|Photo via pxhere|
Microscopes powered by artificial intelligence may help improve diagnosis of deadly blood infections. The enhancement may also help the survival rate of patients’ by receiving an accurate diagnosis, according to the microbiologists at Beth Israel Deaconess Medical Center.
In a demonstration, the scientists reported that the automated and AI-enhanced microscope system can identify images of bacteria more accurately and quickly. The technological upgrade may also address the current issue of the glaring lack of highly trained microbiologists.
“This marks the first demonstration of machine learning in the diagnostic area. With further development, we believe this technology could form the basis of a future diagnostic platform that augments the capabilities of clinical laboratories, ultimately speeding the delivery of patient care,” said Dr. James Kirby, the director of the Clinical Microbiology Laboratory at BIDMC and the senior author of the research.
A class of AI modeled on the mammalian visual cortex called the convolutional neural network was used to properly categorize bacteria strains based on distribution and shape. To train the CNN, the scientists collected and uploaded data of more than 25,000 images from blood samples during routine clinical checkups. After being identified by microbiologists, they managed to generate more than 100,000 cropped training images.
The scientists then obtained blood samples from patients with suspected blood infections. The samples were incubated and later placed on a glass slide and stained with dye to improve the cellular visualization. They tested the ability of CNN without human assistance and the AI had at least 93 percent accuracy in detecting three categories – rod-shaped bacteria including E. coli, round clusters of Staphylococcus, and pairs or chains of Streptococcus.
The new technology will help clinicians in diagnosis, and training and research in microbiology, according to Dr. Kirby.