SALT LAKE CITY — A Utah-based artificial intelligence company and scientists at ARUP Laboratories on Tuesday announced the development of an AI tool that detects intestinal parasites in stool samples quickly and more accurately than traditional methods.
The technology has the potential to transform how labs diagnose parasitic infections across the world.
Before this breakthrough, detecting gastrointestinal parasites meant using traditional microscopy, a labor-intensive process that required highly trained experts to manually scour each sample for telltale cysts, eggs or larvae.
That’s changed, though, with the help of a deep-learning AI model known as a convolutional neural network that is extremely precise, according to a study published Tuesday in the Journal of Clinical Microbiology.
The model, researchers showed, can detect parasites in wet mounts of stool with greater sensitivity than a human observer.
Blaine Mathison, the study’s lead author, ARUP’s technical director of parasitology and adjunct lecturer in the University of Utah’s Department of Pathology, called the discovery “groundbreaking.”
ARUP is an independent nonprofit enterprise of the University of Utah School of Medicine’s Department of Pathology and a leading national reference lab.
“What we’ve accomplished is remarkable,” Mathison said in a release from the university and ARUP. “Our validation studies have demonstrated that the AI algorithm has better clinical sensitivity, improving the likelihood that a pathogenic parasite may be detected.”
Researchers partnered with Orem-based Techyte, a leader in AI-powered diagnostics. Techyte was originally launched as a university startup in 2013 to commercialize discoveries led by Mohamed Salama, then an ARUP medical director and U. pathology professor
The team trained the AI model using over 4,000 parasite-positive samples collected from laboratories across the United States, Europe, Africa and Asia and representing 26 classes of parasites.
“This was really a robust study when you consider the number of organisms and positive specimens used to validate the AI algorithm,” Mathison said in the release.
The tool was also able to identify 169 additional organisms that had been missed in earlier manual reviews, further validating its efficacy and improving diagnosis and treatment options for patients.
Additionally, the study found that AI routinely found more parasites than technologists, even when the samples were highly diluted, suggesting the system can detect infections at earlier stages or when parasite levels are low.
Of course, the discovery wouldn’t even have been possible without humans.
“An AI algorithm is only as good as the personnel inputting the data,” Adam Barker, ARUP’s chief operations officer, said in a release. “We have phenomenal staff who have used their extensive knowledge and skills to build an exceptional AI solution that benefits not just the laboratory, but also patients.”
ARUP became the first lab in the world to implement AI in 2019, collaborating with Techyte to do so.
“ARUP will continue to develop innovative technologies, including AI solutions, to improve diagnostic capabilities and laboratory processes,” said a release from the lab. “ARUP has already implemented an AI solution to improve Pap testing and has other AI innovations in development.”
The Key Takeaways for this article were generated with the assistance of large language models and reviewed by our editorial team. The article, itself, is solely human-written.
