LIRNEasia has been at the forefront of big data analysis for development in Sri Lanka, conducting in-house analysis to generate actionable insights across a range of policy domains. On 6th May 2016, LIRNEasia and the Health Informatics Society of Sri Lanka jointly convened a planning meeting on building better models for forecasting the propagation of infectious disease such as dengue in Sri Lanka. The meeting was intended to lay the foundation for a multi-disciplinary collaboration engaging health informatics specialists, epidemiologists, and data scientists to identify research priorities and opportunities. The participants included the following: Madhushi Bandara, Junior Researcher, LIRNEasia Prof Vajira Dissanayake (Health Informatics Society of Sri Lanka, Biomedical Informatics Programme – Postgraduate Institute of Medicine) Dr. M.
One local telco CEO recently whined about being viewed as a cigarette manufacturer. “Everybody wants to tax us, as if mobiles are a product more hazardous than cigarettes. Tobacco kills, mobiles don’t; communication facilitates better living conditions and saves environment because it reduces transport. It is gross unfair both are seen in the same light.” As Wikipedia tells us, cigarettes are a significant source of tax revenue in many localities.
Assume a scenario where among the chief complaint strings of two unrelated patients in the same District on the same date there was a mention of bloody stools in pediatric cases. The multiple mentions of “bloody stools” or “pediatric” might not be surprising, but the tying together of these two factors, given matching geographic locations and timings of reporting, is sufficiently rare that seeing only two such cases is of interest. This was precisely the evidence that was the first noticeable signal of the tragic Walkerton, Canada, waterborne bacterial gastroenteritis outbreak caused by contamination of tap water in May 2000. That weak signal was spotted by an astute physician, not by a surveillance system. Reliable automated detection of such signals in multivariate data requires new analytic approaches.