SonoSite and Partners | CMS’ Real Transparency

“Publicly disclosing competitively negotiated, proprietary rates will push prices and premiums higher — not lower — for consumers, patients and taxpayers,”

Matt Eyles of AHIP in response to CMS’ latest effort to create real healthcare cost transparency. Safe to say, not everyone agreed with Eyles on this.


Imaging Wire Sponsors

  • Carestream – Focused on delivering innovation that is life changing – for patients, customers, employees, communities and other stakeholders
  • Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound
  • Medmo – Helping underinsured Americans save on medical scans by connecting them to imaging providers with unfilled schedule time
  • Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter
  • Pocus Systems – A new Point of Care Ultrasound startup, combining a team of POCUS veterans with next-generation genuine AI technology to disrupt the industry
  • Qure.ai – Making healthcare more accessible by applying deep learning to radiology imaging

The Imaging Wire


SonoSite and Partners’ AI POCUS Partnership

Fujifilm SonoSite and Partners HealthCare launched a strategic collaboration to develop AI-enhanced point-of-care ultrasound technology. The collaboration will operate through the MGH & BWH Center for Clinical Data Science, while leveraging Partners’ data, computational infrastructure, and clinical expertise.

  • First Up – Partners and SonoSite’s initial collaboration will target “some of the more complex” ED procedures with the goal of building ultrasound systems that can be used by non-experts to shorten diagnosis and treatment times.
  • More Specifically – They’re focusing on developing AI-embedded ultrasound systems that can automate the segmentation of organ boundaries, measure anatomic features, and calculate physiological parameters with the end goal of making POCUS technology more accessible for emergency use and for broader use across the developing world.

With all the ultrasound AI news (here’s one from Monday) and the buzz around newer AI-enabled handheld ultrasound players, it can be hard to differentiate between what’s hot and what’s significant. The combination scale of Partners’ clinical and AI capabilities with Fujifilm SonoSite’s massive POCUS presence makes this alliance significant.



CMS Proposes Real Transparency

CMS took its biggest shot at achieving real healthcare cost transparency this week with a proposal that would force hospitals to disclose payor-specific negotiated rates. The rule focuses on 300 “shoppable services” that patients can schedule in advance, such medical imaging scans, lab tests, or childbirth.

  • Proposal – If approved, in January 2020 all 6,000-plus hospitals that accept Medicare will have to post their negotiated rates for each payor online, while adhering to certain online readability standards. Hospitals that don’t adhere would face a $300 daily fine, which many find to be way too low.
  • Fallout – To the surprise of no one, both hospitals and payors quickly opposed the proposal, suggesting that it’s bad for patients and somehow would reduce competition (debatable), while others argued that it exceeds the executive branch’s authority (more valid). Meanwhile, proponents highlight how it would move the healthcare into an open market system and suggest that opponents to this rule are really “the proponents of the status quo.”

The Wire

  • Research from a MD Anderson team revealed wide variations in breast MRI technical quality across the U.S. following a review of 100 cases by three fellowship-trained breast radiologists and one breast imaging fellow. Of the 88 cases referred from facilities in the United States, 60 (68%) had at least one technical deficiency and 10 (11%) had more than five different technical deficiencies, with the most frequent deficiencies related to artifacts (74%).
  • Dana-Farber researchers developed a deep learning tool that’s able to estimate real-world cancer treatment outcomes (e.g. disease progression and therapy response) as well as trained radiologists. The model was trained on 13,230 imaging reports (1,112 lung cancer patients) and then applied to a 109-image test set, identifying the presence of cancer and improvement/response or worsening/progression results with a >0.90 AUC. A later test using 15,000 different imaging reports was able to predict overall survival with the same accuracy as human reviewers. The researchers noted that the model could eventually annotate 2,000 patients’ imaging reports in 10 minutes, far less than the six months it would take a human curator to perform the same task.
  • ENDRA Life Sciences landed a $2.8 million private placement that it will use to fund ongoing studies, regulatory filings, and eventually the commercial launch of its TAEUS (Thermo-Acoustic Enhanced UltraSound) platform. ENDRA reports that the TAEUS platform will allow clinicians to visualize human tissue composition, function, and temperature in ways that rival CT or MRI, but at lower costs and at the point-of-care.
  • An AHRA survey (n = 300 radiology professionals) explored the use of radiologist assistants (RAs) in the U.S., finding that 26% of organizations currently use RAs, with most organizations employing either one or two full time equivalent RAs (40% one, 23% two). RAs’ most common duties were fluoroscopic and invasive procedures (79% and 66%), while responses were mixed regarding the impact of CMS’ decision to allow billing for diagnostic imaging procedures performed by RAs under direct physician supervision (45% expect related increase in RA usage vs. 40% not sure, 40% believed that they would hire more RAs vs. 40% not sure).
  • Hebrew University of Jerusalem researchers developed a quantitative MRI method that could detect molecular changes in the brain, potentially allowing clinicians to identify early signs of Alzheimer’s or Parkinson’s diseases. The team developed a tissue relaxivity approach that decodes molecular information from an MRI signal, giving insights into the molecular composition of lipid samples and region-specific molecular biomarkers that can currently only be done post-mortem.
  • In his AHRA 2019 session, The Advisory Board’s Stuart Clark forecast that outpatient medical imaging volume will grow by 7% over the next 5 years, but hedged that this growth may be disrupted by reactions from payors and patients (increased deductibles, reduced reimbursements, shifts to lower-cost methods, lower demand). Clark forecast a hefty 16% increase in outpatient ultrasound use, followed by PET (+9%), radiography (+6%), CT (+4%), and MRI and mammography (both +3%), while nuclear medicine would fall 1%.
  • A University of Pennsylvania team using iCAD’s PowerLook Tomo Detection 2.0 found that AI programs can help improve DBT accuracy and reduce radiologist reading times. The study compared the performance of a 24-radiologist group (w/ 13 breast subspecialists) reading 260 DBT examinations (w/ 65 cancer cases) with and without AI support, finding that AI improved malignant lesions detection (0.852 vs. 0.795 AUC), sensitivity (85% vs. 77%), and specificity (69.6% vs. 62.7%). The AI tool also shortened reading times (30.4 seconds vs. 64.1) and decreased recall rate for noncancers (30.9% vs. 38.0%).

The Resource Wire

– This is sponsored content.

  • Four Qure.ai abstracts have been selected for RSNA19 covering: a new metric to evaluate radiology AI models, chest X-ray TB screening, segmenting and measuring ventricular and cranial vault volumes with AI, and how clinical context improves AI performance for cranial fracture detection.
  • In this blog post, Nuance Healthcare’s Diagnostic Division GM and VP, Karen Holzberger, shared some real-world examples of how PowerShare is helping providers #DitchTheDisk and improve patient experiences.
  • In this Carestream video, orthopaedic surgeon Dr. Bryan Den Hartog presents clinical images illustrating traditional CT vs. extremity CT imaging and discusses how the image resolution in the OnSight 3D Extremity System helps in his practice.

3 Responses to “SonoSite and Partners | CMS’ Real Transparency”

  1. The Top Medical Imaging and Radiology News for September 19, 2019

    […] Fujifilm SonoSite partnered with the Allen Institute of Artificial Intelligence (AI2) Incubator to co-develop deep learning-based ultrasound image interpretation solutions. The solutions would be deployed on SonoSite’s portable ultrasound systems and target a wide range of ultrasound interpretation scenarios. Fujifilm SonoSite appears to be leaning-in on an AI-enhanced ultrasound strategy, as the AI2 collaboration comes just a few months after a similar POCUS AI alliance with Partners Healthcare. […]

    Reply

Leave a Reply

Basic HTML is allowed. Your email address will not be published.

Subscribe to this comment feed via RSS

Get every issue of The Imaging Wire

Delivered right to your inbox. We will never sell or share your information.