CarbGeM aims to overcome the problem of antimicrobial resistance through cutting-edge biology and digital technology, cross-industrial and cross-disciplinary collaboration including medicine and engineering, and industry-government-academia collaboration with world leading hospitals and research institutions. Our AI-based product assists physicians and laboratory technicians in estimating the species of bacterial infections and selecting appropriate antimicrobial agents.
CarbGeM’s AI-based image recognition app, Bitte, reads microscopic images of gram-stained samples from patients who might have infection. Bitte then sends the images to our cloud server and obtains the name of the bacteria in the samples. Finally, Bitte presents physicians appropriate antimicrobials for the treatment based on the regional antibiograms. The first version of Bitte, developed in partnership with the National Center for Global Health and Medicine (NCGM) and Kobe University, identifies bacteria from urine samples. Bitte identifies 20 most popular bacteria, covering more than 90% of the infection-causing bacteria found in patients. Bitte will help curb antimicrobial resistance by matching infections to narrow-spectrum antibiotics, which are effective for specific types of bacteria.