Tag: medical imaging

CT Lung Screening Chats Pay Off

Posted on by Brian Casey

Patients who talked about CT lung cancer screening with their doctors were more likely to actually follow through on getting scanned. That’s according to a study this week in CHEST that offers support for shared decision making – a process that some screening proponents have criticized.

The U.S. continues to see disappointing compliance rates for CT lung cancer screening, over 10 years after the USPSTF recommended the exam.

  • Some lung screening proponents suggest that one barrier to screening is a CMS rule requiring a shared decision-making session between patients and doctors before the first scan is performed – a requirement that’s not in place for any of the other major cancer screening tests.

But the new study indicates that shared decision making could actually work to boost compliance. 

  • Researchers from the Harvey L. Neiman Health Policy Institute led by first author YoonKyung Chung, PhD, examined lung screening compliance rates for 22.6k people who had their first CT exam between 2016 and 2019.

Researchers looked at differences in annual follow-up lung screening rates between people who got shared decision-making sessions and those who didn’t, finding… 

  • Only 11% of study participants had a session before their first scan.
  • One year after the initial scan, those who participated in sessions were 27% more likely to get a follow-up exam.
  • Four years later, the compliance rate rose to 33%. 

If CMS requires shared decision-making sessions for reimbursement, why are they occurring so infrequently? 

  • The authors called this phenomenon “puzzling,” and suggested it’s because CMS is not enforcing the mandate through tools like claims denial. CMS could also boost utilization by providing higher reimbursement for the discussions.

The Takeaway

The new findings suggest that shared decision making should be viewed as an opportunity rather than a barrier to convincing patients of CT lung cancer screening’s value. The results track with other studies showing that a high-touch approach with tools like patient navigators can work.

Top 6 Radiology Trends from 2025’s First Half

Posted on by Brian Casey

The first half of 2025 has drawn to a close, and once again it was an eventful period for radiology. As we do every year, we’ve compiled a list of the top six stories – one for each month – to help recap what was important in medical imaging.

Consolidation in Imaging Services

Radiology’s imaging services segment continues to consolidate as smaller providers get gobbled up by larger players. One of the biggest consolidation moves happened in April when Radiology Partners agreed to take over radiology contracts held by Envision Healthcare. Private-practice radiology continues its slow decline, as documented by a study in May that found just 50% of U.S. radiologists in private practice.

DeepHealth Drives AI Consolidation

Imaging services isn’t the only radiology market that’s consolidating – the AI sector is also seeing heightened M&A activity as algorithm developers suffer from a decline in venture capital activity and slower clinical adoption. Many of the deals are being driven by RadNet’s DeepHealth subsidiary, which continues to acquire independent AI developers, such as See-Mode Technologies in ultrasound and its proposed deal for iCAD in mammography.

Radiology VC Funding Goes Boom – Then Bust

Radiology venture capital funding appeared to be gaining momentum in the first few weeks of 2025, as January saw six funding rounds, but the good times didn’t last as economic concerns slowed investment. Radiology startups may have to get used to a more competitive funding environment, although there was positive news in the spring with funding rounds from HOPPR, DESKi, RadAI, Yunu, Aeon, Heartflow, Chipiron, Brainomix, and Brainreader.  

When Will Getting Paid for AI Get Easier?

Reimbursement is one of the major stumbling blocks holding back wider adoption of clinical AI. Legislation that might grease the reimbursement skids was introduced into the U.S. Congress in April, but sources tell The Imaging Wire that it wasn’t included in the big budget bill that was just passed. In the meantime, AI developers and users will have to deal with a patchwork of AI reimbursement pathways.

Concerns Rise about CT Radiation Dose

Several studies were published in the first half of 2025 raising concerns about radiation dose from CT scans. First, researchers in April released a study that they claimed showed that all the CT scans performed in the U.S. in a single year would cause more than 100k cancers over the lives of the patients who got them. That was followed up with a paper in May linking greater CT use in European countries to a higher percentage of patients with five-year cumulative radiation dose of over 100 mSv. 

Can Imaging IT Tools Help Radiologists Manage Rising Volume?

Radiology’s rising scan volume and static workforce have IT developers furiously working on tools to bridge the gap. A March paper listed the half-dozen IT tools radiologists say they want (only one was AI), but another analysis threw cold water on the idea by predicting that AI would actually increase radiologists’ workload, not reduce it. Meanwhile, multiple studies are showing that for applications like breast screening, AI can reduce workload by as much as 41%.

The Takeaway

The midpoint of the year is a great time to take stock of radiology’s progress and the issues that have bubbled to the surface over the past six months. AI dominated radiology for the first half of 2025, and odds are the trend will continue in the back half of the year.

AI and Legal Liability in Radiology

Posted on by Brian Casey

What impact will artificial intelligence have on the legal liability of the radiologists who use it? A new study in NEJM AI suggests that medical malpractice juries may pass harsher judgment on radiologists when they make mistakes that disagree with AI findings.

AI is viewed as a technology that can save radiologists time while also helping them make more accurate diagnoses.

  • But there’s a dark side to AI as well – what happens when AI findings aren’t correct, or when radiologists disagree with AI only to discover it was right all along?

In the new study, a research team led by Michael Bernstein, PhD, of Brown University queried 1.3k U.S. adults on their attitudes toward radiologists’ legal liability in two clinical use cases for AI – identifying brain bleeds and detecting lung cancers.

  • Participants were asked if they felt radiologists met their duty of care to patients across different scenarios, such as whether the AI and the radiologist agreed or disagreed on the original diagnosis. 

Responses were compared to a “no AI” control scenario in which respondents assessed legal liability if radiologists hadn’t used AI at all, with researchers finding …

  • If radiologists disagreed with AI, more respondents found radiologists liable …
    • Brain bleeds: 73% found radiologist liable (vs. 50% with no AI)
    • Lung cancer: 79% found radiologist liable (vs. 64% with no AI)
  • If both radiologists and AI missed the diagnosis, there was no statistically significant difference …
    • Brain bleeds: (50% vs. 56% with no AI, p=0.33)
    • Lung cancer: (64% vs. 65% with no AI, p=0.77)
  • Respondents were less likely to side with plaintiffs when given information about standard AI error rates …
    • When AI agreed with the radiologist diagnosis:
      • Brain bleeds: (73% plaintiff agreement fell to 49%)
      • Lung cancer: (79% fell to 73%)
    • When AI disagreed with the radiologist diagnosis:
      • Brain bleeds: (50% plaintiff agreement fell to 34%)
      • Lung cancer: (64% fell to 56%)

The Takeaway

The new study offers a fascinating look at AI’s future in radiology from a medico-legal perspective. But there’s one question the researchers didn’t address: If AI-supported image interpretation eventually becomes the standard of care, will radiologists be found liable for not using it at all? Stay tuned. 

PET Radiotracers Drive News from SNMMI 2025

Posted on by Brian Casey

The Society of Nuclear Medicine and Molecular Imaging wrapped up its annual meeting in New Orleans this week, demonstrating the growing influence of new PET radiotracers and the rise of theranostics as a discipline that forms the foundation of precision medicine. 

The importance of new PET tracers is evident in the selection of SNMMI 2025’s Image of the Year, a peptide-based PET tracer called fluorine-18 AlF-NOTA-PCP2, developed by researchers from China for imaging patients with head and neck cancers. 

  • The tracer targets PD-L1 expression from tumors, and in an SNMMI 2025 study with 40 patients, it outperformed conventional FDG-PET. Clinical availability is expected in the next 2-3 years.

SNMMI’s Abstract of the Year went to a Canadian study using PET with GE HealthCare’s Flyrcado fluorine-18 flurpiridaz cardiac radiotracer.

  • In 220 patients with coronary artery disease, Flyrcado accurately quantified myocardial flow reserve after exercise and pharmacologic stress, creating the possibility of a first-line test for people with CAD. 

Other SNMMI 2025 highlights included … 

The Takeaway

Nuclear medicine has long been considered one of the less dynamic areas of medical imaging, but that’s changing with its new focus on theranostics. This year’s SNMMI 2025 shows the progress being made, with more advances on the horizon.

Radiology Workforce Shortage Tightens

Posted on by Brian Casey

Radiologist attrition rates have jumped 50% since 2020, and new workforce projections suggest the shortage will only worsen as imaging demand continues to outpace supply. The report – from staffing firm Medicus Healthcare Solutions – projects a worsening supply of radiologists by 2037.

It’s no news to anyone that healthcare is being squeezed by rising volumes from an aging population and chronic staff shortages caused by a training system that simply isn’t turning out enough qualified medical professionals.

  • In radiology, both radiologists and radiologic technologists are in short supply, and there have been only 29 diagnostic radiology PGY-1 training positions added since 2021. 

The Medicus report mostly assembles data acquired from other sources such as a recent study in JACR on radiologist supply, but taken together the numbers paint a sobering picture …

  • Imaging utilization is projected to grow 17-27% by 2055.
  • Radiologist attrition rates have grown 50% since 2020. 
  • Radiologist distribution per 100k population is uneven across the U.S., ranging from 25 radiologists in Minnesota to 9 radiologists in some other states.
  • Reimbursement is falling, with the Medicare conversion factor for 2025 dropping -2.83% for diagnostic radiology and -4.83% for interventional radiology.

What’s to be done? On the positive side, at least one new radiologist residency program started up this year, and legislation was recently introduced that would add 14k residency training slots over seven years. 

  • The report also recommends teleradiology as a possible solution, with 92% of radiologists in a recent survey saying their institution offered remote work options and 73% of radiologists participating in remote work. 

Medicus also advised health systems to take several compensation-focused steps to attract and retain radiologists …

  • Offer flexible, hybrid work schedules.
  • Provide competitive compensation packages and signing bonuses.
  • Improve vacation policies and time-off benefits.

The Takeaway

It’s hard to see short-term Band-Aids like better salary and benefits solving healthcare’s workforce shortage. And some are even questioning whether AI will really help make radiologists more efficient. In the end, systemic changes like a sharp expansion in residency training slots are what’s needed to effect a long-term solution to the staffing dilemma. 

CT Guides Heart Health

Posted on by Brian Casey

People who got evidence of their cardiovascular health from coronary CT angiography scans led healthier lifestyles compared to those who got conventional cardiac risk scoring. That’s according to a new study in JAMA Cardiology that has intriguing ramifications not only for managing heart disease but also for the imaging-based wellness industry. 

Cardiovascular disease is the number one killer globally, accounting for one in seven deaths.

  • The risk of heart disease can be managed through lifestyle changes like better diet and exercise, but getting patients to follow their doctors’ advice can be a challenge.

So researchers in the new study investigated whether data from a patient’s own coronary CT angiography exam would be a better motivational tool compared to simply calculating a risk score based on demographic factors like weight, BMI, and daily step count.

  • They drew 400 participants from the SCOT-HEART 2 study of CT-based cardiovascular screening in Scotland. 

Cardiovascular risk scores were calculated for one group using the ASSIGN criteria for 10-year cardiac event risk, while another group got CCTA scans and were shown their results. 

  • Interventions were recommended for patients in either group based on cardiac risk as calculated by either ASSIGN criteria or CCTA scans, ranging from no interventions to low-level statin therapy to high-intensity statin and enzyme inhibitor treatment.

At six months of follow up, researchers calculated how many participants met the U.K.’s NICE recommendations for diet, body mass index, smoking, and physical exercise, finding …

  • Nearly three times more CCTA patients complied with NICE healthy lifestyle guidelines (17% vs. 6%).
  • Fewer CCTA patients were told to start preventive therapy due to their risk (51% vs. 75%).
  • And of these, CCTA patients were more likely to have followed advice to begin a therapeutic program (77% vs. 46%). 
  • There was no difference in behavior between CCTA patients who saw their own images and those who were told verbally of their results.

In one important fact, the researchers noted that the study was only designed to measure compliance. 

  • It did not assess any change in coronary events over time – these will be addressed in the larger SCOT-HEART 2 study. 

The Takeaway

The new study offers powerful evidence that getting their own medical imaging results can drive patients to adopt lifestyle changes that lower their disease risk. In addition to informing cardiovascular disease management, it’s also possible to see these findings employed as part of the wellness screening programs that are becoming increasingly prevalent. 

All-Star AI for Prostate MRI

Posted on by Brian Casey

An AI model for prostate MRI that combines the best features of five separate algorithms helped radiologists diagnose clinically significant prostate cancer in a new study in JAMA Network Open. 

The Prostate Imaging-Cancer AI consortium was formed to address a nagging problem in prostate cancer screening.

  • Studies have shown that MRI can reduce biopsies and minimize workup of clinically insignificant disease, but it also has high inter-reader variability and requires a high level of expertise. 

The PI-CAI challenge brought together researchers from multiple countries with a single goal: develop an AI algorithm for prostate MRI that would improve radiologists’ performance.

  • Results were presented at RSNA and ECR conferences, as well as in a 2024 paper in Lancet Oncology that showed that individually the algorithms improved radiologist performance and generated fewer false positives.

But what if you combined the best of the PI-CAI algorithms into a single all-star AI model? 

  • Researchers did just that in the new study, combining the top five algorithms from the PI-CAI challenge into a single AI model in which each algorithm’s results were pooled to create an average detection map indicating the presence of prostate cancer. 

To test the new algorithm, 61 readers from 17 countries interpreted 360 prostate MRI scans with and without the model. 

  • Patients in the test cohort had a median age of 65 years and a median PSA level of 7.0 ng/mL; 34% were eventually diagnosed with clinically significant prostate cancer.

Results of PI-CAI-aided prostate MRI were as follows …

  • Radiologists using the algorithm had higher diagnostic performance than those who didn’t (AUROC=0.92 vs. 0.88).
  • PI-CAI working on its own had the highest performance (AUROC=0.95).
  • Sensitivity improved for cases rated as PI-RADS 3 or higher (97% vs. 94%).
  • Specificity also improved (50% vs. 48%).
  • AI assistance improved the performance of non-expert readers more than expert readers, with greater increases in sensitivity (3.7% vs. 1.5%) and specificity (4.3% vs. 2.8%).

The Takeaway

The new PC-CAI study is an important advance not only for prostate cancer diagnosis but also for the broader AI industry. It points to a future where multiple AI algorithms could be combined to tackle clinical challenges with better diagnostic performance than any model working alone.

RadGPT Simplifies Radiology Reports for Patients

Posted on by Brian Casey

When it comes to informing patients of their imaging results, radiologists are caught between a rock and hard place. A new study in JACR shows how generative AI can help by drafting patient-friendly reports that are simple but accurate.

Patients must be informed immediately of their medical results according to a 2021 final rule under the 21st Century Cures Act that prevents medical information blocking. 

  • And while the technology exists to do that through tools like email and electronic patient portals, rapid notification can create confusion because the language physicians use to communicate with each other isn’t easily understood by anyone outside medicine.

Sure, radiology reports could be rewritten manually for patients, who typically read at about the eighth-grade level.

  • But given today’s workforce shortages, who’s going to do that?

Generative AI and large language models offer a solution. In the new JACR paper, researchers from Stanford University led by senior author Curtis Langlotz, MD, PhD, described their development of RadGPT, an LLM designed to improve patient communication.

  • To develop RadGPT, researchers started with OpenAI’s GPT-4 model and the RadGraph concept extraction tool to create an LLM that analyzes patient radiology reports and generates concept explanations and question-and-answer pairs.

How well did RadGPT work? The researchers tested it on 30 radiology reports generated at Stanford from 2012 to 2020, including different modalities and clinical applications. 

  • The LLM was asked to generate reports at a fifth-grade reading level (the level recommended by the Joint Commission for patient-facing healthcare materials).

Five radiology-trained physicians then rated the quality of RadGPT’s responses, finding …

  • The average rating of RadGPT-generated concept explanations was 4.8 out of 5.
  • 95% of concept explanations had an average rating of 4 or higher.
  • 50% of concept explanations were rated 5, the highest possible rating.
  • Questions and answers generated by RadGPT were also rated highly, with an average rating of 3.0 on a three-point scale..

The Stanford researchers told The Imaging Wire that their goal is to make RadGPT more widely available as part of a prospective evaluation with real patients.

  • They are also developing a user-friendly interface in which patients can receive hyperlinked radiology reports.

The Takeaway

RadGPT and solutions like it fill a desperate need for tools that can save time for radiologists while helping patients better understand their reports and get more engaged in their care. The next step is to get technology like this into the hands of practicing radiologists.

Mammo Risk Prediction Improves with AI

Posted on by Brian Casey

Artificial intelligence is beginning to show that it can not only detect breast cancer on mammograms, but it can predict a patient’s future risk of cancer. A new study in JAMA Network Open showed that a U.S. university’s homegrown AI algorithm worked well in predicting breast cancer risk across diverse ethnic groups. 

Breast cancer screening traditionally has used a one-size-fits-all model based on age for determining who gets mammography.

  • But screening might be better tailored to a woman’s risk, which can be calculated from various clinical factors like breast density and family history.

At the same time, research into mammography AI has uncovered an interesting phenomenon – AI algorithms can predict whether a woman will develop breast cancer later in life even if her current mammograms are normal. 

The new study involves a risk prediction algorithm developed at Washington University School of Medicine in St. Louis that uses AI to analyze subtle differences and changes in mammograms over time, including texture, calcification, and breast asymmetry.

  • The algorithm then generates a mammogram risk score that can indicate the risk of developing a new tumor.

In clinical trials in British Columbia, the algorithm was used to analyze full-field digital mammograms of 206.9k women aged 40-74, with up to four years of prior mammograms available. Results were as follows …

  • The algorithm had an AUROC of 0.78 for predicting cancer over the next five years.
  • Performance was higher for women older than 50 compared to 40-50 (AUROC of 0.80 vs. 0.76).
  • Performance was consistent across women of different races.
  • 9% of women had a five-year risk higher than 3%. 

The algorithm’s inclusion of multiple mammography screening rounds is a major advantage over algorithms that use a single mammogram as it can capture changes in the breast over time. 

  • The model also showed consistent performance across ethnic groups, a problem that has befallen other risk prediction algorithms trained mostly on data from White women. 

The Takeaway

The new study advances the field of breast cancer risk prediction with a powerful new approach that supports the concept of more tailored screening. This could make mammography even more effective than the one-size-fits-all approach used for decades.

Cancer Screening at ASCO 2025

Posted on by Brian Casey

The American Society of Clinical Oncology’s annual meeting isn’t usually known for diagnostic radiology research. But this week’s conference in Chicago included a number of radiology-related studies, particularly regarding cancer screening. 

Most ASCO meetings are dominated by new chemotherapy advances. 

  • But oncologists maintain a strong interest in cancer screening as the first step to guiding patients into advanced treatments.

At ASCO 2025, screening disparities were at the top of the agenda, as evidenced by the following presentations …

  • Mobile mammography addressed healthcare disparities for both urban and rural women in a study that analyzed demographics from 8.3k women screened. 
  • Patients served by mobile mammography in Pennsylvania were more likely to be Black (68% vs. 40%), uninsured (71% vs. 2.1%), and live in an economically deprived area (70% vs. 27%), and they also had higher recall rates (19% vs. 9.9%) and twice the median days to case resolution (29 vs. 14 days).
  • U.S. women who didn’t get mammography screening tended to be younger, uninsured, and have issues with medical costs.
  • Farther afield, Uzbekistan’s new breast screening program was described, with 83.6k women screened and 80% of cancers detected at an early stage.
  • The program also uses AI, with AI achieving higher AUC than a three-radiologist average (0.89 vs. 0.82) while reducing workload 41% with 3X lower recall.
  • In Saudi Arabia, AI was used to audit mammography reports for quality and compliance with BI-RADS guidelines. 
  • A virtual-first approach in California successfully reached candidates for colorectal and CT lung cancer screening, using an online platform with educational resources and scheduling. Of 71 people who met lung screening criteria, 24% completed CT scans, and of these 29% had clinically significant findings. 
  • To improve CT lung screening among low-income people of color, Indiana researchers enrolled 89 screening-eligible people in an educational program. Before the program 56% had never heard of lung cancer screening, but afterwards 100% said they believed screening could save their lives.
  • Ohio researchers found that of 116 lung cancer cases in a tumor registry, 24% got low-dose CT lung screening.
  • An IT tool detected patient concerns about screening’s cost – AKA financial toxicity – and assigned financial navigators to help them. 

The Takeaway
This week’s ASCO 2025 sessions demonstrate the synergy between screening and treatment that’s improving survival for a broad spectrum of cancer patients. Continued progress will only serve to benefit both disciplines.

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!