Neusoft’s New Probes | AHA’s Disruption Fight | Another AI “Win”

“We are seeing historic—if not unprecedented—change across the entire healthcare landscape.”

Rick Pollack, president and CEO of the American Hospital Association, on the potentially-disruptive entrance of companies like Amazon and Google into healthcare and the need for traditional providers to adapt.


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.
  • Pocus Systems – A new Point of Care Ultrasound startup, combining a team of POCUS veterans with next-generation technology to disrupt the industry.



The Imaging Wire

Neusoft’s New Probes
Neusoft Medical Systems announced its acquisition of South Korean ultrasound probe company, Humanscan. Besides boasting the most literal company name in the industry, Humanscan is known for its single-crystal cardiac probes, which it supplies to “many global first-class ultrasound manufacturers.” Neusoft expects the acquisition to improve its high-end ultrasound R&D and production capabilities and “enhance the overall competitiveness of (its) ultrasound business” in China and globally. It’s also worth noting that Neusoft is on a bit of an acquisition spree, leveraging investments in Humanscan, Bio Imagem (Brazilian imaging distributor), Whitman Biotech (Chinese in vitro diagnostic company), and an unnamed “manufacturer of core materials on CT detectors,” to expand its product and market coverage over the last two years.

Europe’s Enterprise Imaging Problem
Signify Research shared some tough love with Europe’s radiology departments, criticizing their preferences for best-of-breed radiology IT and calling for a philosophical and structural shift toward true multidepartment enterprise imaging. The firm found that only 17% of Europe’s radiology IT spending was for enterprise solutions in 2018, which was good enough for second place globally, but far behind North America, and not expected to grow significantly over the next five years.

Enterprise Imaging Regional Revenue Share

  • North America – 33% in 2018 / 46% in 2022
  • Western Europe – 17% in 2018 / 31% in 2022
  • E. Europe, Middle East, Africa – 14% in 2018 / 29% in 2022
  • Asia Pacific – 8% in 2018 / 24% in 2022
  • Latin America – 6% in 2018 / 12% in 2022

Signify notes that even this 17% enterprise imaging adoption rate is misleading, as most European EI implementations focus on connecting different radiology departments, and few European providers can actually manage and share information from different clinical departments (oncology, cardiology, etc.). The firm acknowledged the popular belief that this is due in part to Europe’s slow EMR adoption, but placed more of the blame on European providers’ use of proprietary PACS software without solid interoperability capabilities.

Looking forward, Signify suggests that the rise of large European health networks will make the “best-of-breed” radiology IT approach less feasible, while noting that without a major change European hospitals may “fall behind as the shift to interdisciplinary care, centralized procurement, and AI intensifies.”

AHA’s Disruption Fight
The American Hospital Association placed a clear focus on the need for hospitals to evolve at last week’s AHA Forum and in recent blog posts. The industry group specifically warned of the potential disruption brought by the ongoing push from tech and retail leaders into healthcare (Amazon, Apple, Google, Walmart, etc.) as well as from traditionally adjacent companies’ shift towards healthcare delivery (CVS, Optum, etc.). AHA described a healthcare climate that’s ready for disruption with “more engaged consumers, advances in technology and new (well-funded) entrants” and an unfair playing field for hospitals (they care for “everyone” including the uninsured, 24×7).

The AHA called for healthcare providers to fight these disruptors by improving patient and community convenience and shifting to a model that rewards value over volume. AHA appears to suggest that consolidating is one proven path towards this goal, although it’s going to take more than M&A to truly meet patients’ changing needs and fight-off future disruptors.

Fujifilm Continues AI Investments
Fujifilm and Osaka University announced a medical imaging-AI joint research partnership, combining Fujifilm’s funding with Osaka University’s research and image database to develop new AI solutions. Osaka University will build a “massive database of (annotated) medical images” captured at its medical center, using the images to develop advanced AI solutions, including an AI-based medical diagnostic imaging support system and AI solutions that can explain thought processes behind determining lesion locations (Explainable AI). We’re starting to see a pattern here. Fujifilm has signed a number of AI partnerships with both academic institutions and startups over the last year (here’s its partnerships with Lunit, LPixel, Indiana University), making this Osaka University partnership typical for Fujifilm but also a reminder that the company is actively investing in its AI future.

Another AI “Win”
Here’s another “AI beats radiologists” story to get our radiologist readers worked up. A team of South Korea researchers found that a Lunit-developed deep learning algorithm can discriminate between abnormal and normal chest X-rays better than radiologists and can improve performance when used as a second reader.

The researchers trained the automatic detection algorithm (DLAD) using 54,221 normal chest X-rays and 35,613 abnormal X-rays (pulmonary malignant neoplasms, active tuberculosis, pneumonia, or pneumothorax), and then used 486 normal chest X-rays and 529 abnormal X-rays for external validation. The DLAD indeed beat a team of 15 physicians in image classification and lesion localization:

  • DLAD – Image classification 0.979 AUROC / lesion location 0.972 AUROC
  • Thoracic radiologists – Classification 0.932 AUROC / lesion location 0.0907 AUROC
  • Radiologists – Classification 0.896 AUROC / lesion location 0.870 AUROC
  • Non-radiology physicians – Classification 0.614 AUROC / lesion location 0.761 AUROC

The researchers were careful to suggest that DLAD’s primary value is in its ability to help improve radiologist efficiency by prioritizing X-rays with suspicious abnormal findings, which is a pretty proven AI use case. However, the study did note that DLAD has the potential for stand-alone use in select clinical situations, which would be a bit more groundbreaking.


The Wire

  • A team of Swiss researchers found that biplanar hip joint X-rays reconstructed in 3D or 2D can support hip coverage assessments as well as CT-scans, making X-ray a potential alternative given its cost and dosage advantages. The study (n=50) found similar X-ray and CT coverage calculations for anterior 2D (21.1% X-ray, 23.8% CT avg.), posterior 2D (54.2% X-ray, 61.7% CT avg.), and global 3D (46.5% X-ray, 45.6% CT avg.).

  • Ebit (Esaote Group) and DiA Imaging Analysis announced a partnership to offer DiA’s LVivo Cardiac Toolbox as part of Ebit’s Suitestesna CVIS (Cardiovascular Information System ) PACS. LVivo Toolbox (includes EF measurement, biplane wall motion, and automatic strain analysis apps) analyzes cardiac ultrasound images based on more objective and reproducible information (vs. manually & visually). The Suitestesna CVIS integration will allow support of DICOM clips from all ultrasound devices evaluated through the PACS, thus making LVivo AI analysis available to all users. This appears to be DiA Imaging Analysis’ first PACS partnership, although the company does place a focus on integration partnerships, offering its LVivo EF app with GE’s Vscan handheld ultrasound system since last fall.

  • A paper from a NYU Langone Health team detailed how customizations to a new outpatient MRI center improved MRI turnaround times by an average of 5 minutes and 28 seconds per patient, compared to its existing imaging centers. The new imaging center was designed with a number of efficiency-focused elements including: dockable tables, dedicated patient prep rooms, coil storage in prep rooms, two-door MRI rooms, optimized scanner positioning, and duplication of the most frequently used coils. As a result, the new imaging center was able to reduce its MRI room turnaround times to a median of 115 seconds for patients who were ready for their scan (vs. 430 seconds) and 141 seconds for all patients (vs. 481).

  • Bloomberg put a spotlight on the growing appeal of telemedicine among physicians, covering the advantages (schedule flexibility), technologies (apps and platforms), and systemic changes (increased coverage, emergence of specialized telehealth providers) that are making this still-niche method more mainstream. There was also an interestingly high level of emphasis on the insufficient face-to-face interactions in traditional healthcare settings, suggesting that telemedicine is no worse and, in some cases, may be better than the hospital. The article may be more informative for the general public than people in medicine, but coverage like this is evidence of telemedicine’s mainstream shift, whether its news to you or not.

  • Doximity’s third annual Physician Compensation Report (n=90k physicians) ranked radiology as the 10th highest-paying specialty, with a $429,000 average annual salary (neurosurgery was #1 with $616k). However, radiology had the fourth widest gender pay gap, with male radiologists ($442k avg.) earning 21% more than female radiologists ($349k).

  • Life Image announced a partnership with swyMed that combines swyMed’s telestroke solution with Life Image’s clinical image exchange, giving swyMed users access to all relevant medical records, diagnostic imaging, and other clinical data through the Life Image exchange. The partnership also expands swyMed’s ability to connect with neurologists and primary stroke centers that are already part of the Life Image network, while providing Life Image’s hospital customers with a new telestroke solution within the Life Image Interoperability Suite.


The Resource Wire

This is sponsored content.

  • Focus Ultrasound Foundation scientists published a study validating the performance of their Kranion software, intended to allow clinicians to plan and visualize hypothetical transcranial focused ultrasound procedures. Using MRI and CT images from 28 patients, the study found that Kranion was effective at predicting average temperature rise during ablation.

  • Did you know that imaging patients are most likely to no-show for their procedures on Mondays and Saturdays? By partnering with Medmo, imaging centers can keep their schedules full, despite the inevitable Monday no-shows.

  • This Innovatus Imaging blog highlights the company’s Ultrasound School program, designed to train students on managing imaging device inventory, repair methodologies, and device maintenance.



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.