Ovarian cancer cells. Artificial intelligence enhances ovarian cancer diagnostics

Artificial intelligence improves the diagnosis of ovarian cancer

Spread the love

Ovarian cancer affects more than 3,100 Canadian women each year, making it the deadliest of all female reproductive cancers. A new study led by Dr. Ali Bashashati, a researcher at UBC and the Vancouver Coastal Health Research Institute, reveals how artificial intelligence (AI) can aid in ovarian cancer diagnosis to improve patient outcomes. .

The study, recently published in modern pathologyIt is based on the understanding that ovarian cancer is not a single disease, but several different subtypes, called histotypes.

Dr. Bashashati and his team compared ovarian cancer disease classifications made by an AI machine learning-based model with those of a team of expert gynecologic pathologists who specialize in diagnosing female reproductive cancers.

ali bashashati

Dr. Ali Bashashati.

Using a cohort of 948 ovarian cancer tissue samples from Vancouver General Hospital, Dr. Bashashati’s team developed a series of AI computer algorithms that can identify four ovarian cancer histotypes with a high degree of accuracy.

“To the best of our knowledge, our AI model is the first of its kind to achieve the highest performance for diagnosing ovarian carcinoma histotypes, approaching the level of performance of expert gynecologic pathologists,” says Dr. Bashashati, professor at the pathology department at UBC. and laboratory medicine and School of Biomedical Engineering, as well as director of AI research in the Ovarian Cancer Research Program (OVCARE) at BC Cancer.

Each histotype of ovarian cancer has unique characteristics and telltale signs, however, they can be difficult to identify in some cases.

Previous research has shown that general pathologists have weak to moderate interobserver agreement (the degree to which two or more experts agree) when it comes to histotyping ovarian cancer, Dr. Bashashati says. Gynecologic pathologists, on the other hand, have strong to very precise interobserver agreement.

Providing general pathologists with additional tools to achieve higher interobserver agreement would support optimal diagnoses and the best and most timely approach to treatment possible.

“Our AI models can provide a separate set of artificial eyes during clinical evaluations to improve diagnostic performance when used in conjunction with a pathologist’s evaluation.”

Advanced technology to expand the scope of specialized care

The AI ​​algorithms that Dr. Bashashati and his team developed were convolutional neural networks (CNNs), which attempt to mimic the complex neurological connections in the human brain that support thought processes such as pattern recognition and deductive reasoning.

ovarian cancer cells.  Different subtypes, or histotypes, of ovarian cancer shown at high magnification.

Different subtypes, or histotypes, of ovarian cancer shown at high magnification.

The research team’s CNNs overcame some of the challenges that similar CNNs faced before, such as distinguishing between colors and other features on tissue sample slides.

“When we tested our best-performing AI model using another cohort of 60 patients from a different Canadian hospital, the algorithm correctly diagnosed 52 of the cases,” says Bashashati.

Expert gynecologic pathologists re-reviewed eight cases that were misclassified by the AI ​​algorithm. In four of the eight cases, the experts ended up agreeing with the AI ​​classification.

“This means that the accuracy of our best-performing AI model approached 93 percent, which is incredibly exciting.”

Dr. Bashashati’s CNNs were designed to be transferable AI classification aids that could be used in multiple hospitals.

“Many community hospitals don’t have a gynecologic pathologist on staff to help a general pathologist with the staging of an ovarian cancer subtype,” Dr. Bashashati says. “As such, there is a great need to provide general pathologists with additional tools to aid in the diagnosis of ovarian cancer, particularly given the increased demand for specialists in remote communities and the limited number of available gynecologic pathologists in British Columbia.”

Dr. Bashashati is now working to further validate these findings using an international cohort of 8,000 ovarian cancer tissue sample slides.

A version of this story originally appeared on the Vancouver Coastal Health Research Institute website.


#Artificial #intelligence #improves #diagnosis #ovarian #cancer

Leave a Comment

Your email address will not be published. Required fields are marked *