Efficient and Accurate Semi-Automatic Neuron Tracing with Extended Reality

Area of Science:

• Neuroscience
• Computer Vision
• Medical Imaging

Background:

• Neuron tracing reconstructs 3D neuronal morphology from microscopic images, crucial for neuroanatomy and circuit analysis.
• Current methods are labor-intensive, with challenges in user interaction, throughput, and visualization.
• Accurate neuron tracing is vital for understanding brain structure and function at multiple scales.

Purpose of the Study:

• To develop a novel, semi-automatic neuron tracing method using extended reality (XR).
• To enhance user interaction, data throughput, and visualization for 3D neuron reconstruction.
• To validate the method's effectiveness and efficiency through user studies.

Main Methods:

• Implemented a novel XR-based system for semi-automatic neuron tracing.
• Defined intuitive interactors for controllable and efficient interactions within an immersive environment.
• Developed a GPU-accelerated algorithm for real-time automatic tracing and reconstruction.
• Integrated a visualizer for enhanced viewing of volumetric images and 3D objects.

Main Results:

• The XR method demonstrated satisfying results when implemented with virtual reality (VR) and augmented reality (AR) headsets.
• User studies confirmed the effectiveness of the defined interactors.
• The proposed method proved more efficient compared to existing neuron tracing approaches.

Conclusions:

• Extended reality (XR) offers an intuitive and efficient platform for semi-automatic neuron tracing.
• The developed interactors and GPU-accelerated algorithm significantly improve the neuron reconstruction process.
• This novel approach advances neuroimaging analysis by providing a more accessible and effective tool for researchers.