breast imaging Archives - Page 3 of 3

After a long wait, the FDA issued a final rule that adds details on breast density reporting to the Mammography Quality Standards Act. The rule takes effect in September 2024 and should go a long way toward clarifying the issue of breast density for patients.

Breast tissue density is a risk factor for cancer, and dense breast tissue can make it more difficult for radiologists to identify tumors on conventional x-ray mammography. This shortcoming is often not communicated to women who receive “normal” mammograms, but later find out that a cancer was missed.

Prodded by a strong patient advocacy movement, individual states have been passing laws requiring women to be notified of their density status, creating a patchwork of regulation across the U.S.

The FDA in 2018 agreed to set a national standard by rolling breast density reporting into an update of the MQSA. But the long wait has frustrated many in the breast density advocacy movement.

There are several major components to the new rule, which:

Requires breast imaging facilities to provide patients with a summary of the mammography report written in lay terms that identifies whether patients have dense or non-dense breast tissue.
Instructs facilities to include a section in the mammography report explaining the significance of breast density.
Establishes four categories for reporting breast tissue density in the mammography report.
Sets the specific language to be used for reporting density.

The new rules provide much-needed national consistency in breast density reporting, and will replace the patchwork of state regulation that has developed over the years. Developers of breast density software may also benefit from the new federal rules, as they simplify the number of regulations that need to be tracked.

The Takeaway

Better late than never. While the FDA should have signed off on this years ago, now that the rules are issued the breast imaging community can move ahead with integrating them into clinical practice. The new rules should also help density reporting software developers by setting a national standard rather than a patchwork of state regulation.

Radiology Journal detailed a multimodal AI solution that can classify breast lesion subtypes using mammograms, potentially reducing unnecessary biopsies and improving biopsy interpretations.

Researchers from Israel and IBM/Merative first pretrained a deep learning model with 26k digital mammograms to classify images (malignant, benign, or normal), and used these pretraining weights to develop a lesion subtype classification model trained with mammograms and clinical data. Finally, they trained a pair of lesion classification models using digital mammograms linked to biopsy results from 2,120 women in Israel and 1,642 women in the US.

When the Israel AI model was tested against mammograms from 441 Israeli women it…

Predicted malignancy with an 0.88 AUC
Classified ductal carcinoma in situ, invasive carcinomas, or benign lesions with 0.76, 0.85, and 0.82 AUCs
Correctly interpreted 98.7% of malignant mammographic examinations and 74.6% of invasive carcinomas (matching three radiologists)
Would have prevented 13% of unnecessary biopsies and missed 1.3% of malignancies (at 99% sensitivity)

When the US AI model was tested against mammograms from 344 US women it…

Predicted malignancy with a lower 0.80 AUC
Classified ductal carcinoma in situ, invasive carcinomas, or benign lesions with lower 0.74, 0.83, and 0.72 AUCs
Correctly interpreted 96.8% of malignant mammographic examinations and 63% of invasive carcinomas (matching three radiologists)

The authors attributed the US model’s lower accuracy to its smaller training dataset, and noted that the two models’ also had worse performance when tested against data from the other country (US model w/Israel data, Israel model w/ US data) or when classifying rare lesion types.

However, they were still bullish about this approach with enough training data, and noted the future potential to add other imaging modalities and genetic information to further enhance multimodal breast cancer assessments.

The Takeaway

We’ve historically relied on biopsy results to classify breast lesion subtypes, and that will remain true for quite a while. However, this study shows that multimodal-trained AI can extract far more information from mammograms, while potentially reducing unnecessary biopsies and improving the accuracy of the biopsies that are performed.

Get every issue of The Imaging Wire, delivered right to your inbox.