Feasibility and efficacy of a neurovascular ultrasound training course for undergraduate medical students

Abstract

Background

Neurovascular ultrasound is an easily applicable bed-side tool for guiding the diagnosis and monitoring of cerebrovascular disease. Although sonography training is increasingly integrated in medical school curricula, neurovascular ultrasound has been largely neglected in these efforts. We therefore aimed to assess the feasibility and efficacy of a neurovascular ultrasound training course for medical students.

Methods

Five small-group neurovascular ultrasound course offerings were conducted between 2016 and 2019 at the Medical University of Graz, Austria. This study represents a retrospective analysis of prospectively collected educational course evaluation data. Each course offering followed the same course structure and assessment schedule and was offered to 2nd - to 5th -year medical students. Demographics and previous ultrasound experience were documented. To test pre-course knowledge, all participants completed a theoretical test, followed by the 20-hour hands-on training course comprising a standardized step-by-step examination of the extra- and intracranial brain-supplying arteries. Afterwards, all students underwent a practical exam in single-station Objective Structured Clinical Examination format, conducted by neurovascular ultrasound experts, blinded to the study’s scope and data, using a predefined standard protocol.

Results

A total of 51 students (median age: 23 years, IQR: 1; range: 21–28; 24 females, 47%) participated in the courses. Of those, 27 (53%) had previous ultrasound experience. For the practical exam, participants achieved a median score of 56/66 points (IQR: 9; 85%). 44 students (86%) achieved a score above a 70% passing threshold. Of note, results were independent of previous practical ultrasound experience and theoretical pre-course knowledge (p each > 0.1).

Conclusion

This study demonstrates the feasibility and efficacy of a simply designed neurovascular ultrasound course in medical students. It can provide substantial hands-on competence even in students without previous ultrasound experience, supporting its integration into medical school curricula.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-026-08765-z.

Background

Ultrasound is an essential diagnostic tool in modern medicine due to its non-invasive nature, high safety profile, and relative ease of use [1, 2]. Sonographic exploration of the brain-supplying arteries plays a pivotal role in the diagnosis and monitoring of cerebrovascular disease and is employed by physician across various medical disciplines [3–6].

Consequently, the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) and the World Federation for Ultrasound in Medicine and Biology (WFUMB) have both issued statements emphasizing the importance of incorporating ultrasound courses in medical students’ curricula [2, 7]. These recommendations were supported by several studies that have shown feasibility of the implementation of educational point-of-care ultrasound (POCUS) programs during medical school [8–12], which have been increasingly integrated in education routine over the past decade [13]. However, the incorporation of neurovascular ultrasound of the brain-supplying arteries still lacks empirical evidence to support its effectiveness [14]. Existing data are limited to two studies both of which are constrained by a narrow diagnostic focus (i.e., only carotid artery blood flow and intima-media thickness (IMT) measurement) and by the fact that they included only postgraduate physicians [15, 16].

The limited data on neurovascular ultrasound in undergraduate teaching models is further underlined by a recent systematic review by Neuber et al. highlighting the need for a clearly defined methodological framework, specifying how the training is integrated into the curriculum, and using standardized and objective practical assessment tools to evaluate the achievement of predefined learning objectives [17].

We aimed to evaluate whether a structured neurovascular ultrasound course is feasible to implement and whether undergraduate students achieve adequate practical examination performance in a standardized and objective practical assessment at the end of the course. In addition, we examined whether participant characteristics (e.g., previous ultrasound experience, year of study) are associated with practical skill outcomes.

Methods

Study design and course structure

To address the study objectives, we conducted a retrospective analysis of prospectively and systematically collected educational course evaluation data from a structured neurovascular ultrasound training course for medical students at the Medical University of Graz, Austria. The course was introduced in 2016 and subsequently offered on five occasions between 2016 and 2019, each time in an identical format as an elective course within the medical curriculum. Course structure, examination formats, and assessment time points were predefined based on curricular and didactic considerations and were applied uniformly across all five course offerings between 2016 and 2019.

Enrollment was open to 2nd- to 5th-year medical students via an online registration platform. Students registered for the course during the standard university enrollment periods of the respective semesters.

Each course accommodated a maximum of 12 students, with no more than six students assigned per ultrasound station to ensure optimal hands-on learning. The course comprised 20 h and adhered to a standardized teaching protocol. All sessions were conducted by experienced neurosonography specialists with extensive clinical backgrounds (MK, KN).

The course began with a 2-hour introductory session, covering foundational concepts in anatomy, ultrasound physics, machine handling, and general sonographic principles.

The practical component of the course was delivered through five consecutive hands-on training sessions, each lasting approximately 3.5 h. The sessions followed a fixed sequence, with each one building upon the previous and focusing on a specific anatomical region. Each session began with a 30-minute live demonstration by the instructors, who performed a structured examination of the target vessels on volunteers (Fig. 1). In addition, representative ultrasound images of common vascular pathologies in the trained region (e.g., carotid artery plaques) were presented to support interpretation training (see example in Supplemental Fig. 2).

Fig. 1.

Structure of the neurosonography ultrasound training course in 51 undergraduate medical students. Flowchart showing participants, course elements, and assessments. Study year and previous ultrasound experience were recorded as participant characteristics and did not represent separate teaching groups

This was followed by a 3-hour supervised hands-on training session, during which students practiced the techniques on each other under instructor guidance.

Session 5 was exclusively devoted to integrate all previously learned components into a complete, standardized neurovascular ultrasound examination. The ultrasound machines used during the course included the Vivid E90 and LOGIQ 7 systems (GE HealthCare, Illinois, USA). Extracranial vessels were scanned using a linear array transducer (9 L-D, L8-18i-D), and intracranial examinations were performed with a phased-array transducer (3Sc-RS, S4-1) suitable for transtemporal insonation. An overview of the course structure is provided in Fig. 1.

Study population

A total of 51 students participated in the study, of whom 24 (47%) were female. The median age was 23 years (IQR: 1), and 35 students (69%) were in their 4th or 5th year of medical school. Previous practical ultrasound experience was reported by 27 participants (53%).

No formal calculation of sample size was conducted since all the students enrolled in the neurovascular ultrasound course during the five-year study period, from 2016 to 2019, were included. Hence, the sample size was pragmatically determined by course enrollment and operational feasibility instead of consideration for statistical power.

An overview of participant recruitment, course completion and dropouts for each offering of the courses (2016–2019) is given in the participant flowchart (Supplemental Fig. 1).

Ethics

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The project was planned and carried out as an evaluation of a curricular educational activity without intervention, without collection of sensitive data, and without any performance-relevant consequences for participating students. Participant data were collected in a pseudonymized manner as part of regular teaching activities and in accordance with applicable data protection regulations (Austrian Data Protection Act 2000 and 2018; General Data Protection Regulation [EU] 2016/679). Based on these considerations and in line with institutional regulations for educational evaluations, the responsible Ethics Committee of the Medical University of Graz waived the requirement for formal ethics committee approval.

Written informed consent for participation and for the use of pseudonymized data was obtained from all participants at the beginning of each course. (Supplementary Material, Appendix 1) Volunteers who served as models for the practical ultrasound assessments were informed about the procedures and minimal risks of diagnostic ultrasound and provided written informed consent prior to participation. (Supplementary Material, Appendix 1)

Theoretical pre-course assessment

Theoretical pre-course knowledge was tested in writing via multiple-choice and short-answer questions (consisting of 15 items, maximum score: 30 points). This assessment was administered on the first course day immediately before the introductory session (for details, see Supplemental Table 2).

Practical skills evaluation and data collection

At the end of the course, each student underwent a practical assessment, which lasted approximately 30 min and was conducted one day after completion of the last hands-on session. During this assessment, students were required to perform a standardized neurovascular ultrasound examination of the extra- and intracranial brain-supplying arteries on volunteers with at most minor vascular changes (e.g., increased IMT or non-stenosing atherosclerotic plaques). The volunteers were chosen from among the hospital staff. All volunteers underwent a preliminary ultrasound examination to ensure that vascular status was comparable and free of relevant intra- or extracranial pathology. Each examination was performed by two of three independent neurosonography experts (EGR, AG, DT), who were blinded to the study’s objectives and data. The practical assessment followed the core principles of an Objective Structured Clinical Examination (OSCE), including objective scoring based on predefined criteria, a standardized performance task, a structured examination environment, and evaluation by trained and blinded expert examiners. As the assessment included one structured station instead of the usual multi-station setup linked with a full OSCE, it was implemented as a single-station OSCE exam [18]. The assessed arteries included the common, internal, and external carotid arteries (CCA, ICA, ECA), the vertebral and subclavian arteries (VA, SA), and the major intracranial vessels, including the middle, anterior, and posterior cerebral arteries (MCA, ACA, PCA), the basilar artery (BA), and the intracranial segment of the vertebral arteries. Performance was evaluated according to a predefined protocol outlining the essential components for a complete and accurate examination of each vessel (Supplemental Table 1), in adherence to international recommendations [19]. For extracranial arteries (except subclavian arteries), three domains were observed: (1) anatomical localization, defined as the correct identification of the vessel being insonated; (2) vessel tracking, defined as the accurate visualization of the entire visible course of the vessel insonated; and (3) flow velocity measurement, defined correct if appropriate pulse repetition frequency settings were used, the sample volume was correctly placed, and angle correction was properly applied. Moreover, the correct identification of carotid artery plaques (defined as a focal structure that encroaches into the arterial lumen at least 50% of the surrounding IMT value, or demonstrates a thickness > 1.5 mm as measured from the media-adventitia interface to the intima lumen interface) was assessed [20].

Because of their small insonation window, assessment of intracranial and subclavian arteries was limited to anatomical location and flow velocity. The assessment criteria, outlining the required components, are provided in the Supplements, Table 1.

Table 1.

Participant characteristics and correlation with overall practical examination performance (n = 51)

Variable	Total cohort / overall results (n = 51)	Overall score practical post-course exam (rs)	p-value
Demographics
Age, years (Median, IQR)	23 (1)	0.190	0.182
Male, n (%)	27 (52.9)	0.344	0.014
2th and 3th year students, n (%)	16 (31.4)	0.075	0.602
4th and 5th year students, n (%)	35 (68.6)
Previous ultrasound experience, n (%)	27 (52.9)	0.115	0.422
Examination results	Median pts/max pts (IQR)
Theoretical pre-course  exam	11/30 (6)	− 0.010	0.942
Practical post-course  exam	56/66 (9)	-	-
Practical post-course exam > 70%, n (%)	44 (86.3)	-	-

Correlations calculated using Spearman’s rank correlation coefficient

Abbreviations: r_s Spearman rank correlation, IQR interquartile range, pts Points

Apart from practical skills evaluation, data collection included demographic variables of participating students (age, sex), year of study, previous ultrasound experience and theoretical pre-course knowledge. While previous ultrasound experience was self-reported by each participant and defined as extracurricular hands-on ultrasound experience (i.e., participation in peer-teaching programs or clinical clerkships).

Defined endpoints

For the purpose of the scientific evaluation, the primary outcome measure was practical skill acquisition, operationalized as students’ overall performance in the post-course practical examination. This assessment evaluated three core components of a standardized neurovascular ultrasound examination: accurate anatomical vessel localization, complete vessel tracking, and correct flow velocity measurement.

Secondary outcome measures included the individual components of the practical examination (localization, tracking, and flow measurement) as well as factors potentially associated with overall practical performance. These comprised pre-course theoretical knowledge, previous ultrasound experience, year of study, sex, and age, which were examined as potential predictors of practical skill outcomes.

Statistical analysis

All data were entered into Microsoft Excel and analyzed using the SPSS Statistics (Version 28). Testing for normality of continuous variables was performed using the Shapiro–Wilk test and visual inspection of distribution plots (histograms and Q–Q plots). Owing to the relatively small sample size and to ensure consistent reporting across all outcomes, continuous variables were summarized using median and interquartile range (IQR), irrespective of their underlying distribution. Performance outcomes were compared between students with and without previous ultrasound experience. The Wilcoxon signed-rank test was applied for paired comparisons, while the Mann–Whitney U test was used to compare independent groups. Categorical variables were compared using the chi-square test.

Spearman rank correlation was applied to examine the relationship between overall practical examination scores and tested co-factors, including demographic characteristics and pre-course theoretical knowledge scores.

A multivariable linear regression analysis was performed to assess predictors of practical skill assessment scores at the end of the course. Apart from age and gender, study year, previous ultrasound experience and pre-course theoretical knowledge were included in the model [21]. The statistical significance level of p < 0.05 was defined a priori for all secondary analyses examining potential predictors of practical performance, in accordance with standard practice in educational and clinical research and to avoid data-driven analytical decisions.

A threshold of 70% of all available points in the practical skills examination served as threshold to indicate acceptable performance in a manner consistent with curricular standards as a routine passing threshold in practical exams at the Medical University of Graz.

Results

Theoretical pre-course exam

All students completed the theoretical pre-course exam with a median score of 11 out of 30 points (IQR: 6; 37%) (Table 1 and Supplemental Tables 4 and 5).

Practical skills assessment

In the practical post-course exam, students achieved a median score of 56 out of 66 points (IQR 9; 85%). Overall, 44 students (86%) scored above the 70% passing threshold. Participants reached high median scores in the assessment of extracranial carotid arteries (median: 21 out of 24 points [IQR: 3; 86%]) and intracranial arteries (median: 28 out of 31 points [IQR: 7; 87%)]). For vertebral and subclavian arteries, a median of 7 out of 11 points (IQR: 4; 66%) was achieved (Table 1). 47 students (92%) correctly performed plaque assessment.

Accuracy of vessel localization, vessel tracking and flow measurement

Correct anatomical vessel localization was high in the assessment of all extra- and intracranial arteries (carotid arteries: n = 44, 84%; vertebral and subclavian arteries: n = 42, 82%; intracranial arteries: n = 36, 71%). For vessel tracking, 28 (55%) participants achieved successful acquisition of the extracranial carotid arteries, whereas the success rate for the vertebral arteries was low (n = 14, 28%). Correct flow velocity measurement was moderate to low across all extra- and intracranial vessels (extracranial vessels n = 13, 35%; intracranial vessels n = 21, 41%), with the lowest success again in vertebral / subclavian artery assessment (n = 14, 28%; for details, see Table 2; Fig. 2).

Table 2.

Overall post-course exam scores by vessel and correct identification, tracking, and flow measurement (n = 51)

Practical exam results by vessel	Median pts/max pts (IQR, %)	anatomical localization; n (%)	vessel tracking; n (%)	flow velocity measurement; n (%)
Extracranial vessels	28/35 (8, 80)	35 (69)	10 (20)	13 (26)
Carotid arteries	21/24 (3, 88)	43 (84)	28 (55)	18 (35)
Vertebral and subclavian arteries	7/11 (4, 64)	42 (82)	14 (28)	14 (28)
Vertebral arteries	7/9 (5, 78)	44 (86)	14 (28)	25 (49)
Subclavian arteries	2/2 (2, 100)	43 (84)	n.a.	30 (59)
Intracranial vessels	28/31 (7, 90)	36 (71)	n.a.	21 (41)
Transcranial vessels	16/16 (3, 100)	37 (73)	n.a.	30 (59)
Transnuchal vessels	13/14 (8, 87)	42 (82)	n.a.	23 (45)

Transcranial vessels included the anterior cerebral artery (ACA), middle cerebral artery (MCA), and posterior cerebral artery (PCA). Transnuchal vessels included the basilar artery (BA) and intracranial segment of the vertebral artery (VA)

Abbreviations: IQR interquartile range, pts points, “n.a.” indicates not applicable

Fig. 2.

Performance of medical students in the neurosonography course in 51 undergraduate medical students. A Distribution of scores in the practical examination, expressed as percentage of correct performance across all examined vessels. The boxplots show median, interquartile range, minimum/maximum values, and outliers. B Percentage of students successfully completing anatomical localization, vessel tracking, and flow velocity measurement for each vascular region (carotid, vertebral, subclavian, transcranial, transnuchal arteries)

Predictors of practical skills assessment scores

In univariable analysis, male sex was associated with higher post-course practical skills assessment scores (Spearman rank correlation [r_s]: 0.344, p = 0.014), whereas previous ultrasound experience (r_s=0.422), theoretical pre-course test scores (r_s=-0.010) and year of study (4th -5th year students: r_s=0.075) were not related to practical skills performance (all p > 0.1). After adjustment, male sex remained an independent predictor for higher practical skills assessment scores (B 5.7, 95% CI 1.3–10.2, p = 0.013) (Table 3).

Table 3.

Predictors of practical post-course exam scores in multivariable linear regression analysis (n = 51)

	β (SE)	95% Confidence interval	p-value
Age (years)	0.132 (0.71)	-1.30, 1.56	0.853
Sex (male)	5.74 (2.22)	1.27, 10.21	0.013
Previous ultrasound experience (yes)	1.60 (2.37)	-3.17, 6.37	0.500
Study year (4th and 5th )	2.31 (2.29)	-2.30, 6.92	0.319
Theoretical exam pre-course	− 0.47 (0.23)	− 0.93, − 0.01	0.053

Abbreviations: β Beta Coefficient, SE Standard Error

Discussion

This study evaluating the efficacy and feasibility of a neurovascular ultrasound course for medical students shows high overall performance in the post-course practical exam (median overall score: 85%) with 44 out of 51 students (86%) achieving a score above the routine exam passing threshold of 70%. Of note, these results were independent of previous ultrasound experience.

From a methodological perspective, this study represents a borderline case, as performance data were collected prospectively for educational and quality assurance purposes, while the scientific analysis was conducted retrospectively. This approach reflects the pragmatic structure of curricular educational research and does not involve reconstruction of outcome data.

To date, there has been very limited evaluation of neurovascular ultrasound training courses. The only available data emerges from two studies focusing on carotid artery ultrasound hemodynamics [15] and carotid artery intima-media thickness measurement [16]. While both studies demonstrated high success rates for tightly structured ultrasound training among emergency medicine trainees and novice healthcare providers, their scope was limited to carotid artery blood flow and intima-media thickness measurement and did not include undergraduate participants [15, 16].

Our results demonstrate that a comprehensive neurovascular ultrasound course with emphasis on small-group, hands-on training, can be feasibly integrated into undergraduate medical curricula. Furthermore, the high post-course examination scores indicate substantial success in practical skill acquisition among participating students. These results are remarkable as they are based on structured hands-on assessments evaluated by ultrasound specialists blinded to the study’s scope. In contrast, the few previous studies on (non-vascular) ultrasound training in medical students focused only on theoretical knowledge gains and self-reported confidence, without objective evaluation of practical skills [9, 22, 23]. This is of importance, as practical competence has been identified as directly influencing patient care [24, 25].

To our knowledge, this is the first study to systematically report learning outcomes in the ultrasound-based examination of different brain-supplying arteries with respect to anatomical localization, vessel tracking and blood flow measurement. Previous studies and reviews on neurovascular ultrasound have often highlighted varying levels of complexity in assessing different brain-supplying arteries (e.g., greater difficulty with vertebral versus carotid arteries), but these statements were largely based on expert opinion rather than systematic data [26, 27]. In contrast, the standardized assessment approach used in our evaluation allows a detailed analysis of specific learning challenges in neurovascular ultrasound.

We found high success rates in the correct localization of all trained intra- and extracranial arteries, including vessel tracking of the carotid arteries, whereas performance declined for the vertebral and subclavian arteries. This was primarily attributable to suboptimal vessel tracking of the vertebral arteries, likely due to their deeper anatomical course and the challenges of visualization in all extracranial segments, and underlines previously proposed assumptions [26, 27]. A second major challenge was the accurate measurement of flow velocity across all regions. Notably, these assessments were judged correct only if strict evaluation criteria were met, including appropriate pulse repetition frequency, sample volume placement, and angle correction. Taken together, our data suggest that future neurovascular ultrasound training should account for the increased learning demands associated with vertebral artery tracking and the correct application of Doppler techniques.

Of note, our results were independent of previous practical ultrasound experience. To date, no studies have assessed whether such experience affects learning in neurovascular ultrasound, though Knobe et al. found no impact of previous anatomical knowledge on musculoskeletal ultrasound training [21]. Nevertheless, this supports the feasibility of our course, showing that essential neurovascular skills can be acquired without previous experience.

In the multivariable analysis, male sex emerged as the only independent predictor of higher practical exam scores. This finding was unexpected, as previous studies on practical ultrasound courses reported no sex differences in learning outcomes [9, 12]. Nevertheless, research on postgraduate education has demonstrated a gender bias in medical training [28–30]. It was observed that male students more frequently request extra hands-on time, consistent with studies suggesting that men may be more assertive in seeking practice opportunities [31–33]. This may have influenced our results, as we did not control for individual training time. Future research should overcome such a limitation by systematically documenting participants’ hands-on time.

It should be noted that the practical examination was conducted on volunteers without severe vessel pathologies. Although participants were trained to identify plaques and intima–media thickening—skills that were also evaluated in the practical post-course examination—the course was not designed to assess more advanced pathological conditions (e.g., stenosis grading). Nevertheless, representative images of common vascular pathologies were presented for various intra- and extracranial vessels during the sessions (see Supplemental Fig. 2). As pathology recognition was not an intended learning objective of this introductory course, the use of volunteers with at most minor vascular changes was appropriate for the targeted skill training. To better reflect clinical conditions, a future extension of the curriculum may include an additional module focusing on common neurovascular pathologies.

A further limitation is that neither the theoretical test nor the practical skills examination used in this study constitute formally validated assessment instruments. Although the practical examination was structured in a single-station OSCE format and based on a checklist derived from international recommendations, formal procedures to evaluate reliability or other psychometric properties were not conducted. Accordingly, the findings reflect structured expert assessments and should be interpreted with this consideration in mind.

Due to the multi-year design and the inclusion of several consecutive course cohorts, some degree of cohort-related variability cannot be excluded and should be considered as a limitation.

Another limitation of the study is that we did not include either a post-course theoretical assessment or a structured pre-course practical examination, both of which could have provided additional insight into knowledge acquisition and skill development. However, approximately half of the participants had no prior exposure to ultrasound machines or baseline knowledge of vascular scanning, and under these circumstances a pre-course practical test would not have yielded meaningful or interpretable baseline data.

Finally, our data do not permit conclusions about whether the practical skills acquired in this course are retained long term. This highlights the need for further studies to assess skill durability—e.g., through follow-up at 3–6 months.

Conclusion

In conclusion, our study demonstrates that a short, structured neurovascular ultrasound course is both feasible and effective for undergraduate medical students. It can provide substantial hands-on competence even in students without previous ultrasound experience, supporting its integration into medical school curricula.

Abbreviations

ACA

Anterior cerebral arteries

BA

Basilar artery

CCA

Common carotid arteries

ECA

External carotid arteries

EFSUMB

European Federation of Societies for Ultrasound in Medicine and Biology

ICA

Internal carotid arteries

IMT

Intima-media thickness

IQR

Interquartile range

MCA

Middle cerebral arteries

OSCE

Objective Structured Clinical Examination

PCA

Posterior cerebral arteries

POCUS

Point-of-care ultrasound

SA

Subclavian arteries

SD

Standard deviation

VA

Vertebral arteries

WFUMB

World Federation for Ultrasound in Medicine and Biology

Authors’ contributions

KN, SH, CE, and MK contributed to the conception and design of the study.EW, MK, EGR, DT, and AG were involved in data collection.JE and MK performed the statistical analysis and interpreted the results.JE and MK drafted the manuscript.All other authors (TG, SFH, NB, IH, MH, MT) contributed to substantial manuscript revisions.All authors read and approved the final version of the manuscript.

Funding

None.

Data availability

Data from this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The project was planned and carried out as an evaluation of a curricular educational activity without intervention, without collection of sensitive data, and without any performance-relevant consequences for participating students. Participant data were collected in a pseudonymized manner as part of regular teaching activities and in accordance with applicable data protection regulations (Austrian Data Protection Act 2000 and 2018; General Data Protection Regulation [EU] 2016/679). Based on these considerations and in line with institutional regulations for educational evaluations, the responsible Ethics Committee of the Medical University of Graz waived the requirement for formal ethics committee approval.

Written informed consent for participation and for the use of pseudonymized data was obtained from all participants at the beginning of each course. (Supplementary Material, Appendix 1) Volunteers who served as models for the practical ultrasound assessments were informed about the procedures and minimal risks of diagnostic ultrasound and provided written informed consent prior to participation. (Supplementary Material, Appendix 1).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.