Summary:
UCLA researchers in the Department of Bioengineering have developed a neural network-based machine learning system that can use the accurate segmentation of CT scans of bone to generate patient specific VR models for preoperative planning.
Background:
Virtual reality (VR) is a computer-generated simulation in which a person can interact within an artificial three-dimensional environment using electronic devices. Due to its interactive characteristics, VR offers tremendous potential for surgical education. However, image processing hurdles have prevented its practical adoption as a clinically relevant tool since current VR simulators use pre-crafted models that do not allow users to view patient-specific anatomy for preoperative planning. Therefore, there is a need for a method that can adapt case-to-case variabilities within patients that cannot be comprehensively captured by a preconceived set of standardized models.
Innovation:
UCLA researchers in the Department of Bioengineering have developed a method of adapting a neural network-based machine learning method to automate the accurate segmentation of CT scans of bone to generate models for use in VR simulators for patient-specific preoperative planning. The method utilized a machine learning algorithm and high throughput processing to reconstruct into a VR model with clearly identifiable bony elements including fracture fragments. This allows for VR to be accessible for preoperative planning and medical training as models can be easily generated as necessary.
Potential Applications:
- Medical teaching
- Preoperative planning
- Interactive diagnosis
Advantages:
- Patient specific
- Machine learning
- High throughput processing
- Automated segmentation
Status of Development:
First successful demonstration by reconstructed into a VR model with clearly identifiable bony elements and fracture lines