Artificial intelligence (AI) is increasingly being used in medicine, particularly cancer detection. Using AI in breast cancer screening is safe and can cut radiologists’ workload in half, according to a study published in the Lancet Oncology journal.
Breast cancer is the second most common cancer among women, after skin cancer. Nearly 300,000 women are diagnosed with breast cancer annually, according to the American Cancer Society. Regular screening can detect breast cancer early, when it is easier to treat, and reduce mortality.
But reading mammograms takes time. To assess whether AI could assist in this task—and whether it was safe—researchers conducted a randomized, controlled trial.
Interim study results suggest that AI screening is as effective at detecting breast cancer as two radiologists working together, does not result in a higher rate of false positive results and cuts workload in half.
The study followed more than 80,000 women from Sweden with an average age of 54 years and compared AI screening with standard mammography screening, which typically requires two radiologists to do what’s called a double reading.
AI-supported screening detected cancer in more women than standard screening. About 28% (244 women) were found to have cancer according to AI-supported screening, and 25% (203 women) were found to have cancer according to standard screening. The rate of false positives (when a scan is inaccurately determined to be abnormal) did not increase when using AI screening. In fact, both screening methods produced a false positive rate of 1.5%.
“These promising interim safety results should be used to inform new trials and program-based evaluations to address the pronounced radiologist shortage in many countries,” lead study author Kristina Lång, PhD, of Lund University in Sweden, told The Guardian. But, she added, “They are not enough on their own to confirm that AI is ready to be implemented in mammography screening.”
She noted that the implications regarding patients’ outcomes as well as the cost effectiveness of AI technology need to be researched further. “The greatest potential of AI right now is that it could allow radiologists to be less burdened by the excessive amount of reading,” she said. This would allow radiologists to concentrate on advanced diagnostics and result in shorter wait times for patients, Lång said.
To read more about AI, click #Artificial Intelligence. There, you’ll find headlines such as “Training Artificial Intelligence to Diagnose Skin Cancers in People of Color,” “Melanoma Detection in the Digital Age: Total Body Photography” and “How Artificial Intelligence Can Improve Colorectal Cancer Detection.”
For more about breast cancer, click Cancer Health’s Basics on Breast Cancer. It reads in part:
What are the risk factors for breast cancer?
According to the Centers for Disease Control and Prevention, besides being a woman and getting older, other factors that influence the risk of developing breast cancer include:
- Genetic mutations (including BRCA1 and BRCA2)
- Family history of breast cancer
- Early onset of menstruation or early menopause
- No full-term pregnancies or first pregnancy at an older age
- Not being physically active
- Overweight or obesity, especially after menopause
- Use of oral contraceptives or hormone replacement therapy
- Previous radiation therapy
- Drinking alcohol.
What are the symptoms of breast cancer?
The most common sign of breast cancer is a lump or mass. A hard and painless mass is most likely to be malignant, but cancerous tumors can sometimes be tender, soft or painful. Other symptoms may include breast swelling, skin irritation or dimpling, breast or nipple pain, nipple retraction (turning inward), redness, scaliness or thickening of the nipples or skin of the breast and discharge from the nipple.
How is breast cancer diagnosed?
Regular screening for breast cancer can detect the disease early, when it is easier to treat. Professional guidelines vary in how often they recommend screening. The U.S. Preventive Services Task Force now recommends that women at average risk should start screening at age 40. All women and men who notice a new mass, lump or other changes in their breasts should report this to their health care provider.
If a mammogram detects changes in the breasts, an ultrasound may be done for further examination. Once diagnosed with breast cancer, MRI scans are usually done to assess the size of the tumor, look for additional tumors and determine whether the cancer has spread beyond the breasts.
A breast biopsy, or examination of a tissue sample, may be done to determine whether a tumor is malignant. Genomic testing of a tumor sample provides more information about the type of cancer and how best to treat it.