Author: Brian Casey

Top 10 AI Vendors by FDA Approvals

Posted on by Brian Casey

Who are the top 10 radiology AI vendors, based on the number of FDA regulatory authorizations? The agency provided some clarity this week with an update to its list of authorized AI-enabled medical devices through the end of Q1 2026. 

The FDA updates the list on more or less a quarterly basis, and it’s become a closely watched barometer for tracking not only the health of the AI industry but also which companies have received the greatest number of authorizations.

  • As we’ve noted in the past, the list includes both standalone AI algorithms as well as medical hardware that has AI functionality embedded in it, like a mobile X-ray machine with an onboard AI feature for detecting fractures.

The updated list tracks marketing authorizations through the end of March 2026, and shows that the FDA has…

  • Authorized 1,524 AI-enabled medical devices since it began keeping track in 1995, up 5.1% from Q4 2025. 
  • Authorized a total of 1,164 radiology devices, or 76% of all AI-enabled medical authorizations. 
  • In the first quarter of 2026, the FDA authorized 92 AI-enabled medical devices, or 28% more than in the fourth quarter of 2025.
  • For the quarter, 69 authorizations (75%) were for radiology devices, about the same ratio as in Q4 2025 (76%). 
  • GE HealthCare held its lead as the company with the most radiology AI authorizations at 130 (including recent acquisitions that had AI authorizations of their own).
  • Next is Siemens Healthineers at 95, then Philips at 58, Canon at 48, United Imaging at 40, Aidoc at 33, and DeepHealth at 29, with all numbers including acquisitions. 
  • Rounding out the top 10 are Samsung (21), Rapid.ai (20), and Hyperfine (13).

The Takeaway

The FDA’s new numbers on AI marketing authorizations show that the agency is keeping pace with rapid developments in the healthcare AI industry. Indeed, the FDA is even accelerating its pace of product approvals compared to its last update, with radiology still securing the lion’s share of authorizations.

SIIM 2026 Video Highlights

Posted on by Brian Casey

The annual meeting of the Society for Imaging Informatics in Medicine is always one of the highlights on the radiology calendar. SIIM 2026 was no exception, once again underscoring the vibrant community driving advances in imaging IT.

From radiology reporting to enterprise image management, SIIM 2026 highlighted the state of the art in imaging IT. We talked to many of radiology IT’s key opinion leaders in Pittsburgh, and we’re pleased to bring the discussions to you in this newsletter.

We hope you enjoy watching our SIIM 2026 video coverage as much as we enjoyed producing it! 

Check out the SIIM 2026 video links below or visit the Shows page on our website, and keep an eye out for our next Imaging Wire newsletter on Thursday.

– Brian Casey, Managing Editor

Top Trends from SIIM 2026

Posted on by Brian Casey

Last week’s SIIM 2026 conference demonstrated once again radiology’s ongoing evolution, from a discipline once known for big iron to one dominated by software. From radiology reporting to the evolving AI platform segment, below are the top seven trends from Pittsburgh. 

  • Reporting Stays Red Hot: Radiology reporting was the top theme from SIIM 2025, and the segment got even hotter with Microsoft’s decision to sunset its PowerScribe 360 radiology reporting software, which has drawn a host of new competitors into the segment. At SIIM 2026, a common theme was enterprise imaging companies adding reporting modules to their solutions.  
  • AI Adoption Moving Slowly But Surely: Adoption of radiology AI has been frustratingly slow, but it’s moving inexorably toward broader clinical use. At SIIM 2026, some 68% of the radiology-oriented papers focused on AI in some way, especially the new generation of foundation and vision language models that are enabling targeted AI algorithms to be developed more quickly than ever.
  • AI Governance Gets Real: Growing adoption of AI algorithms is creating a new issue: How to manage all this new technology. AI governance therefore was a major issue at SIIM 2026 as healthcare providers debated the legal and ethical necessity to better manage AI adoption, deployment, and utilization.
  • Other ‘Ologies Get into the Act: Radiology likes to think of SIIM as its own conference, but it also encompasses other ‘ologies that are moving into digital image management, like pathology and ophthalmology. At SIIM 2026, several imaging IT vendors showed integration with data from these disciplines, giving healthcare institutions a single source for their healthcare data management.
  • The Rise of All-in-One Vendors: A growing number of imaging IT vendors are rolling out solutions that combine image viewer, worklist, and reporting into a single platform, simplifying purchasing, deployment, and maintenance for radiology customers. Many of these firms seem to be getting traction with potential buyers, indicating the all-in-one concept could be one whose time has come.
  • Agentic AI Takes Shape: Agentic AI is a growing trend in radiology as algorithm developers build solutions to take on mundane tasks and free up radiologists to focus on their primary task: interpreting images. But the question is, will agentic AI work in the real world, or simply pile more technology on clinicians?
  • What Next for AI Platforms? Bayer’s withdrawal from the AI platform market by pulling its support for Blackford in 2025 raised many questions about the platform model that persisted at SIIM 2026. AI platforms seem to be evolving to add additional services like AI monitoring and governance.

The Takeaway

SIIM may not be radiology’s largest show, but for those in the imaging IT space it may be the most valuable one outside of RSNA. SIIM 2026 proved that point, with the top trends from Pittsburgh illustrating the discipline’s direction at the midpoint of the radiology year. For our overview of the top trends at SIIM 2026, check out our YouTube channel or the Shows tab on our webpage.

Radiology’s ‘Zombie Jobs’ Go Unfilled

Posted on by Brian Casey

Is radiology’s workforce shortage really just a matter of “geographic inconvenience?” A new report suggests that job shortages are mostly isolated to areas that radiologists consider to be less desirable geographically, where “zombie jobs” go unfilled for months.

The workforce shortage in radiology (and healthcare for that matter) has become a common refrain, especially since the COVID-19 pandemic. 

  • Exam volumes are rising steadily with an aging population, but the radiology workforce remains static. 

At least, that’s how the story goes. But a new study from RadBoard.io challenges that narrative, claiming that many radiologist openings are going unfilled because they are in geographically undesirable areas. 

  • RadBoard’s Kirill Lopatin analyzed 20.8k job postings in the U.S. over 78 days that represented 11k unique job ads (at least 47% of ads were just re-posts of the same role, but the number is probably much higher). 

The study found that of the postings with complete “lifespans” (from initial posting to deactivation)…

  • 25% were filled in less than one week, and another 12% in 7-14 days.
  • 28% took 31-60 days to fill.
  • 2.3% took 61-90 days (and maybe even longer).

So nearly one-third of radiology job ads were open longer than a month, leading RadBoard to conclude that radiology didn’t have a single fill rate for open positions – “it has two markets layered on top of each other.” 

  • RadBoard called radiologist job ads open for more than 60 days “zombie jobs,” with some markets having higher “stuck rates” as calculated by ads open longer than 60 days divided by total open job ads. 

States with the worst job markets by stuck rate included Nebraska (68%), Minnesota (41%), and Washington state (35%). 

  • At the other end of the spectrum were Florida (18%), Texas (15%), and New York (14%). 

This led RadBoard to conclude that the radiologist shortage was not a national problem – it was concentrated in areas where radiologists didn’t want to live.

  • Also, jobs in stuck markets paid $175k less than those in faster-cycling markets ($550k vs. $725k) – the opposite of what might be expected in a scarce market.

The Takeaway

The new numbers offer an eye-opening look at the narrative around the radiologist shortage, indicating that it may be more nuanced than previously thought. And the subtext to the data hints at the divide in U.S. healthcare between rural areas and metropolitan regions.

SNMMI 2026 News Highlights Theranostics Growth

Posted on by Brian Casey

The growing importance of theranostics was on display at this year’s annual meeting of the Society of Nuclear Medicine and Molecular Imaging in Los Angeles. New data on theranostics agents in development dominated the scientific sessions, while on the diagnostics side a proliferation of new PET radiotracers promises to go beyond FDG. 

The selection of SNMMI’s Image of the Year went to South Korean researchers for their work on the radiotracer 18F-GP1 PET/CT to identify acute lower extremity deep vein thrombosis.

  • In a study with 46 symptomatic patients, the tracer showed high diagnostic accuracy for detecting clots not only in the thigh but also in the calf, and had a high detection rate of pulmonary embolism occurring together with DVT. 

Meanwhile, the Abstract of the Year award was given to a study using PET to link brain metabolism patterns to the effectiveness of treatments for Alzheimer’s disease.

  • UCLA researchers performed FDG-PET brain scans on 124 patients being considered for anti-amyloid therapy. Those whose scans suggested Alzheimer’s disease and who got therapy had higher cognitive scores at one year compared to patients whose PET scans didn’t show evidence of Alzheimer’s. 

Other SNMMI 2026 highlights included…

  • FDG-PET/CT scans showed that patients who got bariatric surgery had metabolic changes across multiple organs that correlated with improved clinical markers. 
  • A new PET tracer, gallium-68 RCC78, was able to detect clear cell renal cell carcinoma and identified additional metastatic lesions missed by standard imaging.
  • A first-in-human study with a novel PET radiotracer, carbon-11 nevanimibe, was presented for imaging patients with overactive adrenal glands. 
  • In patients with metastatic neuroendocrine tumors, a new type of peptide receptor radionuclide therapy with actinium-225 DOTA-LM3 showed promise.
  • PSMA-PET scans showed that prostate cancer patients with just one to five bone metastases had much worse outcomes than patients with no metastases. 
  • A novel approach with two PET radiotracers during cancer treatment detected both tumor progression and cardiac inflammatory response. 
  • A novel PET tracer, fluorine-18 OXD-2314, showed promise for detecting chronic traumatic encephalopathy in living patients.
  • An AI algorithm using data from pre-therapy PET/CT scans predicted radiation dose in lutetium-177 PSMA treatment for prostate cancer.

The Takeaway

This year’s SNMMI 2026 highlighted the exciting evolution of theranostics, from a niche treatment used mostly when other therapies failed to a major step on the road to personalized medicine – and better patient care.

AI Reduces Mammography Workload

Posted on by Brian Casey

Using AI to triage low-risk breast screening exams that don’t need extra review could remove more than three-quarters of mammography cases from radiologists’ workload and allow them to spend more time on high-risk cases. That’s according to a new study in Radiology: Artificial Intelligence that confirms other recent studies. 

Much of recent mammography AI research has focused on its ability to triage low-risk cases to avoid additional radiologist review – saving precious personnel resources.

  • This is particularly valuable in Europe, which uses a double-reading paradigm in which two radiologists review all mammography cases (the U.S. employs single readers but tends to screen women annually rather than every two years). 

The new study comes from France, which employs a slightly different paradigm from the rest of Europe. Double reading is conducted only for lower-risk BI-RADS 1 and 2 cases, while BI-RADS 3-5 go directly to diagnostic workup. 

  • As such, double reading occurs with cases that have low cancer prevalence, which can make it more difficult for radiologists to detect cancers that don’t occur very often.

But what if you offloaded low-risk double reading to AI? 

  • In the new paper, researchers tried that with Therapixel’s MammoScreen AI algorithm, which was employed retrospectively to analyze mammograms from 42.4k women acquired from 2015 to 2019.

AI results were compared to standard radiologist double reading, with the following findings…

  • AI classified 77% of cases as low-risk, meaning these could be safely triaged from the double-reading paradigm.
  • AI missed only one cancer in the low-risk group, a rate the researchers characterized as “small but measurable.” 
  • Eleven cancers were found in the group AI classified as non-low-risk, which would have undergone double reading anyway in the AI triage paradigm.
  • Rates of interval cancer (cancer that occurs between screening rounds) were 5X higher in the cases AI classified as non-low-risk compared to low-risk (2.16 vs. 0.47 cancers per 1k exams). 

Using AI to classify and remove low-risk cases from double reading could therefore save significant resources from the French mammography screening program, with a “small but non-zero risk” of missed cancers.

The Takeaway
The new results track with findings from other recent studies that apply AI to mammography screening, particularly in Europe. While the French reading paradigm is unique, it’s instructive to see that AI maintains its ability to reduce radiologist workload across different types of breast cancer screening programs.

Does AI Still Scare Off Radiology Trainees?

Posted on by Brian Casey

Is AI still scaring off medical students from picking radiology as a specialty? A new study in Academic Radiology found that while prospective radiology trainees don’t seem as worried as they were after radiology AI burst onto the scene in 2015, they still have concerns about how AI will affect the profession. 

Radiology has long been seen as the medical specialty most at risk of broader AI adoption, largely because early AI applications focused primarily on image analysis.

  • These fears led to a widely publicized dip in radiology residency applications after 2016, the year after IBM Watson debuted at the RSNA show and when AI guru Geoffrey Hinton, PhD, issued his famous advice to stop training radiologists. 

But interest in radiology rebounded shortly after that. AI adoption was slower than anticipated, and few hospitals have proven willing to turn over radiologists’ duties to computers. 

  • Given the changes, how have the attitudes of medical students toward AI evolved in the last 10 years? Researchers decided to survey Canadian medical students and residents to find out.

In all, 401 respondents replied to the survey, of whom 13% had ranked radiology as their top specialty choice, with the following findings…

  • Only 2.5% said AI was “extremely influential” in affecting their specialty choice, with 57% saying it had a “slight/moderate impact” and 35% stated “no impact.”
  • AI was more important for those ranking radiology in their top three, with 91% saying AI influenced their decision compared to 54% of those uninterested in radiology. 
  • For those interested in radiology, 33% said AI made them feel discouraged, 13% were encouraged, and 33% reported no AI influence.
  • Those who believed AI would reduce radiologist demand were 50% less likely to be interested in a radiology career.

How to interpret the results? The authors felt the findings showed that AI had either no influence or a slight/moderate effect on specialty choice, but the impact was greater in those who were interested in radiology. 

  • They also saw a “growing polarization” among trainees, in that while many viewed AI as a threat to their job security, some saw it as an opportunity for innovation. 

The Takeaway

Medical students have complex and nuanced attitudes toward AI in radiology, as the new study indicates. But the findings suggest that past fears of radiology AI have evolved into a more measured view that better reflects real-world AI adoption.

AI for Chest X-Ray Varies

Posted on by Brian Casey

Not all AI is created equal when it comes to analyzing chest X-rays. A new study in Radiology found wide variation in performance for seven commercially available chest X-ray algorithms to detect lung cancer. 

X-ray is by far the most widely used imaging modality. Radiography is often the first imaging exam a patient receives, and it frequently serves as a gateway to other more advanced imaging modalities. 

  • But radiography also has well-known shortcomings (which is why advanced imaging is needed for follow-up). Could AI help unlock X-ray’s value and make it more useful?

That’s what a host of AI algorithm developers are banking on, but the wide variety of solutions can create confusion for clinicians.

  • So U.K. researchers decided to hold an AI bake-off, comparing commercially available algorithms from seven developers for detecting lung cancer on chest X-rays. 

The competing companies included Annalise/Harrison.ai, Gleamer, Infervision, Milvue, Oxipit, Qure.ai, and Rayscape. Researchers anonymized performance results from the different products.

In all, chest radiographs from a dataset of 5.2k patients with a real-world lung cancer prevalence rate were included, with researchers finding…

  • Significant variance in algorithm performance by each of the major accuracy measures: sensitivity (21%-78%), specificity (59%-98%), and positive predictive value (1.5%-28%). 
  • All the algorithms increased the number of false positives, and with significant variation. One model generated only 10 more false positives than radiologists, while another produced – wait for it – over 2k. 
  • If used to triage patients for follow-up CT exams, one model would generate $1.6k in additional costs while another would produce $327k.

What accounts for the variation? An underlying factor is most likely differences in the datasets used for model training. 

  • In any event, the study underscores the need for more head-to-head comparisons to determine the strengths and weaknesses of individual AI algorithms. 

The Takeaway

This week’s study on how AI performance varies between commercially available algorithms initially seems disturbing and might suggest a need for stronger regulatory oversight. But AI’s diversity could be its strength in a future where every patient case is analyzed by multiple different algorithms, each with its own advantages. This could ultimately produce a more complete picture of the patient than any one algorithm on its own.

Cochrane Pivots on Prostate Screening

Posted on by Brian Casey

Prostate cancer screening is getting new support from an unlikely source – the Cochrane group, which historically has been skeptical of population-based screening. Cochrane researchers last week published a new report supporting prostate screening, a sharp change from the group’s previous guidance. 

Prostate cancer screening hasn’t achieved the generally accepted status of other cancer screening tests like breast, cervical, colorectal, and lung. 

  • One of the main sticking points has been overdiagnosis. Prostate cancer can often be slow-growing, and many men live for years with prostate disease before dying of other causes.

But that dynamic has been changing in recent years, in large measure due to the ability of MRI to differentiate aggressive prostate cancer from more indolent disease. 

  • Clinicians are incorporating MRI into prostate screening protocols, using it to determine which men with elevated PSA levels should be biopsied and which ones can be followed with active surveillance. 

For its part, Cochrane is an international non-profit research consortium that periodically analyzes the peer-reviewed evidence behind new medical exams and technologies. 

  • But Cochrane’s work has occasionally been controversial: The group last month published a negative review of Alzheimer’s drugs that included treatments that never made it to market. Also, a Cochrane research center in Denmark for years was one of the most vociferous opponents of mammography screening. 

So that’s why last week’s statement on prostate screening is so surprising, especially given that Cochrane’s 2013 review found no evidence to support the claim that screening reduced prostate cancer mortality. 

In the new review, Cochrane analyzed data from six clinical trials in Europe and North America that included 800k men, finding that screening with PSA blood tests…

  • Detected 30% more prostate cancers overall, most at an early stage. 
  • Reduced the relative risk of a metastatic prostate cancer diagnosis by 35%.
  • Reduced prostate cancer mortality by 2 deaths for every 1k men screened (for comparison, mammography’s benefit is estimated to be 6-8 deaths). 
  • For every 1-2 deaths prevented, 36 additional cancers were diagnosed – a possible sign of overdiagnosis. 

What changed? Cochrane researchers said that we now have longer-term data that makes it easier to detect screening’s subtle mortality benefit.

  • They also cited the success of technologies like MRI in reducing unnecessary biopsies – and the harms of overdiagnosis.

The Takeaway

Last week’s news suggests that the ground is shifting under prostate cancer screening in favor of broader use of the exam, potentially with MRI follow-ups. If you can convince a screening-skeptical group like Cochrane of prostate screening’s value, you can convince anyone. 

AI for PE Detection: ‘Selective but Meaningful’

Posted on by Brian Casey

AI made a “selective but meaningful” contribution to radiologist interpretations of CT pulmonary angiography scans for pulmonary embolism. The study, published in Radiology: Artificial Intelligence, offers valuable insights into real-world implementation of AI on a large scale. 

One of the major criticisms of AI is that algorithms used in real-world clinical situations don’t perform as well as they do in the controlled environments that vendors use to acquire data for regulatory submissions.

  • AI performance can drop off as much as 20 to 30 percentage points for important metrics like sensitivity and specificity. 

The new study sought to investigate this phenomenon by analyzing a real-world implementation of Aidoc’s AI algorithm for PE detection. 

  • Researchers assessed the algorithm’s performance for analyzing CTPA exams across a variety of clinical environments in an integrated health network, including the emergency department and inpatient and outpatient settings. 

Scans of 29.5k patients acquired from 2021 to 2023 were included. AI analyzed images in real time, after which exams were interpreted by radiologists who knew the AI findings. Researchers found…

  • Radiologists using AI had higher sensitivity than the algorithm on its own (99% vs. 85%).
  • Specificity was more or less the same (99.8% vs. 99.5%).
  • Agreement between radiologists and AI was high (98%).
  • Agreement was higher when AI assessed cases as negative rather than positive (98% vs. 94%).
  • Radiologists disagreed with AI in 2.2% of cases. The final determination by a panel of expert thoracic radiologists strongly favored radiologists (89%).
  • Of the 3.3k cases positive for PE, 0.81% were detected only by AI – or 26 cases.

In analyzing the results, the researchers characterized AI’s contribution as “selective but meaningful.”

  • AI-positive results meant scans might require more scrutiny from radiologists, while an AI-negative call might be supportive – but not definitive – for negative PE.

The Takeaway

The new study of AI for PE detection is a fascinating look at real-world AI deployment. While the sensitivity, specificity, and agreement numbers are interesting, what draws our attention is the 26 PE cases caught only by AI over 18 months of use. That boils down to 26 patients whose clinical condition wasn’t missed, and 26 potential malpractice lawsuits that were never filed.

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