HomeTechnologyArtificial intelligenceCombined AI system on par with radiologists for detecting breast cancer

Combined AI system on par with radiologists for detecting breast cancer

A research team led by Dr. Tao Tan from the Netherlands Cancer Institute found that an AI multimodal system showed similar sensitivity in identifying breast cancer compared to human readers.

“Multimodal AI systems for detecting breast cancer in women with dense breasts are a potential solution for breast screening in regions where radiologists are scarce,” Tan and co-authors wrote.

The effectiveness of conventional mammography may be limited when it comes to detecting breast cancer in women with dense tissue. Automated 3-D breast ultrasound can help with this as a screening supplement, but it adds more strain to the labor-intensive clinical workload, the researchers noted.

It’s possible that AI could help by combining the automated 3D ultrasound method with tools that would automatically triage women with dense tissue who are being screened for breast cancer, but Tan and colleagues noted that there are no studies that demonstrate the potential of this explored tactics.

Tan’s group sought to compare the detection efficiency of AI systems using mammography data, automated 3D breast ultrasound data and a combination of both modalities, and compared the performance of the respective systems with that of radiologists. To do this, the team conducted a study involving paired digital mammography and ultrasound examinations of a cohort of 430 women with dense breasts; all cases were analyzed by the two AI systems – one for mammography and one for ultrasound and the multimodal AI system; four radiologists read a subset of 152 cases.

The researchers found that the multimodal AI system delivered higher specificity and comparable sensitivity than that of single-modality AI systems. The multimodal system had a total area under the curve (AUC) of 0.865 (with 1 as a reference), while the AI ​​Mammography and AI Ultrasound AUCs were 0.832 and 0.841, respectively.

Must Read
New York City Delays Enforcement of AI Bias Law

The team also found that while the multimodal AI system was less sensitive compared to double read pairs and all four radiologists combined, it was higher or comparable to single readers and showed much higher specificity compared to all other reading techniques.

Comparison of reading techniques for identifying breast cancer in women with dense tissue
Measuring unit Mammography with one reader (average of 2 readers) Ultrasound with one reader (average of 2 readers) Double reading mammography Double reading ultrasound Multimodality double reading (average of 4 reading pairs) Panel of 4 readers Multimodal AI system
Sensitivity 78.5% 89.8% 84.1% 94.2% 94.5% 93.2% 81.1%
Specificity 51.2% 30% 39.7% 15.5% 20.5% 32.7% 95.5%

While the research results are promising, according to Tan and colleagues, more research needs to be done on the effectiveness of AI systems and their combination in multimodal screening settings.

“Clinically relevant endpoints, such as the number of benign biopsies avoided or the improvement in detection sensitivity, need further evaluation,” they concluded.

Copyright © 2023 AuntMinnie.com



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments