|
Ensemble Mammo AI, FDA AI Recalls, and MRI of Body Fat
August 25, 2025
|
|
|
|
|
Together with
|
|
|
|
“Whether AI outperforms the traditional radiologist capabilities in the next 6 months or not for another 20 years, it is important to remain receptive and adaptive as new technology revolutionizes radiology and medicine as a whole.”
|
|
Smith J et al, in a JACR article on AI’s impact on radiology.
|
|
|
|
If one AI algorithm works great for breast cancer screening, would two be even better? That’s the question addressed by a new study that combined two commercially available AI algorithms and applied them in different configurations to help radiologists interpret mammograms.
Mammography AI is emerging as one of the primary use cases for medical AI, understandable given that breast imaging specialists have to sort through thousands of normal cases to find one cancer.
Most of these studies applied a single AI algorithm to mammograms, but multiple algorithms are available, so why not see how they work together?
- This kind of ensemble approach has already been tried with AI for prostate MRI scans – for example in the PI-CAI challenge – but South Korean researchers writing in European Radiology believed it would be a novel approach for mammography.
So they combined two commercially available algorithms – Lunit’s Insight MMG and ScreenPoint Medical’s Transpara – and used them to analyze 3k screening and diagnostic mammograms.
- Not only did the authors combine competing algorithms, but they adjusted the ensemble’s output to emphasize five different screening parameters, such as sensitivity and specificity, or by having the algorithms assess cases in different sequences.
The authors assessed ensemble AI’s accuracy and ability to reduce workload by triaging cases that didn’t need radiologist review, finding…
- Outperformed single-algorithm AI’s sensitivity in Sensitive Mode (84% vs. 81%-82%) with an 18% radiologist workload reduction.
- Outperformed single-algorithm AI’s specificity in Specific Mode (88% vs. 84%-85%) with a 42% workload reduction.
- Had 82% sensitivity in Conservative Mode but only reduced workload by 9.8%.
- Saw little difference in sensitivity based on which algorithm read mammograms first (80.3% and 80.8%), but both approaches reduced workload 50%.
The authors suggested that if applied in routine clinical use, ensemble AI could be tailored based on each breast imaging practice’s preferences and where they felt they needed the most help.
The Takeaway
The new results offer an intriguing application of the ensemble AI strategy to mammography screening. Given the plethora of breast AI algorithms available and the rise of platform AI companies that put dozens of solutions at clinicians’ fingertips, it’s not hard to see this approach being put into clinical practice soon.
|
|
Revolutionize the Reading Experience
With intelligent automation and AI‑powered workflow, PowerScribe One from Microsoft allows radiologists to generate and communicate high‑quality, consistent reports – and get more done in less time.
|
|
Unprecedented Insights Made Possible with AI
With the largest normative dataset of whole-body imaging in the world, Prenuvo’s AI researchers partner with the best academic minds to understand – like never before – what “normal” aging means. Learn about their work today.
|
|
Equitable Study Distribution with an Automated Worklist
Automated worklist functionality can distribute medical imaging studies more equitably than manual study selection. Discover how Merge Workflow Orchestrator enabled one institution to achieve balance in this downloadable white paper.
|
|
- Attention on AI Recalls: As attention turns to the safety of medical AI, a study in JAMA Health Forum examines FDA recall rates for AI-enabled medical devices. Researchers analyzed 950 AI products, almost all cleared through the 510(k) pathway, with recall events occurring in 6.3%. Over 43% of recalls occurred within 12 months of product authorization, almost double the rate for all 510(k) clearances. Recall rates were higher for AI from public companies (OR = 5.9) and algorithms without clinical validation (OR = 2.8).
- Body Fat Linked to Heart Aging: Another study is highlighting the negative health effects of visceral adipose tissue, or fat deep inside the abdomen around organs. Researchers in a new study in European Heart Journal analyzed data from 21.2k participants in the UK Biobank study who got cardiac MRI scans and whole-body MRI adipose tissue assessment. People with higher visceral adipose tissue volume, muscle adipose tissue infiltration, and other factors had a greater difference between the predicted age of their heart and its actual age.
- FDA Clears Coronary Plaque Software: Another company is entering the red-hot arterial plaque analysis segment. Australian software developer Artrya got FDA clearance for its Salix Coronary Plaque AI-powered cloud-based solution, which calculates plaque burden and composition from coronary CT angiography exams. The solution integrates with Artrya’s Salix Coronary Anatomy platform, and would qualify for the Category I CPT code for automated plaque analysis of CCTA scans, which will pay $950 per assessment starting January 2026.
- AI for Knee MRI Predicts Osteoarthritis: Researchers from China developed an AI algorithm that includes MRI scans and clinical data to help clinicians determine whether a patient’s knee osteoarthritis would get worse. In a paper in PLOS Medicine, authors used 1.8k load-bearing knee MRI scans to develop and test their LBTRBC-M model, which also included biochemical test results and clinical data. LBTRBC-M improved resident physicians’ accuracy in predicting whether knee osteoarthritis would worsen over the next two years (65% vs. 47%).
- Active Outreach Improves CT Lung Screening: New research shows the importance of active patient outreach in boosting CT lung cancer screening rates, this time among Asian Americans in California. In a new study in The Annals of Thoracic Surgery, researchers used lay navigators to contact eligible screening candidates and then monitor their progress for up to six months to ensure they got scanned. The process generated higher screening completion rates compared to a passive outreach group that only got emails (11% vs. 4.4%).
- AI-Driven Lung Segmentation: An AI algorithm being developed by Quibim helped radiologists interpret thoracic CT exams by automatically segmenting multiple lung lesions. Quibim’s LLSB-CFPR algorithm segments not only primary lung cancer lesions but also more distant tumors to give a more comprehensive representation of tumor burden, and had lesion detection sensitivity of 85% on internal and external datasets. Quibim offers LLSB-CFPR as part of its QP-Insights research package and plans to bundle it into its QP-Lung clinical product on receipt of marketing authorizations.
- Sarcopenia Predicts COPD Mortality: Critically ill COPD patients who also showed signs of sarcopenia (low muscle mass) on CT scans were 4X more likely to die within a year in a new study in European Journal of Radiology. Researchers from China scanned 148 hospitalized COPD patients with chest CT and used TomoVision’s sliceOmatic software to diagnose sarcopenia at the T12 vertebra level. At one year, sarcopenia patients had higher mortality rates for COPD (19% vs. 4.7%) and for all causes (29% vs. 6.3%).
- AbbaDox Rolls Out AI Upgrade: Radiology workflow software developer AbbaDox is rolling out its AI-powered, cloud-based AbbaDox AI solution after piloting the software at imaging centers. AbbaDox AI includes AI-powered fax intake, conversational voice AI, kiosk and digital patient check-in, and automated follow-up recommendations. Pilot tests found it had 96% accuracy for extracting fax data, placed 2.4k reminder calls to patients and secured 1k appointment confirmations in one week, processed 10k reports, and produced $46k in revenue on a $10k investment.
- Do Doctors Like AI-Drafted Patient Reports? A possible use case for generative AI in healthcare is drafting patient reports with minimal physician effort. But do doctors actually like these reports? In a new study in JAMA Network Open, Stanford Health Care researchers developed an AI patient report tool and then surveyed 244 clinicians from different specialties on their opinions of its output. The highest satisfaction was found for laboratory and imaging reports (72% and 63%), while pathologists were less enthused.
- RADPAIR Signs with PACS Provider: AI-powered reporting company RADPAIR signed an agreement to integrate its technology with AdvaHealth Solutions, a Singapore-based PACS provider. Users of AdvaHealth’s AdvaPACS software will gain access to RADPAIR’s reporting capabilities, with the joint solution to be demonstrated at the Singapore Congress of Radiology August 29-31.
- Surgical Navigation Firm Raises $20M: Cleveland Clinic spinoff Method AI raised $20M in Series A funding to further develop its image-guided surgical navigation platform to improve cancer-free survival by ensuring tumors are completely excised. Method AI’s technology uses continuous 3D ultrasound during surgical procedures to create 3D maps of patient anatomy that can be compared in real time to an AI-generated surgical plan.
|
|
The Benefits of Structured Reporting
Kailo Medical hopes to revolutionize radiology with its structured reporting solutions. At SIIM 2025, we talked to Lauren Therriault and Denholm Rhys about the latest developments at the company and why structured reporting is a benefit to radiologists.
|
|
Purchasing a Cloud-Hosted Enterprise Imaging Solution
The decision to move your enterprise imaging to the cloud is significant. But how do you choose the right solution? Discover the 10 critical questions to ask in this guide from AGFA HealthCare.
|
|
Introducing Voice-Controlled Interventional X-Ray
United Imaging’s uAngio AVIVA is the industry’s first intelligent voice-controlled interventional X-ray system, enabling interventional staff to keep their focus where it belongs: on the patient. Discover AVIVA’s advantages on this page.
|
|
- What’s Next for AI for Cancer Detection? AI is transforming the fight against cancer by enabling faster and more accurate cancer detection. Read this article from DeepHealth to learn how the company is pioneering new ways to advance cancer screening and broader imaging-based care.
- The Benefits of Operational AI: Explore the transformative potential of operational AI in healthcare in this on-demand webinar hosted by Blackford. Learn from the company’s partners how AI can help your practice operate more efficiently.
- The Road to Cloud-Based PACS: Radiology facilities are turning to cloud-based PACS like Visage’s Visage 7 to solve their medical image management needs. Learn about their experiences in this Imaging Wire Show with Amy Thompson of Signify Research and radiologist Marc Kohli, MD.
- Advancing AI-Driven Data Migration: Enlitic has joined forces with GE HealthCare to power the data migration feature in GE’s newly announced Genesis cloud portfolio. Learn how Enlitic’s AI-driven data migration facilitates large-scale transfers of high-quality medical imaging data.
- AI Tools for Lung Cancer Screening: CT lung cancer screening is gaining momentum around the world. Learn about AI-based nodule detection tools that can improve the accuracy of low-dose CT scans in this video from Riverain Technologies.
- Ahead in the Cloud: What do healthcare providers need to consider as they adopt cloud-based solutions for medical imaging? Read this article written for Mach7 Technologies by Eliot Siegel, MD, to learn the important role cloud-based technologies are having in shaping the future of healthcare.
- 2 Questions about AI for Radiology Leaders: Are today’s radiology AI solutions solving the right problems? And are there other solutions available for AI of brain MRI? Read this article from SpinTech MRI to learn how their STAGE solution can optimize MRI utilization.
- AI Applications in Neuroradiology: What are the most common AI applications in neuroradiology? This downloadable e-book from Bayer reviews the most common AI applications for brain imaging and the evidence behind them.
- The Transformative Role of AI in Radiology: In this episode of The Radiology Report podcast, Medality’s Daniel Arnold sits down with Dr. John Simon, who shares his insights into the transformative role of AI in radiology and its ability to enhance efficiency, improve patient care, and unlock new diagnostic possibilities.
- Easily Share Patient Images and Reports: Physicians can easily share patient images and reports from a personal worklist using Intelerad’s InteleShare Physician Portal. Discover how easily it works in this self-guided demo.
- A New Resource for AI of MRI: Gleamer is expanding into AI of MRI with its acquisition of innovative AI developers Pixyl and Caerus Medical. Learn how the company is creating the most comprehensive AI portfolio for medical imaging.
- Validating AI at Scale: Radiology Partners leveraged CARPL to benchmark four MSK AI models, reducing ground truthing workload and enabling same-day validation. Learn how RadPartners benchmarked AI.
- Advanced AI for Prostate MRI: QP-Prostate from Quibim is your advanced solution for detection and diagnosis of prostate cancer from MRI scans. Discover how it streamlines your workflow by detecting suspicious lesions, segmenting the prostate, and ensuring compliance with PI-RADS V2.1 guidelines. Learn more on this page.
- How Sports Can Help Save Lives: The Everton F.C. Premier League club integrated AI-enabled point-of-care diagnostics into a one-stop community hub for early diagnosis of heart failure and COPD using AI echocardiography from Us2.ai. Discover how this project is making early detection more accessible than ever.
- The Future of Fluoroscopy Is Here: The future of fluoroscopy has arrived. The LUMINOS Q.namix fluoroscopy systems from Siemens Healthineers are available on the U.S. market. Discover why they have already earned the prestigious Red Dot Design Award for intuitive design and user-centric innovation.
- Ensuring the Safety and Well-Being of Patients Undergoing MR: It’s important to create a safe environment for every patient, every scan. Unlock how thoughtful planning and innovative technology from Philips can reduce risks and enhance care during MRI procedures.
|
|
|
|
|
