CXR-risk | U.S. Lagging | Konica and Shimadzu’s DDR Deal

“Solve what I whine about – & you’ll be multi-millionaires.”

A tweet from radiologist, Saurabh Jha, giving AI developers millions of reasons to automate radiologists’ pain points, even if it’s not related to interpretation.


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



CXR-risk

A team of Boston and Germany-based researchers developed a deep learning program that can predict a patient’s long-term mortality from a single chest radiograph (no additional demographic or clinical information). Here’s how the new CXR-risk convolutional neural network was able to produce risk scores that correlated with long-term mortality rates and what this may mean for healthcare:

  • Study – The CXR-risk study used 41,856 chest X-rays for CNN development and 10,464 for testing (sourced from the PLCO trial, 55-74 years, asymptomatic nonsmokers and smokers), while also using 5,493 images from the NLST for external testing (55-74 years, heavy smokers).
  • Results – The CNN scored each radiograph based on mortality risk (very low, low, moderate, high, and very high) and found that the “very high-risk” group had a 53% 12.2-year mortality rate with the PLCO set and a 33.9% 6.3-year mortality rate with the NLST set. These mortality rates are 18- and 12-times greater than the “very low-risk” group (3.8% and 2.7% mortality rates).
  • Significance – Noting that chest X-ray is the most common diagnostic imaging test, a CNN like CXR-risk could help patients and physicians understand mortality risks and “consider prevention, screening, and lifestyle interventions,” while healthcare systems could use risk scores to properly allocate resources.
  • Future – The researchers suggested that additional demographic or clinical information could be combined with chest X-rays in future deep learning models to achieve even more precise scores.



U.S. Lags in Digital Health, Leads in DHR

The Philips Future Health Index 2019 reveals that the U.S. is among the top digital health records (DHR) users and is a big AI spender, but is below the 15-country average for most emerging digital health technologies. Here are some of the major takeaways:

  • 84% of U.S. healthcare professionals use DHRs (e.g. EMRs and EHRs), but that’s not necessarily a good thing. U.S. clinicians viewed DHRs negatively from a patient (53%), workload (61%), and professional satisfaction (44%) perspective.
  • Only 46% of U.S. healthcare professionals use telehealth (vs. 61% 15-country avg.).
  • Just 33% of U.S. healthcare professionals use AI-powered solutions (vs. 41% in Germany, 85% in China).
  • Despite lower AI usage, the U.S. spends $0.06 per capita on AI for preliminary diagnosis (vs. $0.03 in Germany, $0.02 in China).
  • As expected, China had a massive 60% share of global funding for AI companies, followed by the U.S. (29%) and India (5%).
  • https://www.usa.philips.com/a-w/about/news/archive/standard/news/press/2019/20190717-philips-future-health-index-2019-finds-us-among-leaders.html



Konica Minolta and Shimadzu’s DDR Deal

Konica Minolta Healthcare Americas and Shimadzu Medical Systems USA announced a partnership to co-market Konica Minolta’s Dynamic Digital Radiography (DDR) technology, which allows clinicians to view anatomical movement from rapidly-acquired X-ray images. The companies will support the DDR image processing technology with Shimadzu’s RADspeed Pro radiographic imaging system, continuing their existing DR partnership in the U.S. and globally-expanding their DDR collaboration that began in Japan.

In addition to ensuring that DDR has a compatible X-ray system, this partnership should significantly expand DDR’s U.S. channel and customer base, both of which would be necessary in order for DDR to become a mainstream healthcare option. Needless to say, this partnership is a major milestone for Konica Minolta, which has placed significant emphasis on DDR’s potential over the last year, and would certainly benefit if DDR breathes new life into digital radiography.



Density Notifications Not Driving Breast Ultrasound, Either

Research from a team of Seattle area radiologists suggests that breast density reporting laws aren’t driving the breast ultrasound use that lawmakers and density notification supporters may have expected.

  • The Study – Using National Ambulatory Medical Care Survey data from 2007 to 2015 (12,787 visits, asymptomatic women, 40-74 years), the team measured state-level breast ultrasound rates before and after adoption of notification laws.
  • The Results – The study found that screening ultrasound orders remained low throughout the 9-year period (just 3.3% of visits vs. 22.1% for mammography) and state-level orders did not change after adoption of density notification laws.

The researchers believe that notification laws still make it more likely that patients and physicians will discuss whether ultrasound screening is necessary, but this study suggests that the laws do little to drive orders, at least initially. Although this study slightly contradicts previous research from Emory that found a 1% increase in breast ultrasound after states adopted density notification laws, it also comes just a few weeks after a separate study found that DBT screening didn’t change after North Carolina adopted its own density notification law in 2014.


The Wire

  • Those wondering how long Radiology Partners could maintain its current expansion pace don’t have to worry about that for a while, as the mega practice just landed $700 million in long-duration capital from Starr Investment Holdings (SHI). To put that into context, RP’s massive growth in recent years was fueled by $234 million and $200 million funding rounds in 2017 and 2018. RP will use its new funding to support “future practice and hospital system partnerships” and “clinically-focused technology and innovation.” SHI’s managing director, Geoff Clark, will become just the second institutional investor on RP’s 11-member board (joining NEA).
  • Hologic announced a partnership with MagView to develop Unifi EQUIP, a forthcoming solution that helps mammography facilities comply with the FDA’s EQUIP guidance and be prepared for inspections. Like Hologic’s other acquisitions and partnerships, the MagView alliance is intended to allow Hologic to support its clients across the breast care continuum.

The Resource Wire

– This is sponsored content.

  • This Qure.ai video details how its qXR TB chest X-ray tool is transforming tuberculosis screening.
  • Carestream’s AHRA booth (number 503) spotlights its ImageView software and mobile imaging portfolio, including its CARESTREAM DRX-Revolution Nano Mobile X-ray System, CARESTREAM DRX-Transportable System/Lite, and new CARESTREAM DRX Plus 2530C Detector.
  • How much does a CT scan cost? According to Medmo, that depends. Scans made with the exact same device on the exact same body part could cost $225 at one facility and $2,500 at another. Medmo also provides some advice to make sure patients don’t pay too much for their scans, including using the Medmo Marketplace, where the average CT costs between $225 and $700.

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.