No AI Resistance

“This is not going to be a resistance kind of thing . . .”

Stanford’s Matthew Lungren, MD MPH on radiologists’ growing embrace of AI (definitely no resistance).




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.
  • 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.



The Imaging Wire


MCG Breakthrough

The FDA granted Genetesis’ CardioFlux magnetocardiography (MCG) imaging system Breakthrough Device Designation for the diagnosis of myocardial ischemia and infarction in patients with suspected acute coronary syndrome. This could be a big deal for Genetesis and the MCG modality, so here are some details:

  • About CardioFlux – CardioFlux leverages room temperature magnetometers to measure the heart’s magnetic fields. The MCG system gained FDA approval as a general tool in 2019, while the new Breakthrough Device Designation (and its reimbursement implications) is a key step towards more widespread clinical use in 2021.
  • Breakthrough Criteria – Genetesis earned its new FDA designation by showing that CardioFlux could be a safer (non-invasive, no radiation, no contrast), more precise, lower cost, and faster (90-seconds) way to diagnose myocardial ischemia and infarction in the emergency room.
  • A Genetesis Milestone – CardioFlux’s FDA designation is a major milestone for Genetesis, after gaining solid fundraising ($36m so far) and research momentum over the last few years. The designation also represents yet another milestone for Dallas Mavericks owner (and Shark Tank shark) Mark Cuban who was part of Genetesis’ Series A and B rounds, and somehow got the Atlanta Hawks to trade him Luka Doncic for Trae Young.
  • The Takeaway – Bringing a new modality into mainstream use isn’t easy, and Genetesis has a lot of work to do, but this Breakthrough Device Designation is a sign that CardioFlux is on the right track.

The Wire

  • Surprise Billing Surprise: It looks like surprise medical billing legislation might finally happen, as a new surprise billing plan reportedly has the support it needs to go through (Dems & GOP, congressional leaders, major committees) plus an inclusion in the 2021 spending bill. The final structure should please hospitals and doctors (it starts with negotiations), while insurers should like the arbitration step for cases that aren’t settled within 30 days (those cases would rely on in-network averages). Now that’s how you compromise! More importantly for patients, the legislation would finally end surprise billing for emergency care, most out-of-network care at in-network facilities (e.g. radiology), and air ambulances (but somehow not ground ambulances).
  • Headache Imaging Way Up: New research out of Emory (n = 18.1m patients) revealed that advanced neuroimaging utilization for ED patients with non‐traumatic headaches nearly doubled from 18.6% of visits 2006 to 34.8% in 2014. Emergency headache imaging utilization rates were highest among adults, commercially insured patients, male patients, urban hospitals, hospitals in the Northeast and Southern U.S., and on weekdays.
  • Butterfly’s Biplane: Butterfly Network’s Butterfly iQ+ is now “the world’s first” POCUS system to support biplane imaging, allowing users to simultaneously view short axis and long axis for plane needle-guided procedures. The new biplane imaging feature bolsters what appears to be Butterfly’s increased focus on needle visualization, as the Butterfly iQ+’s October launch announcement similarly highlighted its new Needle Viz technology.
  • Boston Children’s Explainable COVID AI: DarwinAI and Red Hat are collaborating with Boston Children’s Hospital to develop a suite of open-source explainable AI tools for CXR-based COVID-19 detection and risk stratification. The solution combines DarwinAI’s open-source explainable AI solution (COVID-Net) and Red Hat’s image sharing and GUI technologies to make the tools available across Boston Children’s imaging network.
  • 45M Images Exposed: Digital risk management firm CybelAngel identified over 45 million medical images containing patients’ personal info (identifiable and/or medical info) that are “freely accessible on unprotected servers.” It of course took some work to find all these “freely accessible” images, as CybelAngel scanned about 4.3 billion IP addresses to track down 2,140 unprotected imaging servers. However, CybelAngel still accessed all these files without using any hacking tools and it follows a number of other firms to make similar discoveries (here are other examples).
  • FFRct Advantage: A new JAMA-published study suggests that anatomic testing (CTA, CTA + HeartFlow FFRct) is less costly and more effective for diagnosing patients with low-risk stable chest pain compared to functional testing (stress testing). Using data from the PROMISE trial (n = 10k patients), the study found that these imaging-based anatomic approaches led to an average of 6 more “perfect health” months and were more cost-effective in over 65% of patient scenarios, calling CTA+FFRct the “dominant” option (it was both less costly and more effective). There’s way more to this very detailed study, so check it out if you’re interested.
  • RSIP’s Cardiac POCUS Module: RSIP Vision introduced its new POCUS cardiac diagnostic module, intended to streamline Parasternal Long Axis (PLAX) view for LV size / function measurements and support valvular function evaluations. The new vendor-neutral solution joins RSIP Vision’s relatively wide range of cardiac tools (ultrasound, CTA, MRI, angiography).
  • LCS’s Unintended Consequences: A new JAMA study revealed that 16.6% of patients who underwent invasive diagnostic procedures after LDCT lung cancer screening had complications during that procedure (n = 18,887 patients, 665 w/ invasive procedures). That’s actually lower than a previous study on post-LDCT procedure complications (22.3% – 23.8%), but it’s still way higher than complication rates reported in the famous NLST trial (8.5% – 9.8%).
  • Intelerad Acquires Digisonics: Intelerad continued to add to its enterprise imaging portfolio, acquiring cardiovascular and OB/GYN PACS and structured reporting company, Digisonics. Intelerad highlighted Digisonics’ customer base (over 4k active users) and how the acquisition will help Intelerad’s enterprise imaging platform eliminate specialist silos as key benefits of the acquisition. Intelerad continues to be among the more active enterprise imaging acquirers, buying Clario Medical in 2018 (zero-footprint worklist) and Radius earlier this year (cloud tech), while Intelerad itself gained a new PE parent company at the start of 2020.
  • CT MinIP for COVID: Minimum intensity projection (MinIP) reconstruction improves CT-based COVID-19 assessments and could be suitable for routine use. That’s from a new EJR study that analyzed chest CTs from 185 patients with suspected COVID, finding that MinIP reconstructions assessed ground-glass opacity with significantly higher sensitivity (99.9% vs 95.6%), specificity (95.8% vs 86.1%), and accuracy (99.1% vs 93.8%) than standard multiplanar reformat (MPR) images. A team of six radiologists also gave MinIP reconstructions higher ratings across a range of metrics (e.g. diagnostic confidence, image quality, time efficiency).
  • Rad Tech Assistant Layoffs: Vassar Brothers Medical Center in Poughkeepsie, NY will lay off the majority of its radiologic technologist assistant team (14 out of 19) due to COVID-related financial concerns. The hospital’s imaging technologists will soon be a lot busier, taking over many of the duties previously performed by their ex-colleagues (cleaning equipment, table setup, patient transport, paperwork).
  • Making a Case for Mn-PyC3A: An MGH research team continued to build its case for the new Mn-PyC3A manganese-based MRI contrast agent, with goals of establishing it as a gadolinium agent alternative. In their latest study (this time PET/MRI & retention-focused) the team found that: 1) Mn-PyC3A eliminates from the body more efficiently than the manganese-based liver contrast agent, mangafodipir; 2) Mn-PyC3A is efficiently eliminated from normal rats and rats with renal impairment; and 3) Mn-PyC3A more completely eliminates from renally impaired rats after 7 days than gadoterate (a GBCA agent). Here’s a previous study showing that Mn-PyC3A achieves similar contrast enhancement as GBCAs and faster elimination.
  • Signify’s RSNA Takeaways: Signify Research published its key takeaways from RSNA 2020, which is worth a read. Here are the takeaways from Signify’s takeaways: AI (disappointed exhibitors, focus on AI platforms, more comprehensive solutions), Informatics (major cloud focus, some added platform tools/modules, increased focus on VNA and data management), X-ray (focus on COVID imaging, expanded image-guided lineups, new fluoroscopy additions), Ultrasound (limited launches, some new AI guidance launches but fewer than expected).
  • US DL for Soft-Tissue Mass Classification: A new study in Radiology: Artificial Intelligence detailed a deep learning model that can classify benign and malignant soft-tissue masses in ultrasound scans as well as two experienced musculoskeletal radiologists. The team used US images from 419 patients to train/test one model to distinguish malignant and benign lesions (79% accurate, 0.91 AUC) and then to train/test a second model to differentiate three common benign masses (71% accuracy).

The Resource Wire

– This is sponsored content.

  • Learn how GE-enabled rapid breast cancer diagnosis clinics are reducing diagnoses-to-consultation turnarounds “from weeks to hours,” and how they are helping patients and clinicians in this new GE Healthcare Insight.
  • This Riverain Technologies case study details how Einstein Medical Center adopted ClearRead CT enterprise-wide (all 13 CT scanners) and how the solution allowed Einstein radiologists to identify small nodules faster and more reliably.
  • This Hitachi Healthcare Americas blog details COVID-19’s recent and future impact, warning cardiac practices and clinics of an upcoming wave of patients with cardiovascular issues that worsened due to delayed treatments, followed by a “third wave” of patients who developed heart complications from COVID-19 infections. Hitachi also shared some guidance on how to manage and minimize the impact of these waves.

Get every issue of The Imaging Wire, delivered right to your inbox.

Join thousands of imaging professionals.