Tag: radiologists

Two Views on Lung Screening’s Progress

Posted on by Brian Casey

Two new research studies published this week offer slightly conflicting views on the progress of CT lung cancer screening. One study saw the screening adherence rate rising to nearly 25%, while another revealed that screening might not be as effective in the real world as it was in the landmark National Lung Screening Trial. 

Low-dose CT lung cancer screening was approved for Medicare reimbursement in 2015 following promising NLST results, but the exam was slow to catch on.

  • That’s begun to change as U.S. physicians identify how to get eligible high-risk people into screening, while other countries are launching organized population-based screening programs, Germany being the most recent.

In the first study this week, published in JAMA Internal Medicine, researchers calculated the most recent uptake rate for LDCT lung cancer screening.

  • Data from 26.1k people from the BRFSS survey in 2024 were analyzed and compared to 2022 data.

The rate of up-to-date LDCT screening prevalence in 2024…

  • Rose six percentage points (24% vs. 18%).
  • Differed by just 0.5 percentage points between men and women.
  • Was highest and rose the most for people ages 65-69 (33%, increasing by 9 percentage points).
  • But still lagged screening rates for other cancers like breast (80%), cervical (75%), and colorectal (67%).

This week’s second study was published in JAMA Network Open and offers additional context, finding that real-world lung screening may not reduce all-cause mortality by as much as what was seen in the NLST.

  • Researchers analyzed screening’s effectiveness in a program run by the U.S. Veterans Health Administration, which has one of the more successful lung screening programs in the country.

They compared five-year all-cause mortality rates in the VA program for 732 people who were screened, finding that, compared to NLST…

  • Mortality rates were 2.5X higher in the VA population (24% vs. 9.7%).
  • VA participants had 3X the mortality risk (HR = 2.98).

VA researchers noted that while their population met the same screening eligibility criteria as NLST, the VA patients were sicker and probably less likely to see as much long-term benefit from LDCT screening.

The Takeaway

This week’s studies are a window into the complex changes underway in CT lung cancer screening. While screening’s rising participation rate is good news, the mortality findings show that estimating screening’s effectiveness is more complex than simply overlaying NLST results on real-world populations. 

Mammography AI Improves Breast Screening

Posted on by Brian Casey

Radiologists using a commercially available mammography AI algorithm saw improved diagnostic performance in breast cancer screening, mainly due to better specificity. The study adds to a growing body of research supporting mammography AI.

Mammography screening has been one of the most promising use cases for AI, and recent randomized controlled trials have demonstrated that AI can both improve diagnostic accuracy and speed up workflows. 

  • But RCTs are usually performed under highly controlled conditions in high-income Western countries, and the results might not be generalizable to other countries around the world. 

In the new study in Academic Radiology, researchers in Singapore tested Lunit’s Insight MMG algorithm in a retrospective review of a dataset of 302 digital mammograms that was enriched with 89 breast cancers.

  • Researchers noted that many countries have a high breast cancer incidence-to-mortality ratio due to limitations in population-based screening programs, and AI potentially could help. 

The authors focused on AI’s ability to improve the diagnostic performance of nine breast radiologists from four countries in Asia and North Africa who interpreted the mammograms, finding that AI assistance…

  • Improved radiologist accuracy as measured by AUC (from 0.799 to 0.851).
  • Generated a big jump in specificity (from 77% to 88%). 
  • And significantly reduced per-case image interpretation times (from 122 to 83 seconds per case).
  • Without changing sensitivity at a statistically significant level (83% vs. 82%, p = 0.73).

There were some subtle differences in the current study’s findings relative to previous research, some of which were the result of using a cancer-enriched dataset rather than a screening population as would be the case in an RCT.

  • The specificity improvement with AI would reduce unnecessary recalls in a population-based screening program and make mammography more cost-effective – an important consideration in countries with constrained public health budgets.

The Takeaway

The new study doesn’t have the statistical heft of a large, randomized controlled trial, but it still adds to the body of knowledge supporting AI for mammography, especially at facilities that haven’t been party to the large-scale RCTs.

Physician Burnout Eases – But Remains High in Radiology

Posted on by Brian Casey

Physician burnout rates continue to drop after peaking during the COVID-19 pandemic, but radiology remains one of the medical specialties with the worst burnout problem. That’s according to a new AMA survey that underscores a continuing trend toward improved physician satisfaction.

Physician burnout has been a cause of concern for the past decade. Burnout rates accelerated dramatically during the COVID-19 pandemic as doctors struggled with long hours and crushing patient volume. 

The new AMA survey confirms the trend. The organization surveyed 19k U.S. physicians throughout 2025, finding…

  • An overall physician burnout rate of 42% in 2025, down compared to 2024 (43%) and 2023 (48%), and sharply down from the peak in 2021 (63%).
  • Physician job satisfaction rose to 77%, stable compared to 2024 (77%) and up compared to 2023 (72%) and 2022 (68%).
  • The burnout rate in radiology was 45%, making it the fifth-worst among specialties after emergency medicine (50%), urological surgery (50%), hematology/oncology (49%), and OB/GYN (46%). 

How well do the AMA’s numbers reflect radiology’s reality?

  • Unfortunately, the AMA hasn’t included radiology-specific numbers in previous years of its survey, making it hard to directly compare year-over-year changes.

But in a recent thread on radiology bulletin board RadHQ.net, members anecdotally reported that burnout is becoming less prominent for radiologists, for several reasons…

  • Higher compensation that makes radiologists feel more valued. 
  • More flexible arrangements – like teleradiology – that support working from home or more flexible shifts.
  • Increased job mobility due to the radiologist shortage, with the flexibility to leave toxic practices for new ones.

The Takeaway

So which is it – is radiology burnout higher than other specialties, or is it declining? Ironically, both statements could be true – while radiologist burnout remains high relative to other physicians, at least it’s trending in the right direction.

Radiologist Training Stagnates

Posted on by Brian Casey

The number of new radiologists being trained in the U.S. has largely stagnated, even as concerns grow over workforce shortages. A new analysis in JACR underscores the challenges facing efforts to train more radiologists to meet the rising volume of imaging exams.

The growing imbalance between the radiologist workforce and surging imaging volume has been generating headlines as frontline radiologists struggle with overwork and burnout. 

  • And a recent effort to boost the number of U.S. physicians by adding federal funding for more resident training slots has overlooked radiology, which is getting few of the new positions being offered.

In the new study, researchers analyzed the number of U.S. resident positions in diagnostic and interventional radiology using data from the NRMP, ACGME, and other sources, finding…

  • The number of radiology residency positions grew 33% from 2010 to 2025 (from 1,090 to 1,449), while total medical residency positions grew 69%.
  • The total number of radiology residents rose 23% from 2010 to 2024 (from 4,584 to 5,630). 
  • But the number of practicing radiologists only grew 12% from 2010 to 2022 (from 34.3k to 38.3k). 
  • And the ratio of radiologists per 100k population was stagnant (from 11.1 to 11.5 radiologists).

It’s this last number that’s key to the analysis. Over the last 15 years, the ratio of radiologists to the U.S. population has barely changed, even as imaging exams become more complex.

  • And even more importantly, the Baby Boom generation has aged and now requires more healthcare services per capita – including imaging – relative to when they were younger (a previous study found the average radiologist workload doubled from 2008 to 2018). 

The researchers conclude that the radiologist training pipeline “has barely kept pace” with the increasing U.S. population, and when coupled with growing complexity and per capita imaging use, this raises questions about the sustainability of increasing imaging utilization. 

  • They recommend more federal support for radiologist training positions, perhaps through Medicare programs that specifically target medical imaging.

The Takeaway

The new study on stagnant radiologist training gives ammunition to radiology advocates who are lobbying for more federal funding to expand radiology residency slots. The question is whether anyone is listening.

Radiologist Salaries Grow 9%

Posted on by Brian Casey

It’s not quite double-digit growth, but it’s pretty close. The latest physician salary report from Medscape reveals that radiologist salaries grew 9% in 2025, with medical imaging seeing the third-highest growth rate among the specialties tracked. 

The new Medscape report found that radiologists had an average annual salary of $571k in 2025, up 9% compared to $520k in 2024 and $498k in 2023.

  • Radiology ranked #3 among the 29 physician specialties tracked by the survey, trailing only orthopedic physicians and surgeons ($611k) and cardiologists ($575k). 

Radiology’s numbers mark the second straight year of strong salary growth for the specialty, a welcome trend given anemic growth in 2023, when salaries rose only 3.1% – not enough to keep ahead of inflation.

  • Indeed, industry experts analyzing the new Medscape numbers called the figures a “return to normalization” for physician pay gains. 

Medscape reported a generally “positive vibe” to the survey compared to a year earlier, with more respondents saying they felt fairly paid (53% vs. 48%), and eight specialties reporting average annual compensation over $500k. Other important findings included…

  • Technology-driven efficiency appears to be helping physician productivity, enabling doctors to see more patients and generate more work-related relative value units.
  • Rising demand for services in orthopedics, radiology, and cardiology is driving pay growth, especially given the static supply of physicians.
  • The new data – and improving physician sentiment – could indicate that medicine is climbing out of the trough of doctor burnout that bottomed out during the COVID-19 pandemic. 
  • The gender gap between salaries for male and female physicians was $102k, worsening slightly compared to last year (31% vs. 29%). 
  • The average workweek shortened slightly (49 vs. 50 hours).

The Takeaway
Medscape’s new salary survey findings are positive news for radiologists – and indeed all physicians – and are a welcome departure from the days of burnout and stagnant wage growth. So far, predictions that AI will take jobs from radiologists remain far in the future.

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.

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