutional neural networks that learn from 3D images of proteins.

Abstract:

I will give a short overview of the project that I and other students have been working on under the supervision of Dr. Gil Alterovitz at Harvard/MIT Division of Health Sciences and Technology for the last few months. I will demonstrate how neural networks can learn from static images of drug-protein complexes, introduce multitask neural networks, and show techniques of pre-training on low-quality artificially generated data. After that I will feature microsecond-long live videos of moving protein molecules and show how molecular dynamics can be used to learn protein flexibility in drug design.I hope to receive feedback from the audience before Kaggle challenge we plan to launch in the spring of 2017.

Speaker Bio:

Maksym Korablyov is a software engineer at Boston Children’s hospital and visiting exchange student at Harvard Medical School. His research is supported by JW Fulbright grant. Maksym obtained MS in bioinformatics from Georgia Tech followed by two by courtesy appointments at Georgia Institute of Technology, and at Brandeis University, both in physics.

Event Location:

MIT Stata Center, Building 32, Star Conference Room, 32-D463

32 Vassar St, Cambridge, MA 02139

Directions to Star Conference Room, 32-D463:

The Star Conference room is on the 4th floor of the Dreyfoos Wing in the Stata Center (Building 32). Enter Building 32 at the front entrance and proceed straight ahead; there will be elevators to the right. Take the elevators to the 4th floor; exit to the left and then turn right at the end of the elevator bank. At the end of the short corridor, turn right, just before the R&D Dining room. The Star Conference Room is straight ahead, just past a set of stairs.

See a map of Stata Center Here: http://www.na-mic.org/Wiki/index.php/File:Kiva-Star.jpg

Agenda:

6:30pm : Networking

7:00pm: Fist Talk by Gil Alterovitz

7:40pm: 5 min break

7:45pm: Second Talk by Maksym Korablyov

8:25pm: Networking

9pm: End of Event