Researchers Trained an AI-Equipped Microscope to Diagnose Deadly Blood Infections
A team of researchers from a Harvard University teaching hospital have developed a microscope equipped with AI that can diagnose blood infections. The AI was able to categorize 93 percent of samples without human help.
Microbiologists from Beth Israel Deaconess Medical Center (BIDMC) have developed a smart microscope that employs artificial intelligence (AI) to accurately diagnose deadly blood infections. The microscope is enhanced with machine learning technology in the hopes that it will help prepare the medical field for the upcoming retirements of highly trained microbiologists — which are already in short supply.
The team trained a convolutional neural network (CNN); a type of AI that is built modeling the mammalian visual cortex and is used to analyze visual data. The training began by showing the AI a series of 100,000 images garnered from 25,000 slides that were treated with dye to make the bacteria more visible. With these images, the AI learned how to sort the bacteria based on their visible traits. The system achieved an accuracy of nearly 95 percent.
Next, the researchers showed the system 189 slides without human intervention and it was able to sort 93 percent accurately. With more training, the system has the potential to develop into a reliable, autonomous classification system that could help to save time and lives.
The system will be a great help to microbiology labs that are already struggling to remain staffed with highly skilled lab technologists. The American Association for Clinical Chemistry (AACC) explains how microbiology school closures and the impending retirement of baby boomers is likely to stretch the already limited resource of skilled lab technologists even thinner. Perfecting an AI system that is able to assist in or perform these tasks will go a long way to ensuring patients have the best chance of getting treatment for potentially lethal blood infections.
The CDC notes that these types of infections — typically caused by bacteria like E. coli, Staphylococcus, and streptococcus — result in thousands of deaths each year and billions of dollars in health care costs.
Artificial intelligence is still in the early stages of its relationship with medicine, but great strides are being made: not long ago, a Chinese robot was even able to pass a medical licensing exam. We may not know the full capacity of what the medical field will look like with AI and robots lending a hand but the life-saving potential of non-human helpers is certainly exciting.