Summary: A study of nearly 5,000 screening mammograms revealed that an FDA-approved AI algorithm’s false positive results are influenced by patient race and age, highlighting the need for diverse training data and demographic considerations in AI software for mammography.
Key Takeaways
- Influence of Patient Characteristics on AI Performance: The study found that patient characteristics such as race and age significantly influenced false positive results in mammogram interpretations by an FDA-approved AI algorithm.
- Need for Diverse Training Data: There is a critical need for demographically diverse databases for AI algorithm training, as the FDA does not require diverse datasets for validation.
- AI’s Diagnostic Benefits and Limitations: While AI can improve radiologists’ efficiency and accuracy in detecting breast cancer, its performance varies across different patient demographics, highlighting the importance of considering these factors in AI development and application.
- False Positive Disparities: The study revealed that false positive scores were more frequent in Black and older patients, while less frequent in Asian and younger patients, compared to white patients and those aged 51-60.
————————————————————————————————————————————————————
In a study of nearly 5,000 screening mammograms interpreted by a U.S. FDA-approved AI algorithm, patient characteristics such as race and age influenced false positive results. The study’s results were published in Radiology, a journal of the Radiological Society of North America (RSNA).
Need for Diverse AI Data in Mammography
“AI has become a resource for radiologists to improve their efficiency and accuracy in reading screening mammograms while mitigating reader burnout,” says Derek L. Nguyen, MD, assistant professor at Duke University in Durham, North Carolina. “However, the impact of patient characteristics on AI performance has not been well studied.”
Nguyen says while preliminary data suggests that AI algorithms applied to screening mammography exams may improve radiologists’ diagnostic performance for breast cancer detection and reduce interpretation time, there are some aspects of AI to be aware of.
“There are few demographically diverse databases for AI algorithm training, and the FDA does not require diverse datasets for validation,” he says. “Because of the differences among patient populations, it’s important to investigate whether AI software can accommodate and perform at the same level for different patient ages, races, and ethnicities.”
Evaluates AI in Diverse Mammogram Data
In the retrospective study, researchers identified patients with negative (no evidence of cancer) digital breast tomosynthesis screening examinations performed at Duke University Medical Center between 2016 and 2019. All patients were followed for a two-year period after the screening mammograms, and no patients were diagnosed with a breast malignancy.
The researchers randomly selected a subset of this group consisting of 4,855 patients (median age 54 years) broadly distributed across four ethnic/racial groups. The subset included 1,316 (27%) white, 1,261 (26%) Black, 1,351 (28%) Asian, and 927 (19%) Hispanic patients.
A commercially available AI algorithm interpreted each exam in the subset of mammograms, generating both a case score (or certainty of malignancy) and a risk score (or one-year subsequent malignancy risk).
“Our goal was to evaluate whether an AI algorithm’s performance was uniform across age, breast density types and different patient race/ethnicities,” Nguyen says.
AI’s Demographic Limitations
Given all mammograms in the study were negative for the presence of cancer, anything flagged as suspicious by the algorithm was considered a false positive result. False positive case scores were significantly more likely in Black and older patients (71-80 years) and less likely in Asian patients and younger patients (41-50 years) compared to white patients and women between the ages of 51 and 60.
“This study is important because it highlights that any AI software purchased by a healthcare institution may not perform equally across all patient ages, races/ethnicities and breast densities,” Nguyen says. “Moving forward, I think AI software upgrades should focus on ensuring demographic diversity.”
Nguyen says healthcare institutions should understand the patient population they serve before purchasing an AI algorithm for screening mammogram interpretation and ask vendors about their algorithm training.
“Having a baseline knowledge of your institution’s demographics and asking the vendor about the ethnic and age diversity of their training data will help you understand the limitations you’ll face in clinical practice,” he says.
Featured image: Example mammogram assigned a false-positive case score of 96 in a 59-year-old Black patient with scattered fibroglandular breast density. (A) Left craniocaudal and (B) mediolateral oblique views demonstrate vascular calcifications in the upper outer quadrant at middle depth (box) that were singularly identified by the artificial intelligence algorithm as a suspicious finding and assigned an individual lesion score of 90. This resulted in an overall case score assigned to the mammogram of 96.
