AI outperform humans in identifying ovarian cancer
Researchers at the Karolinska Institutet in Sweden have led a study showing how AI-based models can outperform human experts at identifying ovarian cancer in ultrasound images.
According to Elisabeth Epstein, a professor in obstetrics and gynaecology at Karolinska Institutet, ovarian tumours are common and are often detected by chance.
“There is a serious shortage of ultrasound experts in many parts of the world, which has raised concerns of unnecessary interventions and delayed cancer diagnoses,” Prof Epstein said in a statement. “We therefore wanted to find out if AI can complement human experts.”
The researchers have developed and validated neural network models able to differentiate between benign and malignant ovarian lesions, having trained and tested the AI on over 17,000 ultrasound images from 3,652 patients across 20 hospitals in eight countries. They then compared the models’ diagnostic capacity with a large group of experts and less experienced ultrasound examiners.
MORE FROM ARTIFICIAL INTELLIGENCE
The results showed that the AI models outperformed expert and non-expert examiners at identifying ovarian cancer, achieving an accuracy rate of 86.3 per cent, compared to 82.6 per cent and 77.7 per cent for the expert and non-expert examiners respectively.
Register now to continue reading
Thanks for visiting The Engineer. You’ve now reached your monthly limit of news stories. Register for free to unlock unlimited access to all of our news coverage, as well as premium content including opinion, in-depth features and special reports.
Benefits of registering
-
In-depth insights and coverage of key emerging trends
-
Unrestricted access to special reports throughout the year
-
Daily technology news delivered straight to your inbox
New IET report examines grid transmission costs
Exactly. I generally resist the temptation to reply but can´t always help myself. Aside from Dave´s insistence that the water be pumped using direct...