Nines Redefines | Optical Game Changer

“This is a pretty darn big gap in coverage that we need to make sure we are filling,”


U.S. Rep. Chrissy Houlahan on the DoD’s standard to only cover DBT screening as a secondary option for servicewomen.


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



Nines Redefines

Teleradiology startup, Nines, announced a $16.5 million Series A round and unveiled a new AI triage platform that will combine to support its mission to “redefine radiology.” Here are some details:

About Nines – Calling itself a “first-of-its-kind teleradiology practice,” Nines combines radiologists, “world-class engineering,” and data scientists with the goal of achieving breakthroughs in patient care and “build(ing) the future” of radiology.

More Than Marketing – Although this AI-enabled teleradiology messaging could be viewed as a marketing angle, Nines is backed by some big name co-founders and advisors, including folks from Mount Sinai, Stanford, GE, vRAD, and Udemy.

Nines Labs – The best evidence that Nines actually is unique is the Nines Labs team, which is made up of clinically-focused engineers and data scientists, focused on supporting radiologists’ workflows. It remains to be seen whether DIYing AI is more effective than using the many AI solutions already on the market, but this approach is interesting and could have some upsides.

Nines Software – Nines’ first software product is its Emergent Neuro Suite (in FDA review), intended to help prioritize cases based on the analysis of head CT images (for cytotoxic edema, hemorrhage, and mass effect), while other solutions may be coming.

We’ve certainly seen the major radiology and teleradiology practices increase their focus on AI, but this is a rare case where AI and data science are positioned as a core part of the practice from its inception. That fact and the Nines team’s pedigree makes it worth keeping an eye on.



Optical Game Changer

A MGH-led team developed a new optical imaging technique that can differentiate between potentially life-threatening coronary plaques and less dangerous plaques, possibly improving cardiologists’ ability to predict heart attack risk. Here are some details:

Plaque Characterization – The new intravascular polarimetry imaging technique is intended to help differentiate stable plaques from plaques that are prone to rupture and could cause heart attacks.

The Study – The study (n = 30 patients with CAD) investigated the polarization properties of coronary atherosclerotic plaques and how they correlated with plaque instability using catheterized intravascular optical coherence tomography.

The Results – Of the 30 catheterizations, the team captured 342 cross-sectional plaque images and 244 images from the fibrous caps of the atherosclerotic lesions responsible for stability or risk of rupture. The researchers were able to classify the coronary cross-sections into seven categories (from normal to ruptured) and then define the polarity differences between the different categories.

Significance – The researchers called this technique a potential “game-changer for cardiologists and their patients” due to its ability to provide insights into coronary arterial atherosclerotic lesions with quantifiable imaging data. This first human pilot is a key milestone towards that potential “game-changer” becoming a reality.


The Wire

  • Zebra Medical Vision announced the FDA clearance of its HealthCXR product (its 4th FDA clearance), an AI-based solution intended to identify and triage pleural effusion in chest X-rays and alert physicians if pleural effusion is suspected. Zebra-Med also announced that it is now part of the FDA Digital Health Software Precertification (Pre-Cert) Pilot Program, intended to help support the development of a future regulatory model for software-based medical devices.
  • A pair of U.S. senators recently introduced the Better and Robust Screening Today Act, requiring the Department of Defense to cover DBT as a primary screening option for servicewomen covered by Tricare (DBT currently only covered if something is spotted in an initial DM scan). The Defense Health Agency already plans to begin covering DBT screening on a provisional basis from 2020 to 2024 (with prior authorization) while more research is done on DBT’s benefits, but the new Act intends to make DBT screening a permanent/primary part of Tricare coverage by 2021.
  • Hologic announced the official launch of its Unifi Workspace breast health diagnostics reading solution intended to support informed decision making and improve workflow efficiencies. The multimodality workstation (integrates MRI, US, mammography) features pre-fetching capabilities, MRI tools (MR CAD with subtraction, colorization and wash-in wash-out curve, and MR motion correction), and multimodality-enabled hanging protocols for customization.
  • Arterys officially entered the AI marketplace arena at RSNA this week, launching Arterys Marketplace, an open platform used to develop and distribute imaging AI tools. The Arterys Marketplace allows users to translate models into clinical applications, while giving developers tools to deploy models to an online environment for distribution, testing, or sharing. The Arterys platform has been expected for some time and just added MaxQ AI as its first publicly announced partner.
  • A review article from U of Wisconsin and UCSF detailed recent AI successes in MSK radiographic imaging (estimating pediatric bone age, detecting fractures, and assessing osteoarthritis) that may soon be clinically available. The article also highlighted evidence that other CT and MRI deep learning applications are feasible (internal derangement, metastatic disease, infection, fractures, and joint degeneration), but further development is still required before AI models can reliably interpret these complex modalities for MSK applications. The team also encouraged evaluation through large prospective studies before any MSK AI applications are implemented into clinical practice.
  • GE Healthcare launched its Edison Developer Program, an expansion of GE’s Edison ecosystem that provides developers with a range of services to help speed up product development and then scale and deploy their solutions across GE’s customer base. GE highlighted the deployment and adoption challenges associated with one-off AI applications, suggesting that Edison addresses this issue by providing a range of applications integrated into existing GE Healthcare offerings, which is indeed the main advantage for the big imaging players.
  • Although imaging AI startups landing seven figures in seed funding isn’t all that unique these days, Rad AI’s recent funding round earned some headlines due to the company’s high-profile backer (Google’s AI-focused venture fund Gradient Ventures) and the fact that it’s co-founder (Dr. Jeff Chang) began med school at the age of 16 and was the youngest radiologist in U.S. history. Rad AI is a lightweight app that runs on radiologist workstations, using radiologists’ voice dictation notes to create an automated summary with the same verbiage that radiologists use, saving the typical radiologist an hour of work each day. The company will use the seed round to grow its engineering team and fund market expansion.
  • Research from a Boston University-led team found that gadolinium is retained in parts of the cerebral cortex of humans and rats. The researchers examined the cerebral cortex of two human donors with repeated GBCA exposure (1mo & 3 doses, 5 months & 15 doses) and two human donors without GBCA exposure and similarly examined Sprague-Dawley rats with and without GBCA exposure (n=6 with, 6 without). The researchers identified gadolinium retention in the GBCA-exposed human brains’ cerebral cortex, pia mater, and pia-ensheathed leptomeningeal vessels (but not the GBCA-naive human brains), while similarly finding signs of gadolinium retention in the exposed rats. However, the study still did not provide insights into whether there are confirmed health risks from gadolinium retention.
  • Konica Minolta Healthcare is unveiling a new range of products and solutions at RSNA. The company debuted a new version of its KDR AU Advanced U-Arm equipped with its high-priority DDR moving X-ray technology (FDA pending), its forthcoming Exa-based and web-based Rede PACS2 PACS for specialty practices (orthopedic, urgent care, and family practice), and a new RFID-based solution for secure user authentication to help users securely log into select Konica Minolta systems.

The Resource Wire

  • Here are the top 5 reasons Qure.ai should be on your RSNA must-see list: 1. AI you can trust (just see the peer-reviewed papers and RSNA presentations); 2. AI that works (Qure.ai is integrated with multiple PACS, marketplace, and implementation partners); 3. AI with global users and deployments (solutions in 16 countries); 4. Perspective on what’s new for AI (see what’s next on Qure.ai’s list); 5. See AI built by a cool and fun bunch of innovators (they’re a passionate group of scientists, R&D experts, physicians and data analysts).
  • Catch the Nuance team present at RSNA19 including Dr. Woojin Kim sharing how to spot the AI “hits and misses” for your environment (Monday 12:30pm), Karen Holzberger on the end to end story of AI in Action (Tuesday 2pm), and Dr. William Boonn on the challenges and opportunities of radiology workflow and AI (Thursday 8:30am).
  • By partnering with Medmo, imaging centers can keep their schedules full and their equipment busy. Here’s where to learn more and get started.

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.