|
Automating Stress Echo | AI-Embedded X-Ray May 25, 2022
|
|
|
|
Together with
|
|
|
“If imaging services sneeze, then the whole health system gets the flu.”
|
UCSF’s Christopher Hess, MD on the contrast shortage’s health system impact.
|
|
|
A new JACC study showed that Ultromics’ EchoGo Pro AI solution can accurately classify stress echocardiograms, while improving clinician performance with a particularly challenging and operator-dependent exam.
The researchers used EchoGo Pro to independently analyze 154 stress echo studies, leveraging the solution’s 31 image features to identify patients with severe coronary artery disease with a 0.927 AUC (84.4% sensitivity; 92.7% specificity).
EchoGo Pro maintained similar performance with a version of the test dataset that excluded the 38 patients with known coronary artery disease or resting wall motion abnormalities (90.5% sensitivity; 88.4% specificity).
The researchers then had four physicians with different levels of stress echo experience analyze the same 154 studies with and without AI support, finding that the EchoGo Pro reports:
- Improved the readers’ average AUC – 0.877 vs. 0.931
- Increased their mean sensitivity – 85% vs. 95%
- Didn’t hurt their specificity – 83.6% vs. 85%
- Increased their number of confident reads – 440 vs. 483
- Reduced their number of non-confident reads – 152 vs. 109
- Improved their diagnostic agreement rates – 0.68-0.79 vs. 0.83-0.97
The Takeaway
Ultromics’ stress echo reports improved the physicians’ interpretation accuracy, confidence, and reproducibility, without increasing false positives. That list of improvements satisfies most of the requirements clinicians have for AI (in addition to speed/efficiency), and it represents another solid example of echo AI’s real-world potential.
|
|
|
The Future of Radiology Starts on June 30th
Reserve your spot for AI Visions 2022, featuring live discussions from the top radiology and AI leaders and the global launch of Bayer’s Calantic Digital Solutions AI marketplace.
|
|
- RP’s Follow-up Pilot: Radiology Partners launched a new care coordination pilot program at its ARA Diagnostic Imaging practice, combining RP’s internal best practice guidelines and Rad AI’s Continuity platform to identify patients with incidental findings and initiate follow-up workflows. Radiology Partners will apply its learnings from the ARA pilot and a separate ACR initiative to create a new nationwide RP care coordination program. Although incidental follow-up initiatives aren’t new, this is a great example of non-pixel AI’s value for radiology workflows. It also adds a big logo to Rad AI’s growing list of radiology practice clients.
- ED Stroke MRI ROI: A team of France-based researchers found that dedicating an MRI machine to emergency departments (EDs) for patients experiencing dizziness and double vision may facilitate rapid stroke diagnoses and reduce healthcare costs. Researchers analyzed wait times and patient costs before and after they added a dedicated ED MRI (n = 199 & 181), finding that the MRI reduced time-to-diagnosis (9.8hrs to 7.7hrs) and cost-of-care (€2,701 to €2,389) with these suspected stroke patients.
- Fujifilm & Annalise.ai’s AI Alliance: Fujifilm and Annalise.ai have reportedly partnered to embed Annalise.ai’s CXR solution into Fujifilm’s FDR Nano mobile X-ray system — in Australia and potentially globally. Although we’ve seen a growing number of AI solutions embedded in mobile X-rays, Annalise CXR’s ability to detect 95 different chest findings, in 10 seconds and without accessing the internet, is still very unique and could support a range of point-of-care X-ray use cases.
- MUSC & Butterfly Network: Medical University of South Carolina (MUSC) announced that it will adopt Butterfly Network’s Butterfly Blueprint platform across the health system, with an early focus on its rural clinics. Butterfly Blueprint is intended to make it easier for health systems to adopt the Butterfly iQ+ handheld ultrasound across clinical departments, and it appears to be achieving that goal, scoring enterprise-wide deals at MUSC and University of Rochester Medical Center so far this year.
- CAD AI for IPNs: A new Radiology Journal study found that an AI-based CAD tool improved physicians’ ability to assess the malignancy risk of indeterminate pulmonary nodules (IPNs) in CT scans. The researchers had 12 readers (6 radiologists & 6 pulmonologists) evaluate IPN risks in 300 chest CTs with and without AI assistance, finding that AI significantly improved the readers’ accuracy (AUC: 0.82 vs. 0.89). The AI model also strengthened the readers’ agreement rates for malignancy risk (κ: 0.35 vs. 0.58) and management recommendations (κ: 0.44 vs. 0.52).
- Enlitic Approvals: Enlitic announced that its unique Curie platform and Curie|ENDEX app are now approved in the US and EU, paving the way for a commercial launch just a few months after its pivot from pixel AI to solutions that improve radiology data and workflows. Curie|ENDEX transforms imaging data to a standardized nomenclature and enables relevant clinical content linkage across disparate systems, allowing consistent hanging protocols and improving image routing and AI orchestration.
- HistoSonics & GE Partnership: HistoSonics announced a new partnership with GE Healthcare, combining its Edison liver therapy platform with GE’s LOGIQ E10s ultrasound system. HistoSonics’ Edison System, which is currently in the clinical trial phase, uses histotripsy to target and destroy primary and metastatic liver tumors. This non-invasive and non-ionizing technique, paired with GE’s ultrasound guidance, could be a safe alternative to surgery and radiation therapy.
- Neiman HPI’s Contrast Navigation Guide: A report published by the Harvey L. Neiman Health Policy Institute provided some data-driven guidance to help imaging teams navigate the contrast shortage. The authors analyzed data from 9.6M Medicare CT exams, finding that CE-CTs were most commonly performed in outpatient and emergency settings and for abdominal/pelvic, head/neck, and brain body regions. While the contrast shortage continues, the authors encourage healthcare providers to consider alternative imaging pathways for these high-volume exams, such as MRA or ultrasound for CTA.
- Aidoc & Gleamer’s AI Partnership: Aidoc and Gleamer announced a new partnership that will bring Gleamer’s BoneView X-ray analysis solution onto the Aidoc AI platform. The partnership adds the first bone-imaging solution to the Aidoc platform, which has historically focused on CT-based neuro and cardiopulmonary findings. It also marks a key commercialization step for Gleamer following its March FDA clearance, and represents yet another platform partner addition for Aidoc (joining ScreenPoint, Riverain, Subtle, Imbio, Icometrix).
- Deep Learning Density Predictions: A recent JACR study found that using a deep learning (DL) breast density model might reduce the number of screening mammograms categorized as dense. Researchers analyzed 85k mammograms across three sites, finding that the odds of dense classifications decreased at the two sites using the DL model (aOR=0.94 & 0.81), while odds increased at the site not using the DL model (aOR=1.13). This study shows DL may help clinicians make more accurate breast density assessments, reserving the limited supplemental screening resources for those who really need them.
- RapidAI’s PE FDA: RapidAI announced the FDA clearance of its Rapid PE Triage & Notification AI solution, which analyzes CTPA exams to identify and prioritize patients with suspected pulmonary embolisms (PEs). The new AI solutions will be combined with RapidAI’s Rapid Workflow for PE to support care team coordination for this potentially lethal condition with a history of diagnostic delays. This is a notable extension for Rapid AI, which has primarily focused on stroke and aneurysm solutions.
|
|
Estimating Your MRI ROI
Ready to purchase your first MRI or replace an existing one? Estimate the return on your investment by using Siemens Healthineers’ interactive value calculator. Try it out today and receive a personalized brief with our findings.
|
|
- See how Dubai-based healthcare leader Aster DM Healthcare leveraged the CARPL platform to connect its doctors, data scientists, and imaging workflows, and support its AI projects and development infrastructure.
- Learn how Memorial MRI and Diagnostics’ efforts to improve its MRI patient experience impacted its patient referrals, clinical case mix, throughput, and accessibility in this June 1st webinar by United Imaging and AuntMinnie.com.
- Working on your organization’s AI strategy? This Blackford Analysis post outlines the key considerations for creating your AI goals and strategy, including some you might not have considered.
- Learn how GE Healthcare technologies are enhancing cardiac MR’s role in cardiac structure evaluation by improving tissue characterization and allowing physicians to better understand abnormalities.
- Check out this patient case study showing how the Arterys Chest I MSK AI allowed radiologists at CSE in Paris to identify a fracture that was missed in three previous interpretations.
- PACS efficiency and accuracy can have a major impact on radiologist workflows, but these qualities aren’t guaranteed. Check out this Novarad report detailing how to improve your PACS efficiency and accuracy.
- Did you know one quarter of healthcare organizations have experienced a cyber-attack in the last year? This Change Healthcare animation explains how 3rd-party certified cloud-native enterprise imaging can help secure IT infrastructure that might be exposed with re-platformed imaging systems.
- Discover how Magnolia Regional Health Center started catching more cancers sooner when it adopted Nuance’s PowerScribe Lung Cancer Screening Program and PowerScribe Follow-up Manager.
- With radiologist workloads growing in volume and complexity, having the wrong PACS can lead to radiologist burnout. This helpful Fujifilm post shows how having the right PACS that functions as a centralized and integrated enterprise imaging system can be part of the solution.
- This European Radiology study highlighted Riverain Technologies’ ClearRead Xray – Detect as one of just two imaging AI products to achieve the FDA’s most stringent premarket approval level. See how they measured up against the other 99 AI tools here.
|
|
|
|
|