Author: Brian Casey

Fine-Tuning AI for Breast Screening

Posted on by Brian Casey

AI has shown in research studies it can help radiologists interpret breast screening exams, but for routine clinical use many questions remain about the optimal AI parameters to catch the most cancers while generating the fewest callbacks. Fortunately, a massive new study out of Norway in Radiology: Artificial Intelligence provides some guidance. 

Recent research such as the MASAI trial has already demonstrated that AI can help reduce the number of screening mammograms radiologists have to review, and for many low-risk cases eliminate the need for double-reading, which is commonplace in Europe. 

  • But growing interest in breast screening AI is tempered by the field’s experience with computer-aided detection, which was introduced over 20 years ago but generated many false alarms that slowed radiologists down. 

Fast forward to 2024. The new generation of breast AI algorithms seems to have addressed CAD’s shortcomings, but it’s still not clear exactly how they can best be used. 

  • Researchers from Norway’s national breast screening program tested one mammography AI tool – Lunit’s Insight MMG – in a study with data obtained from 662k women screened with 2D mammography from 2004 to 2018. 

Researchers tested AI with a variety of specificity and sensitivity settings based on AI risk scores; in one scenario, 50% of the highest risk scores were classified as positive for cancer, while in another that threshold was set to 10%. The group found …

  • At the 50% cutoff, AI would correctly identify 99% of screen-detected cancers and 85% of interval cancers. 
  • At the 10% cutoff, AI would detect 92% of screen-detected cancers and 45% of interval cancers 
  • AI understandably performed better in identifying false-positive cases as negative at the 10% threshold than 50% (69% vs. 17%)
  • AI had a higher AUC than double-reading for screen-detected cancers (0.97 vs. 0.88)

How generalizable is the study? It’s worth noting that the research relied on AI of 2D mammography, which is prevalent in Europe (most mammography in the US employs DBT). In fact, Lunit is targeting the US with its recently cleared Insight DBT algorithm rather than Insight MMG. 

The Takeaway

As with MASAI, the new study offers an exciting look at AI’s potential for breast screening. Ultimately, it may turn out that there’s no single sensitivity and specificity threshold at which mammography AI should be set; instead, each breast imaging facility might choose the parameters they feel best suit the characteristics of their radiologists and patient population. 

Headwinds Slow AI Funding

Posted on by Brian Casey

Venture capital funding of medical imaging AI developers continues to slow. A new report from Signify Research shows that funding declined 19% in 2023, and is off to a slow start in 2024 as well. 

Signify tracks VC funding on an annual basis, and previous reports from the UK firm showed that AI investment peaked in 2021 and has been declining ever since. 

  • The report’s author, Signify analyst Ellie Baker, sees a variety of factors behind the decline, chief among them macroeconomic headwinds such as tighter access to capital due to higher interest rates. 

Total Funding Value Drops – Total funding for 2023 came in at $627M, down 19% from $771M in 2022. Funding hit a peak in 2021 at $1.1B.

Deal Volume Declines – The number of deals in 2023 fell to 35, down 30% from 50 the year before. Deal volume peaked in 2021 at 63. And 2024 isn’t off to a great start, with only five deals recorded in the first quarter. 

Deals Are Getting Bigger – Despite the declines, the average deal size grew last year, to $19M, up 23% versus $15M in 2022. 

HeartFlow Rules the Roost – HeartFlow raised the most in 2023, fueled by a massive $215M funding round in April 2023, while Cleerly held the crown in 2022.

US Funding Dominates – On a geographic basis, funding is shifting away from Europe (-46%) and Asia-Pacific (no 2023 deals) and back to the Americas, which generated over 70% of the funding raised last year. This may be due to the US providing faster technology uptake and more routes to reimbursement.

Early Bird Gets the Worm – Unlike past years in which later-stage funding dominated, 2024 has seen a shift to early-stage deals with seed funding and Series A rounds, such as AZmed’s $16M deal in February 2024. 

$100M Club Admits New Members – Signify’s exclusive “$100M Club” of AI developers has expanded to include Elucid and RapidAI. 

The Takeaway

Despite the funding drop, Signify still sees a healthy funding environment for AI developers ($627M is definitely a lot of money). That said, AI software developers are going to have to make a stronger case to investors regarding revenue potential and a path to ROI. 

USPSTF’s Mammography Letdown?

Posted on by Brian Casey

Last year’s relief that the USPSTF would lower its recommended starting age for breast screening to 40 gave way to frustration this week that the group did not go farther in its final decision on mammography recommendations. 

In a series of papers in JAMA journals this week, the USPSTF tackled a range of breast screening issues, from the age at which screening should start to whether modalities like ultrasound and MRI should be used to supplement conventional mammography.

That was the good news. The bad news is that breast screening advocates mostly got shut out on a variety of other issues, with the USPSTF … 

  • Advising that breast screening be conducted biennially (every two years), rather than annually as most women’s imaging advocates would prefer
  • Declining to raise the recommended upper limit for screening from 74 to 79
  • Declining to recommend supplemental screening with MRI or ultrasound for women with dense breast tissue, even as women express frustration with the lack of reimbursement for these exams

On the positive side, the USPSTF finally weighed in on DBT, stating that the 3D mammography technology is equivalent to digital mammography for breast screening. 

  • But in another disappointment, the group said it couldn’t find any studies stating that DBT was better than 2D digital mammography. 

Given the fierce battles that have been fought over screening guidelines in the last 15 years, what made the USPSTF change its mind on mammography’s starting age? 

  • One big factor is the 2% annual rise in breast cancer incidence in women in their 40s from 2015 to 2019; the higher mortality rates among Black women was another issue (see story below in The Wire).

The Takeaway

The USPSTF’s move to lower its recommended starting age for screening mammography is a welcome – if overdue – change for women, who for 15 years have borne the brunt of the group’s conservative approach to guideline formation. The question remains, is the USPSTF making the same mistake all over again when it comes to supplemental imaging and annual screening? And how long will women have to wait this time until it sees the light?

Doctors Work Harder for Less

Posted on by Brian Casey

Medicare reimbursement to physicians per beneficiary has declined over the last 16 years, with radiologists among the biggest losers. That’s according to a new study by the ACR’s Harvey L. Neiman Health Policy Institute, which confirms what many physicians already knew: they are working harder for less money.

It’s no secret that the US government has been struggling to rein in healthcare costs for decades. 

CMS has a number of tools at its disposal for controlling Medicare and Medicaid costs, one of which is the relative value unit (RVU) scale. 

  • RVUs – when multiplied by monetary conversion factors – basically set the amount of money the agency pays physicians per unit of work, with CMS typically reducing the conversion factor when it needs to cut Medicare spending. 

In the new study in the journal Inquiry, Neiman HPI researchers analyzed trends in RVU and conversion factor levels per Medicare beneficiary from 2005 to 2021, analyzing changes to calculate how much work providers have to do to deliver a unit of care. Findings included …

  • Reimbursement per Medicare beneficiary after inflation adjustment fell 2.3% for physicians as a whole
  • Radiology saw one of the biggest declines in MPFS reimbursement per beneficiary, ranking 31st on a list of 39 medical specialties, with a 25% decrease
  • Reimbursement has risen 207% for non-physician practitioners

What’s driving the declines? The Neiman HPI researchers identified the federal government’s budget neutrality rules for Medicare, which stipulate that increases in one area have to be offset by declines elsewhere.

The Takeaway

The new findings confirm what many physicians have suspected – they are not only working harder for less, but non-physician practitioners seem to be getting a bigger piece of the pie. Combined with a recent report showing that radiologist salaries didn’t keep pace with inflation in 2023, it’s not a pretty picture. 

Nuclear Medicine’s AI Uptake

Posted on by Brian Casey

Nuclear medicine is one of the more venerable medical imaging technologies. Artificial intelligence is one of the newest. How are the two getting on? That question is explored in new point-counterpoint articles in AJR. 

Nuclear medicine was an early adopter of computerized image processing, for tasks like image analysis, quantification, and segmentation, giving rise to a cottage industry of niche software developers.

  • But this early momentum hasn’t carried over into the AI age: on the FDA’s list of 694 cleared AI medical applications through July 2023, 76% of the listed devices are classified as radiology, while just four address nuclear medicine and PET.

In the AJR articles, the position that AI in nuclear medicine is more hype than reality is taken by Eliot Siegel, MD, and Michael Morris, MD, who note that software has already been developed for most of the image analysis tasks that nuclear medicine physicians need. 

  • At the same time, Siegel and Morris say the development of AI-type algorithms like convolutional neural networks and transformers has been “relatively slow” in nuclear medicine. 

Why the slow uptake? One big reason is the lack of publicly available nuclear medicine databases for algorithm training. 

  • Also, nuclear medicine’s emphasis on function rather than anatomical changes means fewer tasks requiring detection of subtle changes.

On the other side of the coin, Babak Saboury, MD, and Munir Ghesani, MD, take a more optimistic view of AI in nuclear medicine, particularly thanks to the booming growth in theranostics. 

  • New commercial AI applications to guide the therapeutic use of radiopharmaceuticals are being developed, and some have received FDA clearance. 

As for the data shortage, groups like SNMMI are collaborating with agencies and institutions to create registries – such as for theranostics – to help train algorithms. 

  • They note that advances are already underway for AI-enhanced applications such as improving image quality, decreasing radiation dose, reducing imaging time, quantifying disease, and aiding radiation therapy planning. 

The Takeaway
The AJR articles offer a fascinating perspective on an area of medical imaging that’s often overlooked. While nuclear medicine may never have the broad impact of anatomical-based modalities like MRI and CT, growth in exciting areas like theranostics suggest that it will attract AI developers to create solutions for delivering better patient care.

CT Changes Headache Workup

Posted on by Brian Casey

Recent studies have raised concerns about whether CT is overused in the emergency setting for patients with symptoms like headache, but a new study in JAMA Network Open suggests that higher CT utilization could be contributing to a decline in more invasive procedures like lumbar puncture. 

Earlier this month, we covered a study documenting the rapid rise of emergency head CT for patients presenting with acute-onset headache – which could be an indication of subarachnoid hemorrhage or other serious issues. 

  • Researchers theorized that higher CTA utilization could be a sign of overuse because the rate of positive findings over time fell 38%.

But the new study suggests that the growth in cerebral CTA use could have beneficial effects, by reducing the use of more invasive procedures and by detecting unruptured intracranial aneurysm. 

  • Some 5% of acute-onset headaches in emergency patients are caused by subarachnoid hemorrhage; these cases have a 50% risk of death or serious disability at one year, making accurate detection and workup a serious issue.

Researchers from Kaiser Permanente in Northern California analyzed 198k encounters for patients with headache at 21 community EDs from 2015 to 2021. 

  • They compared multiple workup protocols, ranging from CT only to others in which CT was used first, with nondiagnostic cases sent to either lumbar puncture or cerebral CTA. 

Dramatic changes occurred in headache workup over the study period, including … 

  • Overall use of CT grew at an average annual percent change of 5.4%
  • Cerebral CTA use grew 19% annually
  • Lumbar puncture use fell 11% annually
  • Detection of unruptured intracranial aneurysms grew 33%
  • The ratio of unruptured aneurysms to subarachnoid hemorrhage grew

The authors noted that the findings show clinicians are shifting away from a headache workup protocol that includes lumbar puncture to one that relies more on cerebral CTA.

  • The researchers were equivocal on the increase in detection of unruptured aneurysms; on the one hand, the absolute risk of rupture is low, but on the other, the consequences of rupture are severe.  

The Takeaway

The new study offers a more nuanced – and perhaps more positive – view of growing cerebral CTA use in the ED. In the end, it’s possible that two conflicting statements can be true: CT indeed may be overused in the emergency department, but its growing use is also having a beneficial impact on patient care.

Breast Screening Goes Green

Posted on by Brian Casey

Earth Day will be celebrated on April 22, and the event is a good opportunity to step back and take a look at medical imaging’s (not insignificant) contribution to climate change. Fortunately, a new paper in Health Policy details how one imaging service – breast screening – can be made more environmentally friendly. 

Previous studies have documented that medical imaging is a substantial contributor to greenhouse gas emissions, given the massive energy consumption required to keep all that big iron humming. 

  • Researchers have recommended a variety of solutions to reduce radiology’s environmental footprint, from powering equipment down overnight to switching to alternative energy sources to power medical facilities. 

The new study gets even more specific, analyzing the greenhouse emissions inherent in cancer screening – in particular patient travel – and offering ways to make it more planet-friendly. 

  • Researchers reviewed cancer screening programs in the Italian region of Tuscany, quantifying the CO2 emissions for different screening services. 

Greenhouse gas emissions could be cut dramatically by switching from a provider-centric model that requires patients to travel to centralized screening facilities to one in which mobile vans were sent into the field. Using model calculations for mammography screening, they found that in one district alone …

  • Breast screening was the most polluting cancer screening service, mostly because it had the highest number of invitees (3.4k women) traveling for screening
  • Institution-based breast screening generated CO2 emissions of 35,870 kgCO2-eq/km annually
  • Mobile breast screening had emissions of 805 kgCO2-eq/km – just 2.2% of emissions from site-based screening

The study is unique in that it views sustainability and environmental pollution as a healthcare issue that’s fully within the purview of providers to address. 

The Takeaway

The new study outlines a holistic approach to healthcare services that – right now – many US providers might believe is outside the scope of their operations. But as Earth Day approaches, it’s worth at least considering how in years to come healthcare could be delivered within a broader context of social and environmental stewardship.

Radiologist Pay Grows 3%

Posted on by Brian Casey

Annual salaries for US radiologists grew 3.1% in 2023 in the most recent physician salary report by Medscape. Although radiologist salaries are nearing the half-million-dollar mark, the increase actually represents a slowdown compared to 11% growth last year. 

The Medscape report shows that US radiologists had an average annual salary of $498k in 2023, compared to $483k in 2023 and $437k in 2022.

  • Radiologists ranked sixth on a list of 29 medical specialties, with orthopedic physicians topping the charts at $558k and diabetes and endocrinology doctors at the bottom at $256k. 

The report surveyed 7k practicing US physicians from October 2023 to January 2024, finding … 

  • 61% of those surveyed thought physicians overall are underpaid
  • 49% believe that they themselves are fairly paid
  • 83% said that pay was either no factor or only a minor factor in the specialty they chose, with only 14% saying it was the leading factor
  • Male specialists earned 31% more than female specialists, indicating a widening gender gap compared to 27% last year
  • 57% of radiologists were happy with their compensation, sixth among 29 medical specialties

Physician comments submitted to Medscape focused on the significant stress being experienced by many physicians, a phenomenon that’s been linked to burnout in other surveys. 

The Takeaway

Underlying this year’s Medscape data is an inconvenient truth for radiologists: their 3.1% pay gains for the year were not enough to keep pace with the US inflation rate of 3.4%. In that respect, radiologists find themselves in the same situation as most Americans.

RTs and Radiation Dose

Posted on by Brian Casey

There’s good news and bad news from a new study in Journal of Vascular and Interventional Radiology that tracks 40 years of occupational radiation dose to radiologic technologists who assist with fluoroscopically guided interventional procedures. The good news is that radiation dose is low and trending lower over time; the bad news is that dose to RTs can vary based on work setting. 

As we discussed last month, interventional radiology has delivered major benefits in patient care, replacing invasive surgery for many clinical applications. 

  • But the downside of interventional procedures is that they are performed for extended periods under fluoroscopy guidance, and more complex procedures are requiring longer times with the fluoro beam on – potentially leading to more radiation exposure. 

Researchers from the NIH wanted to investigate how changes in interventional use over the past 40 years affected occupational radiation dose exposure to RTs, while also looking at the impact of radiation exposure control methods. 

  • They reviewed records from 1980 to 2020, starting with RTs participating in the US Radiologic Technologists (USRT) research study who they then linked to data submitted to radiation dosimeter badge maker Landauer. 

In all, 19.7k RTs who reported assisting with fluoroscopically guided interventional procedures over the study period were included, with researchers finding … 

  • Median annual radiation dose of 0.65 mSv, well below the occupational limit of 20 mSv
  • Median doses were highest in the 1980s and decreased over time, reflecting greater awareness of patient radiation dose and better radiation protection gear
  • A second peak in radiation dose happened from 1999-2011, most likely due to more sensitive dosimeters
  • RTs who worked closer to patients (<3 feet) had higher median annual dose, at 1.20 mSv
  • RTs who reported assisting with ≥ 20 procedures per month had higher dose, at 0.75 mSv

The researchers concluded that their findings show that radiation dose control measures are working, and better radiation dosimetry technology offers a far more accurate picture of how much dose RTs are actually exposed to. 

The Takeaway

The study’s findings should give technologists who assist with interventional procedures peace of mind that their radiation dose exposure is well within established limits. But as always with radiation exposure, vigilance is warranted. 

MRI Makes Prostate Screening More Precise

Posted on by Brian Casey

Prostate cancer screening isn’t a guideline-directed screening test yet, but this could change with the use of MRI and other tools. A series of papers published in several JAMA journals late last week indicates the progress that’s being made. 

As we’ve discussed in previous issues, prostate screening with PSA tests hasn’t met the threshold for clinical benefit achieved by other population-based screening exams.

  • PSA-based screening has been characterized by lower mortality benefits and relatively high rates of overdiagnosis and complications from follow-up procedures. 

But some researchers believe that PSA screening could be made more effective by using additional diagnostic tools like imaging and blood tests to focus on potentially high-risk disease for biopsy while active surveillance is used for less threatening prostate lesions. 

In the ProScreen trial in Finland, researchers tested the combination of PSA, a kallikrein four-panel blood test, and MRI in selecting patients for biopsy. 

  • Patients were sent to MRI if they had PSA scores of 3.0 ng/mL or higher and kallikrein scores of 7.5% or higher; those with abnormal MRI scans got targeted biopsy. 

The researchers tested the ProScreen protocol in a study of 61.2k men, with 15.3k invited to screening and 7.7k getting screened. Over a preliminary three-year follow-up period, researchers found …

  • 9.7% of men met the PSA threshold for a suspicious lesion; this fell to 6.8% after the kallikrein test and 2.7% after MRI, illustrating the protocol’s ability to reduce biopsies
  • Biopsy yield for high-grade cancer was 1.7%, which an editorial called a “remarkably high yield”
  • Overdetection of low-grade disease was 0.4%, compared to 3.2% in a comparable previous study

In a second study, this one in JAMA Oncology, researchers performed a meta-analysis of 80.1k men from 12 studies in which MRI was used to direct patients to prostate biopsy after PSA testing, finding that MRI-directed protocols had …

  • Higher odds of detecting clinically significant prostate cancer (OR=4.15) compared to PSA screening alone
  • Lower odds ratio for biopsy (OR=0.28)
  • Lower odds ratio for detecting clinically insignificant cancer (OR=0.34)

Finally, a secondary analysis in JAMA of a large UK trial illustrates the challenges of prostate screening without MRI guidance. Researchers reviewed 15-year outcomes of the Cluster Randomized Trial of PSA Testing for Prostate Cancer (CAP), a study of 415k men,196k of whom were screened from 2002 to 2009 without the use of MRI, finding … 

  • PSA screening increased detection of low-grade cancer (2.2% vs. 1.6%) but not intermediate or high-grade disease
  • Screening reduced prostate cancer mortality by a small amount (0.69% vs. 0.78%)

The Takeaway

Taken together, new studies offer a roadmap toward making MRI an integral part of prostate screening, such that perhaps in years to come it can join other cancer tests as a population-based screening tool.

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!