Tag: medical imaging

Big Prostate AI Study Supports Screening Role

Posted on by Brian Casey

A new international study testing AI’s performance in analyzing prostate MRI exams could lay the groundwork for the technology’s broader role in prostate cancer screening. In the SCARLET-1 study in Radiology, researchers found that an open-source AI algorithm performed well – and consistently – across different research studies. 

Prostate cancer screening stands on the cusp of population-based adoption, but concerns remain about false positives, particularly how to determine which men should undergo biopsy and which ones can be directed to PSA monitoring. 

  • MRI has proven invaluable for working up patients with elevated PSA levels, but it’s expensive and requires a level of radiologist expertise that’s usually only available at high-volume centers. 

AI could help by automating key parts of the MRI process, enabling the modality to be offered more broadly to support prostate cancer screening.

  • The SCARLET-1 study investigated AI’s potential by combining results from research studies conducted in 22 countries around the world: PRIME, STHLM3-MRI, and IP1-PROSTAGRAM.

All three studies used an updated version of the PI-CAI AI algorithm, developed as part of a challenge sponsored by the Prostate Imaging-Cancer AI consortium in 2024.

  • In all, 2k patients received prostate MRI on 1.5T and 3T scanners, and the PI-CAI algorithm was used to analyze exams at different sensitivity and specificity thresholds to detect cases suspicious for prostate cancer rated at Gleason grade group ≥ 2, the threshold for clinically significant cancer. 

When compared to interpretations by local radiologists, prostate AI…

  • Achieved good AUC across the three studies (AUC = 0.84 to 0.89).
  • Had significantly higher agreement with an external reference standard than local radiologists (13 to 16 percentage points).
  • Maintained a positive net benefit across different thresholds, in particular those intended to reduce biopsies (patient must have > 30% chance of cancer).
  • Predictably, sensitivity thresholds affected AI’s performance, with higher sensitivity detecting more cancers but also generating more biopsies. 

The Takeaway

The SCARLET-1 study confirms that AI of prostate MRI performs as well as radiologists in real-world settings across multiple different locations. That’s important in making prostate cancer screening available in areas where specialty expertise may not be available to interpret imaging exams.

Blackford Finds Buyer with Sopris Acquisition

Posted on by Brian Casey

One of radiology’s longest-running mysteries was resolved late last week when a U.S. private equity firm acquired AI platform company Blackford Analysis from Bayer. Sopris Capital said it would fold Blackford into another radiology AI property it owns, Azra AI, as it builds its healthcare technology portfolio. 

The news comes just over a year after Bayer made the stunning announcement that it was backing away from the AI platform business and would “discontinue offerings and services” in the sector, including Blackford. 

  • The move reversed several years of Bayer’s investment in digital platforms, including the company’s own Calantic Digital Solutions division, which had been formed in 2022. 

The announcement left Blackford twisting in the wind. Some industry experts questioned the long-term viability of the AI platform model, in which platform companies act as middlemen between algorithm developers and end-users, adding a markup that made AI deployment more expensive. 

  • But others still see AI platforms and algorithm orchestration as a necessary foundation for successful AI deployment, especially given the explosion of clinical algorithms and the complexity of IT integration.

For its part, early reports following last year’s Bayer announcement suggested that Blackford’s business model was sound, with the company reporting a strong pipeline of prospective customers. 

  • And despite the departure of key executives like founder Ben Panter, Blackford continued operations under new managing director James Holroyd. The company even exhibited at SIIM 2026 – not exactly a sign of a firm about to go under. And not long after SIIM, Blackford announced a new partnership with Azra AI – perhaps a portent of last week’s announcement. 

So what do we know about Sopris? With its main offices in Colorado and Florida, Sopris first invested in Azra AI in 2022 to help the company acquire Digital Reasoning and expand into oncology. 

Sopris sees the Blackford acquisition as adding an AI orchestration capability to its growing imaging IT portfolio, combining Blackford with Azra AI. 

  • Blackford will remain a separate legal entity with operations in Scotland and the U.S., and will continue serving its existing OEM partners and health system customers “with no disruption to current deployments or operations.” 

The Takeaway

The Sopris acquisition of Blackford means that one of the radiology AI sector’s most established companies will continue operations for the foreseeable future. It also raises the intriguing prospect that a new player is entering the game to drive AI consolidation. 

How Common Are Imaging Deserts?

Posted on by Brian Casey

Amid concerns about equitable access to healthcare, questions have arisen about imaging deserts – U.S. regions without any access to radiology services. A new study in Academic Radiology probes the issue, finding that it may not be as bad as originally feared – for reasons unique to radiology. 

Uneven access to healthcare – particularly in rural regions – has become a common refrain in analyses of the U.S. healthcare system. 

  • In radiology, the issue has reached into the debate over workforce shortages as open jobs go unfilled in rural areas while urban positions are snapped up quickly – a trend one job board dubbed “geographic inconvenience.” 

But how common is a complete lack of access to imaging? And what are the correct public policy tools to address it? 

  • In the new study, researchers assessed imaging availability in U.S. counties by determining whether imaging equipment, local radiologists, or both were available. 

In their analysis of 3.1k U.S. counties covering 340.1M residents, researchers found…

  • Only 1.9% of Americans (6.4M people) lived in true imaging deserts, where there was neither installed imaging equipment nor a local practicing radiologist.
  • 7.6% of Americans (25.9M people) lived in counties with an installed scanner but no local radiologist.
  • 0.1% lived in a county with a local radiologist but no equipment.
  • 90% (307.4M) were considered “fully served” by imaging, with local image acquisition available and some kind of remote interpretation possible.

The statistics point to radiology’s advantage over other medical specialties that require doctors to be physically present to deliver care.

  • Thanks to teleradiology, images can be acquired locally with installed equipment and then transmitted to central hubs for interpretation. 

But even if less than 2% of the population lacks access, 6.4M Americans is a lot of people. How does living in an imaging desert affect their access to radiology? 

  • Researchers found that imaging use rates were 6% lower in imaging deserts compared to fully served regions, while Medicare payments per beneficiary were 8.9% lower. 

The Takeaway

The new findings on regional access to imaging show that while most Americans have good access to radiology services, more work is needed to improve care for those who don’t. This is particularly true for services like interventional radiology that require radiologists to be present. 

Is It Time for Mammography AI?

Posted on by Brian Casey

Is AI ready for widespread use in mammography screening? That question was the focus of a point-counterpoint debate in AJR that discussed the pros and cons of broad-based clinical AI deployment for one of radiology’s most high-profile exams. 

Mammography screening is perhaps the clinical use case with the most potential to help radiologists drowning under a rising tide of imaging exams.

  • Several large-scale research studies have been published recently – mostly in Europe – showing that, for most normal mammograms, AI could reduce the need for a second reader, as is common in European screening programs.

Indeed, mammography AI developer Lunit last week announced a contract with a Swedish health system that would do just that, having AI act as an autonomous second reader. 

  • While the U.S. doesn’t typically use second readers for breast screening, AI could still play an important role in assisting radiologists in interpreting mammograms, predicting patient risk, and acting as a sort of clinical spell-check to reduce misses.

So does that mean mammography AI is ready for prime time? Taking the skeptical view were Duke University breast radiologists Eun Langman, MD, and Vilert Loving, MD, who believe caution is warranted, for the following reasons…

  • Most of AI’s reported diagnostic performance gains have been with 2D digital mammography, and may not apply to DBT mammography, which has become the standard in the U.S.
  • While autonomous AI interpretation of mammograms has been touted, there are practical barriers to its widespread application, such as the legal implications of AI misses.
  • There is no evidence showing AI reduces breast cancer mortality, and there are many other interventions that could do so at a lower cost.

Taking the other side of the argument was MGH breast radiologist Manisha Bahl, MD, who noted that…

  • Evidence supporting mammography AI’s effectiveness has come not only from randomized controlled trials like MASAI, but also from studies of routine population-based screening like the PRAIM study in Germany. 
  • Studies of mammography AI in the U.S. – where DBT is routinely used – have shown cancer detection gains comparable to those in Europe.
  • In addition to diagnostic support, AI can speed radiologist workflow, such as through autonomous AI reporting of some mammograms.

The Takeaway

In the debate over mammography AI deployment, both sides present persuasive arguments. But as is the case with AI applications outside healthcare, the technological imperative to leverage this powerful new technology is likely to swamp any attempts to put up guardrails restricting its use.

Measuring AI’s Impact in Real-World Radiology

Posted on by Brian Casey

One of the criticisms of AI in radiology is that there isn’t much data on its effectiveness in real-world environments. Well, a new study in JACR aims to fill that gap by detailing how AI adoption at a large imaging network in Europe affected key radiology performance metrics. 

Plenty of studies have been published on AI’s positive contribution to radiologists’ diagnostic performance. 

  • But in most cases this research was performed under controlled conditions, or involved a single AI algorithm working on a particular task. That’s left many wondering how well AI would perform under actual conditions. 

In the new study, researchers tracked the performance of 10 AI algorithms from seven vendors that were implemented over the Incepto Medical AI orchestration platform at 3R Swiss Imaging Network, a 20-center outpatient radiology network in Switzerland. 

  • Over 4.5 years, researchers tracked the network’s performance before and after AI implementation for the major radiology key performance indicator, report turnaround time (TAT). They also tracked AI adoption rates and infrastructure latency, or how delays in data transfer between PACS networks affected AI usage.

In all, 389k AI-assisted imaging exams were processed over the implementation period from 2021 to 2025, with researchers finding…

  • AI’s biggest impact was in high-volume modalities, reducing TAT for trauma radiography (-26%) and knee MRI (-18%).
  • AI improved the network’s diagnostic capacity, performing the work of almost half a full-time-equivalent radiologist for trauma radiography alone (0.46 FTE). 
  • AI usage rates were high, with 91% reporting active AI adoption and 66% reporting regular use.

On the downside, latency in data transfer between PACS networks slowed down AI performance, and in some cases AI results arrived after reports were finalized – too late to make a clinical impact…

  • Median total latency was 2.06 minutes per exam, with 72% attributable to data routing and tasks like fetching studies from PACS.
  • AI’s “too late rate” was 7.2% overall, ranging from 13% for chest CT to 3% for knee MRI, with 6.8% for trauma X-ray.

How did the network’s radiologists feel about AI? Researchers surveyed over 50 radiologists working for the network, finding that most of the differences in before-and-after opinions weren’t statistically significant. 

  • The only difference was a modest gain in the perception that AI made radiologists more productive (from 2.57 to 2.94 on a five-point Likert scale).

The Takeaway

The new study fills in a major gap in real-world experience regarding radiology AI implementation. It also reveals areas for improvement, as slow data transfer rates between networks can prevent AI results from reaching radiologists in time. 

One-Stop CT Chest Screening

Posted on by Brian Casey

Could opportunistic screening be driving the evolution of low-dose CT from a screening test just for lung cancer into a broader one-stop exam for multiple chest pathologies? A new study in JACR suggests as much – and provides a blueprint for how to do it. 

There’s no question LDCT lung cancer screening is gaining traction around the world. But as screening programs are starting, clinicians are finding that CT scans can detect far more than just lung cancer. 

  • Emphysema, commonly associated with COPD, is one such condition, as it goes hand-in-hand with lung cancer in people with long smoking histories. Another is coronary artery calcium (CAC), a known risk factor for cardiovascular disease. 

In the new study, researchers wanted to investigate the prevalence of COPD and CAC in their LDCT screening program covering three hospitals in the NewYork-Presbyterian system. 

  • In addition to documenting prevalence rates, they wanted to connect clinical outcomes to social determinants of health, which have been demonstrated to affect patient care. 

They tracked results from 2.4k people who got baseline LDCT scans from 2015 to 2022, finding…

  • Emphysema was detected in 77% of patients, and CAC in 65%.
  • 87% of patients with emphysema had no previous diagnosis.
  • The proportion of patients getting pulmonary care after their LDCT scan increased (from 24% to 46%), as did the rate of those getting cardiovascular care (from 43% to 63%).  
  • Lower household income was associated with 57% higher adjusted odds of emphysema, and food insecurity with 52% higher odds. 
  • No social determinants of health were associated with CAC burden.

It’s worth noting that the study did not use any IT tools to classify COPD and CAC, like AI image analysis or longitudinal tracking and reporting. 

  • It’s entirely possible that the results could be even more promising given the plethora of algorithms that have become available for opportunistic screening applications. 

The Takeaway

The new study on incidental CAC and emphysema findings on LDCT exams shows that lung screening has the potential for opportunistic disease detection even if you don’t have fancy AI algorithms at your disposal. All you need are radiologists and clinical staff dedicated to providing better patient care. 

Repeat MRI Scans Take Operational Toll

Posted on by Brian Casey

Nobody likes having to repeat a task, and nowhere is that more evident than in MRI. A new study in JACR quantifies the operational toll of MRI scans that have to be repeated due to patient motion, revealing both good news and bad news. 

Despite its unquestioned diagnostic power, MRI is the imaging modality most susceptible to motion artifacts.

  • That’s because patients have to lie motionless for long periods of time in a confined, noisy metal tube – an environment tailor-made to induce claustrophobia. 

MRI vendors have introduced a variety of new technologies and pulse sequences to speed up exams, but patient motion is still a nagging problem. 

  • To investigate the issue further, researchers from NYU Grossman School of Medicine collaborated with Siemens Healthineers to document MRI exam repeat rates over six months at NYU.

In all, 85.3k MRI exams were included in the analysis, with the following findings…

  • 4.8% of all MRI exams (4k) included at least one repeated sequence.
  • The highest repeat rate occurred in pediatrics and the lowest in breast MRI (9.9% and 3.5%, respectively). 
  • Repeat acquisitions added up to 115 hours of scanner time, equivalent to 268 lost MRI scanning slots. 

To take the study deeper, researchers more closely analyzed brain MRI and MR cholangiopancreatography (MRCP) studies, finding…

  • For 77 patients undergoing brain MRI, repeat acquisitions had better image quality and diagnostic confidence, but 20% of repeat sequences did not improve. 
  • For 27 patients undergoing MRCP, repeat acquisitions were less successful, as they didn’t improve image quality or diagnostic confidence.

So that’s the bad news. What’s the good news? The study’s 4.8% repeat rate was actually lower than those reported in previous research into motion-related repeats, which have ranged from 7.5% to 30%. 

  • One reason for the discrepancy is that the current study relied on exams labeled as repeats by technologists, which could have resulted in an underestimation. 

But another reason is the fact that MRI practices have adopted a wide range of techniques and technologies that reduce the impact of patient motion, like shorter protocols, parallel imaging, compressed sensing, and deep learning reconstruction. 

The Takeaway

The new findings into the operational toll of MRI motion artifacts provide useful insights into the extent of the problem. But they also include a silver lining – that the modality’s relentless pace of technological innovation could make repeat exams less of an issue in the future. 

When Chest CT Needs a Functional Layer

Posted on by Partner Content

Regional lung function is emerging as an important factor in thoracic imaging. While chest CT is central to evaluating lung disease, anatomy (structure) alone does not always explain the full clinical challenge. Fortunately, a new generation of CT-based software tools brings regional lung function into the workflow radiologists use every day – chest CT.

Patients with dyspnea, COPD, emphysema, interstitial lung disease, or pulmonary vascular questions can often have symptoms that are difficult to reconcile with structural findings alone.

Radiologists and clinicians assemble the broader clinical picture from multiple sources: CT findings, pulmonary function tests, and nuclear medicine ventilation/perfusion (V/Q) exams. 

  • Each offers useful information, but each has limits. Pulmonary function tests provide global measures rather than regional detail, while V/Q imaging depends on radiotracer availability, department hours, and nuclear medicine resources.

One example of an emerging technology that provides radiologists with deeper, functional pulmonary detail is CT:VQ from 4DMedical. 

  • CT:VQ, operating in a Software as a Service (SaaS) mode, uses paired inspiratory and expiratory non-contrast chest CT scans to generate regional ventilation and perfusion maps – along with quantitative outputs designed for radiology workflow.

This approach brings functional lung information closer to the CT reading room without requiring injected contrast or radiotracers for the CT:VQ analysis. 

  • It also builds on a modality that is widely available and often operational around the clock.

Adding regional ventilation and perfusion information to CT could help address common thoracic imaging questions…

  • Which regions are ventilating well, and which are not?
  • Is perfusion preserved in areas that look abnormal on CT?
  • Do symptoms align with regional functional impairment?
  • Which lung regions may be most important to preserve before intervention?

The non-contrast approach has compelling clinical advantages… 

  • Contrast can trigger renal screening, prior reaction history, IV access challenges, workflow delays, and patient anxiety. 
  • Avoiding contrast can simplify the pathway for many patients.

CT-based functional imaging is not a replacement for legacy lung function assessments. But it brings quantitative tools that can make routine studies more clinically actionable.

The Takeaway

Technology brings radiologists new opportunities. Chest CT can reveal far more than what the lungs look like. With tools such as CT:VQ, CT can show regional patterns of lung ventilation and perfusion – adding functional context to a radiologist’s toolkit.

More information about CT:VQ and non-contrast functional lung imaging is available at ctvq.com.

What Now for the USPSTF?

Posted on by Brian Casey

After not meeting for over a year and undergoing major turnover, HHS Secretary Robert F. Kennedy, Jr. last week announced new members of the U.S. Preventive Services Task Force – including the USPSTF’s first-ever radiologist. 

USPSTF was created in 1984 as a panel of experts tasked with producing evidence-based advice on preventive medical care, like cancer screening guidelines. 

  • But it’s that last responsibility that’s drawn the USPSTF into controversy. The group in 2009 changed its blanket recommendation regarding screening mammography for women aged 40 to 50, instead recommending women consult with their doctors. This sparked a backlash that persisted until 2023, when it said it would lower the age back down to 40. 

Indeed, the USPSTF has often been accused of being too conservative in recommending new technologies that have life-saving potential. 

  • For example, its implementation of the new mammography screening guidelines neglected to recommend supplemental dense breast screening, despite the demonstrated connection between dense breasts and cancer risk.

Enter RFK, Jr., who as HHS secretary has appointment authority over the panel’s 16 members (who, it should be noted, serve as volunteers rather than as federal employees).

  • He’s held the USPSTF with the same suspicion he reserves for the rest of the U.S. medical establishment, and on his watch the group has gone a year and a half without a meeting. He also removed both of the panel’s vice chairs in May. 

RFK, Jr. hadn’t revealed much about his intentions for the panel, but that ended on September 17, when HHS announced a new slate of appointments. 

  • He tapped eight new members, including a number of medical specialists, addressing a frequent criticism that the USPSTF was too heavily weighted toward primary care and lacked specialty experience. 

And in what is a first, he named a radiologist to the USPSTF: Dennis Wulfeck, MD, of MBB Radiology, a Radiology Partners-affiliated practice in Jacksonville, Florida. 

  • Wulfeck earned his MD at Wright State University, performed his residency at the University of Louisville, and completed a fellowship in nuclear medicine at the University of Cincinnati. Wulfeck has published research in nuclear medicine and has been involved in radiology practice issues. 

What now for the USPSTF? While the new appointments seem to represent physicians from U.S. medicine’s mainstream, their work will be closely scrutinized for signs of politicization, particularly around RFK, Jr.’s pet issue of childhood vaccines. The USPSTF hasn’t said when its next meeting will take place. 

The Takeaway

Radiology has long had a complicated relationship with the USPSTF, mostly due to its conservative stances on cancer screening and the adoption of new imaging technologies. Will that change with a radiologist now serving on the panel? We shall see. 

Cancer Survival Rates Rise as AI’s Impact Looms

Posted on by Virginia Hunt

More people are surviving cancer in the U.S. than ever before, even as the number of new U.S. cancer cases is expected to increase 19% by 2050. The AACR’s 2026 Cancer Progress Report showed rising cancer survival rates – especially for breast and lung cancer – and new technologies like AI could drive survival even higher.

The AACR report revealed big improvements in treatment and clinical outcomes…

  • The overall five-year survival rate for distant-stage cancer is more than double what it was three decades ago (17% to 36%). 
  • Since July 2025, the FDA approved 11 new anticancer therapies and additional uses for five existing interventions.

While that progress is coming fast, it has stalled in some areas… 

  • The pancreatic cancer survival rate remains low at 14%, compared with 92% for breast cancer and 98% for prostate cancer.
  • Disparities persist: Black men and American Indian/Alaska Native women had the highest cancer incidence rates from 2019 to 2023.
  • Across the same period, lung cancer rates were 37% higher in rural versus metropolitan areas.

Breast and lung cancer rank second and third for new diagnoses in the U.S., but advances in imaging-based screening for these cancers are accelerating… 

  • The U.S. breast cancer mortality rate declined by 44% from 1989 to 2024.
  • Breast cancer death rates have been decreasing by 1.2% per year.
  • Efforts to promote earlier detection are occurring, such as the USPSTF’s lowering of the recommended mammography screening age to 40.
  • Declines in lung cancer mortality accounted for the largest share of cancer deaths averted from 1975 to 2020, partly because of tobacco-control programs. 

But screening continues to be an essential tool to drive cancer mortality down even more, and AI could become a handy helper.

  • The MASAI trial found that AI-supported screening was associated with a 12% lower rate of interval cancers.
  • AI-assisted mammography detected early-stage breast cancer with greater sensitivity than standard screening without sacrificing specificity.
  • The report cited two recently cleared AI tools that can help with early detection: RevealDx’s RevealAI-Lung, which reportedly reduced false positives by 18%, and Median Technologies’ eyonis LCS, which showed 93% sensitivity in tests.

The tools could also help radiologists cope with the ongoing workforce shortage by stepping in as second readers or prioritizing high-risk cases. 

  • AI assistance could decrease reading workload by 44%, per the AACR report.

The Takeaway 

The AACR’s report shows encouraging progress in the war on cancer. In the future, AI advances could help radiologists continue to improve screening for lung and breast cancer patients, with some promising early evidence.

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

Like the website? You'll love the newsletter

Completely free. Every Monday and Thursday