Retrieving information, even small bits from vast databases, is something York information technology Professor Jimmy Huang is skilled at doing. He’s now using his knowledge to help make Canada’s health-care system more cost-effective and efficient while boosting the quality of care, with the assistance of several grants, including an Early Researcher Award and, more recently, a Petro-Canada Young Innovators Award.
The Petro-Canada Young Innovators Award, worth $7,500, was awarded this year to help fund Huang’s project, titled “A Hybrid Framework for Discovering Biological Relationships from Large-Scale Biomedical Text Documents.” The Ontario Ministry of Research & Innovation’s Early Researcher Award, worth $150,000 over five years, was awarded to Huang last year for his project “Analyzing and Searching Medical Data for Cost-Effective Health Care.”
Both projects are similar in that they require Huang, a professor in the Atkinson Faculty of Liberal & Professional Studies School of Information Technology, to develop data search tools to mine for relevant information, but their purposes are slightly different. For the Petro-Canada Young Innovators Award, Huang is working on developing software that would search for research that has already been completed, or is in the process of being conducted, on a specific topic. The idea is to develop a search tool that would ultimately eliminate any duplication of research and, at the same time, provide researchers with the most up-to-date information. This would assist them in taking their research to new levels.
Above: Jimmy Huang (centre rear) with his research team
“Right now, everyone does all their research separately,” says Huang, founding director of York’s Information Retrieval & Knowledge Management Research Lab, which is currently supported by the Natural Sciences & Engineering Research Council of Canada and the Ontario Ministry of Research & Innovation. “They don’t share.” One person might find that a certain protein affects lung cancer, while another researcher may discover a particular food affects the way that protein reacts with cancer cells. If there was a tool that searched out meaningful correlations between medical concepts or terms, then researchers could focus on an area not already tackled. That could further increase medical understanding and knowledge, and eventually affect diagnosis and treatment.
With the Early Researcher Award, Huang wants to change the emphasis on keeping paper records of patient information to making everything electronic. Right now, every hospital has its own database. They are not linked to each other, and that, says Huang, limits information sharing and knowledge building. As that changes and the medical databases grow, so too does the need for an advanced information retrieval system.
“How to manipulate, analyze and understand large quantities of complex data becomes extremely important. Currently, most of the valuable information stored in medical information systems is not effectively utilized due to the fact that medical professionals do not know how to use information technology to effectively manage and find useful information,” says Huang. “Those who deal with large quantities of health-care information realize that there is a widening gap between data collection and data comprehension.”
Huang is designing customized search engines that would look for relevant information and hidden patterns in medical databases, including medical research and patient records, while filtering out unnecessary information and maintaining patient confidentiality. It would even analyze image data.
Right: Jimmy Huang
“The doctor bases a diagnosis on all kinds of symptoms, so if we design software that can analyze the symptoms and test results in all the electronic patient records, than they can more accurately decide if someone has a particular illness,” says Huang. “Our main goal is to reduce the cost of health care and at the same time we want to increase the quality.”
The search tools would provide doctors with the best available information on a particular symptom or illness and even the effectiveness of various medications. The idea is to provide doctors and patients with decision-making support so they can choose the best and most affordable treatment plans and care. Eventually, Huang would like to see a health-care network where all databases are linked across the province, and even across the country. This is increasingly important as Canada’s population ages. People are expected to live longer and that means managing chronic illnesses for a greater period of time.
In addition, Huang has been working with the Institute for Clinical Evaluative Sciences (ICES) through an ICES Initiative Collaboration Research Grant of $35,000 in designing software specific to searching through medical databases for information on patients who have had myocardial infarctions or have high blood pressure. He has also received an IBM Shared University Research Award of $172,000 for equipment for his project “Ultra Large-Scale Software Systems with Applications on Health Care,” which looks at best practice recommendations in the area of medical prescription.
Although most of Huang’s many projects are focused on improving health care by ensuring the medical system runs more efficiently and cost-effectively, these same information retrieval systems could be applied to academia, industry or the legal profession. For instance, lawyers could use them to retrieve relevant information on legal cases rather than doing it manually, says Huang.
Huang is also working on several other projects using data mining, and is fast becoming a leader in information retrieval in the Canadian health-care system. If Huang has his way, health care will have better, faster and more effective outcomes in the future.
To watch a video of Huang talking about his work, click here.
By Sandra McLean, YFile writer