• MacKay Medical College M2/M3 students

• Senior students, interns & residents

• Attending physicians (OR planning/CE)

Traditional anatomy education faces three core challenges: scarcity of cadaveric resources, limited spatial comprehension from two‑dimensional materials, and insufficient diversity of clinical cases. To address these bottlenecks, this project leverages the NVIDIA MONAI framework to perform fully automated organ segmentation and high‑fidelity three‑dimensional reconstruction on de‑identified in‑house computed tomography scans, rapidly generating reusable digital teaching assets. The platform supports cross‑device interactive learning in web, virtual reality, and mixed reality environments, following a progressive pathway that integrates normal anatomy, anatomical variations, and clinical pathologies. Since its deployment in March 2025 for second‑ and third‑year medical students at MacKay Medical College, the project has demonstrated notable improvements in academic performance and learner satisfaction.

• Clinically driven three‑dimensional models

• Fully automated artificial intelligence segmentation

• Cross‑platform interaction for web, virtual reality, and mixed reality

• Deployed in March 2025 for second‑ and third‑year courses

• Session duration exceeding thirty minutes in ninety‑two point nine percent of users

• Service uptime of ninety‑nine point nine percent

• Average score increased by three point three (seventy‑five point four to seventy‑eight point seven)

• Pass rate increased from eighty point five percent to ninety‑eight point zero percent

• Low‑score group reduced from nineteen point five percent to two point zero percent

• Strict de‑identification of imaging data

• Encrypted transmission via Hypertext Transfer Protocol Secure

• Role‑based access control and audit logging

• Overall satisfaction of four point eight out of five

• Eighty‑five percent perceive time saving

• Ninety‑two point nine percent are willing to recommend

• Active engagement with the NVIDIA MONAI medical open‑source community

• Planned submissions to international journals on pedagogy and outcomes

• Planned cloud platform for case exchange and resource sharing

• Enhance medical education quality: Integrates real clinical imaging with AI technology to bridge the gap between basic science and clinical practice, strengthening spatial understanding and applied competence.

• Improve learning outcomes: Average score increased by 3.3 points; pass rate rose from 80.5% to 98.0%; low‑score group significantly reduced; overall satisfaction reached 4.8 out of 5.

• Promote healthcare sustainability and digital transformation: Reduces reliance on cadaveric specimens, minimizes printed materials and physical resources, aligning with ESG sustainability goals.