Mammography is a medical imaging technique that makes use of low-dose x-rays to see inside the breast and detect breast cancer changes.
- Mammograms can sometimes yield false-positive results, increasing unnecessary biopsies and financial burden.
- The new artificial intelligence (AI) software accurately interprets mammogram and predicts breast cancer results efficiently.
- The AI software reviews records in a short time by scanning patient charts, collecting diagnostic features and correlating mammogram findings with breast cancer sub-type.
- The AI reviews over 500 records in a few hours saving over 500 physician hours.
Mammogram readings help in the early detection and diagnosis of breast cancer
. Mammogram readings help in the early detection and diagnosis of breast cancer before women experience any symptoms, when it is most treatable.
Use of x-rays involve exposing the body to small dose of ionizing radiation to produce images of the inside of the body. Mammogram x-rays of the breasts show changes in the breast tissue two years prior to it becoming apparent.
‘A new artificial intelligence software can scan millions of mammograms in a short time and accurately predict the risk of breast cancer, decreasing the false positives and need for biopsies.’
Certain limitations of mammography include yielding false-positive or false-negative results.
False positive - When no cancer is present but the mammogram looks abnormal, it is called false-positive.
False negative - When the mammogram fails to detect a genuine breast change, it is called false-negative.
Researchers at Houston Methodist have developed an artificial intelligence (AI) software that interprets mammograms. It assists doctors with a quick and accurate prediction of breast cancer risk.
The new study is published in the journal Cancer
. The computer software intuitively translates patient charts into diagnostic information at 30 times human speed and with 99% accuracy.
"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,"
says Stephen T. Wong, Ph.D., P.E., chair of the Department of Systems Medicine and Bioengineering at Houston Methodist Research Institute.
The research team was led by Wong and Jenny C. Chang, M.D., director of the Houston Methodist Cancer Center and supported in part by John S.Dunn Research Foundation.
Researchers used the new software to evaluate mammograms and pathology reports of 500 breast cancer patients. The software scanned patient charts, collected diagnostic features and correlated mammogram findings with breast cancer sub-type. Clinicians used results, like the expression of tumor proteins, to accurately predict each patient's probability of breast cancer diagnosis.
In the United States, 12.1 million mammograms and over 1.6 million breast biopsies are performed annually,
according to the Centers for Disease Control and Prevention (CDC). According to the American Cancer Society (ACS) 50% of these yield false positive results, resulting in one in every two healthy women told they have cancer and about 20% biopsies are unnecessarily performed due to the false-positive results.
Currently, when mammograms fall into the suspicious category, patients are recommended for biopsies. They include a broad range of 3% to 95% of patients with cancer risk.
The Houston Methodist team hopes this artificial intelligence software will help physicians better define the percent risk of patients requiring a biopsy thus equipping doctors with a tool to decrease unnecessary breast biopsies
Two clinicians manually reviewed 50 charts in 50-70 hours. On the other hand the AI software reviewed 500 charts in a few hours, saving over 500 physician hours.
"Accurate review of this many charts would be practically impossible without AI," says Wong.