“There is a clear market failure . . . and nobody is catching it.”
Emergency medicine and public health professor, Ziad Obermeyer, MD, on how biased algorithms made their way across healthcare.
One of your fellow Imaging Wire readers is investigating radiologist burnout for his PhD paper and he needs your help making it as complete as possible. If you’re a radiologist and willing to contribute, please fill out this quick and anonymous survey.
Imaging Wire Sponsors
Arterys | Bayer Radiology | Canon Medical Systems | GE Healthcare
Healthcare Administrative Partners | Hitachi Healthcare Americas
Novarad | Nuance | Riverain Technologies | Siemens Healthineers
United Imaging | Zebra Medical Vision
The Imaging Wire
Relatively unknown cardiac AI startup, Cleerly, just announced its public “launch” and a sizable $43m Series B round (increasing its total funding to $54m) that it will use to bring its heart attack prevention pathway to market.
We cover imaging AI launches and funding rounds all the time, but this one deserves some extra attention…
- Targeting the Cardiac Void – Cleerly aims to correct the imbalance between cancer screening pathways (numerous, widely adopted) and heart disease screening pathways (not very established or adopted), which is fair considering heart disease’s role as the leading cause of death globally.
- Cleerly’s Background – Even though Cleerly just publicly “launched” this week, the NYC-based startup has been in operation since 2017, and seems to have a solid foundation (team, product, strategy, clinical evidence).
- Cleerly’s Pathway – Cleerly’s digital care pathway analyzes coronary CTA scans to quantify / characterize heart disease and then guides stakeholders (imaging & non-imaging clinicians, patients, payers) through its heart attack prevention pathway. That includes pathway tools that improve understanding of patient conditions (e.g. explain imaging findings to non-imaging clinicians), support collaboration, and track disease changes over time.
- Product Over Tech – Cleerly uses AI to perform its analysis, but it doesn’t seem to define itself as an “AI company” or its product as an “AI product.” That’s still pretty rare (in healthcare, anyway) and it was almost certainly on purpose.
- Population Health AI Momentum – Cleerly’s “launch” and significant funding are the latest signs of AI’s growing momentum in population health, following increased interest from academia and a recent strategic shift from AI heavyweight, Zebra Medical Vision.
Cardiac Dual Source CT’s Potential
Explore untapped potential in dual source CT for cardiac imaging. This new Siemens Healthineers whitepaper showcases the clinical benefits of fast native temporal resolution.
UCH’s Case for ClearRead Bone Suppress
This Riverain Technologies case study details how the University of Colorado Hospital enhanced its chest X-ray workflow with ClearRead Bone Suppress.
- CXR’s Blunt Trauma Shortcomings: New research out of UCSF detailed chest X-ray’s “poor performance” for evaluating blunt thoracic injuries. A review of 4.5k adult blunt trauma patients’ CXRs and chest CTs (1.5k w/ injuries) found that CXR missed 54.7% of injuries (818 patients, including 63 major injuries), suggesting that CXR should only be used with low-risk trauma patients.
- Aidoc & icometrix’s Stroke Suite: Aidoc and icometrix announced their end-to-end stroke AI suite, combining Aidoc’s stroke solution (FDA-cleared, flags & communicates LVO and ICH findings) with icometrix’s icobrain-cva (FDA & CE-cleared, analyzes acute ischemic stroke patients’ CT perfusion scans). It will be interesting to see how this works in the clinic, but it seems to be a good example of how partnerships can make AI workflows much more comprehensive.
- Vetting Techs: A new NHS study found that when radiographers handled incoming requests for routine head CTs and urinary tract CTs (KUB) rather than radiologists, it reduced incoming calls to radiologists for both exams (-30% & -100%) and cut average head CT wait times by 40%. It might surprise some folks in the U.S., but radiographers commonly vet head and KUB scans in other countries.
- AI Bias Playbook: The University of Chicago just released an AI bias “Playbook” detailing widespread algorithmic bias across healthcare and outlining how healthcare organizations can address bias within their own AI portfolios. The playbook suggests that healthcare systems should review their algorithms, check each tool for bias, and either fix or stop using any of their biased AI tools.
- Fujifilm’s First Fluoro: Fujifilm Medical Systems U.S.A. launched its Persona RF PREMIUM radiography & fluoroscopy system, highlighting its patient comfort and operator productivity advantages. The launch marks Fujifilm’s entrance into the U.S. fluoroscopy segment and continues its significant modality expansion that also includes recent / upcoming expansions into the CT, MRI, cart-based ultrasound, and surgical C-arm segments.
- AI’s Bad Hand: The PACSman (aka Michael J. Cannavo) shared more tough love for imaging AI in a recent AuntMinnie editorial. PACSman started off with an overview of AI’s current challenges (slow adoption, rad pushback, limited/difficult reimbursements, lacking ROI/clinical evidence), before listing several ways AI vendors can make the most of their bad poker hands (demonstrate ROI, target high-volume institutions, use per-scan models for smaller hospitals, improve/simplify integration). However, he warned that these challenges will inevitably leave fewer players sitting at the final AI card table.
- Distinguishing Remanufacturing & Servicing: The FDA released its long-awaited medical device remanufacturing draft guidance document, seeking feedback to help better differentiate 3rd party servicing and remanufacturing. The guidance doc currently defines remanufacturing as activities that significantly change a device’s “performance or safety specifications, or intended use” and defines servicing as “repair and/or preventive or routine maintenance” with the goal of bringing its safety / performance back to OEM specifications.
- Sanford & Infervision’s AI Collab: Major U.S. health system Sanford Health (46 medical centers, 1.5k physicians) and AI firm Infervision announced an AI co-development partnership. Through the partnership, Sanford will use Infervision’s InferScholar platform for annotation and training (keeping patient data onsite) and Infervision will use Sanford’s clinical input to create the final products.
- Reduced Dose Gadobutrol Effective: An AJR study found that brain MRIs performed with lower Gadobutrol dosage (0.075 vs. 0.1 mmol/kg) largely matched standard dose MRI, encouraging clinicians to consider reduced-dose scans (especially w/ patients undergoing multiple exams). The researchers performed standard dose MRIs on 141 patients with known / suspected CNS and then performed reduced dose MRIs on patients with brain lesions, discovering that the reduced dose scans were non-inferior for all primary efficacy metrics and identified a similar number of lesions (291 w/ reduced dose vs. 301 w/ standard).
- CRA & HAP Expand: Commonwealth Radiology Associates and Healthcare Administrative Partners (CRA & HAP) expanded their partnership, adding HAP’s practice and financial management services to their 2-year-old revenue cycle management relationship.
- Managing AI Expectations: A new JACR editorial encouraged radiologists to start figuring out how to manage patients’ imaging AI expectations, warning that growing media coverage might lead to patients regularly asking whether their radiologist will be using AI for their diagnosis (including many diagnoses that don’t currently involve AI).
- Inappropriate PCa Imaging’s Impact: A new study in Cancer Reports found that eliminating inappropriate imaging among men with incidentally-identified prostate cancer would reduce imaging rates in this population from 53% to 38% and cut per-person imaging spending from $236 to $157. The study found that 45% of men with low-risk cases still received inappropriate imaging, even though these scans rarely influence treatment decisions and come with a list of downsides (waste, patient concern and inconvenience, radiation exposure).
- Neuroimaging Gamechanger: USC researchers just performed the first human functional photoacoustic CT (fPACT) brain scans, finding that fPACT images are “superior in some ways to 7T fMRI” and could even “change neuroimaging forever.” Although still in its early stages, fPACT could overcome many of fMRI’s imitations (cost, size, portability, support for patients with implants) and its ability to perform 3D blood flow mapping could lead to a range of imaging advancements (vessel & tumor imaging, localizing functions and seizures).
Ditching the Disk with CryptoChart
Do your patients text more than they use CDs? Find out how Novarad’s CryptoChart simplifies image access, combining secure QR codes and text and email communications to help providers and patients ditch the disk.
The Resource Wire
- See how physicians are leveraging GE Healthcare’s Edison Open AI Orchestrator platform and Icobrain’s AI-enabled applications to monitor and treat MS and Traumatic Brain Injury.
- This Arterys white paper details how and why cloud-based AI tools have become essential for streamlining and improving lung nodule detection and tracking.
- Learn how Birmingham Radiological Group-GV used Nuance PowerScribe One to improve its turnaround times, report quality, and workflow efficiency.
- This Diagnostic and Interventional Cardiology article details the unique advantages of cloud-based CVIS systems (off-property access, team collaboration), with insights from one Mississippi-based cardiologist on the benefits of Hitachi’s VidiStar CVIS.
- This AI economics overview from Healthcare Administrative Partners details the various AI ROI scenarios and ways that AI can contribute to radiology practices until reimbursements become more of a reality.
- This Zebra-Med post details how AI is revolutionizing population health programs through automation, workflow integration, and significantly expanding early disease detection.
- United Imaging’s approach to brain imaging in molecular imaging puts the patient first, focusing on reducing scan times and correcting for patient motion to reduce the need to repeat a study — while also providing a high resolution image and quantitative values to aid in interpretation.
- This Bayer Radiology case study details how Einstein Healthcare Network reduced its syringe costs, enhanced its syringe loading, and improved its contrast documentation when it upgraded to the MEDRAD Stellant FLEX CT Injection System.
- Watch CriticalCare’s Dr. Yusuf Karrar, MD discuss how Canon’s CT Fluoro touch interface allows “Simple, Streamlined CT Fluoroscopy.”
Today’s issue was brought to you by Jake Fishman and Jason Barry.