Tag: radiology

AI’s ROI Paradox

Posted on by Brian Casey

As radiology AI slowly moves from pilot projects to widespread clinical adoption, a new survey reveals a paradox: The technology is popular with radiologists, but few imaging facilities using AI have collected hard data showing its return on investment.

AI’s slow clinical adoption has frustrated both clinicians and algorithm developers alike, but the technology is gaining steam.

  • Despite growing clinical evidence, research on AI’s financial value and ROI has been slower in coming. 

To remedy that situation, AI governance startup Croviz.ai conducted a study of 445 radiology AI users on the economics and evaluation of AI. The full report is available here.

  • Survey respondents came from 12 different countries and included a variety of professional roles, including vendor executives, radiologists, and IT and informatics personnel.

Croviz founders Ayman Talkani and AadilMehdi Sanchawala found that while radiology AI power users loved the technology – and some refused to work without it – few had determined a positive financial return from it. Findings included…

  • 95% of sites already using AI had renewed at least one contract with an AI vendor in the last 12 months.
  • But only 30% had quantified a positive financial ROI from AI.
  • 54% cited better quality of life for radiologists as their main reason for renewing an AI contract.

So if AI’s value hasn’t been demonstrated, why are radiology sites renewing AI contracts?

  • The number one reason cited by 54% of those renewing contracts was because their radiologists felt AI improved their quality of life – the only outcome measure leadership could quickly measure with qualitative user feedback.
  • Lower on the scale was reduced turnaround time (18%), more scans per reader (10%), reduced downstream patient costs (10%), and better diagnostic accuracy (8%). 
  • Just 6% paid attention to hard metrics like staff retention rates.

What’s the best way out of the AI ROI paradox? The Croviz researchers recommended more frequent and transparent AI governance.

  • Survey respondents who monitored AI performance more closely – such as more often than once per quarter – exhibited more trust in AI.

The Takeaway

The new survey offers an intriguing look at AI adoption and the question of ROI for the technology. It suggests that – much like another digital technology, PACS – AI adoption is being driven more by its popularity among radiologists than hard ROI considerations.

Imaging Volume Backlash Builds

Posted on by Brian Casey

A backlash is building in response to a controversial paper published last week claiming that growth in U.S. medical imaging volume has slowed over the past several decades. The claims were met with disbelief by many imaging experts who see a growing disconnect between imaging volume and the number of radiologists available to interpret images.

Rising imaging volume has become a mantra within radiology as the field struggles to cope with growing healthcare needs from an aging population and the increasing complexity of imaging technology. 

  • Like other healthcare professionals, radiologists are experiencing rising burnout levels, and a cottage industry of AI and IT solutions has emerged to help them work more efficiently. 

But the new paper challenges many of those assumptions. Published as a commentary in JAMA Health Forum by Harvard University economists David Cutler, PhD, and Lev Klarnet, the article cites previously published research on imaging volume from 2003 to 2016, stating that imaging use per capita stabilized in 2008 and began declining thereafter. 

  • The authors suggest it’s unnecessary to dramatically increase the U.S. supply of radiologists given slowing growth: “The decrease in imaging has allowed the US to meet the need for imaging without an increase in radiologists.”

The paper quickly drew criticism from a number of radiology key opinion leaders, including Radiology Partners Chairman and CEO Rich Whitney (who suggested the authors were doing their research on the moon) and radiologist blogger Ben White, MD, who called some of their claims “nonsensical.” 

Indeed, the major fallacy in the JAMA Health Forum paper comes from its conclusion that a lower per capita imaging growth rate obviates the need to expand the radiologist labor pool. 

Most damning, however, is the paper’s reliance on data that’s nearly a decade old: The Hong et al paper published in Radiology in 2019. 

The Takeaway

There are some valuable (and positive) points made in the JAMA Health Forum paper, such as its contention that medical imaging is used more judiciously now than it was 20 years ago. But to make the leap that radiology’s workforce crisis has been solved simply strains belief. 

Better Radiologist Productivity with Clerical Assistants

Posted on by Brian Casey

What if there was a way to improve your radiologists’ productivity and help them focus on image interpretation without the heavy lift of a massive imaging IT project? Australian researchers found an old-school solution: shifting many clerical tasks to radiology administrative assistants.

The huge – and growing – disconnect between radiologist staffing and imaging volume has imaging managers around the world searching for solutions. 

  • Some are turning to high-tech tools like AI to squeeze more productivity from their radiologists, many of whom are already operating at maximum capacity. 

But lost in the debate is the reality that radiologists perform many functions besides just image interpretation (a fact that seems to have escaped some New York hospital CEOs).

  • These tasks include notifying clinicians of imaging findings, locating prior images, and study protocoling. Previous research indicates that these noninterpretive tasks can consume up to 44% of a radiologist’s workday. 

In the new study, published in Current Problems in Diagnostic Radiology, researchers implemented a system in which radiology administrative assistants were assigned to radiologists at Jones Radiology, a network of 60 radiologists across 30 sites in Australia. 

  • The RAAs worked normal business hours and were assigned tasks through a critical results feature in the PACS. Radiologists could choose if and when they wanted to use the RAA service. 

The main task RAAs handled was communicating critical results to referring physicians. 

  • But they also had other jobs, like finding and importing prior images, flagging scans that needed priority review, and providing research assistance. 

How well did the RAA system work? The researchers tracked its performance over 12 months from 2021 to 2022, finding that RAAs…

  • Were assigned 5.4k tasks during the study period.
  • Saved 679 hours of radiologist time.
  • 50% of the tasks involved communicating significant or unexpected results to clinicians.
  • The remaining tasks were unrelated to results communication, such as sourcing external images, miscellaneous tasks and general inquiries, and supporting radiologists with IT issues.  
  • Over 90% of “important” findings were communicated within the six-hour target turnaround time, but only 55% of “critical” findings met the one-hour turnaround goal.

The Takeaway

The idea of a clerical assistant to take over a radiologist’s noninterpretive tasks isn’t necessarily new, but this study is a great example of how to put it into practice. Radiology administrative assistants could also serve as a bridge to more complex IT-based operational solutions in the future.

The Danger of Incidental Findings in CT Lung Screening

Posted on by Brian Casey

CT lung cancer screening is gaining momentum around the world, but one of the challenges providers face is how to manage incidental findings. It’s especially important given that a new study in JAMA Network Open suggests that incidental findings on screening exams are associated with a higher risk of cancer occurring outside the lung. 

Incidental findings are suspicious areas discovered outside the target region being imaged, and are especially a concern with cancer screening exams.

  • Incidental findings turn out to be normal most of the time, but pathology occurs often enough that most clinicians agree they’re worth investigating. 

The problem is that many providers don’t have a robust system in place for alerting referring physicians to incidental findings and ensuring that patients get the follow-up exams they need.

The new study addresses incidental findings within the context of CT lung cancer screening, specifically in the National Lung Screening Trial, the landmark study that established low-dose CT’s lifesaving benefit.

  • It’s an important question, because chasing down a large number of benign incidental findings would be a resource-intensive task that could alter the cost-benefit ratio of lung screening.

Researchers analyzed significant incidental findings unrelated to lung cancer in 26.4k people across three rounds of LDCT screening who were followed for a year, revealing…

  • Cancer findings outside the lung occurred in 6.8% of people, and 13% of them had multiple cancers. 
  • Patients with significant incidental findings had a higher absolute risk of being diagnosed with extrapulmonary cancer within a year (16 per 1k participants). 
  • Study participants with incidental findings tended to be slightly older (62 vs. 61 years) and more likely to have a history of smoking-related disease (69% vs. 66%).

The findings confirm that having a plan to manage incidental findings should be an important part of any CT lung cancer screening program, especially given previous research showing that 23% of deaths in NLST were due to cancers outside the lung. 

  • In fact, an effective incidental finding program could enhance LDCT screening’s value, especially given that people eligible for screening have heavy smoking histories.

The Takeaway

The new study shows that incidental findings on CT lung screening exams are common and serious enough to warrant further investigation. Screening programs that are able to do so effectively will deliver even more value to their patients than lung screening alone.

Mammography Use Falls after USPSTF 2009 Guideline Change

Posted on by Brian Casey

Mammography use fell after the USPSTF rescinded its recommendation in 2009 of routine breast cancer screening for women in their 40s. The findings, in a new study in JAMA Network Open, confirm the fears of many women’s health advocates following the guideline change.

The women’s health world was shocked in 2009 when the USPSTF pulled its guideline advising women aged 40 to 49 to undergo regular breast screening, instead telling them to consult with their physicians.

  • The group reversed course in 2024, stating that women in their 40s should be screened every two years. Driving the decision were rising cancer rates in younger women, as well as higher mortality rates among Black women.

The new study analyzed data from the Behavioral Risk Factor Surveillance System to find changes in mammography use among 2.6M women, divided into various groups based on age, race, and other demographics.

  • Researchers analyzed self-reported mammography use, focusing on the periods 2000-2008 and 2012 and 2022.

The researchers found that, comparing 2002 to 2022, mammography prevalence fell for…

  • Women aged 40 to 49 (from 70% to 59%).
  • Women aged 50 to 74 (from 81% to 77%).
  • Non-Hispanic Black women in their 40s (from 72% to 65%).

The researchers pointed out that for the above categories, the endpoint comparisons were statistically significant. 

  • But the year-to-year changes in intervening years were not, in particular given a change in BRFSS survey methodology in 2011 that appears to have led to a several-point drop in utilization.

But several subgroups of women saw changes in both endpoint and year-to-year mammography prevalence, with use falling among…

  • Non-Hispanic White women in their 40s (from 71% to 60%).
  • Women in their 40s with insurance (from 74% to 62%) and without (from 47% to 33%).
  • Employed women (from 72% to 61%) as well as in women who classified themselves as homemakers (65% to 55%).

The Takeaway

The new study on falling mammography utilization confirms the fears of many women’s health advocates about the impact of the USPSTF’s 2009 guideline change. While the group righted the ship in 2024, it could take many years to see an effect – as suggested by the new study.

GE’s Photon-Counting CT Clearance

Posted on by Brian Casey

GE HealthCare this week announced FDA clearance for Photonova Spectra, the company’s first photon-counting CT scanner. While GE isn’t the first vendor with a commercially available PCCT scanner, it’s hoping to differentiate the system by highlighting the combination of ultrahigh-resolution scanning with spectral imaging.

Photon-counting CT represents a huge leap forward in CT instrumentation that’s not only driving new clinical applications but is also helping radiologists perform routine CT exams with better resolution and lower radiation dose. 

  • PCCT scanners directly convert photons to digital data, instead of using conventional CT’s two-step energy-integrating technique, resulting in images with less noise and supporting acquisition protocols with lower radiation dose. 

Siemens Healthineers brought the first photon-counting CT scanner to market with the 2021 FDA clearance of Naeotom Alpha.

  • Since then, Siemens has had the market for whole-body PCCT to itself, with only niche photon-counting scanners getting FDA clearance.

But we’re here to talk about GE’s Photonova Spectra, so let’s get to it. The system is based on GE’s Deep Silicon detector technology, which uses a novel semiconductor detector material that’s particularly suited for spectral imaging.

  • Spectral CT acquires images at different energy levels, which is useful for detecting disease because malignant and benign tissue respond differently to different energy spectra.    

GE is highlighting Photonova Spectra’s 8-bin energy resolution, which means the scanner separates incoming photons into eight distinct energy ranges – or bins – rather than grouping them into one or two. 

  • This enables Photonova Spectra to deliver much more precise spectral imaging than previously possible, with better quantitative accuracy and improved differentiation between materials like bone and soft tissue, according to GE CT executive Chad Rowland.

Spectral CT has developed a reputation as a technology that’s powerful but complex, and GE addressed this issue with workflow tools that make spectral imaging “always on” and easier than ever to perform. 

  • GE is banking on the combination of spectral imaging with Photonova Spectra’s ultrahigh-resolution images being a game-changer for many sites considering adopting their first PCCT scanner.

The Takeaway

FDA clearance for GE HealthCare’s Photonova Spectra photon-counting CT scanner is great news for the vendor that puts it on a level competitive footing with Siemens as a CT innovator. But it’s also good news for imaging providers, giving them another option for delivering to patients the benefits of PCCT – lower radiation dose and better image quality. 

The 40 Top Radiology Resources for 2026

Posted on by Brian Casey

Our list of the best radiology news sources last year generated a lot of excitement, so we’re updating the list for 2026 with the people and publications we rely on to find the most interesting medical imaging stories. 

Radiology has seen major changes in social media use since we last updated the list. Key opinion leaders briefly flirted with Bluesky as an alternative to X (formerly Twitter), but as the year went on enthusiasm waned as engagement faltered. Instead, LinkedIn seems to be emerging as the platform of choice for many clinicians and business executives.

Regardless of platform, this list of top radiology resources should keep you well-informed about healthcare’s top medical specialty.

TOP RADIOLOGY SITES

From a radiology newsletter with a laser focus on AI to an educational site with thousands of radiology cases, you’re sure to find something that meets your needs from the list below.

  • AI for Radiology – A great source for news on AI, including the Project AIR testing clearinghouse.
  • CTisUs – Elliott Fishman, MD’s excellent site for educational radiology content.
  • Medality Radiology Report Podcast – Medality CEO Daniel Arnold interviews the biggest names in medical imaging.
  • RadAccess – Newsletter run by Campbell Arnold, PhD, dedicated to improving access to radiology.
  • radHQ.net Forums – Public bulletin board that’s a great place to hear what keeps radiologists up at night. 
  • Radiopaedia – Excellent site for educational radiology content with a global focus.
  • Signify Research – The best radiology market analysis, backed by actual market data.

TOP RADIOLOGY KEY OPINION LEADERS 

Radiology is fortunate to have a wealth of really smart people sharing their thoughts. Here are a few of the best.

AI and Imaging IT

Business and Vendors

  • Jan Beger – OEM executive with insightful high-level thoughts on AI.
  • Morris Panner – Imaging IT executive with cogent takes on the intersection of technology and patient care. 
  • Sebastian Schmidt, MD – OEM executive with thought-provoking analysis of CT lung cancer screening.
  • Reza Zahiri – Detailed LinkedIn posts that deconstruct the financial positions of medical imaging vendors.

Education

  • Gennaro D’Anna, MD – Italian radiologist focusing on education and social media.
  • Francis Deng, MD – Great analysis of radiology education and residency trends.
  • Tan-Lucien Mohammed, MD – Radiologist with educational focus on radiology anatomy.
  • Amy Patel, MD – Tireless advocate for radiology (and the Kansas City Chiefs).
  • Vikas Shah, MD – Radiopaedia managing editor known for high-quality educational content.
  • Chaundria Singleton – Radiologic technologist educator and host of A Couple of Rad Techs podcast. 

Legal and Regulatory 

  • Sandy Coffta – Great source for information on U.S. reimbursement changes.
  • Tobias Gilk – Radiology’s conscience on MRI safety. 
  • Tom Greeson – The authority for perspectives on legal issues in radiology.
  • Hugh Harvey, MD – Excellent source on AI regulation.
  • Mark Weiss – Authoritative voice on legal issues in radiology. 

Practice Management and Leadership

  • Rich Duszak, MD – A superb source for radiology leadership and responsible imaging.
  • Jay Gurney – Executive recruiter and podcaster who hears about industry trends before they make headlines. 
  • Geraldine McGinty, MD – Still the moral compass of radiology.
  • Rasu Shrestha, MD – Radiologist-turned-health-executive.  
  • Ben White, MD – Excellent insights into the vagaries of being a working radiologist.

HEALTHCARE NEWSLETTERS AND WEBSITES 

Looking to get out of the radiology niche and broaden your horizons? Insight Links also offers newsletters covering the cardiology and digital health fields:

The Takeaway

This list should cover all your bases for staying informed about the latest developments in radiology news. Or, just sign up for The Imaging Wire and we’ll do it for you.

Mammo AI Momentum Builds

Posted on by Brian Casey

Momentum is building toward routine clinical use of AI for breast cancer screening. Several new studies offer even more support for mammography AI, including research published today in Nature Medicine in which AI reduced radiologist workload by over 60% by excluding low-risk studies from human review.

Breast screening has become one of the most promising use cases for AI, with the potential to reduce radiologists’ workload while improving their ability to detect cancer. 

  • For example, the recent MASAI study found that ScreenPoint Medical’s Transpara AI algorithm could replace the second human reader in a double-reading protocol, reducing workload by 44% and improving cancer detection rates by 28%.

The new research in Nature Medicine also used Transpara, as part of the AITIC study in Spain with the goal of seeing if AI could triage low-risk studies so they don’t require review by human radiologists. 

  • AITIC had a prospective design, involving 31k women with screening exams split between 2D mammography (17k) and digital breast tomosynthesis (14k). 

Women in the control arm of the study got conventional double reading by two radiologists – the standard mammography paradigm in Europe.

  • The intervention arm used a partially autonomous AI approach: cases that AI interpreted as low risk were classified as normal and were not reviewed by radiologists, while all other cases were double-read by radiologists using AI support.

In analyzing the results, researchers found…

  • Workload in the AI arm was 64% lower than conventional double reading.
  • AI’s workload reduction was similar between DBT and conventional digital mammography (-66% and -62%, respectively).
  • The AI arm’s cancer detection rate per 1k women was 15% higher (7.3 vs. 6.3 cancers).
  • But the recall rate was also 15% higher.

It’s worth noting that the AITIC study differed from MASAI in its inclusion of DBT screening exams, whereas MASAI only included 2D digital mammography. 

  • While 2D mammography is the norm in Europe, much of the U.S. has switched to DBT for breast screening, so the AITIC results offer good news for U.S. breast imaging practices considering AI adoption.

The Takeaway

The AITIC study’s new results are powerful confirmation of findings from the recent MASAI trial and support broader clinical deployment of mammography AI. Taken together with positive findings from last week’s Nature Cancer articles (see The Wire section in this newsletter), they paint a picture of a technology that’s ready for prime time.

Support for Prostate Cancer Screening Grows

Posted on by Brian Casey

Routine prostate cancer screening currently isn’t supported by clinical guidelines. But that could be changing, especially given research presented this week finding that prostate screening – aided by MRI to reduce unnecessary biopsies – was as effective as mammography screening. 

Prostate cancer is one of the leading causes of cancer death, killing some 360k men worldwide every year. 

  • But efforts to develop effective prostate cancer screening programs have been hampered by the challenges inherent in PSA testing, which often identifies indolent disease that may never pose a health risk to patients – the classic definition of overdiagnosis. 

That could be starting to change, however. Researchers are discovering that using MRI to work up patients with rising PSA levels could help identify men with high-risk disease who should be sent to biopsy, while lower-risk patients are monitored with surveillance.

  • New research presented at the European Association of Urology meeting on Sunday supports this idea, showing that – if done right – prostate cancer screening can be as effective as mammography screening.

Researchers from Germany compared prostate cancer screening data from 39.4k men who got PSA tests as part of the PROBASE trial to over 2.8M women who participated in the country’s national breast cancer screening program. 

  • Under the PROBASE protocol, men with confirmed PSA levels ≥ 3 ng/mL underwent MRI and biopsy, while those with lower PSA levels got repeat PSA testing in an effort to keep biopsy rates lower.

Major findings of the study included…

  • Both breast and prostate screening detected a high rate of clinically significant, invasive cancers (73% for mammography vs. 69% for 45-year-old men and 74% for 50-year-old men).
  • False-positive rates were much lower for breast screening (10% vs. 42% and 37%).
  • And fewer indolent cancers were detected with mammography (22% vs. 31% and 26%). 
  • But biopsy rates were comparable (1.1% vs. 0.8% and 2.4%), as the study’s active surveillance protocol limited over-treatment.

While the PROBASE study didn’t use AI as part of its protocol, other research has found that AI analysis of MRI scans can make the modality even more precise, with the PI-CAI study just one worth noting.

The Takeaway

Is it finally time for prostate cancer screening to join breast, colorectal, cervical, and lung among the major population-based cancer screening tests? Results from the PROBASE study at EAU 2026 suggest the time may finally be right.

FDA Updates AI List with New Clearances

Posted on by Brian Casey

The FDA last week updated its list of cleared AI-enabled medical devices, with the new list showing AI marketing authorizations through the end of 2025. The updated list reveals that radiology is maintaining its lead as the medical specialty with the most clearances.

The FDA’s previous update featured data through the end of September 2025, and showed the number of AI-enabled medical devices for radiology crossed the 1k mark. The new numbers show continued momentum for medical imaging.

  • The agency’s data go all the way back to 1995 (the first cleared radiology device on the list was ImageChecker from R2 Technology/Hologic in 1998). 

The new list tracks authorizations through the end of December 2025, and indicates the agency has…

  • Authorized 1,451 AI-enabled medical devices since it began keeping track in 1995.
  • Approved 1,104 radiology devices, or 76% of total AI-enabled medical authorizations.
  • In the fourth quarter of 2025, the FDA cleared 72 AI-enabled medical devices, of which 55 (76%) were radiology devices. 
  • For all of 2025, radiology secured 75% of authorizations, compared to 73% for all of 2024 and 80% for 2023. 
  • GE HealthCare retained the top spot as the company with the most radiology AI authorizations at 120 (including acquisitions Bay Labs, BK Medical, Caption Health, MIM Software, icometrix, and Spectronic Medical).
  • Next is Siemens Healthineers at 89 (including Varian), then Philips at 50 (including DiA Analysis and TomTec), Canon at 45 (including Vital Images and Olea), United Imaging at 38, Aidoc at 31, and DeepHealth at 28 (including Quibim and iCAD). 

As we’ve noted in the past, the FDA’s list includes not only standalone software applications, but also imaging hardware with embedded AI applications, such as a mobile X-ray system with AI algorithms for detecting emergent conditions. 

The Takeaway

The new FDA list shows radiology’s continued dominance when it comes to AI-enabled medical device technology. But an interesting subtext is the ongoing consolidation in the radiology AI space, which could mean that some firms may be climbing the list quickly.

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