Author: Brian Casey

Top 6 Radiology Trends of 2024’s First Half

Posted on by Brian Casey

You can put the first half of 2024 in the books … and it was full of major developments for radiology. What follows are the top six trends in medical imaging – one for each month of the first half.

  • The Rise of AI for Breast Screening – The first half of 2024 saw the publication of studies conducted in Norway and Denmark that underlined the potential role of AI for breast screening, particularly for ruling out exams most likely to be normal. But research conducted within Europe’s paradigm of double-reading workflow for 2D mammograms may not be so relevant in the US, and more studies are needed.
  • Mammography Guideline Controversy – Changes to breast screening guidelines in both the US and Canada were first-half headlines. In the US, the USPSTF made official its proposal to lower to 40 the recommended age to start screening, but many were disappointed it failed to provide stronger guidance on dense breast screening. Things were even worse in Canada, where a federal task force declined to lower the screening age from 50 to 40. Canadian advocates have vowed to fight on at the provincial level. 
  • AI Funding Pullback Continues – The ongoing pullback in venture capital funding for AI developers continues. A study by Signify Research found that not only did VC funding fall 19% in 2023, but it got off to a slow start in 2024 as well. The new environment could be putting more pressure on AI firms to demonstrate ROI to both healthcare providers and investors, while also having broader implications – a major AI conference rescheduled a show that had been on the calendar for May, citing “market conditions.” On the positive side, Tempus AI’s IPO boomed, raising $412M. 
  • Opportunistic Screening Gains Steam – The concept of opportunistic screening – detecting pathology on medical images acquired for other indications – has been around for a while. But it’s only really started to catch on with the development of AI algorithms that can process thousands of images without a radiologist’s involvement. The first half of 2024 saw publication of several exciting studies for indications including detecting osteoporosis, scoring coronary artery calcifications, and predicting major adverse cardiac events. 
  • ChatGPT Frenzy Subsides – The frenzied interest in ChatGPT and other generative AI large language models seen throughout 2023 seemed to subside in the first half of 2024. A quick search of The Imaging Wire archives, for example, finds just four references to ChatGPT in the first six months of 2024 compared to 21 citations at the same point in 2023. LLM developers need to address major issues – from GenAI’s “hallucination effect” to potential misuse of the technology – before LLMs can be used in clinical settings.

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. In 2024’s back half, look for renewed attention on breast screening as the FDA’s density reporting rules go into effect in September, and keep on the lookout for signs that real-world AI adoption is growing, even as AI developers look for consolidation opportunities.

Top 4 Trends from SIIM 2024

Posted on by Brian Casey

SIIM 2024 concluded this weekend, and what a meeting it was. The radiology industry’s premier imaging IT show returned to National Harbor, MD, for the first time since 2018, where the Biosphere-like environment of the Gaylord National Resort and Convention Center offered a respite from the muggy weather outside. 

SIIM is always a great place to check in on new imaging IT technologies like PACS, AI, and enterprise imaging, and hot topics at SIIM 2024 included…  

  • AI Needs to Get Real (World): Research studies showing AI’s value are fine, but developers need to show that AI works in real-world settings before wider adoption will occur. Fortunately that’s started with landmark studies published recently for use cases like breast and osteoporosis screening. Meanwhile, scuttlebutt on the SIIM 2024 exhibit floor reinforced that start-ups are navigating an ugly funding environment, and many industry observers are predicting a wave of AI consolidation. 
  • Outlook Clears for the Cloud: Cloud-based imaging has struggled to catch on for years, but that’s starting to change as healthcare providers warm to the concept of letting third parties oversee their patient data. And there are signs that imaging IT vendors that were quick to develop cloud-based versions of their PACS software are reaping the rewards.
  • Enterprise Imaging Grows Up: This year’s meeting marked the 10-year anniversary of enterprise imaging, as dated from the start of the SIIM-HIMSS collaboration in 2014. The anniversary is a milestone worth observing, but it also raises questions about what the next 10 years will look like, and how AI and data from other -ologies will be integrated into enterprise networks. 
  • Cybersecurity Takes Priority: Several high-profile cybersecurity breaches at healthcare vendors and providers in the last year highlight that not enough is being done to keep patient data secure. Will migrating to the cloud help? Only time will tell.

The Takeaway

SIIM’s collegiality and coziness has always been a selling point for the meeting, even back in the days when it was known as SCAR. This year didn’t disappoint, as deals got done and relationships were built at the Gaylord National.  

Be sure to visit our YouTube channel and LinkedIn page to view our video interviews from the floor of the meeting – it was great seeing you all at the show!

Indies Surge in Imaging IT

Posted on by Brian Casey

The market for medical imaging IT technology continues to shift, with a pair of surging independent players growing rapidly in a sector that’s long been dominated by multinational OEMs. That’s according to the latest report on the imaging IT market by UK market intelligence firm Signify Research. 

The new report is projecting that the global market for imaging information technology will grow 18% over the next few years, from $5.6B in 2023 to $6.6B in 2028. 

  • Radiology will continue to dominate with a majority of sales, with cardiology IT a distant – but growing – second. Advanced visualization and operational workflow tools will make up the rest.

In terms of vendors, the top three market leaders of 2023 were GE HealthCare, Philips, and Fujifilm, but more recently, Visage Imaging and Sectra have been gaining market share. 

  • The report echoes recent news that has seen some of the largest multi-site enterprise imaging installations going to Visage and Sectra; a recent KLAS Research report also showed both companies’ growing momentum. 

Some of the other major points from the report include … 

  • Major growth in cloud deployment will occur – by 2028, 37% of the global imaging IT market will be in either hybrid or fully hosted environments
  • Cloud will represent 44% of the total radiology IT market by 2028
  • On a regional basis, the Middle East will see “significant growth” in imaging IT from 2024 to 2026, particularly in the Gulf States
  • Recovery is expected in China and the ASEAN nations, while India’s growing economy is driving healthcare digitization
  • Latin America is showing rising interest in AI and cloud technologies, but national elections could complicate matters

The Takeaway
The new Signify Research report underscores the evolving nature of the imaging IT market, as independent vendors rise to challenge multinational OEMs that dominated the sector for years. Be sure to check out Signify’s helpful infographic on LinkedIn that succinctly wraps up the changes.

AI Detects Incidental PE

Posted on by Brian Casey

In one of the most famous quotes about radiology and artificial intelligence, Curtis Langlotz, MD, PhD, once said that AI will not replace radiologists, but radiologists with AI will replace those without it. A new study in AJR illustrates his point, showing that radiologists using a commercially available AI algorithm had higher rates of detecting incidental pulmonary embolism on CT scans. 

AI is being applied to many clinical use cases in radiology, but one of the more promising is for detecting and triaging emergent conditions that might have escaped the radiologist’s attention on initial interpretations.

  • Pulmonary embolism is one such condition. PE can be life-threatening and occurs in 1.3-2.6% of routine contrast-enhanced CT exams, but radiologist miss rates range from 10-75% depending on patient population.

AI can help by automatically analyzing CT scans and alerting radiologists to PEs when they can be treated quickly; the FDA has authorized several algorithms for this clinical use. 

  • In the new paper, researchers conducted a prospective real-world study of Aidoc’s BriefCase for iPE Triage at the University of Alabama at Birmingham. 

Researchers tracked rates of PE detection in 4.3k patients before and after AI implementation in 2021, finding … 

  • Radiologists saw their sensitivity for PE detection go up after AI implementation (80% vs. 96%) 
  • Specificity was unchanged (99.1% vs. 99.9%, p=0.58)
  • The PE incidence rate went up (1.4% vs. 1.6%)
  • There was no statistically significant difference in report turnaround time before and after AI (65 vs. 78 minutes, p=0.26)

The study echoes findings from 2023, when researchers from UT Southwestern also used the Aidoc algorithm for PE detection, in that case finding that AI cut times for report turnaround and patient waits. 

The Takeaway

While studies showing AI’s value to radiologists are commonplace, many of them are performed under controlled conditions that don’t translate to the real world. The current study is significant because it shows that with AI, radiologists can achieve near-perfect detection of a potentially life-threatening condition without a negative impact on workflow.

US Tomo for Dense Breasts

Posted on by Brian Casey

What’s the best way to provide supplemental imaging when screening women with dense breasts? A new study this week in Radiology offers support for a newer method, whole-breast ultrasound tomography. 

It’s well-known by now that dense breast tissue presents challenges to traditional X-ray-based mammography.

  • In fact, mammography screening’s mortality reduction is far lower in women with dense breasts compared to nondense breasts (13% vs. 41%). 

A variety of alternative technologies have been developed to provide supplemental imaging for women with dense breasts, from handheld ultrasound to breast MRI to molecular breast imaging. 

  • One supplemental technology is whole-breast tomography, developed by Delphinus Medical Technologies; the firm’s SoftVue 3D system was approved by the FDA in 2021 as an adjunct to full-field digital mammography for screening women with dense breast tissue. 

With SoftVue, women lie prone on a table with the breast stabilized in a water-filled chamber that provides coupling of sound energy between the breast and a ring transducer that scans the entire breast in 2-4 minutes.

  • Unlike handheld ultrasound, the scanner provides volumetric coronal images that provide a better view of the fat-glandular interface, where many cancers are located.

SoftVue’s performance was analyzed by researchers from USC and the University of Chicago in a retrospective study funded by Delphinus. 

  • They performed SoftVue scans along with digital mammography on 140 women with dense breast tissue from 2017 to 2019; 36 of the women were eventually diagnosed with cancer. 

In all, 32 readers interpreted the scans, comparing the performance of FFDM with ultrasound tomography to FFDM alone, finding … 

  • Better performance with FFDM + ultrasound tomography (AUC=0.60 vs. 0.54)
  • An increase in sensitivity in women with mammograms graded as BI-RADS 4 (suspicious), (37% vs. 30%) 
  • No statistically significant difference in sensitivity in BI-RADS 3 cases (probably benign), (40% vs. 33%, p=0.08)
  • A mean of 3.3 more true-positive and 0.9 false-negative findings per reader with ultrasound tomography, a net gain of 2.4

The Takeaway

The findings indicate that ultrasound tomography could become a new supplementary tool for imaging women with dense breasts. They are also a shot in the arm for Delphinus, which as a smaller vendor has the challenge of competing with large multinational OEMs that also offer technologies for supplemental breast screening. 

Better Prostate MRI Tools

Posted on by Brian Casey

In past issues of The Imaging Wire, we’ve discussed some of the challenges to prostate cancer screening that have limited its wider adoption. But researchers continue to develop new tools for prostate imaging – particularly with MRI – that could flip the script. 

Three new studies were published in just the last week focusing on prostate MRI, two involving AI image analysis.

In a new study in The Lancet Oncology, researchers presented results from AI algorithms developed for the Prostate Imaging—Cancer Artificial Intelligence (PI-CAI) Challenge.

  • PI-CAI pitted teams from around the world in a competition to develop the best prostate AI algorithms, with results presented at recent RSNA and ECR conferences. 

Researchers measured the ensemble performance of top-performing PI-CAI algorithms for detecting clinically significant prostate cancer against 62 radiologists who used the PI-RADS system in a population of 400 cases, finding that AI …

  • Had performance superior to radiologists (AUROC=0.91 vs. 0.86)
  • Generated 50% fewer false-positive results
  • Detected 20% fewer low-grade cases 

Broader use of prostate AI could reduce inter-reader variability and need for experienced radiologists to diagnose prostate cancer.

In the next study, in the Journal of Urology, researchers tested Avenda Health’s Unfold AI cancer mapping algorithm to measure the extent of tumors by analyzing their margins on MRI scans, finding that compared to physicians, AI … 

  • Had higher accuracy for defining tumor margins compared to two manual methods (85% vs. 67% and 76%)
  • Reduced underestimations of cancer extent with a significantly higher negative margin rate (73% vs. 1.6%)

AI wasn’t used in the final study, but this one could be the most important of the three due to its potential economic impact on prostate MRI.

  • Canadian researchers in Radiology tested a biparametric prostate MRI protocol that avoids the use of gadolinium contrast against multiparametric contrast-based MRI for guiding prostate biopsy. 

They compared the protocols in 1.5k patients with prostate lesions undergoing biopsy, finding…

  • No statistically significant difference in PPV between bpMRI and mpMRI for all prostate cancer (55% vs. 56%, p=0.61) 
  • No difference for clinically significant prostate cancer (34% vs. 34%, p=0.97). 

They concluded that bpMRI offers lower costs and could improve access to prostate MRI by making the scans easier to perform.

The Takeaway

The advances in AI and MRI protocols shown in the new studies could easily be applied to prostate cancer screening, making it more economical, accessible, and clinically effective.  

Headlines from SNMMI 2024

Posted on by Brian Casey

SNMMI 2024 wrapped up this week in Toronto, Canada, with the conference once again demonstrating the utility of nuclear medicine and molecular imaging for applications ranging from neurology to oncology to therapeutics. 

An annual SNMMI highlight is always the Image of the Year designation, and this year’s meeting didn’t disappoint. 

  • The honor went to a set of ultra-high-resolution brain PET images acquired with United Imaging’s NeuroEXPLORER (NX) scanner, a PET/CT system that the company developed with Yale and UC Davis and introduced last year for research use (although a clinical introduction could be forthcoming). 

The NX system sports a cylindrical design with a 52.4cm diameter and long axial field-of-view of 49.5cm; in the talk presented at SNMMI, researchers compared it to high-resolution research tomograph images with tracers targeting different dopamine receptors and transporters.

  • Researchers said the NX system had “exceptional” resolution in cortex and subcortical structures, with “low noise and exquisite resolution,” and predicted NX would “dramatically expand the scope of brain PET studies.”

Other important presentations at SNMMI included papers finding … 

  • An AI algorithm developed at Johns Hopkins detected six different types of cancer and automatically quantified tumor burden on whole-body PET/CT scans
  • In a study of 10.5k patients, AI that analyzed SPECT/CT images was able to predict all-cause mortality with an AUC of 0.77 by using CT attenuation correction scans to calculate risk factors like coronary artery calcium
  • Cognitive training is less effective in older adults who have beta-amyloid deposits in the brain on PET scans
  • An ultra-low-dose PET protocol presented by researchers from Bern University Hospital in Switzerland and Siemens Healthineers used deep learning reconstruction for a 50X reduction in PET radiation dose, to 0.15 mSv
  • A gallium-68 FAPI-based PET radiotracer was more accurate than fluorine-18 FDG for systemic staging of newly diagnosed breast cancer
  • A new chelating agent that binds radiometals to the parts of molecules that target cancer reduced off-target toxicity in PSMA radiopharmaceutical therapy
  • A combination of alpha- and beta-radionuclide therapy that combined actinium-225 with lutetium-177 worked well for colorectal cancer in a preclinical study
  • Research sponsored by Novartis on radioligand therapy for prostate cancer with lutetium-177 PSMA-617 (Pluvicto) was chosen as Abstract of the Year

The Takeaway

This year’s SNMMI presentations highlight the exciting advances taking place in nuclear medicine and molecular imaging, with the rise of theranostics giving the field an entirely new wrinkle that places it even closer to the center of precision medicine. Perhaps a new letter – T – will need to be added to the conference before too long.

More Backing for CT Lung Screening

Posted on by Brian Casey

Yet another study is showing support for CT lung cancer screening. In a real-world study in Cancer, researchers tracked screening’s impact on military veterans, finding that it contributed to more early-stage diagnoses as well as lower all-cause mortality. 

It’s no secret that uptake of CT lung screening has been disappointing since the USPSTF in 2013 endorsed the test for high-risk people – mostly those with smoking histories. 

  • Uptake rates have been estimated to be under 10% by some studies, although recent research has shown that targeted interventions can improve that figure.

In the new study, researchers described results from the Veterans Health Administration’s effort to provide low-dose CT lung cancer screening to veterans from 2011 to 2018.

  • The researchers noted that smoking rates are higher among veterans, resulting in lung cancer incidence rates that are 76% higher than the general population. 

Researchers tracked outcomes retrospectively for 2.2k veterans who got screening before a lung cancer diagnosis and compared them to those with lung cancer who weren’t screened, finding that screening led to…

  • Higher rates of stage I diagnosis (52% vs. 27%)
  • Lower rates of stage IV diagnosis (11% vs. 32%)
  • Lower rates of cancer mortality (41% vs. 70%)
  • Lower rates of all-cause mortality (50% vs. 72%)

The sharp reduction in all-cause mortality is particularly striking. 

  • As we’ve discussed in the past, most population-based cancer screening tests have been shown to reduce cancer-specific deaths, but it’s been harder to show a decline in deaths from all causes. 

The study also illustrates the advantage of providing lung screening within a large, integrated healthcare system, where it’s easier to track at-risk individuals and direct them to screening if necessary.

The Takeaway

Of all the positive studies published so far this year on CT lung cancer screening, this one is the most exciting. The findings show that even in an environment of low lung screening uptake, dramatic benefits can be realized with the right approach.

Radiology’s Private-Practice Squeeze

Posted on by Brian Casey

It’s no secret that US radiology’s traditional private-practice model has been slowly fading away, but new numbers published in AJR illustrate the magnitude of the shift. The number of radiologist-affiliated and radiologist-only practices has dropped, even as the total number of US radiologists has gone up. 

Radiology has long prided itself on a cozy business model in which radiologists banded together as owner-operators of small private-practice groups that contracted their services with hospitals. 

  • This model has had many benefits for radiologists, but it’s begun to fray in the face of competitive threats like teleradiology providers, health system consolidation, and large national radiology groups like Radiology Partners.

Many radiologists have chosen to switch rather than fight, selling out to national groups or taking positions as employees within health systems.

  • Meanwhile, some practices that want to stay independent are finding strength in numbers by joining with other like-minded groups or seeking out multi-specialty medical groups. 

In the new study, researchers from the ACR’s Harvey L. Neiman Health Policy Institute analyzed CMS data from 2014 to 2023, tracking not only changes in the number of US radiologists but also their type of employment, finding …

  • The number of radiologists grew 17%, from 30.7k to 36k
  • But the number of radiologist-affiliated practices fell 15%, from 5.1k to 4.3k
  • The number of radiology-only practices fell 32%
  • The number of small radiology practices fell, with the decline varying by practice size: 1-2 radiologists -19%, 3-9 radiologists -34%, and 10-24 radiologists -25%
  • The number of large practices jumped, with the biggest increase – 349% – at very large practices (over 100 radiologists)
  • The mean number of radiologists per practice shot up 84%, from 9.7 to 17.9

Why the shift? The researchers theorized that much of it was driven by federal policy and reimbursement changes that incentivize consolidation, mostly to spread the risk and cost of compliance with various regulations like ACA and MACRA.

The Takeaway

There’s no question that radiology is changing – the question is what impact the changes will have on how radiologists perceive their work. The old guard may choose to rage against the dying of the light, while younger generations embrace the new model and its benefits for both professional careers and patient care. 

AI of Cardiac CT Predicts Risk

Posted on by Brian Casey

In a landmark study of 40k patients from the UK published in The Lancet, an AI-derived score that analyzed coronary arterial inflammation on coronary CT angiography scans was effective in predicting future cardiac risk in people regardless of whether they had obstructive coronary artery disease.

CCTA’s power for predicting heart problems has been demonstrated in multiple studies, and it’s now considered a first-line test for individuals with chest pain. 

  • But the situation is trickier in those without obstructive disease – prompting researchers to ask whether CCTA’s ability to visualize subtle changes in cardiac structure and function could be leveraged – such as with AI – to deliver even more prognostic power. 

The Oxford Risk Factors And Noninvasive imaging (ORFAN) study in the UK is addressing that question by conducting CCTA scans in 40k patients as part of routine clinical care at eight hospitals. 

  • Researchers analyzed outcomes in the entire ORFAN population of 40k patients, then followed a subset of 3.4k higher-risk patients for 7.7 years to study the value of a perivascular fat attenuation index (FAI) score. 

FAI scores measure heart inflammation in coronary arteries and are calculated using Caristo Diagnostics’ CaRi-Heart AI software.

  • The scores are combined with other traditional risk factors to create an AI-Risk classification that predicts the likelihood of an adverse event.  

Researchers found that … 

  • Across the entire 40k cohort, patients without obstructive CAD accounted for 64% of cardiac deaths and 66% of MACE – twice as many as those with obstructive CAD
  • In the smaller higher-risk cohort, patients with an elevated FAI score in all three coronary arteries had a higher risk of cardiac mortality (HR=29.8) or MACE (HR=12.6)
  • Elevated FAI scores in any coronary artery also predicted cardiac mortality
  • AI-Risk scores were associated with cardiac mortality (HR=6.75) and MACE (HR=4.68) when comparing very-high-risk versus low- or medium-risk patients 

The first data point is worth noting, as it illustrates the need to improve risk stratification and management in people without obstructive CAD.

The Takeaway
The ORFAN results are an exciting development for cardiac CT AI (in addition to being a major coup for Caristo, which raised $16.3M last year to commercialize CaRi-Heart globally). Measurements of coronary inflammation could give clinicians another tool – in addition to plaque measurements and calcium scoring – to predict cardiac events.

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!