*|MC_PREVIEW_TEXT|*

AI for Brain MRI, Reporting with LLMs, and 4D Mammo
August 21, 2025
site logo

Together with

partner logo

“Medicine moves at the speed of evidence.”

Hugh Harvey, MD.

New automated software solutions are becoming available for analysis of arterial plaque, a major risk factor for heart disease. In this webinar today at 12 p.m. ET, hear clinicians discuss their real-world experiences with Elucid’s PlaqueIQ solution for plaque analysis.

Imaging Wire Sponsors

AGFA HealthCare  •  Blackford  •  Calantic by Bayer  •  CARPL.ai  •  DeepHealth  •  Enlitic  •  Gleamer  •  Intelerad  •  Kailo Medical  •  Mach7 Technologies  •  Medality  •  Merge by Merative  •  Microsoft  •  Philips  •  Prenuvo  •  Quibim  •  Riverain Technologies  •  Siemens Healthineers  •  SpinTech MRI  •  United Imaging  •  Us2.ai  •  Visage Imaging

Brain Imaging

AI for Brain MRI

What if you could speed up brain MRI exams by performing fast scans for most patients, and reserving complex sequences for the patients who need them? A hint of that future comes from a new study in which AI showed progress in helping radiologists interpret scans with fewer sequences.

MRI can visualize minute structures in the body, especially in the brain, but it’s one of the trickiest imaging modalities to operate.

  • There’s an alphabet soup of MRI pulse sequences, and the modality’s complexity is multiplied when contrast has to be used. 

Breast MRI experts have been experimenting with abbreviated scanning protocols that speed up image acquisition and interpretation by using fewer and less complex sequences.

  • Researchers applied that concept to MRI brain imaging in a new European Journal of Radiology paper in which they tested Cerebriu’s Apollo AI algorithm with 414 patients from four hospitals in Denmark.

Apollo processes three brain MRI sequences (DWI, SWI or T2* GRE, and T2-FLAIR) and can detect critical findings like brain infarcts and intracranial hemorrhages and tumors while the patient is still on the table.

  • If an abnormality is detected, Apollo prompts technologists to acquire a fourth sequence, such as T1-weighted imaging.

That sounds great, but how well does Apollo work in the real world? 

  • Researchers compared the algorithm’s performance to that of expert neuroradiologists in multiple workflows, such as reading three- and four-sequence MRI scans with and without AI assistance. 

Compared to neuroradiologists using the four-sequence MRI protocol without AI assistance, they found…

  • Apollo’s sensitivity was better than neuroradiologists for brain infarcts (94% vs. 89%) and intracranial tumors (74% vs. 71%) but slightly lower for intracranial hemorrhages (82% vs. 83%).
  • AI’s specificity was somewhat lower, however, for brain infarcts (86% vs. 99%), intracranial hemorrhages (84% vs. 99%), and intracranial tumors (62% vs. 97%). 
  • When neuroradiologists had AI findings in addition to the four-sequence protocol, tumor detection sensitivity improved slightly, but specificity fell. 

While Apollo’s sensitivity was a benefit, the researchers said its low specificity “presents a challenge” and could result in unnecessary additional sequences or contrast administration. 

  • Specificity could be affected by age-related changes in older patients, as well as differences in MRI scanner models used.

The Takeaway

The new findings show that AI-aided MRI scan assistance still needs refinement. But it’s still early days for Cerebriu and Apollo (which has the CE Mark but not FDA clearance), so watch this space for more updates. 

Patient-Centered Innovation with Photon-Counting CT

Siemens Healthineers is committed to transforming medical imaging with its NAEOTOM Alpha class, now featuring three photon-counting CT systems. Find out how they can enhance your diagnostic results.

sponsor logo

Echo AI – Revolution or Replacement?

How is AI echocardiography viewed by the clinicians who use it? Clinicians note that AI echo can enhance measurement accuracy and improve workflow efficiency, and share their secrets for successful implementation. Learn more on this page from Us2.ai.

sponsor logo

The Wire

  • MRI of Prenatal Pesticide Exposure: Children who experienced prenatal pesticide exposure had brain abnormalities detected on MRI scans. In a new study in JAMA Neurology, researchers recruited 270 children in New York City from 1998 to 2006 whose mothers were exposed to the insecticide chlorpyrifos while pregnant when their apartments were fumigated (residential use of chlorpyrifos was banned in 2001). On MRI scans, children with higher chlorpyrifos exposure levels had brain abnormalities and reduced blood flow that represented the pesticide’s impact on brain function. 
  • FDA Updates Breakthrough List: The FDA this week updated the list of medical devices it is reviewing through its Breakthrough Devices Program. For 2025 through June, FDA granted breakthrough designation to 136 devices, compared to 165 for all of 2024. For the life of the program, FDA has given breakthrough designation to 1.2k devices, with cardiovascular products dominating at 243 submissions compared to 43 for radiology (10th among medical specialties). Twenty-one breakthrough devices have received marketing authorization so far in 2025.
  • Better Radiology Reports with LLMs: Multimodal large language models can be used to dictate radiology reports with performance better than conventional speech recognition. In a new study in European Radiology, German researchers compared 480 reports created with ChatGPT-4o and Claude 3.5 Sonnet to traditional speech recognition software, finding the LLMs produced reports faster (39 vs. 88 seconds) and needed fewer corrections (0.9-1.0 vs. 2.4 corrections), and even lowered costs ($2.30-$2.40 vs. $3.00). 
  • Mammo AI Reduces Screening Workload: Researchers are honing in on the best ways to use AI for breast screening. In a new study in Radiology, Dutch researchers reduced radiologist workload by excluding 38% of mammography exams in which AI had high confidence that cases were normal, while radiologists used double-reading without AI for the rest. In mammograms of 15.5k women, this hybrid AI strategy compared to conventional double-reading had similar cancer detection rates (6.6% for both) and recall rates (24% for both). 
  • Akumin to Deploy Lunit AI for DBT: Lunit scored another important mammography AI contract, signing an agreement to provide its Insight DBT breast cancer AI software to Akumin, which operates 150 fixed-site imaging centers in the U.S. This is Lunit’s second recent victory for Insight DBT with a large U.S. imaging services provider, coming on the heels of an agreement with SimonMed in April. Insight DBT received FDA clearance in November 2023 and is key to Lunit’s focus on penetrating the U.S. market.
  • Ultrasound Guides Thyroid Surveillance: Rather than proceeding directly to surgery, patients with low-risk thyroid cancer can be safely followed with active surveillance including ultrasound, blood tests, and clinical assessment. In a new paper in JAMA Surgery, Canadian researchers used active surveillance for 200 patients with low-risk thyroid cancer (< 2 cm). After a median of six years, there were no thyroid cancer deaths or distant metastatic disease, and conversion to surgery rates were low, especially for patients 45 to 64 (21%) and older than 65 (5.1%).
  • Harrison.ai Touts Model Performance: AI developer Harrison.ai is highlighting the performance of its Harrison.rad.1 radiology-specific foundation model in a large-scale evaluation by ACR and Mass General Brigham. In the Healthcare AI Challenge, Harrison.rad.1 outperformed other multimodal foundation models on medical tasks with 82% accuracy. On another test – differentiating AI-generated from human-written chest radiography reports – the AI model had acceptability rates of 65% versus 80% for clinicians’ reports. The findings show the rapid progress of AI in drafting reports.
  • Sectra Gets Big Canadian PACS Contract: Sectra scored a major victory this week by landing a contract to provide its cloud-based enterprise image management software to a hospital network in Ontario, Canada. Sectra One Cloud will be deployed across 13 hospital sites in ENHANCE Ontario’s regional network, which serves 1.5M people across central eastern Ontario and conducts over 1M imaging exams annually. 
  • Giving ChatGPT-5 a Test Drive: The newest version of ChatGPT – GPT-5 – was released earlier this month, and researchers are already testing it for radiology applications. In a series of papers posted to arXiv, researchers compared GPT-5 to GPT-4o for mammography and brain MRI image interpretation, finding that while GPT-5 showed big improvements over the previous version, it’s still not at a level acceptable for clinical use. GPT-5 performed better in a study for medical decision support by integrating patient narratives, structured data, and medical images. 
  • Radiopharma Firm Taps Former FDA Chief: Radiopharmaceutical firm Nucleus RadioPharma hired former FDA commissioner Stephen Hahn, MD, as CEO. A radiation oncologist by training, Hahn served as FDA commissioner from 2019 to 2021, and prior to that was chief medical executive at MD Anderson Cancer Center in Houston. Nucleus is a contract development and manufacturing organization founded in 2022 that specializes in radiopharmaceuticals, and is building a network of facilities to integrate theranostic radiopharmaceutical R&D and production under one roof. 
  • Startup Images Patient with ‘4D Mammography:’ A startup spun out of Duke University acquired the first human image with a technology it’s calling 4D mammography. Calidar’s technology measures X-ray diffraction, in which X-rays scatter at the molecular level, producing structural signatures that reflect internal breast tissue composition. Calidar acquired the image as part of a new study to assess how well 4D mammography differentiates malignant from benign tissue and how it compares to conventional 2D and 3D mammography systems.

AI-Powered Population Health

DeepHealth closed its acquisition of iCAD, a major step in DeepHealth’s vision to build the future of AI-powered population health. In this Imaging Wire Show, we talked to company executives Kees Wesdorp and Niccolo Stefani about the acquisition and their strategic roadmap.

sponsor logo

Transform Imaging Data into Actionable Predictions

When you choose Quibim, you get more than a partner for detecting and diagnosing prostate cancer on MRI scans. Learn how they can help you transform imaging data into actionable predictions by booking a demo today.

sponsor logo

Rapid AI Deployment in Emergency Care

University Hospitals used CARPL to deploy AZmed’s fracture detection tool directly in the emergency department, reducing interpretation time by 30% without disrupting workflows. Learn how UH accelerated AI deployment.

sponsor logo

The Resource Wire

  • 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.
  • A New Solution for Chest AI: Gleamer’s ChestView AI solution is a computer-aided detection (CADe) model cleared by the FDA for simultaneously detecting multiple findings on chest X-ray. Discover how it enhances explainability and confidence compared to traditional triage-focused CADt solutions.
  • How Intelerad Is Improving the Client Experience: Intelerad has launched a new client obsession initiative to improve the client experience with the company. We caught up with Chief Client Officer Eric Grunden to discuss the project in this episode of The Imaging Wire Show.
  • The Future of Radiology: In this episode of The Radiology Report, Medality’s Daniel Arnold sits down with Marc Gosselin, MD, from Vision Radiology for a thought-provoking conversation on burnout, balance, and the future of radiology.
  • 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.  
  • Diagnostic X-Ray Battery for Life: Discover the Radiography 7000 M from Philips, a premium mobile radiography system offering enhanced care and improved operational efficiency. Plus you’ll never have to pay to repair or replace the main system battery. Terms and conditions apply. 
  • MRI Access and the Cost of Inpatient Stays: Longer inpatient stays due to delayed MRI access are a long-standing and costly issue for hospital systems. Find out how STAGE from SpinTech MRI can reduce your MRI backlog and inpatient stays by shortening brain scan times by 30%.
  • Optimize Radiology Workflows: Harness cloud and AI technology to help your radiology teams unlock insights, increase efficiencies, and improve patient care. Learn more about an integrated approach to AI in radiology in this e-book from Microsoft. 
  • Cutting-Edge PET/CT to Support Theranostics: Theranostics is an exciting new field that combines diagnostics and therapy. Discover how Florida Theranostics is using United Imaging’s uMI Panorama PET/CT scanner to establish a high-quality level of patient care. 
  • 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. 
  • Enterprise Imaging in Action: Find out how Presbyterian Healthcare increased patient engagement and met regulatory requirements by image-enabling their patient portal in this downloadable white paper from Merge.
  • Adding Digital Pathology to Enterprise Imaging: In this on-demand video from SIIM 2025 hosted by Mach7, watch as a panel of industry experts discusses how to add digital pathology images into a healthcare organization’s overall enterprise imaging strategy.
  • Opportunistic Detection of CAC and Pulmonary Nodules: Achieve a newfound certainty of search for thoracic CT when using ClearRead CT from Riverain Technologies. It’s a natural addition for opportunistic CAC scoring and nodule detection, or as part of a CT lung cancer screening program. 
  • Planning for Data Migration Success: When UCSF Health’s enterprise imaging team needed to bring two new hospitals into their network, a new data migration project was born. Watch this on-demand webinar to learn how UCSF ensured a successful migration by working with Laitek, an Enlitic company.
  • AI Acceleration without the Hype: The Visage AI Accelerator program is an innovation magnet – an end-to-end AI solution that bridges research and diagnostic imaging on the same, unified platform. Learn how Visage 7 provides synergistic, smart workflows and flawless third-party AI integration.
  • Discover the Blackford Difference: AI is no longer a futuristic nice-to-have in healthcare, it’s an essential tool. Yet for many providers, the path forward with AI remains unclear. That’s where Blackford comes in – discover the difference today.
  • Synoptic Reporting of Liver Masses: Contrast-enhanced ultrasound exams for focal liver lesions can be challenging to report in at-risk patients. Learn how Kailo Medical’s synoptic reporting solutions can make reporting easier while improving LI-RADS categorization.

The Industry Wire

  1. Doctors are struggling to make ends meet.
  2. Pediatric group recommends COVID vax against federal policy.
  3. Migrating ticks could pose an unexpected challenge for doctors.
  4. Texas measles outbreak is over.
  5. New study casts doubts on RFK Jr.’s gutting of vaccine panel.
  6. Judge dismisses Elevance Health Medicare Advantage stars case.
  7. The future of food as medicine programs.
  8. Rule limiting broker payments in Medicare Advantage overturned.
  9. CMS initiative targets undocumented immigrants in Medicaid.
  10. Large employers forecast 9% hike in healthcare costs next year.