Concurrent AI | Subspecialty Speed | VidiStar Evolves


“. . . the worst radiology department in the country.”

How an unnamed King George Hospital physician ranked his own radiology department.



We’re getting close to the Imaging Wire nomination deadline, so if you know someone who made outstanding contributions to radiology this year, nominate them for an Imaging Wire Award.



Imaging Wire Sponsors

  • Arterys – Transforming medical imaging by reducing the tedium of interpretation and empowering physicians with our web-based AI platform.
  • Bayer Radiology – Providing a portfolio of radiology products, solutions, and services that enable radiologists to get the clear answers they need.
  • GE Healthcare – Enabling clinicians to make faster, more informed decisions through intelligent devices, data analytics, applications and services.
  • Healthcare Administrative Partners – Empowering radiology groups through expert revenue cycle management, clinical analytics, practice support, and specialized coding.
  • Hitachi Healthcare Americas – Delivering best in class medical imaging technologies and value-based reporting.
  • Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter.
  • Riverain Technologies – Offering artificial intelligence tools dedicated to the early, efficient detection of lung disease.
  • Siemens Healthineers – Shaping the digital transformation of imaging to improve patient care.



The Imaging Wire



Concurrent AI Evidence

A study in Radiology: Artificial Intelligence found that using mammography AI systems during the reading process (specifically, Therapixel’s MammoScreen V1) can improve radiologists’ diagnostic performance without slowing them down. Here are some details:

  • The Researchers – The study’s research team largely consisted of Therapixel employees, plus some outside experts.
  • The Study – The retrospective study had 14 radiologists assess 240 digital mammography images. The radiologists initially read half the images without AI and the other half while using the MammoScreen V1 tool. Four weeks later, they read the first set of images with AI (previously without AI) and the second set of images without AI (previously with AI).
  • The Results – The radiologists’ average AUC increased from 0.769 without AI to 0.797 with AI, while sensitivity (0.658 to 0.691) and specificity (0.725 to 0.735) also improved. Reading times were higher with AI assistance but this margin shrank from the first reading session (62.79 vs. 71.93 seconds) to the second (57.22 vs. 62.16 seconds), and the radiologists read images with a low risk of malignancy (<2.5%) just as fast with or without AI.
  • The Takeaway – Some folks could poke holes in this study (performed by Therapixel, focuses in 2D, retrospective), but it still adds new evidence that concurrent-use mammography AI improves radiologist performance without hurting productivity. If you’d like more evidence, this previous UPenn study on concurrent-use DBT came to a similar conclusion.



Imaging Wire Q&A: Evolving With Hitachi VidiStar

The role of imaging in pediatric cardiology has evolved tremendously in recent years, so in order for these practices to operate successfully, their PACS systems have to evolve at the same pace. That can be easier said than done, but it’s exactly what happened with Pediatric Cardiac Care of Arizona and its VidiStar PACS system.

In the latest Imaging Wire Q&A, we sat down with Dr. John Stock of Pediatric Cardiac Care of Arizona (PCCA) to discuss the evolving role of imaging in his practice, how Hitachi’s VidiStar PACS has evolved along with it, and what pediatric cardiology practices should look for in their own PACS systems. Here are some of the big takeaways:

  • PCCA’s Imaging – PCCA performs and interprets about 3,000 pediatric studies per year, with Dr. Stock interpreting all cardiac ultrasounds independently after reviewing and confirming measurements using Hitachi’s VidiStar platform.
  • COVID Impact – COVID-19 has definitely affected his practice, but not how some might think. They experienced a 20% to 30% drop in patient volumes during the shutdown’s peak months, but volumes are rebounding.
  • VidiStar’s Evolution – VidiStar has come a long way since Hitachi acquired the platform two-plus years ago, turning it into a system that is affordable, user friendly, and able to support the simplest and most complex pediatric cases.
  • PACS Advice – Pediatric cardiologists who are considering a new PACS system should evaluate how each system would meet their clinical and workflow needs and whether it fits their budget. They should also confirm its ability to track changes and perform the right measurements, its output, usability, dependability, and connectivity.


The Wire

  • Training AI Leaders: A Brigham & Women’s Hospital and Harvard Medical team detailed the Data Science Pathway (DSP) that they created to help expand their fourth year residents’ imaging AI capabilities. The DSP immersed the med students in a Clinical Data Science work environment, exposing them to all aspects of AI development (data curation, model design, quality control, and clinical testing), and teaching them core AI concepts through didactic sessions and collaborations with data scientists and other staff.
  • UK Contrast Tragedy: An official inquiry into a 91-year-old man’s death at King George Hospital (Ilford, UK) found that an “unnecessary” contrast-enhanced scan damaged his kidneys and led to sepsis and multiple organ failure. The report attributed the death to “systemic medical failings” (nobody flagged his diabetes and kidney issues before the scan) and included testimony from a King George physician that called its radiology department “the worst in the country.”
  • UB Targets MSK MRI: A University at Buffalo-led research team landed a 5-year, $3.7m NIH grant to improve how MRI systems image ligaments, tendons, and bones. The researchers will work to address MRIs’ limitations with connective tissues (low sensitivity, long imaging process) and semi-solid / solid tissues (weak MRI signal) by developing “advanced flexible and wearable imaging hardware” and fast imaging techniques / sequences that operate at 7T (yes, 7T). The team will also research how ultra-short and zero echo time methods might improve semi-solid and solid tissue MR imaging.
  • Subspecialty Speed: Changing radiology reporting workflows from a decentralized / modality-based structure to a centralized / subspecialty-based structure can significantly reduce radiology report turnaround times (RTATs). That’s from a study that reviewed RTATs from a large Swiss hospital system (11 imaging sites), finding that adopting a centralized / subspecialized reporting workflow reduced its overall RTATs for first signatures (82 to 77 min) and second signatures (295 to 238 min). The hospital achieved the greatest RTAT reductions with MRI reports (2nd signature: 1,051 to 401 min), X-ray reports (2nd signature: 278 to 171 min), and at the system’s smaller hospitals (300 to 198 min), while potentially reducing requests for second opinion reads.
  • Groupon Imaging’s Downsides: A new JAMA paper detailed the healthcare downsides of direct-to-consumer (DTC) medical imaging, particularly scans promoted through Groupon. The study (84 companies, 28k Groupon coupons sold) found that 68% the companies made unsubstantiated claims and only one company detailed potential risks, while warning that DTC/Groupon imaging might increase unnecessary testing, incidental findings, false-positives, and radiation exposure. However, they also found that these deals had a relatively high 4.8 / 5 average customer rating, which is just as good as Groupon’s deal for the San Diego Zoo.
  • AI Metrics’ $1.7m: AI Metrics completed a $1.7m seed round (increasing its total to nearly $2.4m) that it will use to expand its AI workflow, product portfolio, and team. Radiologists and oncologists use AI Metrics’ solutions to support their assessments of tumors and other significant findings on CT and MRI scans.
  • Room Temp Terahertz: MIT and University of Waterloo scientists developed a new layering technique that allows terahertz laser systems to operate at higher temperatures and could help make terahertz-based medical imaging systems a reality. The new layering technique could eventually allow terahertz laser sources to operate at room temperature (they’ve made it up to –23°C so far), which when paired with room temperature terahertz detectors, could lead to terahertz-based medical imaging devices.
  • Siemens & U of Iowa: Siemens Healthineers and University of Iowa Health Care announced a 10-year value partnership that will provide the healthcare system with Siemens’ imaging technology/AI and lead to new research/education and workforce development collaborations. Through the value partnership, Siemens Healthineers and UI Health Care will enhance the university’s technologist and sonographer training programs.
  • Patient Centric Return: A survey of 99 of patients who had their elective MRIs postponed early in the COVID emergency found that most patients didn’t view the delays as potentially bad for their health (68%) and most didn’t have safety concerns about going back in for their rescheduled exams (also 68%). So what was of concern to them? The surveyed patients who already completed or rescheduled their exams listed staff friendliness as the most important part of their return (95% fairly or extremely important), followed closely by the staff wearing masks (94%), social distancing policies, short wait times, temperature check screenings, and same-day scheduling.
  • An Innovative IR Expansion: Ohio’s Dayton Interventional Radiology (DIR) will use roughly $100m in funding from dialysis giant Fresenius Medical Care (they control 40% of the U.S. dialysis center market) to expand its operations across Ohio and potentially beyond. This could be a bigger deal than the typical practice expansion. It will bring DIR’s unique outpatient business model (minimally invasive procedures at Medicaid rates) to new regions/patients and allow the company to seek outpatient partnerships with large hospital systems.
  • Novarad’s COVID Solution: Novarad announced the launch of its COVID-19 AI Diagnostic Assistant software, making the Azure/Intel-based COVID diagnostic tool available to radiology groups without charge. Based on Novarad’s CryptoChart product, the new solution combines DICOM routing, anonymization and encryption, an AI processing engine, and secure reporting to help radiologists detect COVID-19 in CT scans “in seconds” and with a <1% false negative rate.
  • Two Sides to CTA DL: A Chinese research team developed a deep learning algorithm that detected cerebral aneurysms on CTA scans with an impressive 97.5% sensitivity (633 of 649 aneurysms), even detecting eight aneurysms that initial radiology reports missed. However, the algorithm’s high sensitivity came with a 13.8 false positive rate per case. A companion editorial suggested that reducing this false positive rate to 4-5 per case would bring the algorithm’s sensitivity down to “the low 90% range,” which isn’t much more valuable to radiologists than if they read the scans on their own.
  • Digirad Sells Mobile Imaging Business: Digirad sold its DMS Health Technologies mobile imaging services division to Knob Creek Acquisition Corp. for $18.75m (well short of the division’s $41.3m 2019 revenue), which it will use to pay down its liabilities and to fund its holding company growth strategy. Digirad will keep its other medical imaging businesses (nuclear imaging cameras & on-site imaging services) although the company has actively diversified outside of imaging for the last year.


The Resource Wire

– This is sponsored content.

  • Learn more about how GE Healthcare helped St. Luke’s University Health Network achieve system-wide ultrasound consistency and $300k in annual efficiency gains.
  • In this Bayer Radiology video, East Texas Medical Center Radiology Director, Bill Tobin, details how they used Bayer’s MEDRAD Stellant Smart Injector and contrast dose management to reduce contrast volumes and repeat scans.
  • Did you know that one in three Americans is obese and obesity is even more prevalent in rural communities? This Hitachi blog shows how its wide aperture CT and MRI systems are the best fit for rural hospitals, helping them care for patients of all sizes and get more ROI from their imaging systems.
  • Learn how one critical access hospital in a California ski mountain town used Nuance PowerShare accelerate care, drive change, and pivot on a dime, and to #ditchthedisk in this upcoming webinar.
  • Radiologists have a new tool available that will potentially allow them to recover some of the added costs related to safety precautions taken during the COVID-19 PHE. Learn more about CPT code 99072 in this blog article from Healthcare Administrative Partners.

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.

Join thousands of imaging professionals.