Application of VR technology in standardized residency training for endoscopic procedures in otorhinolaryngology

Abstract

Objective

To investigate the pedagogical efficacy of virtual reality (VR)-assisted training in the standardized education of otorhinolaryngology residents performing endoscopic procedures.

Methods

Otorhinolaryngology residents were enrolled from Beijing Tsinghua Changgung Hospital Affiliated to Tsinghua University, between January and November 2024. They were randomly divided into two groups: (1) a model simulation (MS) training group utilizing plastic models and electronic nasopharyngoscopes; (2) a VR-based training group integrating theoretical instruction with immersive three-dimensional (3D) virtual simulation for both practice and assessment. Upon completion of the training modules, trainees were evaluated through a standardized knowledge examination, objective structured assessments of endoscopic skills, and satisfaction surveys. Statistical analyses were performed using SPSS 29.0, employing independent-samples t-tests and chi-square tests where appropriate.

Results

A total of 46 resident physicians were enrolled in the present study, including 24 residents allocated to the VR training group and 22 residents assigned to the MS training group. The VR-based training group exhibited better performance in standardized knowledge examination [(74.45 ± 2.55) points vs. (67.38 ± 5.95) points, P < 0.001] and objective structured assessments of endoscopic skills [(11.14 ± 2.05) points vs. (9.29 ± 1.73) points, P = 0.001], compared with the traditional and simulation groups. Additionally, the overall percentage of satisfaction with the training modality was higher in the VR group (66.83% vs.61.33%).

Conclusion

VR-based training enhances cognitive acquisition and technical proficiency in endoscopic procedures among otorhinolaryngology residents. This approach also contributes to greater training satisfaction and self-efficacy in independently conducting endoscopic examinations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-026-09576-y.

Background

Initiated in 2014 under the guidance of the National Health and Family Planning Commission of mainland China, the Standardized Training For Resident Doctors (STRD) was established as a structured form of postgraduate medical education designed to improve the quality of physician training [1, 2]. While the program encompasses a wide range of medical subspecialties and aims to cultivate safe, independent, and proficient junior clinicians, its application remains relatively new in certain disciplines, particularly in otorhinolaryngology.

Otorhinolaryngology–head and neck surgery relies heavily on endoscopic techniques, including nasal endoscopy, otoendoscopy and nasolaryngology. The widespread adoption of endoscopy has significantly enhanced disease detection rates and surgical outcomes [3–8]. However, traditional clinical teaching methods for endoscopy are often time-consuming and ineffective and may pose potential risks to patient safety. VR technology, which integrates advanced multimedia with immersive simulation, has emerged as a valuable tool in medical education. One study has proposed the application of VR temporal bone anatomy teaching in otorhinolaryngology education, and the results showed that 85% of the trainees considered the VR model superior to the traditional model [9].VR applications are now prevalent in areas such as three-dimensional (3D) anatomical visualization, disease modeling and diagnosis, and procedural training, including laparoscopic and simulated surgical techniques. These applications have demonstrated improvements in both educational efficiency and training outcomes while minimizing risks to patients [10]. However, the potential of VR in otorhinolaryngology endoscopic teaching remains insufficiently explored, representing an urgent research gap that needs to be addressed.

To address the specific needs of endoscopic training in otorhinolaryngology, our research group has designed a VR-based training platform, ENDOSCOPE FOR ENT V3.03. This software enables the simulation of clinical scenarios, facilitating mastery of otorhinolaryngology anatomy, recognition of disease presentations, and rehearsal of key endoscopic maneuvers. This study evaluates the educational effectiveness of integrating this VR platform into the STRD curriculum for otolaryngology residents, aiming to support its broader adoption in competency-based surgical education.

Materials and methods

Study design

A randomized controlled trial was conducted to compare the educational outcomes of two endoscopic training modalities for otorhinolaryngology residents: VR-based training (VR group) and physical model-based simulation training (MS group). The primary outcomes assessed were performance on the standardized knowledge examination, proficiency in endoscopic skills, and training satisfaction.

Population

Following approval from the Institutional Review Board of Beijing Tsinghua Changgung Hospital, resident physicians enrolled in the STRD in the Department of Otorhinolaryngology-Head and Neck Surgery of Beijing Tsinghua Changgung Hospital Affiliated to Tsinghua University were recruited between January and November 2024. Inclusion criteria included first-year and second-year residents in standardized residency training, with no prior clinical experience in endoscopic procedures.

Randomization methodology

Participant randomization was conducted via computer-generated random numbers in SPSS 29.0. A fixed random seed was used to generate a non-repetitive allocation sequence, and participants were sequentially assigned without manual intervention to reduce selection bias. Strict allocation concealment was achieved using sequentially numbered, identical opaque envelopes, which were securely stored in a locked cabinet. All research staff remained blinded to group allocation throughout the trial, and envelopes were opened only after baseline data collection. Random sequence generation and envelope preparation were independently performed by an administrator who was not involved in participant recruitment, intervention delivery, outcome assessment, or statistical analysis. No conflicts of interest existed, and the allocation procedures were fully documented to ensure unbiased, transparent group assignment.

Training curriculum

The training curriculum encompassed foundational knowledge in otorhinolaryngology, technical principles of endoscopic examination, and surgical techniques. Trainees were randomly divided using a random number table into the VR group and the MS group. Each group underwent a three-month training program, and all theoretical sessions and practical skills training were conducted by the same associate chief physician.

Training modalities

Both groups followed a unified three-phase standardized training protocol. In Phase 1, trainees received structured instruction in core otorhinolaryngology concepts and the theoretical principles of endoscopic procedures. Advancement to the next phase required successful completion of a theoretical assessment. Phase 2 involved hands-on skills training, with the MS group adopting dedicated plastic simulation models for ear, nose, and throat endoscopic procedures, while the VR group completed corresponding hands-on practice via the virtual reality simulation system. The simulation training utilized high-fidelity plastic anatomical models fabricated from XM550 (provided by Xinman Company). Endoscopic procedures were performed using the Karl-Strotz endoscopy system (KARL STORZ SE & Co. KG, Tuttlingen, Germany) including nasal endoscopes, otoscopes, and nasopharyngeal laryngoscopes. Trainees adjusted a rigid nasal endoscope (diameter 4.0 mm, 0° viewing angle) to standard settings, inserted it into the nasal cavity through the inferior nasal meatus, sequentially navigated the middle meatus, ethmoid sinus, and sphenoid sinus recess, identified key anatomical landmarks, maintained at least 1 mm mucosal distance to avoid trauma, and completed 3 practice cycles with 1-minute rests between cycles. For laryngoscopy, a rigid stroboscopic laryngoscope (diameter 3.5 mm, 70° viewing angle) was connected to the imaging system and stroboscopic light source. Trainees inserted it along the oral midline, examined target anatomical areas sequentially, observed vocal fold mobility via the stroboscopic function, and completed 2 examination cycles and 1 mobility observation cycle with 30-second rests between cycles. For otoscopy, a rigid otoscope (diameter 2.7 mm, 0° viewing angle) was used. Trainees straightened the external auditory canal, inserted the otoscope slowly, identified key ear structures, maintained proper centering of the scope, and completed 4 practice cycles (2 cycles per ear) with 30-second rests between cycles. Following the completion of trainer-supervised hands-on training in Phase 2, eligible trainees progressed to Phase 3, the unsupervised independent practice stage, to consolidate the skills and standardized knowledge examination acquired in the first two phases. During this stage, trainees in the MS group performed the full set of otolaryngology endoscopic procedures independently using the same plastic simulation models as in Phase 2, following the standardized operational steps and technical requirements for nasal endoscopy, laryngoscopy and otoscopy respectively. No real-time guidance or direct intervention from trainers was provided during independent practice; trainees were required to complete a fixed number of standardized practice cycles autonomously, self-monitor operational accuracy and standardization, and record key operational details. This stage aimed to enhance trainees’ operational proficiency, independent performance ability and clinical confidence, laying a solid foundation for subsequent supervised clinical practice.

The VR group underwent the same Phase 1 theoretical instruction and assessment requirements as the MS group. Phase 2 consisted of supervised immersive VR-based hands-on training, while Phase 3 was dedicated to unsupervised independent VR practice, which was tailored to consolidate the skills mastered in the first two phases. Prior to Phase 2 simulated training, the instructor provided standardized demonstrations and detailed orientation to the VR system interface for all trainees. During the supervised VR training in Phase 2, trainees operated a physical endoscope attached with VR controllers, using the ENDOSCOPE FOR ENT (V3.03) simulation system and VR headsets to complete designated operational tasks. The training covered standardized procedural modules for nasal endoscopy, otoscopy, and laryngoscopy. The system ran on the PICO 4 Ultra MR platform, which integrated high-definition visualization, immersive VR headsets, and haptic-feedback controllers to highly replicate the tactile and visual experience of real clinical endoscopic operations. During both supervised Phase 2 training and independent Phase 3 practice, trainees received real-time multi-dimensional performance feedback. The system objectively assessed technical proficiency via indicators including motion smoothness, image stability, and simulated tissue responses. In Phase 3 independent VR practice, trainees completed the full set of ENT endoscopic simulation operations autonomously, without real-time instructor guidance or intervention, following the same standardized procedures as Phase 2. The VR system continuously provided real-time performance feedback and automatically generated and summarized detailed training logs throughout both phases. Instructors then delivered targeted guidance based on these logs and feedback, focusing on key skills including localization and exposure of the pharyngeal orifice of the eustachian tube, navigation through anatomically narrow passages, and laryngeal cavity assessment from different vocal fold positions (Fig. 1).

Fig. 1.

User interface and instructional environment of the VR endoscopy training platform (ENDOSCOPE FOR ENT V3.03). A Login interface with selectable modes for practice and assessment. B Scenario selection interface. C Real-time VR simulation interface illustrating a trainee performing laryngoscopy using the VR module. Upon entering the laryngeal cavity, activation of the controller’s trigger initiates the simulation of vocal cord movement and vibration. D Instructor actively supervising and providing feedback during VR-based training sessions

All training sessions across the three phases were scheduled three times per week for two weeks, with each 60-minute session including 10 min of pre-training review, 40 min of hands-on or independent practice (equally divided among three endoscopic modalities), and 10 min of trainer feedback.

Evaluation of training effectiveness

Standardized knowledge examination

A written examination was conducted to evaluate participants’ professional theoretical knowledge. Developed by the Department of Otorhinolaryngology, School of Clinical Medicine, Tsinghua University, the test items were selected from the National Standardized Residency Training Examination Question Bank established and managed by the Talent Exchange Service Center of the National Health Commission, which guaranteed the authority and standardization of the examination content. The total score of the examination was 100 points, equally divided into five dimensions, each accounting for 20 points: anatomical and physiological foundations (covering sinus, middle ear, pharyngeal cavity, vocal cord anatomy and cervical lymph node classification); disease pathogenesis (including pathogenic mechanisms of chronic sinusitis, chronic otitis media and vocal cord inflammatory disorders); clinical disease manifestations (involving typical symptoms of chronic sinusitis, acute otitis media, acute and chronic pharyngolaryngitis, and laryngeal tumors); diagnostic and differential diagnostic reasoning (including differential diagnosis of nasal inverted papilloma, chronic suppurative otitis media, vocal cord leukoplakia and papillary thyroid carcinoma); and disease treatment principles (covering the management of chronic sinusitis, acute otitis media, acute pharyngolaryngitis, epistaxis, and the classification and therapeutic principles of sudden deafness). Formulated in strict accordance with the core curriculum syllabus, this knowledge assessment has been routinely applied in regular academic evaluations of medical students and further reviewed and validated by a panel of senior medical education experts, confirming its comprehensive content coverage, reasonable difficulty distribution, and good construct validity.

Objective endoscopic skills

Following completion of the training program, trainees performed endoscopic examinations on standardized patients, and two chief physicians who were blinded to the group allocations independently evaluated the trainees’ endoscopic performance and assigned corresponding scores. The evaluation comprised two core components: procedure duration and operational proficiency. The operational assessment focused on five specific dimensions: smoothness of endoscope insertion into the nasal cavity, external auditory canal, and oropharynx; proficiency in navigating anatomical constrictions; visual field stability during operation; degree of mucosal trauma; and clarity of anatomical structure exposure. Each dimension was scored on a 3-point scale, with a total operational score of 15 points. Detailed evaluation criteria are provided in Table 1.

Table 1.

Endoscopic performance scoring rubric

Assessment dimension		1 point	2 points	3 points
Smoothness of endoscope manipulation	Entry into the nasal cavity, ear canal, or oropharynx is difficult; repeated attempts are required		Minor difficulty or brief pauses during insertion	Smooth entry
Operational proficiency	Unable to navigate anatomical constrictions without		Able to pass with occasional pauses	Navigates constrictions smoothly and efficiently
Visual field stability	The visual field frequently deviates from target structures		Occasional misalignment of the visual field	Visual field consistently maintained on target
Mucosal trauma	Frequent contact with the mucosa		Occasional mucosal contact	No mucosal contact
Anatomical structure exposure	Key anatomical structures are inadequately visualized		Most anatomical structures are clearly exposed	All relevant anatomical structures are clearly exposed

Training satisfaction survey

Following the completion of the training intervention, trainees in both groups were invited to complete an anonymous, online satisfaction questionnaire designed by the research team, which was specifically designed for this study. Each item was rated on a 5-point Likert scale, ranging from 1 (very dissatisfied) to 5 (extremely satisfied), yielding a total score of 50 points. Domain-specific maximum scores were 25 points for instructional satisfaction, 15 points for skill enhancement, and 10 points for recommendation intention. This questionnaire was specifically developed for this study and has not yet been subjected to external validation. Detailed evaluation criteria are provided in Table 2. The specific English version of the training satisfaction survey is in the supplementary materials.

Table 2.

Training satisfaction survey structure and rating scale

Domain	Item	Very dissatisfied/‌Extremely Unwilling / Extremely Unlikely (1)	Dissatisfied /Moderately Unwilling / Moderately Unlikely (2)	Neutral (3)	Relatively satisfied/Moderately Willing / Moderately Likely (4)	Extremely satisfied /Extremely Willing / Extremely Likely (5)
Instructional satisfaction	1. Satisfaction with the endoscopy training modality	1	2	3	4	5
	2. Satisfaction with the endoscopy assessment results	1	2	3	4	5
	3. Satisfaction with the training duration and difficulty level	1	2	3	4	5
	4. Clarity and understandability of the training modality	1	2	3	4	5
	5. Confidence in independently performing endoscopy after training	1	2	3	4	5
Perceived skill enhancement	6. Improvement in endoscopic procedural skills	1	2	3	4	5
	7. Mastery of basic clinical knowledge and skills	1	2	3	4	5
	8. Effectiveness of resolving issues encountered during training	1	2	3	4	5
Recommendation intention	9. How willing are you to continue using the current VR-based training modality?	1	2	3	4	5
	10. How likely are you to recommend this VR-based training modality to others?	1	2	3	4	5
	Total score

Statistical analyses

All statistical analyses were conducted using SPSS version 29.0 (IBM Corp., Armonk, NY, USA). Baseline demographic characteristics were analyzed using the chi-square (χ²) test or Fischer’s exact test to assess group comparability. Categorical variables were expressed as frequencies and percentages. Continuous variables were expressed as means ± standard deviation (SD) and compared between groups using independent-samples t-tests. A non-parametric Mann-Whitney U test was used to assess differences in responses to the survey between the VR and MS groups. Since multiple comparisons were performed for three primary outcomes, the Bonferroni correction was applied to adjust the significance level. Accordingly, the two-tailed significance level α was set to 0.05/3 = 0.017 for all comparisons.

Results

Cohort analysis

A total of 46 otorhinolaryngology resident trainees were enrolled and were randomly divided to the MS group (n = 22) and the VR group (n = 24). The study population comprised 27 females and 19 males, with the majority aged between 21 and 25 years old. Of the total participants, 31 were first-year residents and 15 were in their second year of the STRD. Comparative analysis revealed no statistically significant differences between the two groups in terms of gender distribution, age, academic standing, or training year. Baseline demographic characteristics of the involved residents are shown in Table 3.

Table 3.

Baseline demographic characteristics of resident trainees in the VR and MS groups

Characteristic	VR group (n = 24)	MS group (n = 22)	P value
Female, n (%)	13 (54.17%)	14 (63.64%)	0.561
Age (years), mean ± SD	23.88 ± 1.56	23.23 ± 1.34	0.125
Trainee type, n (%)			0.552
Master’s degree students	14 (58.33%)	15 (68.18%)
Standardized residents	10 (41.67%)	7(31.82%)
Training year, n (%)			0.754
First year	17 (70.83%)	14 (63.64%)
Second year	7 (29.17%)	8 (36.36%)

Quality of training

Upon completion of the training, all trainees undertook a standardized knowledge examination. The VR group achieved significantly higher total scores [(74.45 ± 2.55) points] compared with the MS group [(67.38 ± 5.95) points], with the difference reaching statistical significance (P < 0.001). In domain-specific comparisons, the VR group scored significantly higher in anatomical and physiological foundations and treatment principles of diseases (P < 0.001 for both). No significant differences were observed in the domains of disease pathogenesis, clinical manifestations of diseases, and diagnostic and differential diagnostic reasoning (P = 0.670, P = 0.198, and P = 0.195). Detailed results are presented in Table 4.

Table 4.

Comparison of standardized knowledge examination scores between the VR and MS groups (mean ± SD)

Assessment domain	VR group	MS group	t value	P value
Anatomical and physiological foundations	15.82 ± 1.68	13.50 ± 1.93	4.32	< 0.001
Disease pathogenesis	14.50 ± 2.19	14.25 ± 1.75	0.42	0.670
Clinical manifestations	14.27 ± 1.51	14.13 ± 2.27	0.256	0.798
Diagnostic and differential diagnostic reasoning	14.27 ± 1.63	13.42 ± 2.62	1.31	0.195
Treatment principles	15.59 ± 1.53	12.08 ± 2.85	5.11	< 0.001
Total score	74.45 ± 2.55	67.38 ± 5.95	5.15	< 0.001

Performance on the endoscopic skill assessment of the endoscopic skill assessment was significantly higher in the VR group than in the MS group. The total scores of the objective structured assessment of endoscopic skills in the VR and MS groups were (11.14 ± 2.05) points vs. (9.29 ± 1.73) points (P = 0.001), the inter-rater reliability of the objective structured endoscopic skill assessment was evaluated by calculating the intraclass correlation coefficient (ICC). The ICC values were 0.89 in the VR group and 0.87 in the MS group, indicating excellent consistency and reliability among independent blinded raters. The VR group exhibited superior performance in smoothness of endoscopic manipulation, operational proficiency, and visual field stability (all P < 0.01). No statistically significant differences were observed in the domains of mucosal trauma or anatomical structure exposure between the two groups (P > 0.05). Full results are detailed in Table 5.

Table 5.

Comparison of objective structured assessments of endoscopic skills results between VR and MS groups (mean ± SD)

Assessment domain	VR group	MS group	t value	P value
Operation time (min)	12.59 ± 2.26	12.17 ± 3.23	0.51	0.612
Smoothness of endoscope manipulation	2.45 ± 0.59	1.87 ± 0.53	3.47	0.001
Operational proficiency	2.31 ± 0.64	1.79 ± 0.65	2.73	0.009
Visual field stability	2.36 ± 0.49	1.87 ± 0.61	2.96	0.004
Mucosal trauma	1.86 ± 0.77	1.95 ± 0.62	0.45	0.648
Anatomical structure exposure	2.13 ± 0.56	1.79 ± 0.77	1.70	0.094
Total score	11.14 ± 2.05	9.29 ± 1.73	3.30	0.001

Training satisfaction

The reliability of the questionnaire was evaluated using Cronbach’s α coefficient, with a resulting value of 0.92. The validity was supported by a Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy of 0.865 (P < 0.05). In addition, five senior otorhinolaryngology experts confirmed the good content validity of the questionnaire by evaluating the rationality, relevance and clarity of all items. A total of 46 satisfaction questionnaires were distributed via a mobile application (WeChat^®), and all were completed and returned, resulting in a 100.00% response rate. The survey evaluated three core domains: instructional satisfaction, perceived skill improvement, and recommendation intention. The VR group demonstrated higher total satisfaction scores than the MS group (P < 0.001; Fig. 2A). A radar chart was used to illustrate the percentage of trainees in each group who reported being either “relatively satisfied” or “extremely satisfied” across the three measured dimensions (Fig. 2B). Across all dimensions, the VR group consistently demonstrated higher satisfaction levels than the MS group. Further statistical analysis showed that the differences in instructional satisfaction and recommendation intention between the two groups were statistically significant (P < 0.05 and P < 0.001; Fig. 2C and E). However, no statistically significant difference was found in the domain of perceived skill improvement (P > 0.05; Fig. 2D). The satisfaction of residents with their residency training method is presented in Fig. 2.

Fig. 2.

Comparative analysis of satisfaction survey scores between the VR and MS groups. A The difference in the total satisfaction scores between the two groups of trainees. B Radar chart depicting the percentage of trainees in each group who reported being “relatively satisfied” or “extremely satisfied” across the three evaluation dimensions: instructional satisfaction, perceived skill improvement, and recommendation intention. C Comparison of instructional satisfaction scores between groups. D Comparison of perceived skill improvement scores. E Comparison of recommendation intention scores. ns: not significant (P>0.05); *: P < 0.05; ****: P < 0.001

Further analysis of individual questionnaire items revealed notable differences between the VR and MS groups. The largest differences in high satisfaction rates (combining “relatively satisfied” and “extremely satisfied”) were observed in Q2 (satisfaction with the endoscopy assessment results, 77.3% in the VR group vs. 29.2% in the MS group) and Q9 (willingness to continue using the current training modality, 63.6% in the VR group vs.12.5% in the MS group). In contrast, the smallest differences in satisfaction were found in Q5 (confidence in independently performing endoscopy after training, 33.3% in the VR group vs. 36.4% in the MS group) and Q7 (mastery of basic clinical knowledge and skills, 66.7% in the VR group vs. 59.1% in the MS group). With regard to dissatisfaction (combining “dissatisfied” and “very dissatisfied”), the largest group differences were observed in Q9 (4.6% in the VR group vs. 54.2% in the MS group) and Q10 (willingness to recommend the training modality to others, 0% in the VR group vs. 33.4% in the MS group). The smallest differences in dissatisfaction appeared in Q7 (4.5% in the VR group vs. 0% in the MS group) and Q8 (effectiveness of resolving issues encountered during training, 4.5% in the VR group vs. 12.5% in the MS group). These item-specific results are illustrated in Fig. 3.

Fig. 3.

Item-level satisfaction distribution across the VR and MS groups

Discussion

Endoscopic proficiency constitutes a core competency for otorhinolaryngology residents. Nevertheless, conventional training approaches—including textbook-based learning and static two-dimensional anatomical diagrams—exhibit substantial limitations. These traditional methods fail to deliver immersive spatial cognition and hands-on practice, frequently leading to fragmented knowledge acquisition and insufficient technical proficiency, thus widening the discrepancy between theoretical learning and clinical application. Moreover, while traditional physical model simulation provides tactile feedback, it is restricted by constraints such as limited physical consumables, inflexible scenario resetting, and inadequate opportunities for repetitive practice.

Existing evidence has confirmed that VR training for nasal endoscopy can effectively enhance the outcomes of nasal endoscopic surgical training [6–8]. To further expand the role of VR in improving endoscopic skills within standardized residency training, the present randomized controlled trial was conducted to compare the efficacy of VR-assisted training versus MS training. We aimed to explore an optimized training strategy to improve residents’ performance on standardized knowledge assessments, endoscopic operational skills, and training satisfaction.

The primary finding of this study is that otorhinolaryngology residents who received VR-assisted training achieved significantly better outcomes in both standardized knowledge examination acquisition and simulated endoscopic skills than those in the MS group after the three-month training period. Clinically, this indicates that VR-based simulation can effectively accelerate the mastery of core endoscopic-related knowledge and basic operational skills among residents, shortening the preliminary learning curve for endoscopic procedures. For clinical training programs, this suggests that VR integration can reduce the reliance on direct patient-based practice in the early training stage, minimizing operational risks for novice trainees and enhancing patient safety, which aligns with the core goals of standardized resident training. Meanwhile, VR simulation offers distinct advantages in terms of immersive, interactive learning, unrestricted repetition, and automated, real-time performance feedback by VR software. The VR system supports one-click immediate reset, allowing trainees to repeatedly practice without any physical restrictions or delays.

Regarding secondary outcomes, the VR training group also exhibited significantly higher self-reported training satisfaction compared with the MS group, especially in terms of training experience, assessment feedback, and willingness to continue using or recommend the platform. While satisfaction scores related to post-training procedural confidence and basic clinical skill mastery were comparable between the two groups—confirming that both modalities effectively build baseline competency—the superior satisfaction of the VR group reflects its stronger adaptability to modern resident learning habits. This is clinically meaningful as higher training satisfaction correlates with increased learning initiative and long-term skill retention, fostering sustainable learning behavior rather than passive skill memorization. Furthermore, most VR group trainees mastered the system operation within 20 min, demonstrating high user adaptability and making this modality feasible for large-scale standardized residency training promotion.

The study population consisted of full-time otorhinolaryngology residents enrolled in standardized training, all of whom had limited prior endoscopic operational experience and consistent baseline theoretical and skill levels. This cohort is highly representative of the target population for endoscopic skill training, as it reflects the typical learner profile in mainstream otorhinolaryngology residency programs. The homogeneity of the study population reduces confounding factors related to prior experience and learning background, ensuring that the observed differences in outcomes are attributable to the training modalities rather than participant heterogeneity, and supporting the generalizability of the findings to similar residency training cohorts.

These results are consistent with previous studies highlighting that immersive VR simulation outperforms traditional two-dimensional teaching and physical model training in medical procedural education, particularly in enhancing spatial cognition and operational proficiency. Dharmawardana et al. first validated the clinical applicability of VR-based simulations specifically for endoscopic sinus surgery, laying a critical foundation for the use of this technology in otolaryngology training [11]. Similar to prior research on VR-based surgical training, our study confirms that VR’s interactive, repeatable, and risk-free learning environment improves learning efficiency and trainee engagement; Lesch et al. further demonstrated that VR simulation directly enhances procedural confidence among surgical trainees, a finding that aligns with our results on improved self-efficacy in the VR group [12]. Notably, the VR training experience was frequently described by participants as game-like, a perception that is interesting and pedagogically relevant. Drawing on contemporary literature on gamification, learner engagement and motivation in surgical education, this gamified feature of VR is a key educational strength: it boosts intrinsic learning motivation, eases trainee anxiety during procedural practice, and encourages repeated, active training by turning routine skill drills into an engaging learning process. This highlights a unique educational dimension of VR simulation that further differentiates it from conventional teaching methods.

A recent systematic review and meta-analysis by Kim and Kim reinforced these conclusions, showing that VR medical education programs consistently yield positive effects on learning outcomes and learner satisfaction across diverse clinical disciplines [13]. Notably, this study expands existing evidence by focusing specifically on otorhinolaryngology endoscopic training, integrating specialty-specific anatomical structures and procedural pathways—such as sinus drainage pathways and transoral endoscopic approaches—that are rarely covered in general surgical VR simulations. This specialization distinguishes our findings from broader surgical simulation research and strengthens the external validity for otorhinolaryngology residency training. Recent research trends, as summarized by Choi et al., highlight a growing focus on VR surgical simulation in medical education, with increasing emphasis on specialty-specific applications [14]. Hudise et al.’s systematic review further underscored the pivotal role of VR in advancing otolaryngology surgical training, emphasizing its potential to bridge the gap between didactic learning and clinical practice [15]. which directly supports the educational value demonstrated in our study. Our study design employed a randomized controlled approach, with all theoretical and practical training delivered by the same associate chief physician to eliminate instructor-related bias, and objective assessment indicators (theoretical scores, simulated skill scores) combined with subjective satisfaction surveys to ensure result reliability. Nevertheless, the methodological design has notable strengths and areas for reflection. The use of a randomized allocation via random number table ensures group homogeneity at baseline, while the three-month training duration captures short-term learning efficacy effectively.

A notable methodological consideration of this study is that all skill evaluations were conducted in a simulated environment rather than in real clinical practice. Although this design effectively controlled clinical confounding factors, it limited the direct verification of whether training gains could be transferred to authentic clinical scenarios. Furthermore, the VR system adopted in this study provided automated real-time feedback to enhance training standardization, while the conventional model simulation group received no standardized electronic auxiliary feedback. To avoid interventional bias and minimize confounding variables, instructors delivered unified on-site supervision and standardized teaching guidance across both groups, with consistent arrangements in teaching rhythm, training duration and on-site management; no additional instructional reinforcement or preferential intervention was offered to either cohort. Meanwhile, despite its standardized advantages, VR automated feedback cannot fully replace the individualized guidance from senior clinicians when handling complex operative scenarios.

This study has several inherent limitations that need to be acknowledged. First, the sample size was relatively small, and the three-month training and follow-up period only reflected short-term learning outcomes, without relevant data on long-term skill retention and proficiency decay. Considering that procedural operational proficiency naturally declines without continuous, targeted clinical practice, future longitudinal studies with extended follow-up periods and standardized periodic retraining modules are highly warranted to verify the long-term durability and sustainability of the training effects. Second, the current VR training modules are limited in application scope, mainly focusing on routine basic endoscopic operations while lacking high-level and complex clinical scenarios. Therefore, further expansion of module categories and enrichment of scenario-based training content are necessary to cover a wider range of authentic clinical conditions. Third, this study did not incorporate real-world clinical outcome indicators, such as the Objective Structured Assessment of Technical Skills (OSATS) and Direct Observation of Procedural Skills (DOPS). Accordingly, whether the improved simulated operational skills can be effectively transferred to independent and competent clinical performance remains unconfirmed. Fourth, although the MS group provided more realistic tactile feedback, which is theoretically conducive to better skill transfer to real surgical scenarios, this study did not conduct a quantitative comparative evaluation of clinical skill translation efficiency between different training methods, which represents a key research direction for subsequent relevant studies. In addition, the long-term cost-effectiveness of VR-assisted training remains unclear, with potential risks of decreased economic benefits in the later application stage. Further rigorous cost-effectiveness analyses are required to clarify the long-term input–output ratio and comprehensive clinical application value of this immersive training technology. Notably, all outcome evaluation data in this study were exclusively derived from standardized simulated scenario assessments rather than real frontline clinical operational performance, with a clear hierarchical distinction strictly made between simulated operational ability and authentic comprehensive clinical practice competence.

Despite these limitations, this study confirms that VR-assisted training is a highly effective supplementary modality for otorhinolaryngology residency endoscopic training, with clear advantages in standardized knowledge examination mastery, operational skill acquisition, and trainee satisfaction. The gamified, immersive learning experience of VR enhances learning initiative and breaks through the constraints of traditional training, making it suitable for integration into standardized large-scale residency curricula. Future research should focus on long-term efficacy verification, module optimization, and clinical outcome assessment to establish a comprehensive VR-assisted training system that bridges simulation practice and real clinical surgery, ultimately improving the quality of otorhinolaryngology resident training.

Conclusions

Standardized residency training is a crucial transitional phase in the development of clinical physicians, and recent reforms in China’s medical education system have emphasized the need for innovative, competency-based training methods to meet the evolving demands of clinical practice [16]. Despite the aforementioned limitations, this study confirms that VR-assisted training is a highly effective supplementary modality for otorhinolaryngology residency endoscopic training, with clear advantages in standardized knowledge examination mastery, operational skill acquisition, and trainee satisfaction. The gamified, immersive learning experience of VR enhances trainees’ interest and deepens their understanding and recognition of endoscopic procedures, thereby increasing their satisfaction with endoscopic residency training and enthusiasm for incorporating VR into future training. This study offers a structured model for the integration of VR technology into standardized, large-scale residency training programs. By embedding the VR system into the existing curriculum, this study highlights the educational value of immersive simulation as a bridge between laboratory training and clinical application, rather than presenting a mere technological advancement.

Authors’ contributions

Z.Y.Y. helped conceptualize the study, designed the study schedule, helped in data collection, analyzed and interpreted the data and drafted the manuscript. Y.J.Y. conceptualized the study, helped in designing the study schedule, helped in interpreting the data and substantially revised the manuscript. Q.H. conceptualized the study, helped in designing the study schedule, helped in interpreting the data and substantially revised the manuscript.

Funding

None.

Data availability

The datasets analyzed during the current study are available from the corresponding author upon reasonable request. Data of the used questionnaire is provided supplementary information files.

Declarations

Ethics approval and consent to participate

The present study was approved by the ethics committee of the University of Tsinghua (Application No. 202201J056), and all included participants provided informed written consent before inclusion. Research was carried out in compliance with the Helsinki Declaration.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.