Will AI Ease the Adoption of Point-of-Care Ultrasound in Urgent Care?
This article also appears in Urgent Caring, the official publication of the College of Urgent Care Medicine. Subscribe to the Urgent Caring publication at this link.
POCUS has become a staple in emergency medicine, providing real-time insights to improve decision-making and guide procedures. But why has its adoption been so slow in urgent care settings? With advancements in affordable handheld devices, POCUS is well on its way to becoming the standard tool in urgent care. Could artificial intelligence (AI) accelerate this transition by providing real-time guidance for image acquisition and interpretation, making POCUS more accessible and reliable than ever?
The Current State of POCUS in Urgent Care
POCUS in urgent care is still in its infancy. It's mostly used in centers run by emergency physicians who have specialized training. You’re more likely to see it in high-acuity centers or those that perform orthopedics procedures. But the reality is that there are no POCUS guidelines for urgent care, akin to the ones established by the American College of Emergency Physicians (ACEP) for the use of POCUS in emergency departments. [1] Although there are many case reports on the use of POCUS in urgent care, there are no large-scale studies showing the benefits of POCUS in urgent care. [2-4]
Even in the centers that use POCUS, it’s often more of a "quick look" tool—a fast way to check for abscesses, joint effusions, or hydronephrosis. While it can be incredibly useful, this informal use leads to inconsistent documentation, missed billing opportunities, and a lack of structured decision-making. So, how can POCUS become a more integral part of urgent care?
How AI Can Enhance POCUS Adoption
AI might just be the key to unlocking POCUS' potential in urgent care. Let’s break down how it could help:
1. Image Acquisition: AI-guided scanning can help even the least experienced providers get diagnostic-grade images. Imagine a world where providers don’t have to worry about positioning the probe correctly or getting the best view—AI can guide them in real-time. Many device manufacturers have adopted AI not only to assist with probe positioning but also in labeling images and performing calculations. A recent study found that with just 2.5 hours of training, non-ultrasound experts (including nurses and medical assistants) produced diagnostic-quality images 98% of the time. [6]
2. Exam Interpretation: AI-driven software can highlight pathology like pulmonary edema, DVT, or pericardial effusion. For novice users, this is a game changer, enabling them to confidently perform a preliminary interpretation with AI guidance, without needing immediate input from a POCUS expert. [7, 8]
3. Reporting: AI can save time and improve documentation and billing compliance by automating image labeling and the reporting process. [7, 8]
4. Quality Assurance: AI can enhance the quality assurance process by ensuring that exams meet diagnostic accuracy standards and are correctly interpreted. Artificial intelligence (AI) is already being utilized in radiology to improve diagnostic accuracy and facilitate internal review.
5. Standardization: AI brings much-needed consistency, reducing the variability in image acquisition and interpretation that comes from different providers with varying levels of expertise. [7, 8, 9]
6. Impact on Providers:
For Novice Users: AI boosts confidence by offering real-time guidance and feedback, making the technology more accessible and user-friendly. This increased accessibility and support should motivate novice users to practice POCUS more often, accelerating their skill development and making ultrasound a seamless part of their clinical routine.
For Advanced Users: AI is a time-saver. It handles labeling, calculations, and reporting, leaving providers to focus on the more complex aspects of care.
All of these advancements could pave the way for broader adoption of POCUS in urgent care by overcoming the operational and training hurdles that have slowed its progress.
Caveats
But, as with any technology, AI has its limitations. Here’s what to consider:
1. Reliability Issues: AI is only as good as the data it’s trained on, and it can make errors. AI functions as a 'black box,' meaning that even experts may not fully understand how it arrives at specific conclusions. Furthermore, biases in the training data could be introduced and propagated, potentially leading to systematic errors that affect patient care. [9] It’s essential that providers maintain oversight and critically evaluate AI-assisted interpretations.
2. Training Limitations: While AI is a great learning tool, nothing beats hands-on experience. For best results, AI-assisted scanning should complement a formal training program that includes hands-on workshops, asynchronous learning, and exam review by an expert.
3. Operator Dependence: If a provider doesn’t have the right scanning technique, AI can’t fix that. It’s a tool, not a replacement for solid clinical skills. For example, if the operator is not fully compressing a deep vein due to poor technique, the AI software will erroneously interpret the exam to be positive for a DVT.
4. Technology Dependency: Over-relying on AI might erode fundamental acquisition and interpretation skills. Think of how some providers rely on machine-read EKGs without interpreting them themselves—AI could inadvertently lead to the same pitfall.
The Future of AI, POCUS, and Urgent Care: A Game-Changer in the Making?
AI is here to stay, and it is transforming how POCUS is used in urgent care and beyond. By building confidence, improving efficiency, and enhancing diagnostic accuracy, AI is making POCUS more accessible than ever. Increased utilization will allow novice users to develop skills faster, reducing barriers to adoption. At this stage, oversight remains essential to ensure accuracy and prevent biases from being introduced or propagated. However, as AI advances, we may soon find ourselves sitting back, sipping our coffee, and letting AI do the work. And with patients already performing POCUS at home, there’s no reason urgent care providers shouldn’t be doing it—and doing it better! [9-11]
At Hello Sono, we are committed to clinical excellence and efficiency. We build high-quality, compliant, and profitable POCUS programs to improve patient care and save overall healthcare costs. Access the POCUS ROI Calculators to see the financial impact of POCUS. Fill out the contact form to speak to an expert.
Authored by Tatiana Havryliuk, MD
References
"Ultrasound Guidelines: Emergency, Point-of-Care, and Clinical Ultrasound Guidelines in Medicine." Accessed: Feb 10, 2025. [Online]. Available: https://www.acep.org/patient-care/policy-statements/ultrasound-guidelines-emergency-point-of--care-and-clinical-ultrasound-guidelines-in-medicine
Burgin CM, Fredrick JW, Eicken J. "A POCUS-based approach to acute renal colic in the urgent care center." J Urgent Care Med. October 2020. Available at: https://www.jucm.com/a-pocus-based-approach-to-acute-renal-colic-in-the-urgent-care-center/
Burgin CM, Morrow DS. "Utility of POCUS in skin and soft tissue infection." J Urgent Care Med. June 2020. Available at: https://www.jucm.com/utility-of-pocus-in-skin-and-soft-tissue-infection/.
Mati B, Rutherford R. "Point-of-care ultrasound diagnosis of ruptured ectopic pregnancy in an urgent care setting." J Urgent Care Med. 2023;17(7):19-21.
Accreditation Council for Graduate Medical Education. "ACGME program requirements for graduate medical education in family medicine." Effective July 1, 2024. Accessed February 10, 2025. Available at: https://www.acgme.org/globalassets/pfassets/programrequirements/2025-reformatted-requirements/120_familymedicine_2025_reformatted.pdf
Baloescu C, Bailitz J, Cheema B, Agarwala R, Jankowski M, Eke O, Liu R, Nomura J, Stolz L, Gargani L, Alkan E, Wellman T, Parajuli N, Marra A, Thomas Y, Patel D, Schraft E, O'Brien J, Moore CL, Gottlieb M. Artificial Intelligence-Guided Lung Ultrasound by Nonexperts. JAMA Cardiol. 2025 Jan 15. doi: 10.1001/jamacardio.2024.4991. Epub ahead of print. PMID: 39813064.
Najjar R. Redefining Radiology: A Review of Artificial Intelligence Integration in Medical Imaging. Diagnostics (Basel). 2023 Aug 25;13(17):2760. doi: 10.3390/diagnostics13172760. PMID: 37685300; PMCID: PMC10487271.
Najjar R. Redefining Radiology: A Review of Artificial Intelligence Integration in Medical Imaging. Diagnostics (Basel). 2023 Aug 25;13(17):2760. doi: 10.3390/diagnostics13172760. PMID: 37685300; PMCID: PMC10487271.
Saati A, Au A, Joshi AU, Davis R, West FM, Lewiss RE. Can Untrained Patients Perform Their Own Skin and Soft Tissue Ultrasound Examination by Teleguidance? POCUS J. 2023 Nov 27;8(2):159-164. doi: 10.24908/pocus.v8i2.16454. PMID: 38099176; PMCID: PMC10721299.
Duggan NM, Jowkar, N, Ma IWY. et al. Novice-performed point-of-care ultrasound for home-based imaging. Sci Rep 12, 20461 (2022).
Kirkpatrick AW, McKee JL, Couperus K, Colombo CJ. Patient Self-Performed Point-of-Care Ultrasound: Using Communication Technologies to Empower Patient Self-Care. Diagnostics 2022, 12(11), 2884. PMCID: PMC9689087