IBM contributes a tool to the open source community that may help contain the spread of emerging infectious diseases.
by Daphne Lawrence
In an increasingly connected world, the threat of a widespread disease outbreak looms large. On June 8, 2007, an advanced epidemiological modelling tool available freely and openly from IBM (Armonk, N.Y.) and the not-for-profit Eclipse Foundation (Portland, Ore.) was unveiled to help aid the understanding, and possibly preventing, the spread of emerging infectious diseases.
The Spatiotemporal Epidemiological Modeler (STEM) tool is designed to help scientists and public health officials create and use spatial and temporal models of emerging infectious diseases. "This is not a product," says Joseph Jasinski, M.D., program director of healthcare and life sciences at IBM Research. "We're trying to use the open source community to develop a robust model of what the earth looks like from an epidemiological perspective."
The STEM application has built in geographical information system (GIS) data for almost every country in the world, and includes data on borders, populations and airports. Developers, public health officials and researchers can plug their own models in to create a variety of "what if" scenarios to plan for disease outbreaks and share valuable information.
If bird flu became transmissible tomorrow, for example, public health officials would need to know their options. Should they close the airports or quarantine people? And if so, when? "If the model says you need to close your airport within one day of the first outbreak to have an effect, there's not much point in closing it after two days," says Jasinski. "If you can do those scenarios far enough in advance, you can actually develop feasible plans."
Currently in the research domain, STEM is mostly used by the CDC, WHO, individual scientists, and university and academic groups â€“â€“ but anybody can in principal participate in the project. Users can take the basic code, download it, and modify it with their proprietary information — then develop a model and/or publish research. To identify the best models, users can find test cases and compare them against a known data set. Because it's open source, users are welcome to find ways to improve the model, and all new code submitted is reviewed by a panel of experts. "If you see something is wrong, you can contest and try and change it," Jasinski says. "It's a community consensus view of how models should develop."
IBM, which has had a longstanding history with open software like Linux and Java, initiated the model and donated it. Jasinski says that STEM is more than giving something away, because it's also a way to maintain a code base. The project is global — some pieces of it are in more than one language.
From a clinical care or hospital administrator perspective, the STEM project is part of the larger Eclipse open healthcare framework. Eclipse is a not-for-profit, member supported corporation that helps cultivate both an open source community and an ecosystem of complementary products and services. Though STEM is focused on public health modeling of diseases, other parts of the Eclipse project revolve around healthcare interoperability: some of Eclipse's technology was used in the pilot of the national health information network that IBM (and other vendors) delivered to the government in January of this year.
The other parts of the Eclipse project are all open standards based. "What we're building is a community of developers who are contributing applications, pieces of code, pieces of middleware, connectors, that hopefully will ultimately make it very easy for both clinical and public health systems to operate and share data," says Jasinski. "That's the bigger goal of the whole project."
To download STEM, go to http://www.eclipse.org/ohf/components/stem/