Hospital acquired infections (HAIs) due to drug resistant organisms are a looming morbidity problem in modern healthcare. Costing millions of dollars per year to treat, and ending many lives, they represent a widespread problem that needs immediate attention. In this SBIR project we define a clinical informatics application that can identify drug resistant microbes as they appear, as well as predict any normal infectious outbreak. The basis of the project is a comprehensive curation of antimicrobial susceptibility breakpoint data, algorithms which map these quantitative data individually for each microbe and antimicrobial combination, allow the infection control practitioners to set alert breakpoints, and which informs the correct use of antimicrobials if deployed properly. By generating a real time software application that monitors the clinical microbiology """"""""HL7"""""""" data stream of ICUs, we propose a method to reduce the number of HAIs seen every year. The method also involves creation of biometrics-based predictive algorithms that infer an impending infection and alert the Infection Control Practitioner (ICP). This research will increase our understanding of microbial drug resistance, enable the tracking of new drug resistant organisms, and provide a measurement and surveillance utility to keep track of microbes in the clinic. Success will be measured by the reduction of cases and the decline of disease
The goal of this SBIR project is to track infection in the healthcare clinic. The goal is to reduce hospital acquired infections, by developing new software that can i) Predict that a microbe outbreak is about to occur and ii) Determine whether microbes are becoming antibiotic resistant. We propose new methods to monitor a clinical informatics data stream, with software that can detect when an outbreak of normal or drug resistant microbes is about to occur.
