Breast cancer is the most common cancer among women in the world (after skin cancer). Although both sexes are susceptible to abnormal breast tissue development, women are most commonly affected by this disease.
Advances in radiology and self-detection awareness campaigns have contributed to the earlier detection of breast cancer in recent years. Knowing that it is a hereditary disease also helps doctors distinguish higher risk patients who need regular screenings.
These assessments usually depend on mammogram analysis. According to conventional knowledge, the denser the breast tissue appears on the mammogram, the higher the risk of cancer.
However, many factors besides tissue density can be involved in neoplastic growth, according to the University of Hawai’i‘s John Shephard. The researcher and his team at the cancer center have recently published a study in Radiology, outlining the results of their research that involved teaching an artificial intelligence model to detect breast cancer risk.
A new way to detect breast cancer risk
The researchers trained the algorithm using 25000 digital mammograms taken from 6369 women between 2006 and 2014. The AI learned how to pick up subtle details that are difficult to recognize. However, these details may be the key to detecting an increased risk of cancer.
The algorithm did not perform well in detecting the risk for cancers developing between screenings. However, it was better at determining screening-detected cancer risk as compared to clinical risk factors.
“The results showed that the extra signal we’re getting with AI provides a better risk estimate for screening-detected cancer than other forms of assessment,” said Shepherd. “It helped us accomplish our goal of classifying women into low risk or high risk of screening-detected breast cancer.”
“By ranking mammograms in terms of the probability of seeing cancer in the image, AI is going to be a powerful second reading tool to help categorize mammograms.”
Seeing this success, the researchers now plan to enroll more underrepresented groups like Native Hawaiian and Pacific Islander women, so that their cancer risk can also be assessed. Furthermore, they want to use the AI technology to assess women’s risk of developing different grades of cancer.
Source: University of Hawai’i news
https://www.hawaii.edu/news/2021/09/30/ai-can-predect-breast-cancer-risk/
