Category: Artificial Intelligence

Prostate AI Improves Biparametric MRI

Posted on by Brian Casey

Researchers continue to hone in on the best way to use MRI for patients suspected of having prostate cancer, and AI is helping the effort. A new study in AJR shows that AI can improve the diagnostic accuracy and consistency of prostate MRI – while making it easier to perform.

Multiparametric MRI is the gold standard for prostate cancer imaging, but requires the use of three different MRI sequences as well as contrast administration, making it more complex and time-intensive to perform. 

  • On the other hand, biparametric MRI uses just two sequences – T2-weighted and diffusion-weighted imaging – and omits the contrast entirely, leading to shorter scan times and lower cost.

But what are you losing with bpMRI – and can AI help you get it back? Researchers addressed this question in the new study in which six radiologists interpreted bpMRI scans of 180 patients from multiple centers. 

  • Radiologists used a deep learning algorithm developed at the NIH to interpret bpMRI scans acquired on 3T scanners. The open-source algorithm generates binary prostate cancer prediction maps that are overlaid on T2-weighted images.

Researchers found that radiologists using the bpMRI AI algorithm to detect clinically significant prostate cancer had…

  • An increase in lesion-level positive predictive value (77% vs. 67%).
  • But lower lesion-level sensitivity (44% vs. 48%). 
  • And no statistically significant difference in patient-level AUC (0.82 vs. 0.83, p = 0.61).
  • While inter-reader agreement scores improved for lesion-level and patient-level PI-RADS scores and lesion size measurements. 

What to make of the numbers? The authors pointed out that the study design – in which AI was used as a first reader – may have reduced AI’s performance.

  • In real clinical practice, AI would most likely be used as a sort of clinical spell checker, with AI results overlaid on images that radiologists had already seen. 

The researchers said the results on improved positive predictive value and inter-reader agreement show that AI can improve the diagnostic accuracy and consistency of bpMRI for prostate cancer. 

The Takeaway

The new findings echo other research like the PI-CAI study highlighting the growing role of AI in prostate cancer detection. If validated with other studies, they show AI-assisted bpMRI could be ready to take on mpMRI for a broader role.

RP Builds AI Mosaic as Company’s IT Foundation

Posted on by Brian Casey

Radiology Partners announced a new initiative to guide the rollout of AI across its nationwide network of radiology practices. The company’s new MosaicOS will be the IT foundation that connects RP practices and supports clinical uses from AI-assisted reporting to report generation and even image management.

Radiology Partners has grown since its founding in 2012 to become the largest privately held provider of imaging services in the U.S. and a major force behind the consolidation of private-practice radiology groups.

  • RP has always maintained a heavy technology investment, and has been looking closely at the rise of AI in radiology.

That’s because the growth in imaging volume is so massive that clinicians will no longer be able to care for patients adequately without AI’s assistance, at least according to RP’s Associate Chief Medical Officer for Clinical AI Nina Kottler, MD.

RP laid the groundwork for MosaicOS in 2020 by first migrating its technology stack to a cloud-native infrastructure. 

  • This frees RP from reliance on on-premises legacy software and enables the company to push out updates that can be adopted quickly across its network.

RP’s Mosaic rollout includes the following components as the company…

  • Forms a new division, Mosaic Clinical Technologies, to oversee its AI activities.
  • Debuts MosaicOS, a cloud-native operating system that combines AI support with workflow and other IT tools.
  • Launches Mosaic Reporting, an automated structured reporting solution that combines ambient voice AI with large language model technology.
  • Develops Mosaic Drafting, a multimodal AI foundation model that pre-drafts X-ray reports that radiologists can review, edit, and sign. 

Mosaic Reporting is already in use at some RP sites, and the company is pursuing FDA clearance for broader use of Mosaic Drafting. More Mosaic applications are on the way.

  • Mosaic tools will be disseminated to RP centers using the cloud-native infrastructure, and MosaicOS will include image management functions that providers can choose to use in place of or alongside existing tools like viewers and archives. 

Kottler told The Imaging Wire that RP has de-emphasized individual pixel-based AI models in favor of foundation models that have broader application.

  • What’s more, RP CEO Rich Whitney said the company has chosen to develop AI technology internally rather than rely on outside vendors, as this gives it greater control over its own AI adoption.

The Takeaway

The launch of MosaicOS marks an exciting milestone not only for Radiology Partners but also for radiology in general that could address nagging concerns about clinical AI adoption on a broad scale. RP has not only the network but also the technology resources to make the rollout a success – the question is whether outside AI developers will share in the rewards.

Radiology AI Approvals Near 1k in New FDA Update

Posted on by Brian Casey

The FDA last week released the long-awaited update to its list of AI-enabled medical devices that have received marketing authorization. The closely watched list shows the number of AI-enabled radiology authorizations approaching the 1k mark.

The FDA has been tracking authorizations of AI-enabled devices going back to 1995, and the list gives industry watchers a feel for not only how quickly the agency is churning out reviews but also which medical specialties are generating the most approvals.

  • But the last time the FDA released an updated list was August 2024, and recent turmoil at the agency had some observers wondering if it would continue the tradition – as well as whether it could stay on pace for new approvals.

Those fears should be assuaged with the new release. The numbers indicate that through May 2025 the FDA has…

  • Granted authorization to 1.2k AI-enabled medical devices since it started tracking.
  • Approved 956 AI-enabled radiology products, or 77% of total medical authorizations.
  • Radiology’s share of overall authorizations from January to May 2025 ticked up to 78% (115/148), compared to 73% in the 2024 update, and 80% in all of 2023.
  • GE HealthCare remains the company with the most radiology AI authorizations, at 96 (including recent acquisitions like Caption Health and MIM Software), with Siemens Healthineers in second place at 80 (including Varian). 
  • Other notable mentions include Philips (42 including DiA Analysis), Canon (35), United Imaging (32), and Aidoc (30). 

In a significant regulatory development, the FDA said it was developing a plan to identify and tag medical devices that use foundation models, including large language models and multimodal architecture. 

  • The agency said the program would help healthcare providers and patients know when LLM-based functionality was included in a medical device (the FDA has yet to approve a medical device with LLM technology). 

In another interesting change, the FDA dropped “machine learning” from the title of its list, apparently with the idea that “AI” was sufficient as an umbrella term. 

The Takeaway

The FDA’s release of its AI approval list is a welcome return to past practices that should reassure agency watchers that recent turmoil isn’t affecting its basic operations. The LLM guidance suggests the agency may be changing its approach to the technology in favor of disclosure and transparency instead of more stringent regulation that could delay some LLM solutions from reaching the market.

AI-Driven Diagnostics Detects Multiple Chest Diseases from Single CT Scan

Posted on by Partner Content

A new generation of AI solutions is enabling clinicians to detect multiple chest pathologies from a single CT scan. Lung cancer, cardiovascular disease, and chronic obstructive pulmonary disease (COPD) can all be detected in just one imaging session, ushering in a new era of more efficient imaging that benefits both providers and patients. 

Advances in CT lung cancer screening have been generating headlines as new research highlights the improved clinical outcomes possible when lung cancer is detected early. 

  • But lung cancer is just one of a “big three” of thoracic comorbidities – the others being cardiovascular disease and COPD – that can result from long-term exposure to toxic substances like tobacco smoke. 

These co-morbidities will be encountered more often as health systems ramp up lung cancer screening efforts, creating challenges for radiologists in managing the many incidental findings discovered with chest CT scans.

  • And it’s common knowledge that radiologists already have their hands full in an era of personnel shortages and rising imaging volumes. 

Fortunately, new AI technologies offer a solution. One of these is Coreline Soft’s AVIEW LCS Plus, an integrated three-in-one solution that enables simultaneous detection of lung cancer, cardiovascular disease, and COPD from a single chest CT scan. 

  • AVIEW LCS Plus is the only solution adopted for national lung cancer screening projects across key countries, including Korea (K-LUCAS), Germany (HANSE), Italy (RISP), and the pan-European consortium (4ITLR). 

Coreline’s solution is widely recognized as a pioneering AI platform for an era where unexpected findings can save lives, gaining increasing attention in academic journals and health policy reports alike.

  • In the U.S., AVIEW LCS Plus has been adopted by Temple Health, and the Pennsylvania system’s use of the solution in their Temple Healthy Chest initiative has been recognized as an innovative healthcare model within the Philadelphia region. 

Temple Health clinicians are finding that AI contributes to early detection of incidental findings, improved survival rates, and more proactive care planning.

  • AVIEW LCS Plus is streamlining lung cancer screening, helping identify chest conditions at earlier stages, when treatment is most effective. It is finding not only lung nodules but also undetected comorbidities that were often missed with conventional CT workflow. 

Coreline Soft will be presenting AVIEW LCS Plus in collaboration with Temple Health at the upcoming American Thoracic Society (ATS 2025) international conference in San Francisco from May 16-21. 

  • Attendees will be able to learn how AI in medical imaging can establish new standards, not just in diagnostics, but across policy, patient care, and hospital strategy. 

Getting Paid for AI – Will It Get Easier?

Posted on by Brian Casey

Reimbursement is one of the major stumbling blocks holding back wider clinical adoption of artificial intelligence. But new legislation was introduced into the U.S. Congress last week that could ease AI’s reimbursement path. 

For AI developers, getting an algorithm approved is just the first step toward commercial acceptance. 

  • Perhaps even more important than FDA clearance is Medicare reimbursement, as healthcare providers are reluctant to use a product they won’t get paid for. 

Reimbursement drives clinical AI adoption, as evidenced by a 2023 analysis listing the top algorithms by CPT claims submitted (Heartflow Analysis topped the list). 

  • But CMS uses a patchwork system governing reimbursement, from temporary codes like New Technology Add-On Payment codes that expire after 2-3 years to G-codes for procedures that don’t have CPT codes, on up to the holy grail of medical reimbursement: Category I codes. 

The new legislation – S.1399 or the Health Tech Investment Act – would simplify the situation by setting up a dedicated Medicare coverage pathway for AI-enabled medical devices approved by the FDA (called “algorithm-based healthcare services”), as follows … 

  • All FDA-approved products would be assigned a Category III New Technology Ambulatory Payment Classification in the HOPPS program.
  • NTAPC codes would last for five years to enable collection of cost data before a permanent payment code is assigned. 
  • Payment classifications will be based on the cost of service as estimated by the manufacturer. 

The bill at present has co-sponsors from both political parties, Sen. Mike Rounds (R-SD) and Sen. Martin Heinrich (D-NM). 

  • The legislation has also drawn support from industry heavyweights like GE HealthCare and Siemens Healthineers, as well as industry groups like AdvaMed and others.

The Takeaway

The new bill sounds like a great idea, but it’s easy to be skeptical about its prospects in today’s highly charged political environment – especially when even bipartisan compromises like the 2025 Medicare fix got scuttled. Still, S.1399’s introduction at least shows that the highest levels of the U.S. government are cognizant of the need to improve clinical AI reimbursement.

Radiology’s Rising Workload

Posted on by Brian Casey

If you think new imaging IT technologies will reduce radiologist workload in the future, you might want to think again. Researchers who analyzed hundreds of studies on new scientific advances predicted that nearly half of them would increase radiologists’ workload – especially AI. 

Radiologists are desperately in need of help to manage rising imaging volumes during a period of global workforce shortages. 

But how true is that belief? In the new study in European Journal of Radiology, radiologists Thomas Kwee, MD, and Robert Kwee, MD, from the Netherlands analyzed a random sample of 416 articles published in 2024 on imaging applications that could affect future radiologist workloads, finding …

  • 49% of the articles on applications that had the potential to directly impact patient care would increase radiologist workload in the tertiary care academic setting. 
  • Studies on AI-focused applications were 14X more likely to increase workload compared to research that didn’t.
  • Similar numbers were found for non-academic general teaching hospitals.
  • The findings are largely similar to a 2019 study by Kwee et al that used the same methodology.  

Why don’t new imaging applications show more potential to reduce radiologists’ workloads? 

  • The Kwees found that image post-processing and interpretation times have grown for both existing and new applications. 

In the specific case of AI, they cited an example in which a deep learning algorithm was introduced to analyze CT scans to segment and classify features of spontaneous intracerebral hemorrhage and predict hematoma expansion.

  • The model successfully predicted hematoma expansion and automatically segmented lesions, but CT images still had to be post-processed with a separate workflow. This required additional radiologist interpretation time and extended their workload.

The Takeaway

The new study throws cold water on the idea that AI will be able to solve radiology’s workload dilemma. It’s possible that AI will have an impact on radiology that’s similar to that of PACS in the 1990s in making radiologists more productive, but we’ll need new efficiency-oriented changes to achieve that goal.

VC AI Funding Plummets

Posted on by Brian Casey

If you thought venture capital funding of AI firms was lower last year, you weren’t wrong. A new report from market analysis firm Signify Research found that VC funding of radiology AI firms dropped by nearly half in 2024 compared to the year before. 

VC funding has become a closely watched barometer of the radiology AI segment’s overall health. 

  • As most AI developers haven’t generated significant cash flows from product revenues yet, VC money is the oxygen that keeps AI firms breathing. 

And there are signs that after peaking in 2021, that oxygen is coming into short supply. 

  • Signify’s report last year documented a 19% drop in VC AI funding in 2023, a development attributed to tighter access to capital due to high interest rates. 

Those trends continued into 2024, with the new Signify report finding …

  • Total VC funding was $335.5M, down 48% compared to $645.6M in 2023.
  • The number of funding rounds fell 35% (20 vs. 31), to the lowest level since 2015.
  • Average deal size fell 16% ($16.8M vs. $20.1M).
  • Cleerly raised the most in 2024 with $106M in funding, followed by Qure.ai with $65M (putting Qure into Signify’s elite “$100M Club”). 
  • Funding declines were even worse in the Asia-Pacific (-84%) and the Europe, Middle East, and Africa regions (-76%) compared to the peak in 2021. 

Signify analyst Umar Ahmed noted that 2025 got off to a strong start, with $100M in funding rounds announced in January.

  • This could either represent a rebound in investor confidence, or indicate that the AI funding cycle is getting longer as VC firms put developers under the microscope and demand better ROI for their investments. 

The Takeaway

So it’s agreed that 2024 was a wash for VC radiology AI funding – what about 2025? The year’s strong start appears to have petered out as we approach the spring quarter, and ongoing regulatory turbulence and economic uncertainty in the U.S. isn’t likely to help. Stay tuned. 

New Mammography AI Insights

Posted on by Brian Casey

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. 

Will FDA Staff Cuts Slow AI Adoption?

Posted on by Brian Casey

The Trump Administration’s campaign to cut the federal workforce arrived at the FDA last weekend – in particular its division regulating AI in healthcare. Multiple staff cuts were reported at the Center for Devices and Radiological Health, which had been in the midst of a major overhaul of AI regulation. 

A February 15 article in STAT News first reported the layoffs, which as with other recent staff reductions concentrated on FDA employees with probationary status and was part of a larger initiative that has also affected the CDC and NIH. 

The rapid growth of medical AI has had a major impact on the center, which as of its last report had given regulatory authorization to over 1k AI-enabled devices (76% of which are for radiology). 

  • To deal with the deluge, CDRH reportedly had been hiring many new staffers who were still on probationary status, making them targets for layoffs (permanent federal employees have civil service protections that make them harder to fire). 

FDA also has been retooling its regulatory approach to AI with new initiatives that reflect the fact that AI products continue learning (and changing) after they’ve been approved, and thus require more aggressive post-market surveillance than other medical devices…

So what impact – if any – will the layoffs have on the rapidly growing medical AI segment? 

  • The FDA may simply scale back its new AI initiatives and regulate the field under more traditional avenues that have served the medical device industry well for decades.

In another scenario, the FDA’s frenzied pace of AI approvals and initiatives could slow as the agency struggles to handle a growing number of product submissions with less staff. 

The Takeaway

The FDA layoffs couldn’t have come at a worse time for medical AI, which is on the cusp of wider clinical acceptance but still suffers from shaky confidence and poor understanding on the part of both providers and patients (see story below). The question is whether providers, organized radiology, or developers themselves will be able to step into the gap being left.

AI As Malpractice Safety Net

Posted on by Brian Casey

One of the emerging use cases for AI in radiology is as a safety net that could help hospitals avoid malpractice cases by catching errors made by radiologists before they can cause patient harm. The topic was reviewed in a Sunday presentation at RSNA 2024. 

Clinical AI adoption has been held back by economic factors such as limited reimbursement and the lack of strong return on investment. 

  • Healthcare providers want to know that their AI investments will pay off, either through direct reimbursement from payors or improved operational efficiency.

At the same time, providers face rising malpractice risk, with a number of recent high-profile legal cases.

  • For example, a New York hospital was hit with a $120M verdict after a resident physician working the night shift missed a pulmonary embolism. 

Could AI limit risk by acting as a backstop to radiologists? 

  • At RSNA 2024, Benjamin Strong, MD, chief medical officer at vRad, described how they have deployed AI as a QA safety net. 

vRad mostly develops its own AI algorithms, with the first algorithm deployed in 2015. 

  • vRad is running AI algorithms as a backstop for 13 critical pathologies, from aortic dissection to superior mesenteric artery occlusion.

vRad’s QA workflow begins after the radiologist issues a final report (without using AI), and an algorithm then reviews the report automatically. 

  • If discrepancies are found the report is sent to a second radiologist, who can kick the study back to the original radiologist if they believe an error has occurred. The entire process takes 20 minutes. 

In a review of the program over one year, vRad found …

  • Corrections were made for about 1.5k diagnoses out of 6.7M exams.
  • The top five AI models accounted for over $8M in medical malpractice savings. 
  • Three pathologies – spinal epidural abscess, aortic dissection, and ischemic bowel due to SMA occlusion – would have amounted to $18M in payouts over four years.
  • Adding intracranial hemorrhage and pulmonary embolism creates what Strong called the “Big Five” of pathologies that are either the most frequently missed or the most expensive when missed.

The Takeaway

The findings offer an intriguing new use case for AI adoption. Avoiding just one malpractice verdict or settlement would more than pay for the cost of AI installation, in most cases many times over. How’s that for return on investment?

Get every issue of The Imaging Wire, delivered right to your inbox.

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!