|
New Mammo AI Insights | AI Hype at HIMSS
March 17, 2025
|
|
|
|
|
Together with
|
|
|
|
“[M]ost of HIMSS was about AI and a lot of hype, and despite the many presentations and success stories, there is definitely still a lot of work to do to create the level of trust that is needed for healthcare providers and patients to embrace this technology.”
|
|
Herman Oosterwijk, in a review of AI at HIMSS 2025.
|
|
|
In this episode of The Imaging Wire Show, we talk to radiologist Hari Trivedi, MD, of Emory University about disparities in breast cancer detection and how artificial intelligence (AI) can possibly reduce them.
|
|
|
|
Breast screening is becoming one of the most promising use cases for AI, but there’s still a lot we’re learning about it. A new study in Radiology: Artificial Intelligence revealed new insights into how well mammography AI performs in a screening environment.
As we’ve reported in the past, mammography is one of radiology’s most challenging cancer screening exams, with radiologists sorting through large volumes of normal images before encountering a case that might be cancer.
In the new study, researchers applied Lunit’s Insight MMG algorithm to mammograms in a retrospective study of 136.7k women screened in British Columbia from 2019 to 2020.
- Canada uses single reading for mammography, unlike the double-reading protocols employed in the U.K. and Europe.
AI’s performance was compared to single-reading radiologists using various metrics and follow-up periods, finding …
- At one-year follow-up, AI had slightly lower sensitivity (89% vs. 93%) and specificity (79% vs. 92%) compared to radiologists.
- At two-year follow-up, there was no statistically significant difference in sensitivity between the two (83.5% vs. 84.3%, p=0.69).
- AI’s overall AUC at one year was 0.93, but this varied based on mammographic and demographic features, with AI performing better in cases with fatty versus dense breasts (0.96 vs. 0.84) and cases with architectural distortion (0.96 vs. 0.92) but worse in cases with calcifications (0.87 vs. 0.92).
The researchers then constructed hypothetical scenarios in which AI might be used to assist radiologists, finding …
- If radiologists only read cases ruled abnormal by AI, it would reduce workload by 78%, but at a price of reduced sensitivity (86% vs. 93%) and 59 missed cancers across the cohort.
It’s worth noting that Insight MMG is designed to analyze 2D digital mammography exams.
The Takeaway
While the new findings aren’t a slam dunk for mammography AI, they do provide valuable insight into its performance that can inform future research, especially into areas where AI could use improvement.
|
|
|
The Leader in Molecular Imaging
United Imaging’s uMI portfolio of solutions is designed to help you lead the way in molecular imaging. From digital PET/CT systems designed to stand the test of time to the cutting-edge uEXPLORER total-body PET scanner, discover the uMI difference today.
|
|
Unlock Next-Generation AI with Foundation Models
Learn about Microsoft’s new family of cutting-edge multimodal medical imaging foundation models designed for healthcare organizations to test, fine-tune, and build tailored AI solutions specific to their needs, while minimizing extensive compute and data requirements.
|
|
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.
|
|
- Herman O’s Take on HIMSS AI Hype: Imaging IT consultant Herman Oosterwijk serves up a cogent analysis of HIMSS 2025, particularly focusing on the hype around AI and healthcare in Las Vegas. He believes that despite AI’s tremendous opportunity to transform healthcare, physicians at HIMSS expressed major trust issues with the technology, especially with the recent removal of federal regulations that might have ensured patient safety. He also covers AI accountability, governance, archiving results, and ethical and equity issues in AI deployment.
- Welcome to Match Week 2025: March 17 marks the start of Match Week, when U.S. medical students will learn from the Main Residency Match the programs where they will train for their chosen specialty. Residency applicants will learn if they’ve matched on March 17, and will be notified of their specific programs on Friday March 21, an occasion marked by festivities at most medical schools. Radiology has enjoyed close to a 100% fill rate for its residency programs in recent years.
- Korea Reports Lung Screening Success: As more countries ramp up CT lung cancer screening programs, they should examine a new report from South Korea on the first two years of the country’s national screening program. From 2019-2020 some 150k eligible people were screened, 98% of whom were men, with 83% sensitivity, 92% specificity, an 8% false-positive rate, and a 0.5% true-positive rate. Some 2.7% of participants got a radiation dose higher than the standard of 3 mGy CTDIvol, but overexposure rates fell 48% in the second year of screening.
- Better Lung Nodule Follow-Up: CT is great at detecting pulmonary nodules that could be precursors to lung cancer, but that won’t help patients much if nodules aren’t followed appropriately. A new literature review in CHEST tracked different methods for nodule follow-up across 31 published studies, finding nodule tracking systems performed the best, with clinical decision support tools close behind. Better nodule tracking is needed to fully appreciate the benefits of CT lung cancer screening.
- ACR Expands Lung Screening Registry: Better tracking of incidental pulmonary nodules is the goal of ACR in expanding its Lung Cancer Screening Registry to include diagnostic performance feedback on managing IPNs. ACR notes 60% of patients with IPNs are lost to follow-up, but a new registry module will help facilities document their performance in tracking and completing follow-up recommendations. The registry is being renamed the Early Lung Cancer Detection Registry as part of the change.
- Qure Launches AI Platform for LMICs: Qure.ai launched an AI orchestration platform called QureOS that’s targeted at radiology sites in low- and middle-income countries. The company notes that LMICs are experiencing medical personnel shortages that are even more extreme than in more developed countries, and access to AI can help improve care quality in resource-constrained areas. QureOS gives healthcare providers a way to test and deploy quality AI algorithms, while Qure will continue to market its algorithms through platform partners in the U.S., U.K., and European Union.
- AI Denoising Slashes CT Radiation Dose: A specialized CT protocol with AI-based denoising reconstruction slashed CT radiation dose for immunocompromised patients with suspected pneumonia. In a study in Radiology: Cardiothoracic Imaging, Israeli researchers used a CT protocol at 100 kVp with fixed tube current based on BMI (10 mA for <29 and 20 mA for ≥29), with a CNN-based reconstruction algorithm to remove noise from ultralow-dose images. In 54 patients, the protocol achieved radiation dose just 2% of normal-dose CT (0.12 vs. 6.15 mSv).
- Ziosoft Lands Olympus Partnership: Advanced visualization developer Ziosoft signed an agreement to supply its Revoras platform to endoscopy manufacturer Olympus for creating 3D models to aid with surgeries. Olympus wants to provide surgeons with additional AI-powered tools to plan surgeries such as by visualizing vessels and tumors, and the partnership is the first in a series of new innovations. Initial applications will be in thoracic, liver, and urologic surgery.
- 30-Minute Whole-Body MRI Outperforms PET: A 30-minute whole-body MRI protocol performed better than gallium-68 PSMA-PET/CT for local and general prostate tumor staging. In a study in European Journal of Radiology of 52 patients at high risk for metastases, Belgian researchers tested their biparametric prostate assessment protocol based on T2 and DWI-MRI sequences. Whole-body MRI had higher sensitivity for local tumors (87% vs. 77%) and oligometastatic cases (19% vs. 11%), while PSMA-PET/CT detected more small metastatic nodes <5 mm (40% vs. 29%).
- New Canon Clearances: Canon Medical Systems USA announced FDA clearances this month for two new imaging systems. The company got the FDA’s nod for enhancements made to its Aquilion ONE / Insight Edition flagship CT scanner, including the company’s Precise IQ Engine for deep learning reconstruction as well as SilverBeam Filter for optimized CT lung cancer screening at lower radiation doses. Separately, Canon began commercial sales of its Adora DRFi hybrid radiography and fluoroscopy system after getting FDA clearance in December.
- Vave Launches Whole-Body POCUS: Handheld ultrasound developer Vave Health launched Universal Wireless Probe, a new POCUS scanner that supports whole-body imaging in a single PZT probe. The transducer provides both linear and phase imaging and uses a swappable battery for longer imaging times in remote locations.
- Spectrum Dynamics Taps CEO: Digital SPECT camera developer Spectrum Dynamics Medical tapped Tomer Gabay as its new CEO. Gabay previously was the company’s CFO, and has been with the firm for the past nine years. He replaces Gilad Yoeli, an 18-year veteran with the company who had been CEO for the past six years.
|
|
Perspectives on Cybersecurity in Medical Imaging
Healthcare organizations are a prime target for cyberattacks, so how can you proactively defend yourself against these threats? Learn how to protect your operations – and your patients – in this March 27 webinar from Merge by Merative.
|
|
Reach New Heights with Enterprise Imaging Cloud
Embrace the potential of AGFA HealthCare’s Enterprise Imaging Cloud, a fully managed Software-as-a-Service (SaaS) solution that will transform the handling, storage, and accessibility of medical imaging data. Learn how EI Cloud can help you today.
|
|
- 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.
- Time to Address Radiologist Burnout and Inefficient Workflows: Mach7’s new UnityVue radiology solution is a comprehensive reading platform combining enterprise imaging, workflow orchestration, AI-driven decision support, and the tools your team needs to thrive. See for yourself on March 26 in a live product demonstration by the Mach7 team.
- An Update from Blackford at ECR 2025: What’s the latest news at AI platform company Blackford? Check out this video interview with David Hilderbrand from ECR 2025 for recent highlights, including its expanding relationships with medical imaging OEMs and AI developers.
- Leading the Way in AI Transparency: There’s a need to better inform radiologists about AI’s role when interpreting images and generating measurements. Visage Imaging’s Visage 7 can display text in the viewer indicating that AI was used as a diagnostic aid – find out how it works today.
- Simplifying Complex Image Exchange Workflows: Guadalupe Regional Medical Center and Methodist Hospital implemented a PocketHealth Community Gateway that saved over 1,700 staff hours. Read how they streamlined bidirectional image exchange, created operational efficiencies, and improved continuity of care for patients.
- 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.
- Automated Measurements of Standing Radiographs: Precise lower limb measurements are crucial for assessing musculoskeletal health. Learn how measurements from Gleamer’s BoneMetrics AI solution offer a reliable alternative to manual assessments.
- Start at the Source to Improve MRI: Looking for ways to improve MRI speed and image quality while addressing broader concerns in healthcare? The answer may lie in proven MRI physics in your existing scanner – learn how to unlock it with STAGE from SpinTech MRI.
- The Transformative Role of AI in Radiology: In this episode of The Radiology Report Podcast, 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.
- Feel the Freedom of Helium-Free MRI: Lift limitations and experience MRI excellence with Philips BlueSeal, the industry’s lightest, vent pipe-free, high-performance, helium-free 1.5T scanner. Save on helium and energy costs, achieve precise AI-enhanced diagnoses, enjoy faster scans, and optimized workflows. Learn more today.
- Discover AI Apps Curated by Body Regions: Discover how Calantic Digital Solutions by Bayer supports you in tackling radiology challenges through multiple stages of the patient’s journey by offering AI solutions curated for specific body regions. Learn more about Calantic today.
- Join Clearpath at RBMA 2025 in Nashville: Discover why RBMA attendees trust Clearpath to streamline access to images and documents for legal requestors, patients, and healthcare providers. Don’t miss the chance to learn how they can help your organization – schedule a meeting or visit them at booth #211.
- Enterprise Imaging in the Cloud: What are the benefits of cloud-based enterprise image management? Check out this Imaging Wire Show interview with Tracy Byers, CEO of enterprise imaging at Optum, on the importance of the cloud in radiology.
- An Update from DeepHealth: What are the latest developments at DeepHealth? Check out this video interview from ECR 2025 with company executives Kees Wesdorp and Niccolo Stefani, who discuss the company’s recent highlights.
- Echo AI for Cardiac Amyloidosis: Echo can be used to diagnose patients with cardiac amyloidosis and differentiate them from those with left ventricular hypertrophy. Learn how Us2.ai’s novel solutions improve diagnosis of this underdiagnosed condition.
- Quality in Photon-Counting CT: Quality is the cornerstone of Siemens Healthineers’ photon-counting CT technology. They’ve invested in every step of the process, from X-ray tubes to detectors and workflow. Discover how they strive for the highest levels of quality in photon-counting CT.
- Presenting Unboxing AI: Check out CARPL’s video series, Unboxing AI, featuring experts discussing AI and its future in radiology. The next episode on March 20 features Namita Gandhi, MD, of Cleveland Clinic – reserve your seat today.
- A Step-by-Step Guide to InteleShare Viewer: How can Intelerad’s InteleShare Viewer help you handle the demands of enterprise healthcare with a diagnostic-quality viewer? Learn about some of the tools available to you in this self-guided demo.
|
|
|
|
|
