Stanford Seminar - Computational Epidemiology: The Role of Big Data and Pervasive Informatics

"Computational Epidemiology: The role of big data and pervasive informatics" - Madhav Marathe of Virginia Tech

Colloquium on Computer Systems Seminar Series (EE380) presents the current research in design, implementation, analysis, and use of computer systems. Topics range from integrated circuits to operating systems and programming languages. It is free and open to the public, with new lectures each week.

Learn more: http://bit.ly/WinYX5

0:00 Introduction
0:46 Acknowledgements
1:07 Objectives for today's lecture
2:17 What is computational epidemiology
5:19 Epidemics in history
7:21 Recent example: Ebola outbreak in Africa
9:30 Goal: Real-time epidemic science
20:20 Mass action compartmental models
23:52 Pros and cons of compartmental models
25:37 An alternative approach: Networked Epidemiology
27:55 Amathematical framework: Graphical Dynamical Systems (GDS)
29:06 Epidemiological problems reduce to reasoning over the phase space P(G,F)
30:45 Pros and cons of networked epidemiology
31:51 Simdemics: A computing environment for real- time networked epidemiology
33:42 Elements of networked epidemiology
34:06 Realistic synthetic contact networks
35:05 Big-data challenge
36:29 Networks are dynamic & relational
39:33 Disease progression models
39:52 HPC simulations
42:10 Selected case studies
46:55 ILI prediction pipeline: Data driven statistical models
53:32 Vaccine allocation
54:31 Strategies for targeted vaccination
55:46 Performance of group based strategies
58:02 Summary and key insights
59:21 References