Private equity deals in healthcare hit nearly $90 billion in 2022, the second largest annual tally on record, Bain & Company reported Monday.

The number represents a drop from $150 billion invested in the industry in 2021 but is still $10 billion more than any other year on record. Investors continue to pour funds into radiology and other medical markets, seeking shelter from the volatility seen in other market segments.

“Healthcare private equity has earned a recession-proof reputation, typically outperforming overall private equity activity during economic downturns,” Kara Murphy, co-lead of healthcare private equity at Bain & Company, said in a statement. “While the space is resilient, investors will face continued challenges ahead as interest rates and labor costs continue to climb and credit continues to be tight.”

A recent report from Bloomberg highlighted some of the headwinds facing radiology and other specialties. Five companies in the healthcare space defaulted on their loans last year (compared to a historic average of one), while 33 saw their credit ratings cut by S&P. Standard & Poor’s revised its outlook for private equity-backed industry giant Radiology Partners to “negative” back in December, citing a tight labor market, delayed cash collections due to pressure from payers and “persistent” negative free cash flows. Moody’s, meanwhile, downgraded Rad Partners in November while noting some of the same factors.

“Notwithstanding the economic and policy headwinds that radiology practices face and must be addressed, we remain committed to adapting to the current environment by reducing debt and strengthening our balance sheet through organic EBITDA growth,” an RP spokesperson told Bloomberg at the time.

Specialties with favorable payer mixes, attractive consumer demographics and limited reliance on patient financing are most likely to remain insulated and able to maintain margins, the Bain & Company report noted. Radiology, alongside oral surgery and vision care, has historically been less affected in economic downturns, “making them interesting investment opportunities this go around.”

In the report, Murphy and co-authors highlighted two radiology-related private equity deals executed in Europe. Those included GBL inking a deal to acquire a majority stake in Affidea, one of the continent’s leading providers of imaging services. And EQT signed an agreement to acquire Meine Radiologie and Blikk, which together operate more than 65 radiology and radiotherapy locations in Germany.

Monday’s Bain & Company report also mentioned artificial intelligence as one “sector to watch” in 2023, given recent breakthroughs.

“2022 was a monumental year for generative artificial intelligence, with new services emerging in imaging and text generation,” the report noted. “Use cases for generative AI are just emerging. Stakeholders are watching closely and are ready to adapt when the time is right,” it added.

See the full report from Bain & Company for free here.

Advanced artificial intelligence (AI) models can help cardiologists identify patients who may require a permanent pacemaker (PPM) after transcatheter aortic valve replacement (TAVR), according to new findings published in World Journal of Cardiology.[1]

PPM implantation, the study’s authors explained, remains a relatively common complication after TAVR.

“PPM is associated with increased length of hospital stay and mortality,” wrote first author Pradyumna Agasthi, MD, a cardiologist at Mayo Clinic in Phoenix, and colleagues. “Additionally, advanced conduction defects requiring PPM implantation have been demonstrated to lead to worse functional capacity and clinical outcomes in patients with aortic stenosis (AS). The PPM requirement rate in TAVR is two to five-fold higher than in surgical aortic valve replacement.”

Agasthi et al. examined data from 1,200 patients who underwent TAVR due to severe symptomatic AS from 2012 to 2017. After excluding 236 patients who already had a pacemaker prior to undergoing TAVR, the group used supervised machine learning to run a PPM risk prediction analysis on the remaining 964 patients. The post-TAVR PPM rate among this cohort was 19.6% after 30 days and 26.7% after one year, numbers considered “similar to previous trials.”

The AI model used gradient boosting to focus on certain patient characteristics—including age and the presence of right bundle branch block, for example—associated with increased PPM rates.

Overall, the team’s AI model was found to be superior to an existing PPM risk score. The AI model’s area under the ROC curve (AUC) for predicting a patient’s 30-day and one-year risk of PPM were 0.66 and 0.72, respectively. The existing PPM risk score, meanwhile, had AUCs of 0.55 for 30-day risk and 0.56 for one-year risk.

Also, the team’s research found that “brachiocephalic to annulus distance-to-height ratio” was the single most significant predictor for post-TAVR PPM implantation, a finding that “has not previously been described in the literature.”

“This was primarily a feasibility study and is retrospective in nature, which restricts our ability for defining causal associations,” the authors added. “There is a need for prospective validation with an external cohort.”

Read the full analysis here. 

Experts at the University of Manchester in the United Kingdom recently developed and trained a deep learning model capable of estimating breast density. 

Nearly 160,000 full-field digital mammography images with assigned density values were used to develop a procedure that estimates a density score for each image [1]. When tested, the model achieved a performance comparable to that of human experts, lead researcher Professor Susan M. Astley noted. 

“The advantage of the deep learning-based approach is that it enables automatic feature extraction from the data itself,” explained Astley. “This is appealing for breast density estimations since we do not completely understand why subjective expert judgments outperform other methods.” 

Instead of building the model from the ground up, the team used two pretrained deep learning models that had been previously trained on a network of more than 1 million images—a process known as transfer learning. This enabled the team to train the new model with less data. The team’s goal was to be able to feed the model a mammogram image as input and have it produce a density score as output. 

This involved preprocessing images to make the task less computationally intensive, the experts explained. The deep learning models would then extract features from the processed images and map them out with a set of density scores. The models’ scores were then combined to give a final estimate of breast density. 

Not only did the final product offer accurate breast density assessments, but it also required less computation time/resources and memory. This approach could make future model development a less time intensive process that requires significantly less data, the team explained. 

“... we have demonstrated that using a transfer learning approach with deep features results in accurate breast density predictions,” the authors wrote. “This approach is computationally fast and cheap, which can enable more analysis to be done and smaller datasets to be used.” 

Of note, the two pretrained models used in this work were trained on a nonmedical data set. Model performance would likely show improvement with the inclusion of medical images, the team suggested. 

The study is available in the Journal of Medical Imaging. 

Having identified an “urgent” need for guardrails to keep healthcare AI from veering into an avoidable ditch, the Coalition for Health AI has put together a 24-page guide applicable to numerous groups of stakeholders.

CHAI released its “Blueprint for Trustworthy AI Implementation Guidance and Assurance for Healthcare” April 4.

The document is intended to steer the technology along an open-ended journey in which a focus on “fairness, ethics and equity” positions tech-enabled medical progress to remain accessible, or become so, for all populations.

Or, as the authors of the paper put it, their mission is to help U.S. healthcare navigate “an ever-evolving landscape of health AI tools to ensure high-quality care, increase trustworthiness among the healthcare community, and meet the needs of patients and providers.”

‘Need for a common, agreed-upon set of principles’

The CHAI blueprint fleshes out attributes healthcare AI algorithms, software and systems should possess before being implemented in clinical settings.

Along with safety and efficacy, these include usefulness, reliability, testability and ease of use.

Further, the blueprint maintains, adopted AI iterations should be explainable and interpretable, fair (“with harmful bias managed”), secure and resilient, and privacy-enhanced.

In a section on next steps, the authors write:

Each healthcare institution may use different kinds of AI tools. However, there is a need to use a common, agreed-upon set of principles to build them and facilitate their use. Through an assurance lab, health systems as well as tool developers and vendors can submit processes and tools for evaluation to ensure readiness to employ AI tools in a way that benefits patients, is equitable and promotes the ethical use of AI.”

The authors note that large medical centers may already have such measures in place.

Access to trustworthy health AI shouldn’t depend on patient location

In any case, institutions of any size may do well to form or refresh an advisory committee to “advance the field and ensure equity so that, for a given patient, access to trustworthy health AI would not depend on where they live or with which health system they are interacting.”

The authors state the document builds on and aligns with the White House Office of Science and Technology Policy’s “Blueprint for an AI Bill of Rights” and the National Institute of Standards and Technology’s “AI Risk Management Framework.”

Stakeholder groups the CHAI blueprint considers within its target audience include data scientists, informaticists, software engineers, vendors, end users, patients, professional societies that publish clinical practice guidelines, hospital and health-system leadership, researchers and research funders, educators and medical trainees.

CHAI says the guide incorporates input from founding members of the coalition itself. These include representatives of academia, healthcare and industry who work with governmental observers from AHRQ, CMS, FDA, ONC, NIH and the White House Office of Science and Technology Policy.

Read the full blueprint here and a CHAI news item on it here.

The European Congress of Radiology (ECR) 2023 meeting in March saw an attendance increase of 14% over ECR 2022, but it fell short of the pre-pandemic ECR headcount of just over 23,000. This increase with in-person attendance was similar to what was seen at the Radiological Society of North America (RSNA) meeting last fall. This also failed to reach pre-pandemic numbers but was a clear sign of a gradual return to normalcy. ECR had 17,262 participants from 122 countries this year. 

“After all the disruptions to lives in recent years, it was wonderful to see the halls and rooms of the Austria Center crowded with busy, enthusiastic, happy attendees, enjoying all the educational and social opportunities provided by ECR 2023," said ECR 2023 Congress President Professor Adrian Brady, Mercy University Hospital, Cork Ireland, in a statement. "Onsite feedback from Congress participants was very positive, and watching attendees, I sensed that the future of radiology is in excellent hands, with young colleagues ready to continue the success and constantly-increasing clinical relevance of and delivery of value by our specialty."

Observations from the ECR show floor

The ECR industry exhibition featured more than 250 medical imaging vendors. Bhvita Jani, research manager at the healthcare market analysis firm Signify Research, attended the meeting and offered several takeaway observations.

"There was notable presence of new exhibitors on the show floor, such as Chinese vendor United Imaging, as well as an absence of conventional exhibitors, such as Shimadzu and Hologic," she said. Speculation loomed that the disappointment in attendance, as well as the revised layout from ECR 2022, could have resulted in some reluctancy from vendors to commit to taking up large booths, especially as economic headwinds continue to challenge the medical imaging market," she Jani said.

Environmental sustainability is a growing trend among radiology vendors

The global healthcare market contributes 4% of total CO2 emissions, with imaging a substantial contributor, Jani explained. 

"With healthcare providers increasingly concerned about the energy crisis and environmentally sustainability, many medical imaging vendors focused their efforts on how best to decarbonize radiology," she said. 

Rising energy prices, which increase operational costs for healthcare providers, resulted in modality vendors wanting to showcase new sustainability features on their imaging equipment. For example, energy saving modes on large new imaging modalities, when the systems are idle between examinations, could save up to 40% of energy consumption per scan compared to the existing fleet of installed systems, she explained. Moreover, new research initiatives were announced, including “metering” CT systems in the U.K., thereby better understanding real-world application and identifying where energy consumption could be improved.

Other themes included continued demonstrations of how healthcare providers can benefit from long-term value of their investments such as smart subscriptions, or lifetime value through long-term managed contracts, in which core system components may not require replacement. New MRI technology at the show touted up to a 70% reduction in helium consumption, Jani said. She added MRI vendors also focused on ensuring the long-term viability of coils and re-use of MRI magnets when providers upgrade to new systems. New manufacturing initiatives were also highlighted, with vendors showcasing green credentials in terms of production of equipment, design, materials and carbon-footprint.

Connected systems enable remote servicing of medical imaging equipment

On the back of a frenetic period during COVID-19, many healthcare providers have had limited time to focus on modernization strategies for radiology services, especially as pent-up demand and staff-shortages have put huge pressure on providers. Medical imaging vendors at the ECR tried to position themselves as long-term, collaborative partners, whether this be through their service offerings, ongoing training services, or strategic consulting support, Jani said.

A key theme tied to this partnership approach was “connected systems”. One foundational facet of this was supporting the ongoing shift to remote servicing of modalities. COVID accelerated the number of servicing cases done remotely, with an estimated 40% of servicing performed remotely today. 

Analytics software was also shown by vendors to provides real-time monitoring of modality and staff performance across multiple sites. Jani said this is intended to help imaging service stakeholders better manage their fleets and ensure increasingly overstretched resources are deployed most effectively. Some systems included predictive tools, helping imaging service-line managers to support operational planning and better react to peaks of demand. Analytics also can help optimization by identifying patterns causing patient “no show” patients. 

AI creates smarter radiology workflows

Jani said artificial intelligence (AI) continues to be integrated across the entire workflow, from examination preparation to acquisition and from reconstruction and post-processing. While much has been debated over the role of AI in post-processing and image analysis, which was also on show in the dedicated AI showcase and forum, AI was also a headline at ECR in terms of supporting and improving imaging acquisition.

Demonstrations at ECR 2023 included AI being used to track patient breathing and movement within the bore or gantry itself, use of AI to reduce scatter in X-ray imaging, and automation of patient positioning to provide consistent outcomes and deep learning reconstruction, Jani observed. 

AI is being used by modality vendors to improve the quality and speed of imaging services, reducing the need for re-scans and improving the quality and safety of imaging, especially for scans requiring radiation or contrast agents. AI is increasingly being viewed by modality vendors as an important differentiator alongside the traditional hardware-based innovations, Jani said.

In particular, protocol management solutions to harmonize protocols across a fleet of systems was a prominent theme. “One-click” scan set up was also a widely touted feature of new systems. 

"Typically, setting up CT and MRI systems for scans, especially more complex protocols, can take substantial technician time, sometimes as much as 15-20 minutes per scan. Leveraging information from the RIS and automated patient positioning technology, new modality workstation and on-gantry software can support much faster set-up, in many cases down to a single click. This not only saves time technicians time, but also allows scheduling scan appointments closer together, providing a tangible improvement in scan-throughput for a provider," Jani explained.

Retaining and supporting the radiology workforce healthcare providers against the backdrop of staff shortages was also a big theme for the use AI. This included examples of assistance in caseload prioritization through embedded AI at the point of care, one-click examinations or by AI powered smart protocoling. Jani said intelligent systems help limit the reliance on the operator experience or expertise.

What does the future hold for radiology?

As well as demonstration of the above across imaging product lines, medical imaging vendors are predicted to further showcase how their systems can be even more efficient, ergonomic, and intelligent this year and into RSNA, Jani said.  

"AI is expected to not only proliferate in diagnostic imaging, but also in image-guided therapy applications, such as in interventional cardiology. Within the computed tomography market, photon-counting CT is expected to have strengthened further clinical validation and evidence with growing commercial traction and more installs," Jani said.

She also expects to see an increased focus from healthcare providers and vendors on screening, exemplified by the growing focus at ECR 2023. Lung screening in particular is gathering momentum in Europe, which will have substantial impact on X-ray, CT and AI industry segments in the mid-term future Jani said. Additionally, new research supporting the use of imaging and the potential of AI deployment in screening programs to not only support earlier detection of diseases, but also highlight incidental findings and early biomarkers of disease, will go a long way to reshaping the medical imaging industry in the long-term future. 

HeartFlow, the California-based healthcare technology company known for its FFR-CT technology, has announced the completion of a Series F funding round worth $215 million by its parent company, HeartFlow Holding.

The funding round was led by Bain Capital Life Sciences. A new investor, Janus Henderson Investors, also participated, as did Baillie Gifford, Capricorn Investment Group, Hayfin Capital Management, HealthCor, Martis Capital, USVP and Wellington Management.

HeartFlow plans to use the new financing to help keep up with its rapid growth and expand sales of its artificial intelligence-powered solutions, including RoadMap Analysis and Plaque Analysis, which were both cleared by the U.S. Food and Drug Administration in October 2022.

“The oversubscription of our Series F funding round, particularly in the current market backdrop, is a strong validation of our technology, our team and the opportunity in front of us,” HeartFlow CEO John Farquhar said in a prepared statement. “We appreciate the support of our investors, both existing and new, who share HeartFlow's vision to build a new standard of care for people at risk of heart disease.”

“HeartFlow is a leader in precision heart care and its AI-enabled products promise to help physicians more effectively diagnose and treat heart disease, which continues to be the leading cause of death in the U.S.,” added Nicholas Downing, MD, a managing director at Bain Capital Life Sciences, in the same statement. “We look forward to supporting the company’s commitment to improving cardiovascular care for patients as it heads into this exciting next chapter of growth.”

HeartFlow’s growth reached new heights after its FFR-CT technology was included in the 2021 American College of Cardiology/American Heart Association chest pain guidelines. Watch this video from ACC.22 for additional context.

Radiology by far has the largest penetration of artificial intelligence (AI) in medicine, but with minimal or no reimbursement, hospital administrators may ask where the return on investment (ROI) is to justify AI spending. Radiology Business spoke to several key opinion leaders to find out their thoughts on the value of AI.

The U.S. Food and Drug Administration (FDA) has now cleared well over 500 market-cleared AI medical algorithms as of January 2023. The vast majority of these are related to medical imaging. This includes 396 radiology algorithms. Cardiology has the second largest number of approvals at 58, many of which also are specific to cardiac imaging. 

"We have seen from the U.S. FDA the number of cleared algorithms available for commercial use are escalating," explained Bibb Allen, MD, FACR, chief medical officer of the American College of Radiology (ACR) Data Science Institute and a past ACR president. But, he said, the adoption rate has been very slow. This is partly due to hospitals wanting assurances that the AI will work as intended and has clear evidence it will improve workflow, reduce costs and improve patient care. There also is a lack of reimbursement to cover the cost of adopting AI.

"I think finding the right AI tools and the right value proposition for the institutions is the main thing," Allen said. "On the payment and policy side, there are two reasonable arguments. As a radiologist, I believe it will provide us with safer and more effective care, whether that comes from decreases in turnaround times or decision support capabilities, or the identification of critical findings. On the other hand, the payer might say 'well wait a minute, we are already paying you, the radiologist and the expert, to make those same conclusions, so why should we on a fee for service basis provide additional payment?' And I think that is going to be a struggle for health policy makers."

Should AI gain reimbursement or is it just a practice expense?

The decision to adopt AI is not really up to the radiologist. Administrators or entire group practices must first figure out if AI adoption is worth the cost, explained Keith J. Dreyer, DO, PhD, chief science officer of the American College of Radiology (ACR)  Data Science Institute. Often, that decision lies with hospital or health-system executives who do not understand the day-to-day work of a radiologist. 

"There are not a lot of use cases today that have proven to a payer that AI is of value such that they are willing to reimburse the clinician who uses it. There is no reimbursement for PACS either, or for radiology information systems (RIS) or electronic health records (EHR), it's just a practice expense," Dreyer said. "So thinking of AI as a full category of its own, but just as a technology, then the question is does this technology make you more efficient? Does it bring is some new action that can improve care or make it faster so it is worth paying for it?"

Dreyer said the lack of reimbursement for AI might also dissuade health systems from spending more to purchase the technology. However, health systems need to consider if the technology can help enhance the workflow of radiologists to make them more efficient, or if AI could help improve patient outcomes. 

"Is AI saving time for the radiologist such that you make them more efficient and it makes the AI worthy of purchasing?" Dreyer said. "I don't see any published evidence that would demonstrate that as 100% yes. I hear anecdotal stories and I have seen some small studies that have shown this, but I have not seen by and larger that this is the case. But, this will be on a case-by-case basis, because it is not just the use case as to the accuracy of the algorithm, it is also how it was implement that it can make you faster or slower."

AI may help address the growing shortage of radiologists

The role of AI is becoming more important as the U.S. faces a growing shortage of radiologists, explained Charles E. Kahn, Jr., MD MS, Editor of the the Radiological Society of North America (RSNA) journal  Radiology: Artificial Intelligence, and professor and vice chair of radiology at the University of Pennsylvania Perelman School of Medicine. He said the technology can help augment radiologists to improve their workflow and make them more efficient. As an example, he said AI and do a first pass read to triage exams into suspected normal and suspected disease. This helps allow the radiologist  to concentrate on reading cases with suspected disease or more complex cases. 

This augmentation to help offset the shortage of radiologists was echoed by Ed Nicol, MD, consultant cardiologist, cardiac imager and honorary senior clinical lecturer with Kings College London. He is also president-elect of the Society of Cardiovascular Computed Tomography (SCCT). He said AI to do the simple but time consuming tasks and to help prioritize exams based on complexity is worth the cost to get more of what really matters from the human exam readers.

"We don't pay radiologists or cardiologists to draw lines around things, my 7 and 10 year olds can probably do as good a job on that. We also have AI systems that can do a first read and determine the first 10 cardiac CT scans are normal, and these 12 are abnormal. So you can get a cup of coffee and blast through the normals and then determine when you are fresh, or in the morning, go through the difficult ones. These technologies exist already, but we are not leveraging them. What we are really pay a radiologist or cardiologist for is to put the findings into context," Nicol explained. 

What are the value propositions for radiology AI?

Allen outlined several key AI value propositions that hospitals should be considering. 

"AI models can find things that radiologists cannot find, or we as radiologists just can't see," Allen said. This can includes figuring out a phenotype of a brain tumor using radiomics so the appropriate therapy can be chosen.

Population health is another area where AI can sift through vast amounts of imaging data to identify patients with key incidental findings for things like pulmonary embolism, coronary disease, pulmonary emphysema and hepatic steatosis. 

"All of these things no one is going to care about when a patient is in the emergency department for diverticulitis. They are going to get antibiotics, and that is really the end of their episode of care," Allen described. "And the fact that they have hepatic steatosis, or they have coronary artery calcification, and they are only 40-years-old gets lost, even if we say it it just gets buried in the report or it does not make it to the problem list for follow-up. So this opportunistic AI screening for population health has a great chance for ROI."

For AI that can act as a second set of eyes for radiologists, that might be an area where the radiologists figure out what the value is on that type of algorithm for themselves. Even AI that can help detect cancers, lung nodules or other conditions in the non-acute setting may have benefits to prevent missing things in scans, or to act as a second opinion. As a second set of eyes for radiologists, AI can double check datasets to make sure nothing was overlooked, or to get a second opinion about a questionable area of an image. AI might also help detect rare conditions a radiologist might only come across a few times in their career. 

"If we have that second set of eyes that helps us find more breast cancers, maybe that is valuable enough to us to give us that extra peace of mind," he said. Valuable enough to invest in the technology, even if there is no reimbursement. 

But, opportunity for AI may also rest in conversion from a fee-for-service model to a value-based payment model.

"You can imagine in a value-based payment system, any tool that makes you more efficient, if you are getting paid the same, you can pay a little more for AI to become more efficient and it helps your bottom line," Allen explained.

AI that can automatically identify a pneumothorax on a mobile digital X-ray system, or alert clinical care teams of a suspected stroke, pulmonary embolism or other emergency conditions also could potentially improve patient outcomes.

Many AI algorithms also perform complex measurements that are reproducible, image reformatting, anatomical labeling, contouring anatomy, and other tasks that are time consuming. AI in these instances can help reduce the tedious manual processes and may help improve radiology report information and accuracy, while reducing the time it takes to read an exam, especially amid falling reimbursements to read exams. 

The business case for AI to make radiology workflows more efficient

Anyone who has attended the large RSNA or Healthcare Information and Management Systems Society (HIMSS) meetings has seen PACS, enterprise imaging, workflow management and advanced visualization vendors adding complementary AI solutions that can enhance the radiology workflow. This includes AI for things like better organization of DICOM worklists in which critical findings may automatically move the exam to the top of the reading list, explained Sanjay Parekh, PhD, senior market analyst with Signify Research.

AI is also helping enable things like single-click marking of exams for follow-up for things like incident findings, automated measurements, automating things like anatomical countering and cardiac strain, or automatic ally calling up relevant priors for comparisons for tracking changes in tumors. 

“It’s not just the image analysis, it’s also the workload balancing, the fleet management, or how do you get all these resources onto one platform to deliver value to the healthcare system, not just the radiologist,” Parekh said. 

This includes looking at both upstream and downstream workflows and what clinicians need and how AI can be implemented to help facilitate better care by reducing bottlenecks or areas where there is a lot of manual data entry. In acute care this includes several AI vendors to aid acute care teams with alerts, immediate access to imaging and other patient data, and the ability to message everyone at the team before the images are even read by the radiologist.

How will AI change the future of radiology?

Parekh said there is a trend toward “opportunistic screening” by AI to search for incidental findings in all types of medical imaging that could help improve care by catching diseases before they become more advanced and symptomatic. This includes things like lung screening CT or virtual colonoscopy exams where AI algorithms look for calcium in the arteries to estimate cardiac risk scores, or scans the images for abnormal findings in the background unrelated to the purpose of the exam.

“This is looking downstream from radiology and saying, OK, we are picking something up opportunistically, but then we don’t want the patient to fall off the radar. You want that to be followed up,” Parekh says. 

Opportunistic screening may offer ways for hospitals to better care for patients preemptively rather than being reactive to acute episodes or long after a cancer had become a serious problem and is more difficult to treat. 

Allen said AI that can help facilitate that followup process also could be worth the expense if it brings in additional business for the healthcare system. 

Parekh said the American Medical Association (AMA) now has CPT codes to track usage of AI technology, which is a first step when evaluating if new technologies should receive reimbursement.

Even if AI is not reimbursed, Parekh said healthcare providers need to look at whether AI can help improve patient outcomes and save money in other areas, which could make the investment in AI worthwhile. Beyond opportunistic screening, if AI apps can help save radiologists enough time to be able to read a few more exams each day, there could be business ROI over the long term.

AI may help address health disparities 

"We have not seen good payment models, so one of the fears that we have is that larger academic research centers will be able to adopt AI into clinical use, while smaller practices, particularly in underserved areas, are going to struggle without a reimbursement plan and create health disparities that I don't think anyone wants to see," Allen explained. 

Kahn also said AI may play a key role in the coming years of addressing health disparities. 

"At some level, we need to find ways to where we can deliver care that is cost-effective, reaches all the people we need to reach and provided equitable healthcare, and the hope is that we can use AI to expand the reach of what we o and improve the quality of it," Kahn said. 

Allen agreed AI can help close areas of health disparities, which is of interest for health system administrators and possibly to payers. 

U.S. health systems are increasingly leveraging digital health to conduct their operations, but how health systems are using digital health in their strategies can vary widely.

That’s according to a new report from Finn Partners and Galen Growth that examined the growing digital health portfolios of U.S. health systems, hospitals and clinics. Digital health covers a wide range of technologies and applications that enable remote patient monitoring, physician-to-physician expertise remotely, remote care and use of data to make informed decisions about patient care. 

In the COVID-19 pandemic, digital health also offers people the ability to access healthcare right from their home, alleviating many of the issues with accessibility in remote or rural areas, as well as for those who can’t take time away from work or have mobility issues.

Health systems in the United States are encompassing more and more of the digital health space, with 31% of all global partnerships with digital health ventures founded in the U.S. built within hospitals and health systems in 2022. Globally, there have been more than 1,400 partnerships between digital health ventures and U.S. health systems since 2012, according to Galen Growth. However, the spread between health systems is big, with a 2.1x difference in the digital health portfolio size between the most active health systems and the largest health systems. 

Here are some of the U.S. health systems with the largest digital health portfolios and the number of partnerships, according to the report:

• Mayo Clinic (77)
• Mount Sinai Health System (34)
• Cleveland Clinic (32)
• Boston Children’s Hospital (24)
• Memorial Sloan Kettering Cancer Center (23)

The ranking comes at a time when upwards of 50% of health systems are reputed to be struggling financially due to the tight labor market. As health systems face ongoing challenges, digital health will continue to play a crucial role in their success.

“We are likely to see more partnerships between health systems and digital health innovators in the months and years to come because close relationships between these groups are vital as we seek ways to leverage technology to enhance patient care and outcomes and improve both providers’ and patients’ experiences,” Aaron Lewis, executive vice president, growth and integrated solutions of Lifepoint Health, said in the report. “The reality is that health systems and digital health innovators can sometimes have very different views on the challenges and opportunities that exist in healthcare. Working with health systems allows innovators to understand the realities of care delivery and working with innovators allows health systems a view into the possibilities of new technology. Collaboration is how we make a difference faster.”

An advanced artificial intelligence (AI) algorithm interpreted transthoracic echocardiograms (TTEs) with more accuracy than trained sonographers, according to new research published in Nature.[1]

The study was a blinded, randomized trial that asked cardiologists to review 3,495 TTEs from 3,035 patients. The cardiologists were asked if they agreed with these initial assessments and if they had any corrections to offer.

Overall, cardiologists made corrections to 16.8% of AI-generated assessments and 27.2% of sonographer assessments. Also, an AI-guided workflow was associated with saving both sonographers and cardiologists time compared to a more traditional workflow.

“The results have immediate implications for patients undergoing cardiac function imaging as well as broader implications for the field of cardiac imaging,” senior author David Ouyang, MD, a cardiologist with the Smidt Heart Institute at Cedars-Sinai in Los Angeles, said in a prepared statement. “This trial offers rigorous evidence that utilizing AI in this novel way can improve the quality and effectiveness of echocardiogram imaging for many patients.”

Another key takeaway from the group’s findings was that cardiologists were unable to consistently distinguish between AI- and sonographer-generated assessments. Cardiologists correctly guessed the source 32.3% of the time, incorrectly guessed the source 25.2% of the time and said they were “unsure” for all other studies.

“This speaks to the strong performance of the AI algorithm as well as the seamless integration into clinical software,” Ouyang added. “We believe these are all good signs for future AI trial research in the field.”

“This work raises the bar for artificial intelligence technologies being considered for regulatory approval, as the U.S. Food and Drug Administration has previously approved artificial intelligence tools without data from prospective clinical trials,” added co-senior author Susan Cheng, MD, director of the Institute for Research on Healthy Aging at the Smidt Heart Institute. “We believe this level of evidence offers clinicians extra assurance as health systems work to adopt artificial intelligence more broadly as part of efforts to increase efficiency and quality overall.”

Read the team’s full analysis here.

An artificial intelligence work list prioritization tool can reduce both turnaround times and exam wait times for patients with CT-detected pulmonary embolus (PE). 

In turn, the tool's use could result in patients with positive CT pulmonary angiography (CTPA) exams undergoing expedited treatment, according to experts involved in new research published in the American Journal of Roentgenology. 

“By assisting radiologists in providing rapid diagnoses, the artificial intelligence tool could potentially enable earlier interventions for acute pulmonary embolus,” concluded lead researcher Kiran Batra, MD, from the Department of Radiology at University of Texas Southwestern Medical Center in Dallas. 

The FDA-approved tool works by reprioritizing CTPA exams to the top of a radiologist’s work list when the scan is positive for PE. To assess the tool’s utility, experts used timestamps from electronic medical records and dictation systems to compare wait times (exam completion to report initiation), read times and report turnaround times prior to and following its implementation. 

The team used a total of 2,501 exams—1,166 pre-AI and 1,335 post-AI—for their analysis. When comparing the two time periods, researchers noted that the tool’s use resulted in 12.7% of CTPA exams being moved to the top of the reading radiologist’s list. For PE-positive exams, this move rendered significantly shorter report turnaround times (47.6 vs 59.9 minutes) and mean wait times (21.4 vs 33.4 minutes) but did not impact mean read times. 

This finding was even more pronounced in wait times during regular operational hours, with a difference of 28.4 minutes between pre- and post-AI waits. 

The results suggest that the use of similar work list prioritization tools could improve outcomes in patients with PE by enabling providers to initiate treatments, such as anticoagulation therapy, sooner, the study authors noted.

To learn more about the work, click here. 

Massachusetts is the best state in the nation when it comes to children’s healthcare, according to a new ranking from WalletHub.

The ranking looked at all 50 states plus the District of Columbia and compared 33 key indicators of cost, quality and access to children’s healthcare. While roughly 95% of children between 0 and 18 have health insurance to protect them if they become ill, parents still face high healthcare costs. And the disparity across states when it comes to cost and access is wide. 

On average, American workers pay $6,100 per year toward employer-sponsored family coverage. Other families may qualify for health insurance through the Children’s Health Insurance Program (CHIP) or Medicaid, but those without government assistance may find the cost of healthcare for their families a challenge. 

WalletHub’s ranking underscores the differences between states and the impact on children’s health.

Here are the top 15 best states for healthcare:

1. Massachusetts
2. District of Columbia
3. Rhode Island
4. Vermont
5. Hawaii
6. New York
7. Maryland
8. New Jersey
9. Oregon
10. Minnesota
11. Connecticut
12. Delaware
13. Pennsylvania
14. Iowa
15. Washington

The top state, Massachusetts, was No. 1 when it came to kids’ nutrition, physical activity and obesity. The state also scored fifth overall for kids’ health and access to healthcare. In addition, Massachusetts was No. 1 when it came to the percentage of uninsured children. By comparison, Texas, at the bottom of the ranking, had eight times more uninsured children than Massachusetts. Massachusetts also had one of the lowest infant death rates in the nation (No. 3 overall).

In comparison to the top states, those ranked among the worst for children’s healthcare had higher percentages of overweight and obese children, a lower percentage of children with excellent/very good health, higher infant death rates and fewer pediatricians and family doctors per capita.

Mississippi, which was ranked at the worst state in the nation for children’s healthcare, came in last for the percentage of children in excellent or very good health.

Here are the 10 worst states for children’s healthcare:

1. Mississippi
2. Texas
3. Louisiana
4. Wyoming
5. Indiana
6. West Virginia
7. Kentucky
8. Oklahoma
9. Arkansas
10. New Mexico

Leadless pacemakers may be a safe and effective short-term treatment option for pediatric patients presenting with bradycardia, according to new research published in Circulation: Arrhythmia and Electrophysiology.[1] Additional research is still required to determine the long-term effectiveness of these devices.

Researchers tracked data from 63 bradycardia patients between the ages of 4 and 21 years old treated with a leadless pacemaker from 2016 to 2021. The average patient age was 15 years old, and patients were treated at one of 15 centers in the United States, United Kingdom or Italy. Thirty-two percent of patients presented with congenital heart disease.

All patients were treated with a Micra transcatheter leadless pacemaker from Medtronic. Data came from a registry managed by the Pediatric and Congenital Electrophysiology Society.

Overall, implantation was successful in all but one patient. While 16% of patients experienced a complication, a majority of those issues were minor bleeding events that could be treated quickly. There were three more serious complications, including a blood clot in the femoral vein in one patient, cardiac perforation in one patient and suboptimal pacemaker functionality, which required it to be removed after one month, in one patient. There were no deaths, pacemaker-related infections or device embolizations.

After an average follow-up period of 9.5 months, the research team noted that all successfully implanted pacemakers still had a strong battery life.

“The leadless pacemaker works very well in children, just like it does in adults,” lead author Maully J. Shah, MBBS, director of cardiac electrophysiology in the Cardiac Center at Children’s Hospital of Philadelphia and a professor of pediatrics at the Perelman School of Medicine at the University of Pennsylvania, said in a prepared statement. “We found it may be safely implanted in select pediatric patients that need pacing. Our study’s results indicate select children may be considered candidates since they may benefit greatly from leadless pacing. However, because of the current technology, which uses a very large catheter designed for adults to place the leadless pacemaker and lack of reliable future extractability of the pacemaker, the wider pediatric population is not able to benefit from this device.”

The group noted that smaller catheters are needed to get the best possible clinical outcomes when treating these younger patients. However, better catheters are just one piece of the puzzle.

“Leadless pacemaker technology is the wave of the future,” Shah added. “This is an excellent technology that may be offered to a wider pediatric population. However, techniques and tools to place the device must be designed for smaller patients, specifically children, and there needs to be a mechanism to remove and replace this pacemaker without surgery when the battery runs out since pediatric patients will likely require pacing for the rest of their lives, which is several decades after implantation.”

This study does have certain limitations, including its small sample size and relatively short-term follow-up period. Shah et al. plan to keep tracking these patients for five years.

Shah does work as a consultant for Medtronic, the company behind the technology being examined, but Medtronic did not fund this analysis.

Read the full study here.

Transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) are associated with similar short-term outcomes among patients with pure native aortic regurgitation (AR), according to new data published in The Annals of Thoracic Surgery.[1]

The study’s authors noted that patients with pure native AR are often excluded from clinical trials comparing TAVR and SAVR. In addition, they noted, the few studies that have examined this issue have focused on “relatively small” patient populations.

Hoping to provide a more comprehensive look at this topic, first author Amgad Mentias, MD, a cardiologist with the Heart, Vascular and Thoracic Institute at Cleveland Clinic, and colleagues tracked data from more than 11,000 Medicare patients. All patients presented with pure AR and underwent TAVR or SAVR from 2016 to 2019. Patients presenting with a history of aortic stenosis were excluded from this analysis.

While 89.5% of patients underwent SAVR, the remaining 10.5% underwent TAVR. SAVR patients tended be younger and less frail than TAVR patients. SAVR patients also presented with fewer comorbidities.

Overall, TAVR and SAVR patients were linked to comparable in-hospital mortality, 30-day mortality and 30-day stroke rates. After a median follow-up period of 31 months, however, TAVR patients were associated with a higher mortality rate and higher risk of redo aortic valve replacement.

“The poorer intermediate-term outcomes with TAVR in our study could potentially be due to residual unmeasured differences and surgical risk between the two groups, but could also be related to the anatomical differences between pure severe AR and aortic stenosis,” the authors wrote. “The increased prevalence of bicuspid leaflets and annular/aortic root dilation in AR patients, with relatively less leaflet and annular calcification, pose a challenge with transcatheter heart valve anchoring and adequate positioning and increase the risk of paravalvular leak and device embolization. Abnormal hemodynamics across prosthetic valves and paravalvular leakage, whether mild or moderate/severe, are known predictors of poor outcomes, including death. The recommendation to oversize the THV during implantation is also associated with increased risk of aortic rupture and conversion to open heart surgery, a complication that was observed in 1% of TAVR patients in our study.”

TAVR was also associated with a much lower risk of in-hospital acute kidney injury and new-onset atrial fibrillation. In addition, TAVR patients were less likely to require a blood transfusion and, as one might expect, they had shorter lengths of stay.

“Randomized trials of SAVR compared with new transcatheter heart valves dedicated for pure AR with extended follow-up are awaited,” the authors concluded.

Though a majority of the study’s authors were from the Cleveland Clinic, co-authors from The Warren Alpert Medical School of Brown University, Baylor College of Medicine and University of Iowa Carver College of Medicine also contributed.

Read more here.

A grand jury has indicted Mission Viejo, California, dermatologist Yue “Emily” Yu, MD, PhD, for allegedly poisoning her radiologist husband using liquid drain cleaner.   

She faces four felony counts and more than eight years in prison if convicted, the Orange County District Attorney’s Office announced April 5.

Jack Chen, MD, began noticing a strange taste in his daily cup of tea in April 2022. When the problem persisted, he set up a camera in the family’s Irvine, California, kitchen. On three separate occasions in July 2022, the device captured Yu pouring a substance out of a bottle of drain cleaner into the beverage while it rested on the counter. Chen suffered from stomach ulcers resulting from the alleged poisoning, according to local authorities.

“Our homes should be where we feel the safest,” Orange County District Attorney Todd Spitzer said in an announcement of the indictment. “Yet, a licensed medical professional capitalized on her husband’s daily rituals to torment her husband by systematically plying his tea with a Drano-like substance intending to cause him pain and suffering.”

Chen collected samples of his beverage and shared them with the Irvine Police Department. The FBI later confirmed that the substance contained such household chemicals. Authorities arrested 45-year-old Yu in August 2022 under suspicion of poisoning, but released her when she posted the required $30,000 bond. She has denied the allegations, according to previously published reports.

Under terms of the indictment, Yu must report to the Medical Board of California, which will decide whether she can continue practicing medicine. She is scheduled to be arraigned on April 18.

Chen, who was 53 at the time of the incident, has since filed for divorce following 10 years of marriage, the New York Post reported. He also has been granted full custody of their two children. Chen has previously practiced interventional and diagnostic radiology in Tarzana, California, and was affiliated with multiple medical centers in the area, according to Doximity.

As the use of digital breast tomosynthesis (DBT) has markedly increased in recent years, so too have radiologists’ interpretive performances. 

A new paper published in Radiology compares radiologist performance when reading 2D and 3D mammography screenings. Based on more than 2 million screening exams included in the retrospective analysis and multiple measures, experts concluded that radiologists’ performance has significantly improved with the growing use of DBT. 

“Our study demonstrated that more radiologists in U.S. community practice are meeting recommended performance standards with digital breast tomosynthesis than they did with digital mammography,” said lead author Christoph I. Lee, MD, professor of radiology at the University of Washington (UW) School of Medicine in Seattle and director of the Northwest Screening and Cancer Outcomes Research Enterprise at UW. “This is good news for women and breast cancer screening, as digital breast tomosynthesis has quickly become the most popular breast cancer screening modality in the U.S.” 

Since receiving approval from the U.S. Food and Drug Administration in 2011, DBT has become the most common method for breast cancer screening and as of September 2022, 84% of all U.S. mammography screening facilities housed DBT units. 

Experts sought to assess whether the growing popularity of DBT screening has had any impact on radiologist performance benchmarks established by the Breast Cancer Surveillance Consortium (BCSC). To do this, the team analyzed DBT screening exams from five BCSC registries that occurred between 2011 and 2018. Altogether, this resulted in a total of 2,301,766 screening exams—458,175 DBT and 1,843,591 2D digital mammograms. 

Radiologist performance measures were based on abnormal interpretation rate (AIR), cancer detection rate (CDR), sensitivity, specificity and false-negative rate (FNR). The team found that use of DBT resulted in improved performance in every category except sensitivity and FNR, which were both in line with 2D performance measures. 

Lee signaled optimism for how such improved performance also could positively impact screening outcomes in the future.  

The study abstract is available here. 

A bipartisan group of physician lawmakers have proposed legislation to fix what they say is an outdated Medicare payment system that’s threatening patients’ access to care.

House representatives first introduced the Strengthening Medicare for Patients and Providers Act on April 6, drawing praise from the American College of Radiology and other professional associations. The bill proposes to provide a permanent annual inflationary physician payment update tied to the Medicare Economic Index.

Doc groups have long championed such a change. When adjusting for inflation in practice costs, Medicare physician payments have actually declined by 26% since 2001, the American Medical Association estimates. Failing to take action will result in further consolidation, physician burnout, and medical practice closures, members of Congress contended.

“I am deeply concerned about the impact the outdated Medicare physician payment rate is having on healthcare access for my constituents,” Rep. Raul Ruiz, MD, D-Calif., who previously practiced in emergency medicine, said in a statement. “That is why I am announcing legislation that will move us away from a system where every year seniors’ access to care is threatened due to uncertainty over potential cuts.”

The proposal comes after the Medicare Payment Advisory Commission just recently recommended a physician payment update tied to half of the economic index. ACR, the Society of Interventional Radiology and more than 130 other professional groups also wrote to leaders of the House and Senate in March, asking them to address lagging pay rates.

“As the only major fee schedule without a built-in inflationary update, Medicare reimbursement for providers billing under the MPFS has failed to keep pace with the true cost of practicing medicine,” the college said in an April 6 news update praising HR 2474, later adding: “The ACR will continue to advocate for short- and long-term payment stability for our members and the patients they serve.”

“We are deeply worried that many practices will be forced to stop taking new Medicare patients—at a time when access to care is already inadequate,” AMA President Jack Resneck Jr., MD, said in a separate statement. “Congress often waits until problematic situations become full-fledged crises. Members need to hear from their hometown physicians that we are nearing a crisis. Congress needs to pass this bill stat.”  

Others sponsoring the legislation included Reps. Larry Bucshon, MD, R-Ind., Mariannette Miller-Meeks, MD, R-Iowa, and Ami Bera, MD, D-Calif.