News | Artificial Intelligence | May 21, 2024

In a new study of nearly 5,000 screening mammograms interpreted by an FDA-approved AI algorithm, patient characteristics such as race and age influenced false positive results. The study’s results, “Patient Characteristics Impact Performance of AI Algorithm in Interpreting Negative Screening Digital Breast Tomosynthesis Studies” were published today in the RSNA journal Radiology.

According to a newly-published study of nearly 5,000 screening mammograms interpreted by an FDA-approved AI algorithm, patient characteristics such as race and age influenced false positive results.

In a newly-published study of nearly 5,000 screening mammograms interpreted by an FDA-approved AI algorithm, patient characteristics such as race and age influenced false positive results. The study’s results, “Patient Characteristics Impact Performance of AI Algorithm in Interpreting Negative Screening Digital Breast Tomosynthesis Studies” were published today in the RSNA journal Radiology. Image courtesy: RSNA Radiology (Nguyen, DL and Ren, Y et al)


May 21, 2024 — According to a newly-published study of nearly 5,000 screening mammograms interpreted by an FDA-approved AI algorithm, patient characteristics such as race and age influenced false positive results. The study’s results, “Patient Characteristics Impact Performance of AI Algorithm in Interpreting Negative Screening Digital Breast Tomosynthesis Studies” were published today in Radiology, a journal of the Radiological Society of North America (RSNA).

“AI has become a resource for radiologists to improve their efficiency and accuracy in reading screening mammograms while mitigating reader burnout,” said Derek L. Nguyen, M.D., Duke Health, breast radiologist and assistant professor at Duke University in Durham, NC. “However, the impact of patient characteristics on AI performance has not been well studied,” added Nguyen.

Key Results

- In a retrospective study including 4855 breast screening patients, false-positive case scores were more likely in Black patients (odds ratio [OR] = 1.5) and less likely in Asian patients (OR = 0.7) compared with White patients.

- False-positive risk scores were more likely in Black patients (OR = 1.5) and patients with extremely dense breasts (OR = 2.8) compared with White patients and patients with fatty density breasts, respectively.

- The influence of patient characteristics on algorithm performance necessitates more demographically diverse data sets for testing and training and greater transparency.

In a written summary of the findings issued by RSNA, Nguyen said 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,” Nguyen said, adding: “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.”

Collaborating with Nguyen were Lars J. Grimm, M.D., M.S.,  and Joseph Y. Lo, PhD, both with Duke University School of Medicine Department of Radidology; Tyler M. Jones, B.S., Duke University Pratt School of Engineering; Samantha M. Thomas, M.S., Duke University Department of Biostastics and Bioinformatics, and for iCAD Senior Research Scientist, Yinhao Ren, Ph.D.

AI Breast Imaging Study Details

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 said.

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 said. Notably, he added, “Moving forward, I think AI software upgrades should focus on ensuring demographic diversity.”

Nguyen said 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 said. 

More information: www.rsna.org

Reference: https://doi.org/10.1148/radiol.232286


Related Content

News | Mobile C-Arms

June 19, 2024 —Royal Philips, a global leader in health technology, today announced the launch of Philips Image Guided ...

Time June 19, 2024
arrow
News | Artificial Intelligence

June 18, 2024 — The advancement of Artificial Intelligence (AI) in healthcare to support diagnostic decision making ...

Time June 18, 2024
arrow
News | PET Imaging

June 18, 2024 — Positron Corporation, a leading molecular imaging medical device company offering PET and PET-CT ...

Time June 18, 2024
arrow
News | Lung Imaging

June 18, 2024 — A new study led by American Cancer Society (ACS) researchers shows less than one-in-five eligible ...

Time June 18, 2024
arrow
News | Magnetic Resonance Imaging (MRI)

June 17, 2024 — MRI has transformed neuroscience research over the past 50 years, but research participants have had to ...

Time June 17, 2024
arrow
News | Radiology Business

June 17, 2024 — Strategic Radiology welcomed Northwest Radiologists, Inc, a 20-radiologist practice based in Bellingham ...

Time June 17, 2024
arrow
Feature | Imaging Technology News - ITN

Dear Friends and Readers of ITN, Can you spare 3 minutes today to give us some feedback? Please share your insight with ...

Time June 17, 2024
arrow
News | Pediatric Imaging

June 13, 2024 — Cervical spine injuries in children are relatively rare but can have serious consequences, like ...

Time June 13, 2024
arrow
News | Neuro Imaging

June 12, 2024 — Brainet, a developer of cutting-edge diagnostic tools for assessing brain health, and SimonMed Imaging ...

Time June 12, 2024
arrow
News | SPECT-CT

June 11, 2024 — A newly developed radiotracer can generate high quality and readily interpretable images of cardiac ...

Time June 11, 2024
arrow
Subscribe Now