Clinical teaching remains both a cornerstone and persistent challenge in medical education. High-quality, efficient clinical learning is decisive for accumulating clinical experience, enhancing clinical skills, and cultivating clinical thinking (1, 2). As a critical subspecialty of orthopedics, arthroplasty exhibits distinctive teaching characteristics due to its broad disease spectrum, technically complex procedures, and high surgical difficulty (3). However, traditional teaching methods fail to meet contemporary educational demands. On one hand, orthopedic clinical instruction predominantly relies on skeletal models and patient imaging as teaching aids (4). Yet skeletal models largely demonstrate only normal anatomy (5), while 3D images constrained by 2D display interfaces cannot authentically reconstruct the three-dimensional spatial conformation of the musculoskeletal system, thereby limiting students' spatial cognition and precise measurement of key clinical parameters (e.g., anteversion angle, abduction angle, teardrop sign, lower limb mechanical axis, flexion-extension gap, retroversion angle, prosthesis sizing and positioning), which ultimately constrains teaching effectiveness (6, 7). On the other hand, scarcity of educational resources, strained doctor-patient relationships, and limited hands-on opportunities create systemic challenges for medical students and junior surgeons in developing clinical reasoning and operative skills (8, 9). Consequently, actively exploring more effective instructional strategies has become imperative.

Teaching methods and tools are fundamental pathways to achieving educational objectives. With the rapid advancement of computer science and related disciplines, 3D printing and digital virtualization technologies have become widespread across various industries and are increasingly applied in the medical field (10). This offers significant opportunities to reform traditional medical teaching models and innovate clinical teaching practices. Specifically, 3D printing technology enables the creation of highly realistic physical models that replicate the features of actual cases (11). It concretizes abstract and complex anatomical structures, greatly enriches teaching resources, and allows trainees to engage with numerous case models before entering clinical practice (12). Simultaneously, Mixed Reality (MR) technology, as an advanced three-dimensional visualization tool, integrates both “augmented reality” and “augmented virtuality” functionalities (7, 13). It can stereoscopically display intricate multi-layered structures, seamlessly merge the real physical world with virtual information, and support multi-party real-time interaction (14, 15). Within the teaching context, MR technology converts patient 2D imaging data into intuitive 3D images and accurately superimposes/projects them onto real anatomical areas or operational spaces, achieving a “transparent visualization” effect (virtual anatomical structures are semi-transparently overlaid on physical models via MR to enable “perspective” observation of deep structures) that significantly deepens trainees' understanding of anatomical structures and surgical procedures while improving learning efficiency (15, 16). Critically, these two technologies demonstrate significant synergistic potential: 3D printed models provide a stable anatomical reference base and tactile feedback, while MR technology dynamically overlays simulated pathological variables, creating an immersive interactive experience. Together, they construct a comprehensive educational ecosystem that fuses physical reality with virtual information.

To fully integrate and leverage this synergistic “virtual-physical integration” technological advantage, this study innovatively proposes a “MR Immersive Classroom with 3D Printed Tactile Enhancement” teaching model. This model aims to systematically evaluate its practical efficacy in enhancing the effectiveness of orthopedic clinical teaching.

Orthopedics, as a highly specialized clinical discipline with a complex knowledge system and constrained instructional hours, features content intrinsically linked to anatomy and imaging sciences (17). It demands rigorous comprehension of biomechanical principles and spatial cognition in three dimensions. Traditional didactic teaching proves inadequate for deep mastery of core knowledge—particularly in knee/hip arthroplasty education, where theoretical instruction must be synergistically reinforced with hands-on procedural training (18). Currently, most medical schools adhere to conventional pedagogy: didactic lectures using radiographic data (x-ray/CT) coupled with slide presentations, supplemented by limited small-group practical sessions. This model exhibits significant limitations: monotonous formats, inadequate spatial representation, low student engagement, and passive knowledge assimilation, collectively compromising educational effectiveness (19). To address these challenges, our study innovatively integrates 3D printing and MR technologies to establish a novel clinical teaching paradigm for joint surgery. By integrating virtual and physical elements, this approach overcomes persistent practical constraints and demonstrably enhances instructional efficacy.

Recent years have witnessed the deep integration of 3D printing, virtual reality (VR), and augmented reality (AR) technologies into medical education, driving systemic pedagogical transformation (13, 20). Within orthopedic training, while these technologies demonstrate significant potential, each exhibits inherent structural limitations when deployed independently. VR excels in providing exceptional 360° anatomical visualization and surgical planning within immersive environments, yet fails to support tactile-dependent procedures like implant placement due to the absence of haptic feedback (21). Conversely, AR enables interactive guidance through real-world overlays but lacks realistic force feedback mechanisms critical for simulating orthopedic interventions (7). Meanwhile, 3D printing addresses tactile needs via patient-specific 1:1 models that facilitate hands-on practice of plate/screw fixation, though its static nature restricts dynamic visual guidance for spatial navigation (22). Recognition of these complementary strengths has catalyzed integrated solutions, as demonstrated by Jade et al.'s AR-coupled 3D-printed ankle/foot model which requires further validation of anatomical accuracy and educational efficacy (23). This is further evidenced by the AEducaAR project confirming enhanced learning motivation, long-term knowledge retention, and 3D spatial comprehension through physical-AR fusion (24), alongside documented cases where patient-specific VR with 3D models significantly strengthened preoperative confidence (25). Collectively, hybrid approaches synergizing digital immersion with physical interactivity demonstrate superior pedagogical effectiveness over single-modality methods, establishing foundational groundwork for exploring deeply integrated paradigms like MR.

As an emerging holographic imaging modality, MR combines the strengths of AR and VR to enable real-time coexistence and interaction between physical and digital objects, demonstrating significant potential in medical instruction (7, 14, 15). MR-based holographic visualization renders complex 3D anatomical structures with exceptional fidelity, allowing less-experienced clinicians to intuitively comprehend disease-specific morphological characteristics and pathological evolution. Concurrently, its capacity for real-time interactivity and seamless virtual-physical integration establishes an innovative platform for high-fidelity procedural simulation (26). Our prior research has validated MR's substantive advantages in clinical applications including physician-patient communication, preoperative planning, intraoperative navigation, and remote surgical consultations (7). However, current MR technology remains constrained by reliance on purely visual simulation without tactile feedback—a critical limitation that particularly impedes skill acquisition for procedures requiring haptic perception. The advent of 3D printing addresses this sensory gap by enabling localized reality enhancement within virtual environments: patient-specific CT imaging data is utilized to fabricate photorealistic 1:1 physical models, thereby establishing a foundation for tactile interaction and simulated operative training (10).

This study integrates 3D-printed anatomical models with MR technology to advance clinical education in joint surgery. The 3D models establish a static anatomical reference framework, while the MR system introduces dynamic pathological variables, synergistically resolving the dual challenges of spatial cognition deficits and scarce practical resources inherent in traditional teaching methodologies. This integrated approach bidirectionally enhances students' visual-tactile perception. Results demonstrate that students trained with this 3D-MR paradigm significantly outperformed conventionally trained peers in both theoretical knowledge assessments (P < 0.05) and comprehensive practical evaluations, validating its efficacy as an instructional tool. This finding aligns with Li et al.'s research demonstrating that MR-integrated 3D models enhance trainees' mastery of complex surgical pathologies while significantly boosting learning engagement and self-directed initiative (27). Furthermore, this innovative virtual-physical integration surpasses traditional methods across pedagogical metrics including learning efficiency, instructional interactivity, knowledge mastery, and clinical competency development. Its operational practicality substantially stimulates learning motivation while strengthening concentration and comprehension.

Key advantages manifest through four interconnected dimensions: First, the fusion of MR holographic visualization and tangible 3D models transcends spatiotemporal limitations, enabling persistent 3D visualization of orthopedic pathologies. This immersive environment facilitates multidimensional understanding of disease characteristics, igniting curiosity that drives knowledge internalization (28). Second, for complex surgical concepts like fracture classification systems and associated procedures (e.g., arthroplasty or internal fixation), the technology provides comparative virtual-physical integration modeling and imaging analysis, enabling intuitive differentiation of pathological variations while accelerating knowledge acquisition. Third, MR's visual immersion coupled with 3D models' tactile feedback generates spatio-haptic convergence that elevates spatial reasoning and anatomical comprehension—particularly for critical parameters including femoral valgus angle, anteversion angle, teardrop sign, and prosthetic positioning. Fourth, direct manipulation of patient-specific 3D models enables synchronous knowledge application, where progressive analysis across 2D imaging, 3D visualization, and physical specimens cultivates diagnostic reasoning. This closed-loop training enhances clinical problem-solving capabilities through systematic integration of patient data, ultimately improving diagnostic accuracy and therapeutic planning. Notably, the strong positive correlation between OSCE scores and questionnaire results (r = 0.82, P < 0.001) confirms both instrument validity and enhanced learning experiences. The highest correlation between learning efficiency and OSCE performance suggests knowledge assimilation mediates efficacy gains.

Furthermore, this study has several limitations that warrant acknowledgment. First, the participant pool was confined to orthopedic trainees from a single institution with a relatively small sample size, potentially limiting the generalizability of findings. Single-center study (n = 36) has limited generalizability and multi-center validation (n > 200) is required in the future. Second, the non-blinded design coupled with post-randomization informed consent procedures may have introduced unavoidable selection bias. Third, implementing the integrated 3D printing and MR approach requires significantly greater preparation time and effort from instructors compared to conventional teaching methods. Finally, the primary reliance on examination scores as evaluation metrics lacks multidimensional assessment approaches, potentially failing to comprehensively reflect students' actual clinical competencies.

In conclusion, the virtual-physical integrated pedagogical paradigm—combining tangible 3D models with persistent MR holographic visualization—demonstrates distinct advantages in medical education. It effectively stimulates learner engagement and motivation, enhances instructional interactivity, and facilitates holistic understanding of representative orthopedic pathologies. This approach fosters a deeper comprehension of musculoskeletal disorders and merits further exploration and broader implementation.