Development of digitization workflow for 3D heart anatomy models applied in medical education
Main Article Content
Abstract
Objective: To develop a digitization workflow for 3D heart anatomy models using mobile scanning technology and to assess the geometric accuracy of the digital product compared to the physical model.
Subjects and Methods: The study subject was a plastic heart anatomy model from 3B Scientific at the Vietnam Military Medical University. A descriptive experimental method was employed: 3D model creation using KIRI Engine app (LiDAR/Photogrammetry), post-processing in Blender 4.1, and dimensional measurement of 25 specified anatomical indices - the physical model measured directly with a digital vernier caliper (gold standard) and the digital model measured in Blender after scale normalization. Accuracy was validated using Cosine Similarity index and R² correlation coefficient.
Results: A successful digitization workflow comprising three stages (scanning, technical processing, and publishing) was established. Quantitative results indicated a very high similarity between the 3D and physical models, with a mean Cosine Similarity of 0.999 and R² = 0.997. The average dimensional deviation of structures was less than 5%.
Conclusion: The digitization of heart models using low-cost tools (smartphones, open-source software) is feasible and achieves high geometric accuracy - a prerequisite for medical education application, providing the basis for further studies evaluating actual pedagogical effectiveness in students.
Keywords
3D Model, Heart Anatomy, KIRI Engine, Blender, Digitization, Cosine Similarity