Philips Acquires Carestream HCIS | Ultrasound Hack | FDA’s Alternative Summary

“I don’t want to sound overdramatic here, but it seemed like a cover-up.”

Surgeon, Dr. Douglas Kwazneski, after learning discovering the FDA’s hidden database of a million-plus adverse medical device events and malfunctions.


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


Philips Buys Carestream Health IT
Royal Philips acquired Carestream’s healthcare information systems (HCIS) business for an undisclosed sum, combining two of the segment’s top brands, and bolstering Philips’ enterprise imaging customer base, geographic reach, and technology portfolio. This is a pretty big deal, as it combines two of the enterprise imaging market’s leading players (#s 4 and 5) and represents a major consolidation milestone for the segment. The deal (if executed well) could significantly improve Philips’ already strong health IT business, as the company specifically highlighted Carestream’s cloud-based enterprise imaging informatics platform and complementary geographic footprint as the prizes of the acquisition. Meanwhile, Carestream emerges from the acquisition with a much more-exclusive focus on its medical imaging hardware business (along with its dental and industrial films, non-destructive testing, and precision coating operations), while earning itself and Onex an unspecified amount of cash from selling HCIS.

Ultrasound Hack
Last week’s RSA security conference revealed a proof-of-concept cyberattack where researchers were able to easily hack into a hospital’s ultrasound, download and manipulate files, and execute ransomware. This medical hacking version of hitting for the cycle came as a result of a collaboration between Check Point Research and a major Tel Aviv hospital. The Check Point hackers entered the hospital’s network, located and accessed the ultrasound, and exploited a well-known Windows 2000 vulnerability to gain control of the device. Check Point suggested that the growing number of connected medical devices, slow equipment replacement cycles, lagging hospital IT resources, and lucrative market for patient data have created a perfect storm, calling medical devices the “weakest link” in healthcare cybersecurity.

The FDA’s Alternative Summary
A Kaiser Health News expose revealed that the FDA has maintained an “alternative summary reporting” repository of over 1.1 million adverse medical device events and malfunctions, while keeping these reports out of its MAUDE public database. The FDA assured that the “well known” alternative database is used to review adverse events internally. However, a statement from former FDA deputy commissioner, Dr. Robert Califf, suggests that it’s been pretty secret until now (he “never heard anything about it”) and some FDA watchers suggest that this is evidence that the agency “handed over their ability to oversee the safety and effectiveness of these devices” to device makers.

AI’s Noninferiority Complex
Here’s another “AI versus radiologists” story that’s going to upset and interest an equal number of readers. A team of Dutch researchers found that an AI algorithm can detect breast cancer in digital mammograms as well as many trained radiologists, with performance that was “statistically noninferior” to the average radiologist, but not as good as the best radiologist. The researchers tested the AI program against 101 actual human radiologists who read 2,652 exams (653 malignant) for a total of 28,296 independent interpretations. The AI solution beat the radiologist group’s AUC score (0.840 vs. 0.814 avg, out-performing 62 individual rads) and achieved higher sensitivity than 55 radiologists, but was “consistently lower” than the best radiologist. These results led the researchers to suggest that AI may have the greatest benefits serving as first or second reader for less experienced radiologists, which is a logical starting point, but unlikely to be the end point for AI.

Hitachi Healthcare’s 2019 Rollout
Hitachi Healthcare’s ECR 2019 booth brought the European debut of a range of new systems from its MRI, Ultrasound, and CT product lineups, continuing global rollouts that began in Japan last year.

  • Hitachi unveiled the new ECHELON Smart Plus 1.5T MRI, bringing improvements to patient comfort (reduced noise via SmartComfort), image quality (via SmartQuality), operational productivity (via SmartSpeed and SynergyDrive), and footprint (SmartSpace). The ECHELON Smart Plus will likely make its way to the US later this year, where Hitachi already has a pair of Echelon family 1.5T MRIs.
  • Several months after updating its ultrasound lineup in Japan, Hitachi Healthcare introduced the entry-level ARIETTA 50/50LE series and mid-range ARIETTA 65 to the European market. The ARIETTA 65 combines many of the features found in Hitachi’s flagship ARIETTA 850 (image processing tech, ergonomic design, probes) at a lower cost, while the ARIETTA 50 (standard monitor arm, 3 transducer connectors) and better-equipped ARIETTA 50LE (flexible monitor arm, 4 transducer connectors) lead with their ease-of-operations, compact form factor, and entry-level pricing.
  • After launching in Japan last year and making its US debut at RSNA, the 64/128-slice SCENARIA View CT reached Europe last week, leading with its 80cm wide bore (vs. 75cm) and Hitachi’s IPV image reconstruction solution (improves chest resolution at an unchanged dose).



The Wire

  • A team of University of Zurich researchers found that Advanced Breast-CT’s recently CE-cleared nu:view spiral CT breast scanner’s ability to produce high-quality images at low radiation doses and without compression, could make it a superior option for breast lesion detection and characterization. In the study (n=12 women, 15 breast CT exams), a pair of radiologists identified the patients’ three masses and 11 of the 12 existing calcifications. The team also measured a 5.08 mGy mean radiation dose (vs. DM’s 3.7), while 75% of the patients ranked the breast CT exam as a 4 out of 5 for comfort (5 = most comfortable).

  • A team of MIT researchers developed a near-infrared optical imaging technique that can spot cancer cells “far deeper than any existing biomedical optical imaging technique” (8cm vs. 3.2cm) and can identify much smaller cancerous tumors than traditional modalities such as CT and MRI (0.1cm vs. 1cm). This combination may make the new technology (code named Dolphin) ideal for the diagnosis of cancers that are typically difficult to detect in their early stages, such as ovarian cancer. The new technique inserts a 0.1mm fluorescent probe into the patient’s digestive tract (initially rats) and uses hyperspectral and diffuse imaging to capture high-resolution data from the entire body.

  • A University of Alberta study put Microsoft’s HoloLens to the test, finding that doctors using the AR system to superimpose spinal X-rays onto patients’ backs were able to locate vertebrae levels with 73% accuracy and 77% repeatability (n = 13 patients), suggesting that AR may represent an effective way to identify underlying anatomy. The researchers highlighted AR/MR’s potential impact on clinical care, including improving doctor-patient interactions, but also suggested that these tools could prove useful in educational settings.

  • Varex Imaging launched its new Service Solutions Group, staffed with teams of account managers and support engineers focused on supporting 3rd party servicers. The new group will roll-out a range of benefits in the coming months including new warranty and service packages, technical and installation support, expanded logistical support (regional inventory and repair), lower-cost DR and DRF upgrade packages, wider X-ray tube assortments, and expanded technical support.

  • Researchers from Rensselaer Polytechnic Institute (RPI) developed a deep learning approach that could improve the speed and quality of molecular imaging. The team leveraged compressed sensing-based imaging, a signal processing technique that creates images based on a limited set of point measurements, combined with a new CNN algorithm (Net-FLICS) to improve image reconstruction and enable almost real-time visualization of molecular events. The RPI team believes that this new approach will allow doctors to monitor the intracellular delivery of drugs by measuring the decay rate of fluorescence in the images.



The Resource Wire

This is sponsored content.

  • This Medmo video details how its healthcare marketplace platform and network of participating radiologists help underinsured patients pay as little as possible for their imaging procedures.



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.