AI Solution Provides Radiologists with 'Second Pair' Of Eyes to Detect Breast Cancers

By MedImaging International staff writers
Posted on 22 Feb 2024

Mammography, a key tool in breast cancer screening, has proven effective in improving patient outcomes and reducing mortality by identifying breast cancers at an earlier, more manageable stage. Despite its effectiveness, it's estimated that 20-30% of interval cancers, which should have been detected during the previous mammogram, are missed. Additionally, many suspicious findings are ultimately found to be benign. To enhance sensitivity in detecting disease, European guidelines recommend that two radiologists independently review each screening mammogram. However, there's a notable shortage of breast radiologists globally, and training a radiologist to proficiently interpret mammograms can take more than a decade. In response to these challenges, Artificial Intelligence (AI) has been suggested as a potential automated second reviewer for mammograms. This could not only reduce the radiologists' workload but also enhance screening accuracy. AI has shown promising results in retrospective studies, where it was used to decide whether mammogram examinations should be read by one or two radiologists, and in providing radiologists with computer-aided detection (CAD) markings to highlight suspicious areas, thereby reducing the incidence of false negatives. Now, an advanced AI solution provides radiologists with a 'second pair' of eyes to help detect cancers earlier and reduce recall rates.

ScreenPoint Medical’s (Nijmegen, Netherlands) Transpara breast AI is designed to assist radiologists in the early detection of breast cancers and enhance the efficiency of breast screening programs. Utilized by numerous leading institutions across over 30 countries, Transpara is integrated into the radiologists' workflow. Studies have shown that Transpara can identify up to 45% of interval cancers earlier, while also reducing the workload and optimizing workflow processes. Transpara demonstrates consistent performance across various countries, patient ages, breast densities, and ethnic backgrounds. To date, Transpara breast AI has analyzed over five million mammograms, including more than one million Tomosynthesis (3D) exams, supporting radiologists in mammography examinations. The system has received FDA clearance and European regulatory approval (CE Mark) for use with both 2D and 3D mammography from multiple manufacturers.


Image: AI markers highlight suspicious calcifications and soft tissue lesions together with an objective Region Score (Photo courtesy of ScreenPoint Medical)

Transpara continues to deliver proven clinical and workflow benefits in mammography screening in global practice and clinical research. A first-of-its-kind randomized controlled trial, the MASAI-trial, investigated cancer detection rates and the types of cancers detected using AI-supported screening. This trial found notable improvements in cancer detection when AI support was used, compared to traditional double reading methods without AI. Another prospective clinical trial explored the use of AI in safely reducing the radiologists' workload. It focused on using AI to exclude low-risk cases from human review and applying double reading to the remaining cases.

"We are glad that users and researchers continue to see value from Transpara in improving the mammography screening process,” said Mark Koeniguer, ScreenPoint Medical CEO. “It's important to note that these studies reflect global consistency in Transpara's performance. In fact, we now have retrospective, prospective and randomized controlled trials all showing that Transpara provides radiologists with the ability to effectively detect cancer early while keeping recall rates consistent. Women should not have to compromise."


Latest Radiography News