AI model helps sort breast MRI scans in women with dense breasts – Consumer Health News


TUESDAY, October 19, 2021 (HealthDay News) – A deep learning (DL) model can rule out nearly 40% of magnetic resonance imaging (MRI) breast exams without lesions without missing any malignancies in women with dense breasts, according to a study published online on October 5 in Radiology.

Erik Verburg, Utrecht University Medical Center in the Netherlands, and his colleagues examined the feasibility of an automatic triage method using LD to reject the greatest number of lesion-free MRI scans without affecting the identification of lesions. malignant diseases. A model was developed to differentiate between breasts with and without lesions, and was trained using data from seven hospitals and tested on data from an eighth hospital. This secondary analysis used data from 4,581 MRI scans of 4,581 women with extremely dense breasts (mean age, 54.3 years) who were included in the DENSE trial.

The researchers found that 838 of 9,162 breasts had at least one lesion, of which 77 were malignant, and 8,324 had no lesions. At a threshold set at 100% sensitivity for malignant lesions, the DL model considered 90.7% of MRI scans with lesions as non-normal and sorted them for X-ray examination. Overall, 39.7 percent of lesion-free MRI scans were rejected by the DL model. To differentiate between normal breast MRI scans and lesion MRI scans, the DL model had an average area under the operating characteristic curve of the receptor of 0.83.

“We have shown that it is possible to safely use artificial intelligence to rule out breast cancer screening MRIs without missing any malignant disease,” Verburg said in a statement. “The results were better than expected. Forty percent is a good start. However, we still have 60 percent to improve.”

An author has disclosed financial ties to Bayer Pharma.

Summary / Full text


About Author

Leave A Reply