The work of Penn State's Multidimensional Image Processing Laboratory spans many areas of imaging science: digital image processing, computer vision, pattern recognition, scientific visualization, and virtual reality. We are particularly driven by problems in high-resolution 3D medical image processing and multimodal video analysis, with a focus on image-guided endoscopy systems.
The laboratory has made numerous theoretical contributions to the general field of imaging science, including:
	1. Virtual endoscopy, multimodal image registration, and data fusion
	2. Volume visualization, dynamic navigation, and surgical guidance
	3. Procedure planning, 3D axes definition, and hierarchical tree analysis
	4. Texture analysis and mathematical morphology
	5. Pattern recognition and machine learning for multimodal image/video analysis
	6. Image segmentation, interpolation, and enhancement
These contributions have had a substantial impact on clinical practice. Of special note, they have been integrated into numerous FDA-approved systems used in hospitals worldwide for image-guided bronchoscopy and procedure planning.
Our current research focuses on devising large-scale multimodal bronchoscopy systems for lung-cancer assessment. This work is applying concepts from image processing, machine learning, and virtual reality to 3D CT/PET images, endobronchial ultrasound, and multimodal bronchoscopic video. Over 80 students have worked in the laboratory over the years.