NeuroVase: A Tangible Mobile Augmented Reality Learning System for Neurovascular Anatomy and Stroke Education

Abstract: Stroke remains a leading cause of mortality and disability worldwide, requiring rapid and informed clinical decision-making. A solid spatial understanding of cerebrovascular anatomy and vascular territories in relation to stroke symptoms and severity is critical for timely clinical decision and patient care. However, this knowledge is typically conveyed through static 2D diagrams and printed materials, which can hinder mastery of the complex neurovascular system and their clinical implications. Mobile augmented reality (AR) offers an accessible medium for delivering intuitive 3D anatomical education, yet applications focused on the neurovascular system and stroke remain limited despite the demand. To address this, we propose NeuroVase, a tablet-based mobile AR platform within a structured pedagogical framework that enhances stroke-related neuroanatomy learning by providing an interactive, engaging, and accessible alternative to traditional methods. NeuroVase features a dual-mode setup, using tangible cue cards as standalone study aids while also serving as interactive markers for AR content delivery. A custom learning curriculum focused on cerebrovascular anatomy and stroke supports exploration of vascular territories, stroke syndromes, and arterial occlusions, in the context of annotated 3D anatomical models in NeuroVase. A controlled user study with 40 participants revealed that NeuroVase is an effective and user-friendly AR platform to facilitate complex anatomical and physiological education, compared with traditional learning.

• The paper introduces NeuroVase, a tangible mobile AR system that fuses interactive 3D neuroanatomical visualizations with physical cue cards for enhanced neurovascular and stroke education.
• The paper demonstrates a dual-mode curriculum leveraging AR gestures and cue cards, achieving high usability (SUS=90.0±5.7) and superior user engagement compared to traditional methods.
• The controlled study reveals significant learning improvements and user satisfaction while highlighting technical challenges such as cue card tracking under variable conditions.

NeuroVase: Tablet-Based Tangible AR for Neurovascular Anatomy and Stroke Education

Introduction

NeuroVase introduces a tangible mobile AR system designed to address deficiencies in current neurovascular and stroke education. The platform delivers interactive three-dimensional neuroanatomical and cerebrovascular visualizations, leveraging physical cue cards for both augmented interaction and offline study. The design aims to overcome the spatial and clinical complexities that impede traditional 2D neuroanatomy curricula, providing context-rich and scalable pedagogical interventions on consumer-grade tablets.

System Architecture and User Interaction

The NeuroVase platform utilizes a dual-mode curriculum integrating AR visualizations and tangible cue cards. The cue cards, using carefully selected high-contrast anatomical images with QR codes, serve as both physical AR targets and standalone learning aids. When the master card is positioned in the camera field, the core 3D neuroanatomical model is rendered and anchored within the AR interface. Trigger cards update content contextually, allowing modular progression through lobar anatomy, arterial vasculature, and arterial territories.

Figure 1: The AR user interface leverages tangible cue cards for content-triggering and spatial interaction, seamlessly integrating physical and virtual content delivery.

The interface supports canonical AR gestures (pinch/zoom/rotate/translate), with a rigid assignment of cue cards to content modules to maintain linear pedagogical flow. The design supports structured, progressive learning aligned with clinical reasoning relevant to stroke management.

Figure 2: The main menu allows users to select between core modules and engage with high-fidelity MRI overlays in the Vascular Anatomy section.

Figure 3: Example module states showing pre- and post-trigger activation for lobar, arterial, and territory-specific content.

Pedagogical Content and Model Construction

The content framework was constructed with domain expert curation, integrating three neuroanatomy modules of increasing complexity:

1. General lobar anatomy: Extracted and segmented from the BCI-DNI atlas for high anatomic fidelity.
2. Cerebral arterial system: Modeled using multi-modal MRI/MRA data and refined using Frangi-based vessel enhancement and ITK-SNAP segmentation.
3. Vascular territories: Registered from a large-scale clinical stroke atlas (1,298-patient cohort) for territory boundary definition and pathophysiological correlation.

MRI overlays and vessel renderings are fused through Unity’s rendering pipeline, importing mesh data in .obj format for volumetric and surface renderings.

Controlled Study Design and Results

A parallel-arm randomized study ($n=40$; 20 AR, 20 control) with STEM participants of low cerebrovascular expertise assessed usability, enjoyment, and educational effectiveness. The AR cohort used the complete NeuroVase platform, while controls used matched paper materials with identical text and 2D images.

Key findings:

• System Usability: The AR platform achieved a SUS of $90.0 \pm 5.7$, statistically well above the “good usability” threshold of 68 ($p < 0.001$).
• Learning Outcomes: Both groups showed significant pre–post improvement ($p < 0.0001$). The AR group improved 29.45% (from 40.83% to 70.28%), and the control group 26.67% (from 33.89% to 60.56%), with no significant between-group difference in gain.
• User Engagement and Satisfaction: The AR group outperformed controls on UX dimensions (engagement: $4.4 \pm 0.7$ vs $3.3 \pm 1.0$; enjoyment: $4.5 \pm 0.6$ vs $3.4 \pm 1.2$; both $p < 0.01$).

  Figure 4: Customized UX questionnaire responses demonstrate significant preference for AR-based engagement and satisfaction relative to control.

• Qualitative Feedback: 3D visualizations were consistently cited as promoting spatial understanding and recall. Physical–virtual cue card integration was well-rated for interaction and review utility, although some tracking glitches were reported (6/19 AR users).

Implications and Theoretical Considerations

NeuroVase’s design addresses three core deficits in neuroanatomy technologies:

1. Direct targeting of cerebrovascular territories and syndromic correlates, domains largely ignored by existing AR/VR systems.
2. Integration of tangible interaction workflows (cue cards), bridging embodied cognition theory and digital visualization.
3. An extensible, modular pedagogical scaffold, with open-source release enabling community adaptation.

The measured superiority in engagement and system usability, despite parity in raw knowledge gains, suggests differentiated impact on motivation and depth of spatial learning. The lack of significant delta on post-intervention testing aligns with prior AR neuroanatomy studies, but the high user-perceived value and lower cognitive friction observed reinforce AR as an effective supplement to didactic approaches.

Technically, NeuroVase demonstrates the practical feasibility of hybrid tangible–AR interaction paradigms deployable on consumer hardware, which is a critical criterion for broad clinical and educational adoption. The pipeline for converting MRI/MRA clinical data to AR-renderable assets establishes groundwork for future AI-driven automation in medical education content generation, and the structured cue card system provides a template for gamified or adaptive curriculum extensions.

Limitations and Future Directions

Key limitations include sensitivity to physical cue card tracking under variable lighting, modest sample diversity (mostly STEM students), and possible content complexity overload for novice users. Planned directions involve:

• Augmenting model fidelity and interaction fluidity (e.g., stabilized gestures, annotation overlays, blood flow animation).
• Formal integration into clinical and undergraduate medical curricula, followed by scaled multi-institutional evaluation.
• Investigation into AI-augmentation for personalized feedback, automatic content creation from clinical neuroimaging repositories, and ML-driven assessment of spatial learning outcomes.

Conclusion

NeuroVase advances mobile AR-based neurovascular education by merging tangible cue card interfaces, high-fidelity anatomical modeling, and a robust, clinician-informed didactic sequence. The system’s high usability and user engagement, combined with its open-source approach and modular extensibility, support its utility as a scalable supplement to traditional neuroanatomy and stroke teaching. Theoretical impact includes validation of embodied learning in AR-enhanced curricula, with practical implications for rapid knowledge transfer in time-sensitive clinical training.

(2604.00296)