Towards Estimating the Uncertainty Associated with 3D Geometry Reconstructed from Medical Image Data
Marc Horner, Karim O. Genc, Stephen M. Luke, Todd M. Pietila, Ross T. Cotton, Benjamin Ache, Kevin C. Townsend, Zachary H. Levine
Patient-specific computational modeling is increasingly being used to assist with the visualization, planning and execution of medical treatments. This trend is placing more reliance on medical imaging to provide an accurate representation of anatomical structures. Digital image analysis is used to segment (extract) patient anatomical data before it is used in clinical assessment/planning. However, the presence of image artifacts, whether due to the physical object or the scanning process, can degrade image accuracy. The process of extracting anatomical structures from medical images introduces additional sources of variability, e.g., when thresholding or eroding a biological interface. An estimate of the uncertainty associated with extracting anatomical data would therefore assist the segmentation process.
Journal of Verification, Validation, and Uncertainty Quantification