Imaging’s Security Moment | WhatsApp AI

“Everything is hackable,”

The FDA’s medical device cybersecurity director, Kevin Fu, on which healthcare technologies are vulnerable.

Imaging Wire Sponsors

Arterys | Bayer Radiology | Canon Medical Systems | GE Healthcare
Healthcare Administrative Partners | Hitachi Healthcare Americas
Novarad | Nuance | Riverain Technologies | Siemens Healthineers
United Imaging | Zebra Medical Vision

The Imaging Wire

Imaging’s Security Moment

Cybersecurity emerged as one of the world’s most pressing healthcare issues over the last month after ransomware attacks crippled operations across several major health systems.

These events certainly affected imaging, but they didn’t seem to affect the radiology media’s editorial focus (including ours) or radiology social media’s discussion focus. However, a series of separate-but-symbolic events from the last week suggests that imaging security deserves more of our attention.

  • Fujifilm Hack – Fujifilm’s Japan headquarters fell victim to a cyberattack, which didn’t seem to directly affect its medical devices or healthcare customers, but served as a reminder of how hacks can impact healthcare OEMs. The attack forced Fujifilm to shut down parts of its global network, created a range of operational outages (email, phone, order accepting/processing), potentially exposed company info, and might force it to pay-off the ransomware gang responsible for the attack.
  • Hoboken Hack – New Jersey practice, Hoboken Radiology, revealed that hackers had access to its servers between June 2019 and December 2020 (maybe longer), potentially exposing its patients’ personal and medical information. This is a much smaller scale event than what’s happening at Fujifilm or Scripps Health, but it’s certainly a big deal for Hoboken and its clients, and it’s a reminder of how vulnerable smaller practices can be (we cover hacks at smaller practices / centers pretty regularly).
  • Public Pressure – The massive outages at Scripps Health and the Irish Health System turned healthcare cybersecurity into a “dinner table” conversation topic and brought a wave of articles and editorials highlighting healthcare’s cyber vulnerabilities. Considering imaging tech’s central role in medical operations and its possession of patient information, that’s going to place a lot more pressure on imaging vendors and departments to keep things secure too.

All-In with United Imaging

United Imaging’s “all-in” approach means that every system ships with its entire suite of features and capabilities (no options), giving its clients more clinical flexibility and predictability.

– Sponsored.

Reducing ED X-Ray Errors with Arterys Chest MSK

See how ED physicians at France’s Hospital of Maubeuge reduced emergency imaging error risk by 75% using Arterys Chest MSK.

– Sponsored.

The Wire

  • Mayo & Visage’s AI Alliance: Mayo Clinic and Visage Imaging entered a multi-year research collaboration, allowing Mayo to leverage the Visage AI Accelerator platform for AI development and potentially commercialization. Noting that Mayo Clinic is an established Visage 7 user (live since 2018), this alliance and Visage’s recent AI Accelerator partnership with NYU are solid examples of how enterprise imaging can play a key role in both AI delivery and development.
  • MLB FEVER: An AJR study found that the flexed elbow valgus external rotation (FEVER) MRI view can improve ulnar collateral ligament (UCL) evaluations among Major League Baseball pitchers (a group known for UCL injuries). The researchers performed standard and FEVER elbow MRIs on 44 MLB pitchers, finding that the FEVER view increased UT joint space width (mean increase: +1.80 mm) and improved radiologist confidence for three out of five UCL-related findings (mean increase: -0.14 to 0.98).
  • AI Over WhatsApp: Indian researchers unveiled their X-Ray Setu platform, which detects COVID in chest X-rays with the help of a WhatsApp-supported AI algorithm, and could “revolutionize” COVID testing in rural areas with insufficient RT-PCR tests and/or radiologists. This resourceful program requires a clinician to upload a chest X-ray using the X-Ray Setu WhatsApp bot, which analyses the image and generates a report in 10-15 minutes. The team plans to deploy X-Ray Setu to 10k Indian doctors over the next two weeks.
  • Siemens Gets in To Penn State: Siemens Healthineers just entered a 10-year imaging services agreement with Penn State Health (2,300 physicians, 125 medical locations) that will replace 70% of Penn State’s imaging equipment over the next five years and allow it to adopt a “more unified and system-level” approach to imaging.
  • B Reader AI: A team of MDs and JDs proposed integrating imaging AI into the NIOSH’s 50yr-old B Reader Program, which originally launched to screen for black lung disease, and has become increasingly challenged (broadened scope, declining doctor participation, financial conflicts of interest). The group suggests that imaging AI tools could help improve B Reader physicians’ productivity and accuracy, while preventing potential bias due to conflicts of interest.
  • POCUS Space Race: Butterfly Network’s Butterfly iQ is headed to the International Space Station where the Translational Research Institute for Space Health (TRISH) will research how handheld ultrasound devices could improve astronauts’ medical self-reliance. Butterfly will have company, as UltraSight announced last week that its cardiac ultrasound guidance software is also coming aboard the ISS.
  • Expert-Level Fetal Ultrasound AI: UCSF researchers developed a deep learning system that achieved “expert-level” detection of prenatal congenital heart disease (CHD) in fetal ultrasound screening exams. The model (trained w/ 1,326 echocardiograms & screening ultrasounds, tested w/ 4,108 ultrasound screenings) distinguished normal hearts and complex CHD with a 0.99 AUC, 95% sensitivity, 96% specificity, and a 100% negative predictive value (comparable to physicians).
  • One-Month Notices: A new AJR study showed how simply notifying referring physicians of overdue radiology recommendations can improve follow-up rates. The Washington University-led team (the WU in St. Louis) identified 680 one-month-old imaging reports with follow-up recommendations and sent notifications for the 177 (26%) overdue cases, resulting in 36 additional follow-ups and reducing non-compliance risk by 20.4%.
  • Knee X-Ray Tilt Correction AI: Japanese researchers developed an AI system that could reduce lateral knee X-ray retakes, and alleviate retake-related patient and technologist burdens, by classifying knee tilting direction in initial X-rays. The lateral knee radiograph retake support system (trained w/ 11.5k raysum images from 60 CT exams) correctly classified between 73.3% and 88.5% of test images into four tilting direction categories.
  • Lucida’s CE Mark: Just two months after completing its Seed round, Lucida Medical secured CE Marking for its Prostate Intelligence (Pi) prostate cancer detection software. Pi combines radiogenomics, machine learning, and image processing to help automate prostate MRI interpretations (generates risk scores, identifies biopsy targets, creates segmentations).
  • Smart-C’s CE Mark: Turner Imaging Systems’ Smart-C Mini-C Arm fluoroscopy and X-ray device just gained CE Marking (already FDA-cleared), expanding the portable system (just 16 lbs.) to Europe. Turner Imaging might not be a household name, but it comes to Europe with extra credibility given that it’s funded by RadNet and lists Siemens Healthineers as a U.S. reseller.
  • COVID LUS Score AI: Italian researchers developed a deep learning model that analyzes lung ultrasound video frames and automatically produces COVID LUS scores. The algorithm analyzed ~315k frames (from 1,488 videos, 82 COVID patients) and produced LUS risk scores that had an 85.96% agreement rate with lung ultrasound experts, highlighting AI’s potential to support the very manual LUS scoring process.
  • contextflow’s Series A: Austrian AI startup, contextflow, wrapped up a “mid-seven figures” Series A round that it will use to fund its regulatory and commercial expansion in Europe and the U.S. contextflow’s “SEARCH” 3D image-based search engine detects patterns in CT and MRI scans, providing radiologists with reference information to support diagnosis.

HAP’s Practice Benchmarks

Know how your practice measures up? In this post, Healthcare Administrative Partners details the key benchmarking quality metrics and how they can help radiology practices improve.

– Sponsored.

The Resource Wire

  • 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.
  • Learn how Yale New Haven Health improved its radiology efficiency, communications, and turnaround times when it adopted Nuance’s PowerScribe Workflow Orchestration and PowerConnect Communicator solutions.
  • With Turbo Suite Excelerate by Siemens Healthineers, you can reduce MRI exam times by up to 50%. See how it’s possible in these videos featuring example hip, knee, and brain scans.
  • See how and why Zebra Medical Vision sees a much bigger future for public health AI than many of us imagine in this Imaging Wire Q&A with company CEO, Zohar Elhanani.
  • This Hitachi Healthcare blog outlines the criteria providers should consider for their image and reporting platforms, and how the Hitachi VidiStar platform’s features, service, and vendor collaboration meet providers’ needs.

You might also like

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!