Abstract: Generating a crown model using a statistical method (such as k-means clustering, principal component analysis (PCA), or similar statistical methods) can result in a crown model that is missing sharpness details below the threshold of the statistical technique. A method is provided to add back the sharpness to the resulting model by combining a single full-featured example into the algorithms generating the statistical model. The end result is a crown model that is relatively simple to produce and manipulate in real time, yet maintains the anatomical sharpness of a natural tooth.

Abstract: A flexible arch model (FAM) is computed to capture and parameterize the variations of the multiple real dental arches in a training set to reconstruct missing teeth in a patient's dental anatomy. Building the FAM includes acquiring multiple sets of digitized dental arches with a pair of maxillary (upper) and mandibular (lower) jaws in the right relative position and gathering a pre-defined set of landmark points on the occlusal surface of each arch all in the same order and same corresponding positions across multiple samples. The gathered vectors of landmark points are used to perform statistical modeling (e.g. Principal Component Analysis) to create a linear subspace of the feature points with the basis of principal components (when PCA is used) found during the procedure. An arbitrary set of landmark points on a pair of upper and lower arches can be reconstructed by a linear combination of the principal components within a reasonable range of variations captured from the training samples.

Abstract: A flexible arch model (FAM) is computed to capture and parameterize the variations of the multiple real dental arches in a training set to reconstruct missing teeth in a patient's dental anatomy. Building the FAM includes acquiring multiple sets of digitized dental arches with a pair of maxillary (upper) and mandibular (lower) jaws in the right relative position and gathering a pre-defined set of landmark points on the occlusal surface of each arch all in the same order and same corresponding positions across multiple samples. The gathered vectors of landmark points are used to perform statistical modeling (e.g. Principal Component Analysis) to create a linear subspace of the feature points with the basis of principal components (when PCA is used) found during the procedure. An arbitrary set of landmark points on a pair of upper and lower arch can be reconstructed by a linear combination of the principal components within a reasonable range of variations captured from the training samples.

Abstract: Generating a crown model using a statistical method (such as k-means clustering, principal component analysis (PCA), or similar statistical methods) can result in a crown model that is missing sharpness details below the threshold of the statistical technique. A method is provided to add back the sharpness to the resulting model by combining a single full-featured example into the algorithms generating the statistical model. The end result is a crown model that is relatively simple to produce and manipulate in real time, yet maintains the anatomical sharpness of a natural tooth.