Overview
These renderings come from RENCI Senior Visualization Researcher David Borland’s Ph.D. thesis in the computer science department at UNC Chapel Hill with advisor Russell M. Taylor II and collaborator John P. Clarke of UNC’s radiology department in the School of Medicine.
These renderings use a novel technique called Flexible Occlusion Rendering (FOR), an extension to standard volume rendering techniques, that can render 3D views of MRI and CT data and remove occluding material to show views that would otherwise not be possible. Image 1 shows a standard volume rendering of the shoulder joint (left). Using FOR, (right) the user can position the virtual camera within the humeral head (upper area) of the shoulder and see the entire shoulder socket, removing material of the humeral head that obstructing the view of the shoulder socket. This technique can also be used to look for tumors in the urinary tract and inspect knees for fractures, among other uses.
Viewing 3D reconstructions of MRI and CT is often a more natural way to view human anatomy than looking at slices through the data. However, the complexity of the human body makes navigating the 3D data difficult, and sometimes the best view for pathology is obstructed by other tissue and/or bone. FOR enables physicians to automatically remove material occluding the view of interest, enabling improved views for diagnoses.
Overview
These renderings come from RENCI Senior Visualization Researcher David Borland’s Ph.D. thesis in the computer science department at UNC Chapel Hill with advisor Russell M. Taylor II and collaborator John P. Clarke of UNC’s radiology department in the School of Medicine.
These renderings use a novel technique called Flexible Occlusion Rendering (FOR), an extension to standard volume rendering techniques, that can render 3D views of MRI and CT data and remove occluding material to show views that would otherwise not be possible. Image 1 shows a standard volume rendering of the shoulder joint (left). Using FOR, (right) the user can position the virtual camera within the humeral head (upper area) of the shoulder and see the entire shoulder socket, removing material of the humeral head that obstructing the view of the shoulder socket. This technique can also be used to look for tumors in the urinary tract and inspect knees for fractures, among other uses.
Viewing 3D reconstructions of MRI and CT is often a more natural way to view human anatomy than looking at slices through the data. However, the complexity of the human body makes navigating the 3D data difficult, and sometimes the best view for pathology is obstructed by other tissue and/or bone. FOR enables physicians to automatically remove material occluding the view of interest, enabling improved views for diagnoses.
Project Team
David Borland
Partners
Dr. John P. Clarke, UNC School of Medicine
Dr Julia Fielding, UNC School of Medicine
