Diagnostic Imaging's Weekly Scan: Nov. 6, 2020

November 6, 2020
Whitney J. Palmer

Cerebral Aneurysms and CT Angiography; MammoScreen and Breast Cancer Detection; Low-Dose Lung Cancer Screening Program Performance; and Breast Cancer Screening Advocacy Efforts

Welcome to Diagnostic Imaging’s Weekly Scan. I’m senior editor, Whitney Palmer.

Before we get to our featured interview with Dr. Amy Patel about the importance of increasing women’s access to breast cancer screening services and the advocacy efforts underway at the legislative level to increase that access, here are the top stories of the week.

Cerebral aneurysms are responsible for between 80 percent and 90 percent of all non-traumatic subarachnoid hemorrhages – half of which are fatal, and 10 percent to 20 percent of which cause permanent disability. CT angiography is the go-to modality for detecting these aneurysms, and this week, the journal Radiology published a study from investigators in China who have developed a 3D-detection algorithm based on a convolutional neural network that can help improve a radiologist’s ability to pinpoint these problems, including the smaller aneurysms that can go undetected. The team used CT angiograms showing 688 cerebral aneurysms from 534 patients to train their algorithm. One validation set from another 534 patients also included 649 aneurysms, and an external validation set of 400 CT angiograms included another 206 aneurysms. Using their algorithm, they detected 633 of 649 aneurysms, reaching 97.5-percent sensitivity. In addition, using the tool, radiologists were able to detect eight additional aneurysms that had previously gone unidentified – six of which were smaller than 3 mm. Less experienced radiologists saw the most benefit from using the algorithm. And, although the algorithm had a high rate of false-positive findings per case of 13.8, the results of this study still highlight its utility as a support tool, they said.

Earlier this year, the U.S. Food & Drug Administration awarded 510(k) clearance to MammoScreen™, an artificial intelligence tool from the company Therapixel that is designed to be used with 2D mammography to evaluate their likelihood of malignancy. Now, this week, investigators from Therapixel published a study in Radiology: Artificial Intelligence that shows using this tool can somewhat improve sensitivity and provide a better false-negative rate. Using a cadre of 14 radiologists as interpreters, the team assessed a dataset comprised of 240 2D digital mammography images conducted between 2013 and 2016. Half were read once without MammoScreen™, and the other half were read once without the tool and once with it. According to their analysis, they found the average area under the curve across readers was 0.769 without AI and 0.797 with it, and average sensitivity increased by 0.033. In addition, for 11 of 14 radiologists, the tool created an average improvement of 18 percent in the false-negative rate, and the false-positive rate dropped by an average of 25 percent for eight radiologists. Overall, the team said, MammoScreen™ creates these improvements without impeding workflow. In fact, for cases where the likelihood of malignancy is low – less than a 2.5-percent chance – it produces slightly faster reading results.

Low-dose CT screening for lung cancer might not be as beneficial as has been believed. In a study published in JAMA Network Open, investigators from China set out to assess both the participation and detection rates of population-based screening programs. What they found was the lung cancer detection rates are not statistically significant between high-risk individuals who go through screening and those who do not. The rates were 0.35 percent and 0.38 percent respectively. Using a study population of 55,428 individuals who were identified to benefit from low-dose CT screening, the team found that only 40 percent -- 22,260 individuals – got the scan. Overall, 78 cancers were identified in screened patients, while 125 were pinpointed in patients who did not undergo screening. Among the patients, former smokers and patients with a family history of lung cancer were more likely to get screened. So were women, but men were more likely to have cancer.

And, finally this week, Diagnostic Imaging spoke with Dr. Amy Patel, medical director of women’s imaging at Liberty Hospital in Missouri. Dr. Patel is a leading voice in the industry for expanding access among women to breast cancer screening services. In our conversation, she discussed her advocacy efforts and shared guidance and insights on what radiologists can do to participate in these legislative activities. Here’s what she had to say.

For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.