R2 Technology is leveraging knowledge gained in developing computer-aided detection to make products easier and more efficient to use.
R2 Technology is leveraging knowledge gained in developing computer-aided detection to make products easier and more efficient to use.
The company, acknowledged as a pioneer of CAD, unveiled at the RSNA meeting two new technologies: Citra Mammography Applications Suite, a software toolset developed in concert with Sectra, and ImageChecker Version 8.1, the latest revision of its CAD mammography product. The two reflect a new corporate philosophy to get the most from the underpinnings of CAD.
Citra includes software tools that automatically scale mammograms to each other, sizing the previous mammogram with the current image to the same dimensions, regardless of the system used to capture the images or how much compression was applied to the breasts.
"By virtue of being able to do CAD, we have learned a lot about the digital image, and we can use that to do a lot to improve the customer experience," said John Pavlidis, president and CEO of R2 Technology.
Version 8.1 allows users to adjust the sensitivity of the machine to flag more or fewer anomalies. Separate controls are available for microcalcifications and for masses. This choice relates to the different sensitivities for each of these functions: CAD has a 98% sensitivity for microcalcifications and 86% for masses, according to peer-reviewed research conducted with ImageChecker.
"This gives users the opportunity to get the highest sensitivity for microcalcifications, which leads to very few false positives, and, depending on their needs, allows maybe a lower sensitivity setting for masses so as to reduce the number of these false positives," Pavlidis said.
The latest release, which is integrated into new ImageChecker mammography products and available as an upgrade to the company's installed base, includes two other productivity tools. Malc Mark identifies regions of breast tissue that contain both a mass and calcifications. EmphaSize assigns variable-sized CAD marks to indicate the potential for malignancy. Small marks reflect a lower chance of malignancy, and large marks indicate more suspicious findings.
"We are using CAD to add intelligence to the raw data to help smooth the transition to digital," Pavlidis said.
This transition has involved a complete change in the workflow that physicians had relied on for decades. Those making the transition to digital mammography may have to deal with two types of media, analyzing films from prior examinations and soft-copy display from recent digital exams. Citra is an example of how digitized films can be brought more efficiently into the review process.
"We can use CAD intelligence to improve the display and workflow for physicians and help them regain some of the productivity they have lost by going to digital," Pavlidis said.
R2 focused initially on the development of productivity tools regarding mammography, but more are coming for other modalities. Some have already arrived.
ImageChecker CT Lung offers an algorithm called AutoPoint, which highlights nodules, then compares new and past images to demonstrate how they have changed over time. This tool, available as an option on Version 2.0, provides a tabular report featuring information on nodule volume and density changes and predicts, based on its history, the time needed for each nodule to double in size, an important indicator of whether the nodule is malignant.
The company has also developed a tool to help in the diagnosis of pulmonary embolism. This algorithm helps detect obstructions, called filling defects, in the pulmonary artery. R2's PE tool helps evaluate suspicious regions by calculating vessel diameter, percent occlusion, and size of the detected obstruction, typically either a tumor or a blood clot.
Research indicates that the PE tool has a sensitivity of 88% for the detection of segmental pulmonary embolism and 78% for the detection of subsegmental pulmonary emboli. It is not as specific, however, as users would like, a problem the company is working on.
"There are always technical challenges," Pavlidis said. "That is what keeps our engineers and scientists busy."
The PE tool addresses a huge problem. The National Heart, Lung, and Blood Institute estimates that 600,000 U.S. residents may be affected each year by pulmonary emboli, leading to 60,000 deaths annually. The signs of these emboli can be found in the results of multislice CT scans, but may they become apparent only after plowing through hundreds or thousands of slices. R2's PE tool helps dig out the evidence of such problems.
The need for such assistance in medical imaging has never been greater. Radiology is bursting at the seams with data coming from megaslice CT, data-intense MR applications, and digital replacements of film-based radiography. Computer intelligence can help.
"There is a need for CAD across many applications," Pavlidis said. "Our challenge is to find the biggest opportunities."
CAD will be effective, however, only if it is part of the IT environment. The last 18 months have seen an emphasis on integration with PACS: first RIS, most recently 3D visualization tools. CAD, with its increasing popularity, has fallen into this latter category.
"Customers ask us, 'Does it work on my PACS?'" Pavlidis said. "This is why for us connectivity was a major theme at this RSNA meeting."
R2 has developed the means for transferring results from ImageChecker products, including those focused on digital mammography and CT, into PACS. Connectivity is a natural extension of R2's strategy to make life easier for its customers and an essential part of increasing the reach and popularity of its products.
Photon-Counting Computed Tomography: Eleven Takeaways from a New Literature Review
May 27th 2025In a review of 155 studies, researchers examined the capabilities of photon-counting computed tomography (PCCT) for enhanced accuracy, tissue characterization, artifact reduction and reduced radiation dosing across thoracic, abdominal, and cardiothoracic imaging applications.
Emerging AI Mammography Model May Enhance Clarity for Initial BI-RADS 3 and 4 Classifications
May 21st 2025In a study involving over 12,000 Asian women, researchers found that an artificial intelligence (AI) model converted over 83 percent of false positives in patients with initial BI-RADS 3 and 4 assessments into benign BI-RADS categories.
Can AI Predict Future Lung Cancer Risk from a Single CT Scan?
May 19th 2025In never-smokers, deep learning assessment of single baseline low-dose computed tomography (CT) scans demonstrated a 79 percent AUC for predicting lung cancer up to six years later, according to new research presented today at the American Thoracic Society (ATS) 2025 International Conference.