Researchers at the Houston Methodist Research Institute have developed artificial intelligence (AI) software that can interpret mammograms and identify breast cancer risk with 99% accuracy.

In the US, 12.1 million mammograms are performed annually and, according to the American Cancer Society (ACS), 50% yield false positive results.

As a result over 1.6 million breast biopsies are performed annually in the country, of which around 20% are unnecessarily performed due to false-positive mammogram results.

The Houston Methodist team hope their new AI software will help physicians better define the percent risk requiring a biopsy, thereby helping to decrease the amount of unnecessary breast biopsies.

According to a new study published in the Cancer journal, the AI translates patient charts into diagnostic information at 30 times the speed of a human.

“This software intelligently reviews millions of records in a short amount of time, enabling us to determine breast cancer risk more efficiently using a patient’s mammogram. This has the potential to decrease unnecessary biopsies,” said Chair of the Department of Systems Medicine and Bioengineering at Houston Methodist Research Institute, Dr Stephen T. Wong.

During the study, which was supported by the John S. Dunn Research Foundation, Dr Wong and Director of the Houston Methodist Cancer Center, Dr Jenny C. Chang, led a team of researchers in using the AI software to evaluate mammograms and pathology reports of 500 breast cancer patients.

The AI software was able to review the 500 charts in a few hours, which is more efficient than the time it would take clinicians to do it manually. According to the study, it can take between 50 and 70 hours for two clinicians to manually review 50 charts. The study found the AI software saved over 500 physician hours.

The software scanned patient charts, collected diagnostic features and correlated mammogram findings with breast cancer subtype. Clinicians were then able to use the results to accurately predict each patient’s probability of breast cancer diagnosis.

“Accurate review of this many charts would be practically impossible without AI,” concluded Dr Wong.

For more information contact news@eHealthNews.co.za, like us on Facebook or tweet us @eHealthNewsZA.