(NewsNation) —AI may be the next frontier in cancer detection as a new study shows that the technology can outperform human experts at identifying ovarian cancer in ultrasound images.
A study conducted by researchers at Karolinska Institute in Sweden shows that Artificial Intelligence (AI) based models achieved an accuracy rate of 86% for detecting ovarian cancer, compared to 82% for human experts and 77% for those with less expertise.
Researchers trained an AI program to be able to tell the difference between benign and malignant ovarian lesions, using more than 17,000 ultrasound images from nearly 3,700 patients across 20 hospitals in eight countries.
In a retrospective triage simulation, AI-driven diagnostic support reduced referrals to experts by 63% while significantly surpassing the diagnostic performance of the current practice, according to the study which was published in Nature Medicine.
“Ovarian tumors are common and are often detected by chance,” Elisabeth Epstein, a senior physician with the Stockholm South General Hospital’s Department of Clinical Science and Education, said in a news release from the Karolinska Institute in Sweden.
“There is a serious shortage of ultrasound experts in many parts of the world, which has raised concerns of unnecessary interventions and delayed cancer diagnosis. We therefore wanted to find out if AI can complement human experts.”
While the study noted that further prospective and randomized studies are needed to validate the clinical benefit and diagnostic performance of the AI models, the study offers insights into the applicability of AI-driven diagnostic support systems for ovarian cancer detection.
“The models’ consistent superiority to human assessment and robust performance under comprehensive evaluation indicates that they are ready for prospective clinical implementation studies, bringing us closer to the adoption of AI-assisted diagnostics in clinical settings,” the study stated.
