|
AI Experiences & Expectations | Echo Goes Home
June 26, 2022
|
|
|
|
|
Together with
|
|
|
|
“Compared with initial predictions and expectations, the overall impact of AI-based algorithms on current radiological practice is modest.”
|
|
The European Society of Radiology
|
|
|
|
The European Society of Radiology just published new insights into how imaging AI is being used across Europe and how the region’s radiologists view this emerging technology.
The Survey – The ESR reached out to 27,700 European radiologists in January 2022 with a survey regarding their experiences and perspectives on imaging AI, receiving responses from just 690 rads.
Early Adopters – 276 the 690 respondents (40%) had clinical experience using imaging AI, with the majority of these AI users:
- Working at academic and regional hospitals (52% & 37% – only 11% at practices)
- Leveraging AI for interpretation support, case prioritization, and post-processing (51.5%, 40%, 28.6%)
AI Experiences – The radiologists who do use AI revealed a mix of positive and negative experiences:
- Most found diagnostic AI’s output reliable (75.7%)
- Few experienced technical difficulties integrating AI into their workflow (17.8%)
- The majority found AI prioritization tools to be “very helpful” or “moderately helpful” for reducing staff workload (23.4% & 62.2%)
- However, far fewer reported that diagnostic AI tools reduced staff workload (22.7% Yes, 69.8% No)
Adoption Barriers – Most coverage of this study will likely focus on the fact that only 92 of the surveyed rads (13.3%) plan to acquire AI in the future, while 363 don’t intend to acquire AI (52.6%). The radiologists who don’t plan to adopt AI (including those who’ve never used AI) based their opinions on:
- AI’s lack of added value (44.4%)
- AI not performing as well as advertised (26.4%)
- AI adding too much work (22.9%)
- And “no reason” (6.3%)
US Context – These results are in the same ballpark as the ACR’s 2020 US-based survey (33.5% using AI, only 20% of non-users planned to adopt within 5 years), although 2020 feels like a long time ago.
The Takeaway
Even if this ESR survey might leave you asking more questions (What about AI’s impact on patient care? How often is AI actually being used? How do opinions differ between AI users and non-users?), more than anything it confirms what many of us already know… We’re still very early in AI’s evolution, and there’s still plenty of performance and perception barriers that AI has to overcome.
|
|
|
Nuclear Medicine and Pediatric Orthopedic Surgery Converge
Pediatric patients can’t always accurately describe their orthopedic-related pain. Read how Lorenzo Biassnoi, MD, describes how SPECT/CT can help in this SPECT/CT and pediatric orthopedic surgery story.
|
|
RMI Sees Clearly and Decides Confidently
See how adopting ClearRead CT allowed Michigan’s Regional Medical Imaging’s radiologists to complete their chest CT reads faster and more accurately in this Riverain Technologies case study.
|
|
- Caption’s Home Echo Service: Ultrasound AI startup Caption Health unveiled its new Caption Care service, which provides in-home cardiac wellness assessments through a partnership with major home care company Portamedic. The Caption Care service will allow Portamedic technicians (who probably aren’t sonographers) to perform AI-guided echo exams in patient homes across America. This is the first nationwide home echo service we’ve seen, but it isn’t completely unexpected given the advancements in ultrasound accessibility (portability, cost, AI guidance/reporting, connectivity) and healthcare’s continued shift towards patient homes.
- Vanderbilt Follows-Up: A Vanderbilt University Medical Center ED team detailed their homegrown workflow for emergency imaging incidentals management. Vanderbilt’s informatics team used an EHR-based workflow to improve radiology incidental alert visibility, and created follow-up reporting requests that are completed by ED clinicians and then routed to nurse case managers who arrange follow-ups. As a result, 95.3% of ED patients with imaging incidentals during the workflow’s first 13 months (888 of 932 patient visits) were notified of their incidental findings and surveillance plans.
- Magnetic Insight’s MPI Funding: Magnetic Insight completed a $17M Series B round (total funding now $41.5M) that it will use to bring Magnetic Particle Imaging (MPI) into clinical use for the first time. Magnetic Insight is developing an MPI-based method to directly monitor cell therapy in cancer patients, noting that no current imaging methods support cell therapy localization for solid tumor treatment planning and monitoring.
- Clinical Robustness: A study published in JMIR found that many digital health startups aren’t “clinically robust,” with very few or nonexistent regulatory filings and clinical trials. The analysis examined FDA data on 510(k) and De Novo approvals, as well as clinical trials listed on ClinicalTrials.gov, then assigned each company a “clinical robustness score” that was a simple count of their regulatory filings and clinical trials. Of the 224 companies included in the study, 98 had a clinical robustness score of 0, only 45 received above a 5, and the median score was 1.
- Blackford Adds Rad AI: Blackford Analysis will offer Rad AI’s Omni solution to its AI clients, expanding its sizable portfolio beyond pixel-based AI, while introducing Rad AI to Blackford’s large customer base. Rad AI’s Omni solution automates the creation of radiology report impressions and has achieved solid radiology practice adoption due to its ability to streamline radiologist workflows.
- Preoperative FDG PET/CT for NSCLC: A team of Taiwan-based researchers endorsed preoperative 18F-FDG PET/CT for patients with advanced resectable non–small cell lung cancer (NSCLC), noting its ability to avoid unnecessary surgeries and improve staging accuracy. Analysis of 6,754 NSCLC patients who underwent preoperative PET/CT and 6,754 patients who didn’t (procedures 2009-2018, follow-up thru 2019) found that PET/CT patients with stage IIIA and IIIB NSCLC were less likely to die during the study period (Hazard ratios: 0.90 & 0.80). However, NSCLC stage I–II patients who underwent preoperative FDG PET/CT had higher mortality rates (HR: 1.19) due to PET’s lower sensitivity and specificity at earlier stages.
- Siemens & Tellica’s 10yr Partnership: Siemens Healthineers announced a 10-year value partnership with Intermountain Healthcare’s new Tellica outpatient imaging subsidiary. The alliance launches through Tellica’s three initial Utah locations, although Siemens will also provide medical imaging systems and software to future Tellica locations that open during the next decade.
- Kids ≠ Small Adults: A recent study showing that Lunit’s Insight CXR accurately interpreted pediatric X-rays gained headlines across radiology news outlets, who largely touted the study as a sign that we might be able to fix our pediatric AI deficit with adult-trained AI tools. However, a Twitter thread from Duke’s Walter Wiggins, MD, PhD accurately reminded us that “kids are not simply little adults,” and this approach doesn’t take into account AI’s unpredictable performance when used on out-of-distribution data (like using adult AI on kids) or AI’s potential to harm vulnerable populations (also like kids).
- Elucid’s Series B: Cardiac imaging AI startup Elucid locked in $27M in Series B funding to support the development of its FDA-cleared and CE-marked software, Elucid Vivo. The solution uses AI and CT angiography to identify and quantify heart attack-causing tissue in the arterial wall, potentially helping physicians diagnose the direct cause of chest pain and determine if patients have early-stage heart disease. The fresh investment brings Elucid’s total funding to nearly $50M.
- Rad+AI Synergies: A new Lancet study detailed a mammography AI workflow that could improve radiologist accuracy, while reducing reading labor. The researchers had Vara’s AI tool triage 82,851 digital mammography screenings from an external dataset (2,793 w/ cancer), creating automated reports for exams that Vara was confident are “normal” and routing the remaining exams for radiologist interpretations. This AI + radiologist “decision-referral” approach achieved 89.8% sensitivity and 94.4% specificity, surpassing both the AI tool and the radiologists on their own (AI: 84.6% & 91.3%; Rads: 87.2% & 93.4%).
- AZmed’s FDA: French AI startup AZmed announced the FDA clearance of its Rayvolve X-ray AI fracture detection solution, which already has its CE Mark and is in use across 300 medical centers in 21 countries. In its FDA validation study at Cleveland’s University Hospital, Rayvolve increased the UH physicians’ accuracy by 5.6% and reduced their interpretation times by 27%.
|
|
- Precision medicine startup BAMF Health just installed United Imaging’s uEXPLORER scanner, making it the first total-body PET/CT used for theranostics in the US. See how this combination will allow BAMF Health to deliver more effective and efficient theranostics treatments.
- Faced with two aging legacy PACS systems, South Jersey Radiology Associates moved to Intelerad’s IntelePACS, allowing its 12 sites to operate as a single, more-efficient entity. See how SJRA has since improved its radiologist efficiency by 8% to 10% and achieved a unified experience across its locations.
- Evaluating your patient engagement strategy? Check out this Imaging Wire Show featuring Novarad’s Paul Shumway for a great conversation about how new technologies are helping imaging providers safely and securely improve patient engagement.
- See why radiologist Dr. Eleanna Saloura called Arterys’ Lung AI solution “a fast and reliable second opinion” for chest CT lung nodule analysis and tracking, allowing “more accurate diagnostic and treatment decisions.”
|
|
|
|
|
