“Some say data is the new oil. Others say the algorithm is the gold. Actually, it’s the labels.”
Dr. Hugh Harvey on the value and scarcity of quality imaging labels.
Imaging Wire Sponsors
- Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound.
- GE Healthcare – Providing point of care ultrasound systems, from pocket-sized to portable consoles, designed to support your clinical needs and grow along with your practice.
- Healthcare Administrative Partners – Empowering radiology groups through expert revenue cycle management, clinical analytics, practice support, and specialized coding.
- Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter.
- Qure.ai – Making healthcare more accessible by applying deep learning to radiology imaging .
- Riverain Technologies – Offering artificial intelligence tools dedicated to the early, efficient detection of lung disease.
The Imaging Wire
Three AI Perspectives
We hear a lot about radiologists’ perspectives on AI, but the last week brought us a rare trio of studies revealing how med students, patients, and healthcare executives view the technology. Here they are:
- Students on AI – A survey of UK medical students (n = 484 students, 19 med schools) believe healthcare AI is growing in importance (84%), AI training will help their careers (89%), and AI should be part of their curriculums (78%). However, only 9% have received any AI training and 49% are avoiding radiology specialization due to AI-related fears (especially those who haven’t been educated on AI: ~63%). This is consistent with a previous Swiss survey that found 26% of surveyed med students were steering away from radiology due to AI concerns.
- Patients on AI – As patients become more aware of the role of imaging AI, a paper out of Canada (n = 17 patient discussions) reveals that patients are anxious about how the technology will impact their care, whether it can be trusted, and if AI will change the human components of healthcare. The patients also had surprisingly-relevant preferences for how AI should be used including: 1. Reducing wait times; 2. Cutting time to diagnosis; 3. Expanding regional access; 4. Improving accuracy; and 5. Improving how clearly results are communicated.
- Execs on AI – A KPMG report found that 89% of healthcare executives say that AI is already creating efficiencies in their system and 90% believe AI will improve the patient experience in the future (with the greatest impact on diagnostics, 47%), while 69% wish their organizations would speed up AI adoption. However, only 37% responded that AI is “at least moderately functional” in their organizations and many still think AI is more hype than reality (52%).
More Coronavirus Imaging
Coronavirus imaging was a major industry news topic for the third straight week (here’s the last two weeks), as CT gained more credit for its role in diagnosing the disease and manufacturers continued to highlight their contributions to help fight it.
- CT Positive – A 167-patient study in China revealed that chest CT can detect novel coronavirus, even when the initial RT-PCR mouth swab DNA tests are negative. Although this only occurred with five of the patients, the study authors suggest that CT may help identify patients who should receive a follow up RT-PCR test to confirm their coronavirus status. News also emerged that the ‘thermometer guns’ used on the coronavirus front lines are ‘notoriously not accurate,’ making diagnostic certainty even more important.
- CT Surge – The Chinese government’s move to expand the diagnostic criteria for coronavirus to include CT scans (vs. only RT-PCR mouth swab tests) led to a surge in confirmed new cases last week (+15k in 24hrs vs. +2k typical daily avg).
- Coronavirus Business – There was another round of coronavirus imaging business and PR announcements, as Micro-X announced new coronavirus-related orders from Asia for its mobile X-ray system and Konica Minolta announced a donation of eight ultrasound systems to six Chinese hospitals to help diagnose the disease.
Share The Wire
Please take a moment to consider your colleagues who might find this newsletter valuable and let them know about it. Once again, we’re donating $2 to these Australian Bushfires charities for each new subscriber that you refer.
If you’re a subscriber, please consider using your personal referral link to introduce your industry friends to The Imaging Wire (You can find your personal link in your latest newsletter).
If you haven’t received a referral from one of your colleagues yet, just use this link to sign up.
- Kenyan Waste: A widely-covered Reuters article detailed how the country’s $625 million imaging device “binge” since 2015 far exceeded the bandwidth of its radiology workforce, leaving these new devices idle in one-third of hospitals. Kenya’s healthcare critics believe this money should have been spent on the country’s “basic healthcare” needs, while estimating that Kenya would have to add a whopping 10,000 new radiology professionals to staff all of its new equipment.
- Siemens’ Reinvents Service: Siemens Healthineers announced the U.S. launch of its CentriCare service portfolio. The new service contract portfolio “reinvents” traditional imaging service models by performing scheduled maintenance/repairs during customer-defined periods (7-day x 16hrs) to help keep machines available during key scanning hours.
- Simplified Authorizations: CMS head, Seema Verma, revealed plans to simplify the prior authorization processes, potentially reducing burdensome PA tasks that have been found to delay treatment, lead to burnout, and increase administrative costs.
- Students Look Ahead: Vanderbilt researchers found that med students prefer a ‘look ahead’ approach to radiology training, which exposes students to more realistic radiology work (e.g. interacting with PACS, making observations and conclusions), rather than watching the instructors observe images and then discuss their observations/teachings. A survey (n = 34 students, agreement responses scored 0 to 100) revealed that the ‘look ahead’ method increased students’ interest (92.5 vs 75.1 avg. scores), was more engaging (94 vs 70.3 avg.), had more educational value (92.5 vs 73.2 avg.), and was more memorable (88.5 vs 73.2 avg.) than the standard method.
- New KM USs: Konica Minolta Japan launched its new SONIMAGE HS2 (diagnostic) and SNiBLE2 (orthopedic) ultrasound systems, replacing the internationally-available SONIMAGE HS1 and SNiBLE 1. The new models are highlighted by the SONIMAGE HS2’s expanded support for the dialysis market (added Vascular NAVI scanning for shunt management), the NiBLE2’s improved operational response, and their new probes and processing features.
- Fast & Furious: A recent brain bleed misdiagnosis settlement in Florida has many folks on the radiology message boards up in arms after the defendant radiologist’s “too fast” reading speed (691 CT images in 6.5 minutes) was used as prosecutorial evidence. The Imaging Wire covered this story without questioning whether 6.5 minutes was actually too fast (thanks for all the correction emails) and the overall consensus among upset rads is that malpractice cases should be judged by whether an at-fault misdiagnosis occurred rather than the reading time (especially since 6.5 minutes isn’t that slow).
- Phone Waste: Emergency radiologists are wasting time by discussing negative results with referring physicians over the phone. A study of emergency radiologists’ result-reporting communication events over a 168-hour period (n = 286 communication events, 232 via telephone) found that the radiologists spent 10% of their working time communicating results and that time was split evenly between routine/negative and positive/abnormal results (139 vs. 147 interactions).
- Viz.ai Upgrades: Viz.ai had a big week, launching the second generation of its Synchronized Stroke Care Platform and adding new algorithms and features to its Viz RECRUIT Clinical Trial Platform. The updated Stroke Care Platform adds a range of communication improvements (e.g. stroke code notifications, messaging, alerts) and makes a number of viewing and analysis improvements (e.g. <60 second viewing, mobile PACS, in-app image reconstruction) among other upgrades. Viz RECRUIT launches with a number of new algorithms (ICH, aneurysm detection and ischemic Core volumes) along with improved image sharing and collaboration software.
- AI Label Fundamentals: As noted in today’s quote of the issue, labeled images are both crucial and hard to come by. The good news is, a new paper in Radiology Journal laid out some steps to start addressing that issue. Among other things, the paper’s team of influential authors detailed these fundamental steps for preparing imaging data for AI development: get ethical approval > get access > query > de-identify > transfer > QC > structure > label the data.
- Aquilion ONE / PRISM Edition: Canon announced the FDA clearance of its Aquilion ONE / PRISM Edition spectral CT, highlighted by its included AiCE technology (previously limited to high-end CTs) for improving image quality and scan speeds, as well as its support for Canon’s new Deep Learning Spectral Imaging (supports diagnosis with spectral insights).
- Home Ultrasound Apps: German researchers are using €3.2 million in government funding to research and potentially develop apps that expectant mothers can use to guide home prenatal ultrasound exams. The funding will be used for a 500-woman, 2.5-year clinical trial that will provide each with portable ultrasounds, contraction monitors, and smartwatches and/or smartphones with apps.
- RSNA Reactions: A Reaction Data survey of radiology professionals (n = 192) suggested that RSNA attendance might fall in 2020 (29% attended in 2019, 25% plan to return), with the most common complaints targeting the show’s costs and timing (18%, 13%). RSNA’s AI and MRI technology received the most tech-based praise (32%, 22%), GE had the show’s most impressive tech (24%), and Siemens provided the greatest booth experience (27%).
- NIH Budget Plea: The Ad Hoc Group for Medical Research’s 335 members (including 24 imaging groups) urged Congress to block $3 billion in proposed NIH budget cuts and instead increase the agency’s funding by $3 billion to $44.7 billion. NIH funding is often used for radiology-related research.
The Resource Wire
– This is sponsored content
- Interested in bringing AI into your practice? Catch Nuance’s upcoming panel discussion where AI experts and radiology practice users detail the necessary steps to successfully adopt AI.
- Riverain Technologies is dedicated to providing enterprise software tools to aid clinicians in the efficient, effective, early detection of lung disease. Learn more.
- The 2020 CPT code updates brought a number of changes to diagnostic radiology, interventional radiology, and nuclear medicine. This Healthcare Administrative Partners blog details the 2020 MPFS CPT code changes and what they mean to your practice.
- In this recent Imaging Wire Q&A, Qure.ai and MEDNAX detailed how their partnership helped Qure.ai “prove and improve” its qER algorithm at high volumes and across a wide range of variables.
- The latest Focused Ultrasound Foundation newsletter details an early-stage clinical trial using histotripsy for liver cancer and a range of other FUS advancements.
- GE Healthcare’s Venue Go ultrasound system has a uniquely adaptable design, built to go from cart to wall to table so you can take it almost anywhere at the point of care.