| |
CS 5150
Software Engineering
Fall 2012
Project Suggestion: Bacteria Database
|
|
Bacteria Database Client Craig Altier, Associate Professor, College of Veterinary Medicine Background The Animal Health Diagnostic Center (AHDC) of Cornell University is among the largest veterinary diagnostic laboratories in the country. A primary function of the Microbiology section of the AHDC is to diagnose diseases in animals caused by bacteria and to determine which antibiotics are appropriate to treat these infections. We receive samples from a wide variety of animals thought to have bacterial infections and perform culture and susceptibility testing, which first identifies the bacterial species present and then tests that bacterium against a panel of antibiotics to identify which can kill the organism.Problem The testing we perform requires several days. In the meantime, animal suffering would be reduced if our veterinary clients could make educated guesses about which bacterial species are likely to be the cause of infection and which antibiotics might be effective, thus allowing them to initiate effective therapy sooner.Project Over many years, we have maintained a database of the bacteria we have cultured and the antibiotics effective against them, correlated by animal species and the body site of the infection. Those data, however, have not been in a format that is readily accessible to us or our clients. The goal of this project, therefore, is to make this information rapidly available to our clients, allowing them to help their patients more quickly.The database is currently maintained on a proprietary system (Sensititre, Trek Diagnostics), but can be downloaded onto Excel spreadsheets. The objective will be to establish a searchable database using these data that will allow users to query by animal species and/or body site. This will thus allow them to determine which bacteria most commonly cause infection at a specific site, and then to assess which antibiotics might be most effective, based upon their past performance. This software could take the form of a web-based application or a smartphone/tablet app, depending upon feasibility. |
[ Home ]