#389 – The Wire

  • Xenon MRI for Long COVID: UK-based researchers discovered that Xenon gas-enhanced MRI could help scientists understand and develop new treatments for long COVID-related breathlessness. The researchers performed Xenon MRI, CT, and lung function tests on 35 individuals (11 w/ long COVID but never hospitalized, 12 w/ long COVID who were hospitalized, 13 controls). They found that Xenon gas transferred less effectively among the long COVID patients, even though their CT scans didn’t show lung abnormalities and their lung function tests matched the controls.
  • An Exchange Ultimatum: A JACR editorial by some of medical imaging’s biggest influencers called for image exchange companies to start working together or face consequences. The authors blamed competitive motives (not technology factors) for the fact that images are still mainly exchanged via physical media, proposing a vendor-to-vendor exchange model to solve this problem. They also asked exchange vendors to “immediately commit” to this model, while encouraging healthcare institutions to only contract with committed exchange vendors by 2024.
  • Lung-RADS Underestimates: A new AJR study revealed that Lung-RADS category 3 and 4 nodules might be more prevalent and have higher malignancy rates than current ACR estimates suggest. The researchers reviewed 9,148 LCS exams, finding that a far higher percentage of patients had category 3 (9.4% vs. 5%) and category 4 nodules (7.9% vs. 2%) than the ACR Lung-RADS estimates. These patients also had far higher malignancy rates for category 3 (3.9% vs. 1-2%), 4A (15.5% vs. 5-15%), 4B (36.3% vs. >15%), and 4X nodules (76.8% vs. >15%) than the Lung-RADS estimates.
  • AI in the NHS: The UK NHS provided an interesting breakdown of its AI use, while highlighting imaging’s central role in its AI rollout. Of the NHS’s 240 active AI technologies, Diagnostic AI tools were the most prevalent ‘type’ (34%), radiology and cardiology were the top two ‘clinical areas’ (11% & 9%, after ‘multiple’ at 23%), and radiology, radiographers, and cardiology were among the ‘most affected workforce groups’ (15%, 8%, 8%). 
  • Butterfly & Ambra Integrate: Butterfly Network and Ambra Health will launch an integrated solution, combining Butterfly’s Blueprint enterprise platform with Ambra’s image management technology. The solution would bring POCUS exam data into clinical workflows, representing a significant expansion for a modality that has “point-of-care” in its name.
  • Info-Blocking Influx: Current Problems in Diagnostic Radiology once again lived up to its name, detailing how complying with the 21st Century Cures Act’s information-blocking rules can lead to a surge in questions to referring physicians. A survey of 249 MGH referring physicians found that the majority were contacted by patients about their radiology reports (195, 78%) and received a higher volume of patient calls (165, 66%) during the 60 days after MGH began giving patients immediate access to their radiology reports.
  • Vara & Mamotest’s Mexico Venture: Vara and Mamotest launched a breast cancer screening alliance in Mexico that combines Vara’s mammography AI with Mamotest’s tele-mammography service, and could significantly improve breast screening accessibility in the country (~$15 vs. ~$100 normally). The partners launched their first breast screening unit in Reina Madre private hospital near Mexico City and plan to open “many” other units across Mexico. This is a great example of how AI can make an impact in countries without formal screening infrastructure.
  • Public Dataset Disparities: The publicly available CXR datasets that we cover in AI studies nearly every week might contain significant demographic imbalances. That’s from a recent JACR study review of 23 public CXR datasets, finding only 13 datasets detailing patient gender, just two reporting patient race or ethnicity, and only one citing coverage status. Across the dataset with gender details, there were far more CXRs from male patients (avg: 55.2%; range: 47.8%-69.7%).
  • Five Years for Kickbacks: California imaging center CEO, Sam Sarkis Solakyan, found out the hard way that crime doesn’t pay (at least not in the long run) after being sentenced to 5 years in prison and ordered to pay $28M for operating an MRI “cross-referral scheme.” Solakyan paid recruiters and clinicians over $8.6M to refer workers comp patients to his nine imaging centers where he generated over $250M in kickback-related imaging procedure billings.
  • Cautiously Optimistic & Needing Education: A new JMIR study of 2,167 Canadian healthcare students (10 fields, 18 universities) revealed that most students expect AI to have a major impact on their careers (78.7%) and are cautiously optimistic about AI’s role in their fields (74.5%). However, only half of the students knew what AI was (49%) and even fewer had an accurate understanding of AI (31%). We’ve seen similar AI opinion surveys, although this study’s diverse student population is notable (included: dentistry, nursing, dietetics, others) considering that many see radiologists as just the first step towards enterprise AI adoption.
  • VIA Acquires: Alabama-based practice Vulcan Imaging Associates (>30 radiologists, founded in 1969, serves AL, GA, FL) expanded its breast imaging capabilities with its acquisition of local tele-mammography practice, Women’s Imaging Associates. This is one of the smaller practice acquisitions we’ve covered, but it still helps solidify VIA’s presence and capabilities.

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team