GE Wireless | COVID Correlations | Flawed AI

“Ultrasound makes bad doctors good and good doctors even greater,”

Dr. Yale Tung-Chen of Madrid’s Hospital Universitario Puerta de Hierro Majadahonda with high praise for handheld ultrasound.

Imaging Wire Sponsors

  • Arterys – Reinventing imaging so you can practice better and faster.
  • Bayer Radiology – Providing a portfolio of radiology products, solutions, and services that enable radiologists to get the clear answers they need.
  • GE Healthcare – Enabling clinicians to make faster, more informed decisions through intelligent devices, data analytics, applications and services.
  • Healthcare Administrative Partners – Empowering radiology groups through expert revenue cycle management, clinical analytics, practice support, and specialized coding.
  • Hitachi Healthcare Americas – Delivering best in class medical imaging technologies and value-based reporting.
  • Novarad – Transformational imaging technologies that empower hospitals and clinicians to deliver clinical, operational and fiscal excellence.
  • Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter.
  • Riverain Technologies – Offering artificial intelligence tools dedicated to the early, efficient detection of lung disease.
  • Siemens Healthineers – Shaping the digital transformation of imaging to improve patient care.
  • United Imaging – Our mission, Equal Healthcare for All, pushes us beyond conventional boundaries to help clinicians expand modern, digital, intelligent care to more people within their communities.
  • Zebra Medical Vision – Transforming patient care with the power of AI.

The Imaging Wire

GE Goes Wireless

GE Healthcare’s handheld ultrasound lineup took a major step this week with the launch of the Vscan Air, marking GE’s expansion into the wireless handheld ultrasound segment.

  • About the Vscan Air – The Vscan Air ($4,495 Basic, $5,245 w/ Sonosim software) combines a dual-probe whole-body transducer (high frequency linear & convex) with GE’s Vscan visualization software, allowing point-of-care clinicians to wirelessly perform a range of exams using their mobile device screens (e.g. cardiac, lung, COVID). However, GE’s messaging placed just as much focus on how the Vscan Air adds value in-between scans (portability, durability, setup efficiency, cleanability, security) and how the device supports a wider range of clinicians and clinical workflows.
  • Startup Disruption – The ultrasound startups who’ve run the wireless handheld segment so far will face a new level of competitive pressure from GE, which is targeting a 30% handheld share by 2025. It appears that GE Healthcare is emphasizing the Vscan Air’s image quality and security as its main competitive differentiators, which makes sense given the concerns about some current wireless models, but GE’s massive ultrasound client base and sales force will almost certainly be the Vscan Air’s greatest advantage.
  • Small Device, Big Marketing – GE’s marketing team leaned-in with the Vscan Air launch, including a social media launch countdown, some solid media coverage, and a dedicated Vscan Air product page with way more multimedia content than we typically see in this industry. Given the handheld ultrasound’s role as a “personal” medical device, and considering how high Butterfly has raised the marketing bar in this segment, the Vscan Air’s comprehensive and consumer-style marketing makes a lot of sense.
  • The Takeaway – It’s been impressive to see how the wireless ultrasound segment has evolved without the presence of a major OEM, and although we’re still a long way from wireless handhelds becoming as ubiquitous as the stethoscope (as some have envisioned), the entrance of GE is almost certain to help drive the next phase of this segment’s evolution.

The Wire

  • COVID CT’s Neuro Correlation: Physicians might be able to use COVID-19 patients’ CT severity scores to predict the diseases’ neurological impact and help guide early neurological care. That’s from a new University of Cincinnati-led multicenter study (n = 135 patients w/ COVID & neurological symptoms) that found that the 49 patients with acute abnormal neuroimaging findings also had far higher CT severity scores than patients with normal brain MRIs (9.9 vs. 5.8 avg. scores).
  • Debating Screening Motivations: A recent study found that the vast majority of U.S. breast imaging center websites (n = 606 sites) encourage women to start breast cancer screening in their 40s (77.2% at 40yrs, 7.2% at 45yrs) and continue their exams annually after that (72.7%). The authors suggested that this discordance with USPSTF guidelines (50-75yrs & biannual) creates confusion and might not have entirely patient centric motivations. The major imaging societies quickly responded to this “outrageous and insulting” study, clarifying that these imaging centers do indeed follow (other) national guidelines and arguing that the USPSTF guidelines aren’t sufficient.
  • Check-Cap Milestone: Digestible imaging capsule startup, Check-Cap, landed FDA approval to begin a “pivotal” study to measure its C-Scan colorectal cancer screening system’s safety, performance, and patient compliance. Check-Cap already has solid research momentum, and if this study proves successful, it could demonstrate C-Scan’s clinical (and commercial) potential. This announcement also comes just a few days after the UK NHS revealed plans to study the similar PillCam system, suggesting that the overall imaging capsule segment is headed in a very positive direction.
  • Micro-Ultrasound, an mpMRI Alternative: Micro-ultrasound–guided prostate biopsies detect cancer comparably to multiparametric MRI-guided biopsies (the current standard) and could be “an attractive alternative.” That’s from a recent Journal of Urology study review (18 qualitative, 13 quantitative, 1,125 participants) that found micro-US and mpMRI achieved similar prostate cancer detection rates (0.99 overall, 1.05 for grade ≥2 PCa, 1.25 for grade ≥3 PCa).
  • Samsung & Intel’s Nerve Guidance Tool: Samsung Medison and Intel announced a collaboration to develop NerveTrack, an Intel OpenVINO toolkit-based ultrasound tool that helps anesthesiologists identify nerves in a patient’s arm and could reduce scan time by 30%. This is Samsung’s fourth OpenVino-based ultrasound solution, following a trio of obstetric solutions announced with Intel last fall.
  • Flawed COVID AI: The COVID pandemic brought a wave of imaging AI studies focused on diagnosing and/or measuring COVID-19 infections, but none of these studies are likely to make it to clinical use. That’s from a study review published in Nature Machine Intelligence (n = 62 COVID AI papers, selected from 2,120 total papers) that found no studies that met all of their criteria (1. Documented and reproducible methodology; 2. Followed ML development best practices; 3. Sufficient external validation). These methodology lapses led to “highly optimistic reported performance” but no clinically valuable products.
  • PA’s Suboptimal Notifications: A new UPenn study found that the language used in Pennsylvania’s abnormal imaging results notifications are “suboptimal” for patients of any education level and should be improved. The researchers produced four notifications (6th & 12th grade reading level, with & without infographics) and randomly delivered each of the notifications to 909 adult volunteers. Of the 821 volunteers who participated in a follow-up survey, only 60% understood all three questions related to the notifications’ subject, next steps, and process for obtaining reports and just 52% understood that the notifications conveyed abnormal imaging results.
  • Imbio RV/LV Analysis FDA: Imbio announced the FDA approval of its RV/LV Analysis solution, which analyzes CTPA exams for potential right ventricular dilation by measuring heart ventricle diameter ratios. The RV/LV Analysis solution adds quantitative findings to patient imaging studies that can be used to assess various pulmonary conditions.
  • Questioning AI: There’s only “sparse evidence” that machine learning-based clinical decision support systems (CDSSs) actually improve clinicians’ diagnostic performance, but there’s plenty of evidence of AI bias, so healthcare providers should remain cautious when considering adopting CDSSs. That’s from a JAMA study review (n = 37 studies) revealing that out of 107 statistically significant study results, only 54 improved with CDSSs (4 decreased, 49 no/unclear change), while none of the studies that were performed in representative clinical settings showed that CDSSs actually improved diagnosis. The authors clarified that ML still could improve diagnosis, but it will take more and better studies to scientifically prove it.
  • Viz.ai’s $71M: Viz.ai just announced $71m in Series C funding (bringing total to >$150m) that it will use to expand its Intelligent Care Coordination A.I. platform to new acute care areas (e.g. cardiology, pulmonary, and trauma) and to bring the company into new global markets. Although Viz.ai became best known for its 2020 NTAP, Viz.ai’s strategy is all about using AI and mobile technology to coordinate care, and it seems it’s VC partners see big potential in expanding this strategy beyond stroke care.
  • Contrast & Car Crashes: A new JACR editorial criticized the policy requiring radiologists to be onsite at outpatient imaging facilities to monitor contrast administration for severe reaction. Because severe contrast reactions only occur in up to 0.04% of patients (1 in 170k risk of death), most radiologists have little or no hands-on experience handling contrast reactions, but they do face increased risk of dying themselves while driving to these outpatient shifts. The authors suggest that radiologic technologists could and should be trained to handle contrast emergencies, allowing radiologists to focus on interpreting exams and stay off the roads.
  • RADLogics Results: When the Moscow Center for Diagnostics & Telemedicine integrated RADLogics’ COVID-19 CT image analysis algorithm, they reduced radiologist interpretation times by an average of 7 minutes (30%) without negatively affecting accuracy. That’s from their large-scale study (128k chest CTs, 36k CTs w/ RADLogics solution, 570 radiologists, 130 hospitals) comparing the results and workflow of radiologists with and without RADLogics AI support.
  • Delayed CPT Adoption’s Big Costs: A Duke University study detailed how one academic radiology practice’s delayed moderate sedation (MS) CPT code adoption cost it $363k in lost revenue. The researchers reviewed billing and reimbursement data from January 2017 through April 2019 (28 months), finding that MS wasn’t appropriately coded until June 2018 (17 months at $21k/mo). The study attributed the mishap to lags in education and coordination across the reporting chain, calling for radiologists to intervene to prevent future CPT code adoption delays.
  • ProFound AI 3.0: iCAD announced the FDA approval of ProFound AI Version 3.0 for DBT, adding up to 10% higher specificity, 1% greater sensitivity, and 40% faster processing. iCAD emphasized how the new ProFound AI version was able to significantly improve specificity (fewer false positives) without sacrificing sensitivity (no increase in false negatives).

The Resource Wire

– This is sponsored content.

  • The phrase “prevention is better than cure” was made famous 520 years ago, and it’s a core part of Zebra Med’s strategy today. Learn how and why Zebra Med is shifting its emphasis to population health as we head into 2021.
  • If you missed United Imaging a couple of months ago at RSNA, you missed their well-attended virtual presentation, Beyond the Status Quo. In part one of that video, they cover their product launches.
  • CMS continues to modify its Quality Payment Program (QPP) and MIPS policies due to the COVID-19 public health emergency, but these changes and their timelines can be hard to manage. Check out this Healthcare Administrative Partners post detailing how these policies have changed and how radiology practices can react.
  • This Bayer case study details how radiation benchmarking programs can help push CT dose exposure reduction initiatives from achieving compliance to driving quality.

You might also like

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!