STFM Announces New Point of Care Ultrasound Task Force and Initiative on POCUS Family Medicine Education

Dear Editor,

We read with great interest the article by Paulus and Davies announcing the new STFM Point-of-Care Ultrasound (POCUS) Task Force and initiative on family medicine ultrasound education (1). As Spanish family medicine researchers who have collaborated on the European Space Agency’s ALISSE project to develop an AI-supported ultrasound system for astronauts (2), we wholeheartedly agree that ultrasound is becoming an indispensable tool in primary care and even more exotic environments, where health care and monitoring are a top priority. We wish to expand on the points raised by Paulus and Davies, discussing the importance of ultrasound in frontline practice, its advantages over other imaging modalities, and the potential of artificial intelligence to accelerate POCUS training (3), all within a framework of rigorous evidence and prudent educational strategy.

Family physicians are increasingly integrating ultrasound at the point of care to enhance diagnostic accuracy and patient management. Numerous studies have demonstrated that POCUS can improve clinical decision-making, leading to faster diagnoses and treatment while reducing unnecessary referrals. In their narrative review, Carrera et al (4) describe how POCUS applications span from head to toe, aiding in the evaluation of everything from inflammatory conditions and acute abdominal pain to cardiopulmonary dysfunction and musculoskeletal injuries, all of which are commonly encountered in primary care. Importantly, many investigations link POCUS use to reductions in morbidity, mortality, and healthcare costs. For example, a recent systematic review of ultrasound in resource-limited settings (5) found that POCUS frequently changed physicians’ diagnostic impressions (in 15–52% of cases) and altered patient management plans (in 17–87% of cases). The ability to make immediate, accurate assessments at the bedside is very helpful in general practice, where timely decisions can greatly influence outcomes. We have seen first-hand, in remote and extreme environments like spaceflight, that access to point-of-care imaging can be critical when other diagnostic options are absent, a lesson equally applicable to rural clinics and underserved communities on Earth.

Ultrasound offers several comparative benefits that make it especially suited for primary care settings. First, it is radiation-free, which is a significant advantage over X-rays or CT scans, particularly for pregnant patients and children. This safety profile means ultrasound can be used liberally as an extension of the physical exam without cumulative harm. Second, POCUS is portable and brings imaging directly to the patient. Handheld ultrasound devices now can fit in a coat pocket, allowing family physicians to perform real-time imaging during a clinic visit or even in a patient’s home. In contrast, modalities like CT or MRI require patients to travel to specialized facilities and involve delays in obtaining results. Third, ultrasound enables dynamic, real-time assessment. Clinicians can observe organ motion, blood flow, and immediate responses to clinical maneuvers, which static X-rays cannot capture. For instance, lung auscultation has long been a staple of primary care, but lung ultrasound can visually confirm conditions such as pleural effusions or pulmonary edema at the bedside with high accuracy. A recent study in general practice (6) demonstrated that lung ultrasound performed by family physicians could detect pulmonary edema and pneumonia with sensitivities above 85% and specificities above 96%, performance metrics that rival or exceed those of chest radiography. Indeed, evidence suggests that ultrasound can be more sensitive than chest X-ray for certain diagnoses like heart failure-related edema, while offering comparable specificity (7). Fourth, POCUS can increase diagnostic confidence and potentially reduce the need for costly downstream tests. Andersen et al’s systematic review noted that focused ultrasound exams by general practitioners tended to have high diagnostic accuracy with minimal harms, especially when used to answer specific clinical questions, in contrast to comprehensive scans that can pick up incidental findings. Moreover, because POCUS is performed by the treating physician who understands the clinical context, the results can be immediately integrated into management decisions, streamlining care. These advantages underscore why many have begun to call ultrasound the “stethoscope of the 21st century,” and why family medicine, as the point of first contact, stands to gain enormously from its judicious use.

As Paulus and Davies highlight (1), the growing demand for POCUS in clinical practice necessitates robust training in family medicine programs. We concur that simply acquiring ultrasound machines is not enough, clinicians must be properly trained to use them effectively (8–10). The move by the ACGME to require POCUS experience in family medicine residency (effective July 2024) is a key step in acknowledging ultrasound as a core competency. However, implementing this mandate comes with challenges. Early surveys of residency programs revealed a paucity of formal ultrasound training; in 2014, only around 2% of U.S. family medicine residencies had an established POCUS curriculum (11). By 2019, that figure had improved (53% of programs reported having a core ultrasound curriculum) yet nearly one in ten still had no plans to introduce POCUS training (12). Encouragingly, this represents significant progress over just five years, reflecting rapid adoption of ultrasound in training. But improved training uptake has not automatically translated into routine clinical use. Hall et al found that despite more residencies teaching POCUS, the actual use of ultrasound by residents in patient care remained sporadic in many programs; in fact, most surveyed programs performed the majority of POCUS applications fewer than a couple of times per year. This gap between training and practice may be due, in part, to persistent barriers that make it difficult for new POCUS-trained physicians to maintain and apply their skills.

Foremost among these barriers is the lack of trained faculty to teach and supervise ultrasound skills. Family medicine has historically had few ultrasound mentors, as this technology was traditionally the domain of radiologists and cardiologists. As a result, many residency programs struggle to find instructors confident in POCUS, and faculty development in ultrasound lags behind learner interest. The recent CERA survey confirmed that the top obstacles to integrating POCUS in residency training are a dearth of qualified faculty, limited access to equipment, and providers’ discomfort interpreting images without specialist backup. These findings echo earlier reports and have remained unchanged even as POCUS enthusiasm has grown. Clearly, a concerted effort is needed to “train the trainers” and establish standards so that every graduating family physician is competent in fundamental ultrasound applications. In this light, the STFM’s multi-year initiative, Project FOCUS (1), is both timely and important. By convening experts to develop a consensus curriculum and competency-based assessments, this initiative directly targets the current heterogeneity in POCUS education. A standardized curriculum will not only delineate what skills family medicine residents should acquire, but also help define the proficiency benchmarks they must meet. We particularly applaud the plan to use Delphi methods and a broad stakeholder summit to build consensus, as this inclusive approach will ensure the curriculum is realistic and applicable across diverse practice settings. Such standardization efforts, backed by the American Board of Family Medicine, will lay the groundwork for POCUS to become a routine part of primary care practice rather than a niche interest. In contrast, while Spain has several local initiatives, family medicine residency training programs have yet to reach this level of maturity.

In promoting ultrasound training, it is important to remain scientifically rigorous and avoid hyperbole. POCUS is not a panacea or a replacement for all other diagnostics; rather, it is a complementary skill that augments clinical evaluation when used appropriately. The literature still calls for high-quality studies to quantify the impact of POCUS on patient-oriented outcomes in primary care (12,13). As educators, we must be mindful to teach not just the technical skills of image acquisition, but also the critical thinking to know when ultrasound is indicated and when it is not. Overuse or misuse of POCUS could lead to false reassurance or unnecessary anxiety from incidental findings. Thus, integrating ultrasound into family medicine should go hand-in-hand with teaching evidence-based use and maintaining a patient-centered approach. In this context, a competency-based curriculum (as envisioned by STFM) is ideal, because it focuses on achieving concrete skills and decision-making abilities rather than just completing a set number of scans. Early evaluations have shown that even a few hours of focused training can allow generalist physicians to become competent in certain POCUS applications. For example, brief courses on focused cardiac or obstetric ultrasound for general practitioners have yielded respectable diagnostic accuracy with no increase in adverse events, particularly when practitioners stick to targeted questions. These findings support the notion that a well-structured, focused curriculum can produce safe and effective POCUS users without requiring radiologist-level training. The key is ensuring quality and consistency in training – exactly what the STFM initiative aims to achieve. As Paulus and Davies noted (1) , implementation and faculty development will be needed. We agree and further suggest that ongoing mentorship and refresher opportunities (e.g. via online modules or mini-fellowships) be built into the plan so that faculty across the country can confidently teach and sustain POCUS skills in the long run.

One exciting development that can bolster these educational efforts is the advent of artificial intelligence in ultrasound. AI has the potential to function as a virtual instructor, helping to overcome some of the faculty limitations mentioned above (14,15). Our experience with the ALISSE project (2), founded by the European Space Agency, which involved guiding non-expert operators (astronauts) to perform ultrasound in space, has shown us firsthand how powerful AI guidance can be. In that project, a deep-learning system provided real-time feedback on probe position and image quality, effectively “coaching” novices through the scan. The results were striking: even completely untrained individuals were able to capture diagnostic-quality images of organs like the kidney and bladder with high success rates, approaching the performance of experienced radiologists. In fact, in a preliminary study, 90% of AI-assisted scans by novices were judged to be clinically acceptable, a level of quality nearly on par with experts. These findings, now published, validate that intelligent software can dramatically shorten the learning curve for ultrasound beginners (16). The system works by recognizing anatomical structures in the live ultrasound video and guiding the user (through visual cues) on how to adjust the probe position to obtain the standard views. Essentially, it encapsulates the expertise of an instructor into an algorithm available on-demand. This kind of AI-driven guidance could be a game-changer for POCUS training in family medicine: a resident in a rural program, for instance, might practice scanning with an AI tutor when no radiologist is on site, receiving instant corrections and tips.

Early independent research likewise supports AI’s role in accelerating POCUS proficiency. In a recent controlled study, novice operators (nurses and EMTs with no prior ultrasound experience) were randomized to perform trauma ultrasound scans with or without real-time AI guidance (17). Those assisted by the AI had significantly higher rates of clinically acceptable scans (84% vs 68%) compared to the unassisted group. Not only was image quality better, but their confidence improved as well, because the AI could confirm when a correct view was obtained. Interestingly, the total scanning time in the AI group was only marginally longer in the initial attempts, and with practice the AI users became just as efficient as those scanning without assistance. The authors concluded that while AI guidance may add a few minutes up front, it markedly lowers the learning curve for novices by preventing frustration and repeated trial-and-error. A study employing a comparable AI-guided system showed that nurses were able to acquire standard cardiac views with similar success (18,19). We believe this is highly relevant to family medicine training. An intelligent ultrasound system integrated into residency could allow learners to practice autonomously and attain basic competency faster, freeing human instructors to focus on more advanced interpretation and clinical integration. It is conceivable that in the near future, residency clinics might have an AI-enabled ultrasound device that not only helps residents acquire images but also provides preliminary interpretations or flags concerning findings for supervisor review. Such a synergy between physician and AI aligns well with the vision of taking advantage of technology to enhance, not replace, clinical skills (3,10).

While we are optimistic about AI-facilitated POCUS education, we echo the need for measured implementation. Any AI tool introduced into medical training should be thoroughly validated and its limitations clearly understood by users. The goal is to support learning and improve consistency, not to create over-reliance or a false sense of security. For example, an AI might be very effective at guiding image acquisition for certain standardized views, but it may not easily handle atypical anatomies or rare pathologies, which are less represented in the training datasets. Trainees must therefore still learn to recognize when an image “doesn’t look right” and seek expert help, rather than trusting the AI unconditionally. Additionally, ethical considerations around patient data privacy and algorithmic bias need to be addressed when adopting these technologies (20). Fortunately, POCUS datasets used to train AI (such as in ALISSE) often encompass a wide range of normal variants and pathology, helping ensure the tool is robust across populations (21). Ongoing research is exploring how AI can not only guide image capture but also perform real-time image analysis – for instance, automatically calculating bladder volume or detecting an abdominal aortic aneurysm. Such features could augment a family doctor’s capabilities in clinic, but they should complement, not replace, the physician’s judgment. In the end, the integration of AI into POCUS training and practice should be pursued in the same spirit as the STFM initiative itself: thoughtfully, based on evidence, and with the ultimate aim of improving patient care.

The STFM POCUS Task Force initiative comes at a watershed moment for primary care ultrasound. POCUS has proven its worth in a variety of settings, showing advantages in safety, speed, and diagnostic yield that can greatly benefit family medicine. However, to fully realize these benefits, we must build a solid foundation of training and competency among primary care clinicians. Standardizing education and assessment will address the current inconsistencies and ensure that the next generation of family physicians is uniformly prepared to employ ultrasound where appropriate. At the same time, embracing innovative solutions, such as artificial intelligence support, can amplify these educational efforts and help overcome longstanding barriers like limited faculty expertise. Our perspective from working on advanced AI-guided ultrasound systems reinforces the idea that technology, when properly applied, can democratize ultrasound knowledge and make skills acquisition more efficient (22). Going forward, collaboration will be key: among family medicine educators, across specialties (borrowing ultrasound insights from radiology, emergency medicine, and others), and between clinicians and technologists. By cultivating POCUS proficiency through evidence-based curricula, and possibly bolstering it with AI tools, family medicine can continue to strengthen its role at the forefront of accessible, high-quality care. This measured, rigorous approach will ensure that the expansion of ultrasound in primary care remains safe, effective, and aligned with our profession’s commitment to patient-centered practice.

REFERENCES:

1. Paulus R, Davies A. STFM Announces New Point of Care Ultrasound Task Force and Initiative on POCUS Family Medicine Education. Ann Fam Med. 2025 Mar 1;23(2):173–4.

2. Miraut D, Reale G, Illana C, Runge A, Tenajas R, others. Autonomous uLtrasound Image improvement SyStEm (ALISSE) for guiding astronauts to take clinically valuable images in future long manned Space missions. In: Proceedings of 17th International Conference on Space Operations, Dubai, United Arab Emirates. 2013.

3. Tenajas R, Miraut D. Echoes in space: Online training and AI’s potential in advancing ultrasound competency. WFUMB Ultrasound Open. 2023 Dec 1;1(2):100015.

4. Carrera KG, Hassen G, Camacho-Leon GP, Rossitto F, Martinez F, Debele TK. The Benefits and Barriers of Using Point-of-Care Ultrasound in Primary Healthcare in the United States. Cureus. 2022 Aug;14(8):e28373.

5. Baloescu C, Chen A, Schnittke N, Hicks B, Zhu M, Kaili M, et al. Development and interobserver reliability of a rating scale for lung ultrasound pathology in lower respiratory tract infection. WFUMB Ultrasound Open. 2023 Dec 1;1(2):100006.

6. Halata D, Zhor D, Skulec R, Seifert B. Accuracy of point-of-care ultrasound examination of the lung in primary care performed by general practitioners: a cross-sectional study. BMC Prim Care. 2025 Apr 8;26(1):99.

7. Maw AM, Hassanin A, Ho PM, McInnes MDF, Moss A, Juarez-Colunga E, et al. Diagnostic Accuracy of Point-of-Care Lung Ultrasonography and Chest Radiography in Adults With Symptoms Suggestive of Acute Decompensated Heart Failure: A Systematic Review and Meta-analysis. JAMA Netw Open. 2019 Mar 15;2(3):e190703.

8. Tenajas R, Miraut D. Challenges in standardizing image quality across diverse ultrasound devices. Intell Med. 2024 Nov 1;4(4):275.

9. Tenajas R, Miraut D. Rethinking ultrasound probe maintenance in the era of AI. WFUMB Ultrasound Open. 2023 Dec 1;1(2):100014.

10. Tenajas R, Miraut D, Tenajas R, Miraut D. El pulso de la Inteligencia Artificial y la alfabetización digital en Medicina: Nuevas herramientas, viejos desafíos. Rev Medica Hered. 2023 Oct;34(4):232–3.

11. Hall JWW, Holman H, Bornemann P, Barreto T, Henderson D, Bennett K, et al. Point of Care Ultrasound in Family Medicine Residency Programs: A CERA Study. Fam Med. 2015 Oct;47(9):706–11.

12. Hall J, Holman H, Barreto T, Bornemann P, Vaughan A, Bennett K, et al. Point-of-Care Ultrasound in Family Medicine Residencies 5-Year Update: A CERA Study. Fam Med. 2020;52(7):505–11.

13. Baloescu C, Parhar A, Liu R, Wanjiku GW. Effect of Point-of-Care Ultrasound on Clinical Outcomes in Low-Resource Settings: A Systematic Review. Ultrasound Med Biol. 2022 Sep 1;48(9):1711–9.

14. Tenajas R, Miraut D. Ecografía Inteligente. Rev Cuba Informática Médica. 2023;15(2).

15. Tenajas R, Miraut D. El renacimiento tecnológico de la Radiología: la revolución open source y la inteligencia artificial. Rev Cuba Informática Médica. 2023;15(2).

16. Ossaba S, Diez Á, Marti M, Parra-Gordo ML, Alonso-Gonzalez R, Tenajas R, et al. Validation of a deep-learning modular prototype to guide novices to acquire diagnostic ultrasound images from urinary system. WFUMB Ultrasound Open. 2024 Dec 1;2(2):100049.

17. Chiu IM, Lin CHR, Yau FFF, Cheng FJ, Pan HY, Lin XH, et al. Use of a Deep-Learning Algorithm to Guide Novices in Performing Focused Assessment With Sonography in Trauma. JAMA Netw Open. 2023 Mar 28;6(3):e235102.

18. Narang A, Bae R, Hong H, Thomas Y, Surette S, Cadieu C, et al. Utility of a Deep-Learning Algorithm to Guide Novices to Acquire Echocardiograms for Limited Diagnostic Use. JAMA Cardiol. 2021 Jun 1;6(6):624–32.

19. Burke D, Corrigan N, Herlihy M, Nasaj O, Dickson J, Delaney D, et al. Real world evaluation of artificial intelligence echocardiography image guidance and acquisition with novice scanners in multiple clinical settings. Eur Heart J - Cardiovasc Imaging. 2022 Feb 1;23(Supplement_1):jeab289.011.

20. Tenajas R, Miraut D. The Hidden Risk of AI Hallucinations in Medical Practice. Ann Fam Med. 2025 Jan 19;23(1):eLetter.

21. Parra-Gordo ML, Ossaba S, Diez Á, Alonso-Gonzalez R, Marti M, Garzón G, et al. Sistema de guiado mediante inteligencia artificial para la realización de ecografías en las futuras misiones espaciales tripuladas. In Santiago de Chile: Sociedad Chilena de Radiología; 2023. Available from: https://congresochilenoradiologia.cl/envio-de-posters/sistema-de-guiado-...

22. Tenajas R, Miraut D, Illana CI, Alonso-Gonzalez R, Arias-Valcayo F, Herraiz JL. Recent Advances in Artificial Intelligence-Assisted Ultrasound Scanning. Appl Sci. 2023 Jan;13(6):3693.