“There is currently no scientific basis for policy makers to use AI as a reason to refrain from expanding the radiology workforce.”
Thomas Kwee MD, PhD and Robert Kwee MD, PhD after their research found that most imaging AI academic studies would have added to radiologist workloads.
Imaging Wire Sponsors
Arterys | Bayer Radiology | Canon Medical Systems
Fujifilm Healthcare Americas | GE Healthcare |
Healthcare Administrative Partners | Novarad | Nuance
Riverain Technologies | Siemens Healthineers
United Imaging | Zebra Medical Vision
The Imaging Wire
Radiology Research’s Efficiency Gap
The majority of innovations analyzed in radiology-focused academic studies would improve patient care, but most would also increase radiologist workloads.
- The Study – That’s from an Insights into Imaging review of 440 studies that evaluated each innovation’s potential effects at an academic medical center and a general hospital.
- Patient Care Impact – When the studies’ techniques / technologies were applied to imaging workflows at the two hospitals, ~64% were found to directly contribute to patient care.
- Radiologist Workload Impact – However, only ~4.5% of the studies that improved patient care would also reduce radiologist workloads and ~48.5% would create more work for radiologists.
- AI Influence – The proposed AI innovations were about 10.5 times more likely to increase radiologist workloads than studies that didn’t involve AI (> 86% of AI studies increased rad workloads). That might relieve some folks concerned about AI’s threat to radiologists’ job security, but adding work surely isn’t AI’s goal and most workload increases would have to be justified by greater patient care improvements.
- The Takeaway – The authors positioned these results as a sign that forthcoming innovations will increase radiologist workloads, encouraging radiology groups / departments to expand their teams accordingly.
HAP’s Guide to Staying Private
Independent and staying that way? Healthcare Administrative Partners just released a helpful set of guidelines that radiology practices can follow to stay private despite ongoing consolidation pressures.
United Imaging’s All-In
United Imaging already announced that it’s “all-in for Nashville” this summer. After exhibiting at HealthTrust University in late July, United Imaging’s U.S. leadership will stay in Nashville as the company exhibits in-person at AHRA just a few days later. United Imaging is a Gold sponsor and promises to talk more at AHRA about what “all-in” means to healthcare administrators, clinicians, and patients.
- Philips’ Ultra-Fast Cardiac MRI: Philips and a team of Spanish researchers developed a new ultra-fast cardiac MRI protocol that could reduce scan times to “a few minutes,” and might help make MRI a mainstream cardiac modality. The researchers scanned over 100 cardiac patients using both conventional MRI and the new Enhanced SENSE by Static Outer-volume Subtraction technique (ESSOS), finding excellent agreement between the two protocols for heart function measurements and characterizing heart muscle tissue damage.
- Patients’ Report Feedback: Allowing patients and providers to submit feedback about radiology reports can lead to report quality improvements. That’s from a JACR study (n = 219 patients, 148 providers) that embedded patient and physician report evaluations in their EMR (a star system and text-based feedback). They found patients were more likely to provide feedback than providers (71% vs. 50%) and were more satisfied with their radiology reports (76% vs. 65%), while both noted a similar presence of errors (8% patients, 9% providers).
- The ABR’s Extended Leave: The American Board of Radiology continued its resident and radiologist-friendly pivot, announcing that it will allow residents to take eight weeks of annual “Time Off” during their residencies (including vacation, family leave, medical leave, sick days, etc.). That’s a lot more flexible than the ABR’s initial proposal of “up to six weeks” of total family medical leave throughout an entire four-year residency.
- A Ransomware Recovery Framework: After experiencing a ransomware attack in 2020, a Cleveland Clinic team published a 4-step framework that imaging departments can use to prepare for their own ransomware recoveries. The imaging recovery framework phases include: 1) Ensuring patient care, evaluating damage, setting up communications; 2) Transitioning to analog imaging workflows w/ focus on emergency patients; 3) Working with the IT team to begin recovery; 4) Rebuilding a ransomware-free environment and reconciling imaging data accrued during downtime.
- EchoNous’ $60M: Point-of-care ultrasound startup EchoNous announced a hefty $60m funding round (increasing its total to $95m) that it will use to support the commercial launch of its flagship Kosmos POCUS platform. Those are some big funding numbers, but EchoNous has a solid pedigree (led by SonoSite’s former CEO/founder), and Kosmos boasts a number of unique image quality and data collection / transmission capabilities.
- Cardiac POCUS AI Evidence: A new study in Circulation: Cardiovascular Imaging detailed a Caption Health POCUS ML algorithm that accurately quantified left ventricular ejection fraction (LV EF) in 4-chamber and parasternal long-axis views (typical views for PoC settings). The researchers compared the ML estimates against 166 reference EF measurements performed by echocardiographers and against 10 physicians’ visual estimates. The automated ML measurements achieved “excellent” agreement with the cardiographers’ reference measurements and were as accurate as 85-90% of the physicians’ visual estimates.
- Northern Ireland Reporting Review: The NHS Northern Ireland Trust is reviewing over 9k patients’ radiology exams performed by a locum consultant radiologist between July 2019 and February 2020 (13k exams) after an initial review created “serious concerns.” The Northern Ireland Trust will initially prioritize higher-risk cases, estimating that up to 7% of these patients could be at risk of delayed diagnosis.
- Finland’s Diagnostic Delays: Speaking of diagnostic delays, a new Insights into Imaging study revealed that Finland’s imaging-related malpractice cases (n = 1,054, 1991 to 2017) were most commonly due to diagnostic delays (38.3% of claims), followed by malpractice, infection or complication (33.7%), and incorrect / inadequate diagnosis (16.8%). More than half of the delayed diagnosis claims were related to either breast (30%) or skeleton exams (30%), followed by diagnostic and therapeutic vascular radiology (14%).
- Introducing Fujifilm Healthcare: On July 1st, Hitachi Healthcare officially became Fujifilm Healthcare, marking a major step in the integration of Hitachi’s imaging business into Fujifilm Corporation. Fujifilm Healthcare will initially lead with the portfolio that Hitachi is known for (CT, MRI, Ultrasound, VidiStar), although its US announcement noted the future integration of Fujifilm’s AI technology and its European announcement highlighted the various Fujifilm groups’ combined portfolios (CT, MRI, X-ray, AI, PACS, endoscopy, cart & POC ultrasound).
- HCC Ultrasound Variability Challenges: Ultrasound-based hepatocellular carcinoma (HCC) exam quality can vary widely from scan-to-scan among cirrhosis patients. UT Southwestern researchers reviewed scans from 2,053 patients with cirrhosis, finding that 18% of the exams had “suboptimal” LI-RADS visualization scores (B or C). However, 53% of patients with B/C baseline visualization scores had improved visualization scores in their second exams, while 19.6% of patients with A baseline visualization scores had B/C scores in their second scans.
- VUNO & Samsung’s AI Alliance: VUNO and Samsung announced a global alliance that will allow Samsung to embed VUNO’s Med-Chest AI solution into its GM85 mobile X-ray system, which flags thoracic abnormalities in real time. VUNO’s Samsung alliance comes just one week after a similar alliance with LG Electronics (embedded in LG’s X-ray detector), suggesting that (at least for now) embedded AI alliances with S. Korean conglomerates will play a key role in VUNO’s strategy.
- CT AI for Kidney Stone Detection: Turkish researchers trained a DL model that can automatically detect kidney stones on CT images. The model (trained w/ 1,799 images, tested w/ 146) identified kidney stones with high accuracy (96.82%) and sensitivity (95.76%), even with small stones. The model was also able to accurately identify areas of interest to further support diagnostic decision making.
- Avicenna.AI Joins Nuance AI Marketplace: Avicenna.AI’s CINA-ICH and CINA-LVO AI tools (for stroke triage/prioritization) are now available on the Nuance AI Marketplace. Avicenna.AI’s addition to the Nuance AI Marketplace further expands its platform / partner network, which also includes Arterys, TeraRecon, Blackford, Deepc, Olea Medical, Dicom Systems, and Canon.
Calculating LV & RV with GE and Arterys
See how GE Healthcare’s 2D FIESTA Cine cardiac MRI acquisition technique works with Arterys’ VF workflow to calculate LV and RV function and cardiac output.
The Resource Wire
- See how Novarad’s CryptoChart solution allowed Central Ohio Primary Care (COPC, 70 practices, 400 physicians) to make the transition to digital imaging sharing in this Healthcare IT News case study.
- Concerns about being sued were recently found to be radiologists’ top workplace challenges, but it doesn’t have to be. This Nuance blog details how adopting the right mix of structured reporting, clinical decision support, and follow-up tracking can help radiologists improve patient outcomes and reduce legal exposure.
- Take the AiCE challenge and see why half the radiologists in a recent study “had difficulty differentiating” images from Canon Medical Systems’ Vantage Orian 1.5T MR using its AiCE reconstruction technology compared to standard 3T MRI images.
- 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.
- 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.
- With radiation dose management now largely considered best practice, this Bayer white paper details the top five benefits of adopting contrast dose management.
- See how VidiStar users have benefitted from the Hitachi cardiovascular information system’s flexible SaaS-based model and leveraged its productivity advantages to increase reimbursements.
- See how St. Olav’s Hospital in rural Norway worked with GE Healthcare to create an in-house PET radiopharmacy.
- Cardiovascular disease is the number one global cause of death, but it’s also preventable, which is one of the reasons Zebra-Med views AI-powered cardiovascular screening as the next frontier in population health.