Abstract:A fast, robust and accurate method for femur segmentation in digital AP pelvic radiographs was developed by combining SSM and SAM with dynamic programming. This method can be extended to segmentation of other bony structures such as the pelvis.
“…We divided our framework for the complete‐pelvis segmentation into two separate modules as shown in Figure : left hemi‐pelvis segmentation and right hemi‐pelvis segmentation. By methods presented in , contours of the left hemi‐pelvis were segmented in the first module and then taken as the input to the second module to initialize the right hemi‐pelvis segmentation. All details including the clinical study will be described in the following sections of this article.…”
Section: Methodsmentioning
confidence: 99%
“…The SAM of the left hemi‐pelvis was constructed as follows . For a landmark (depicted as the red dot in Figure (a)) in a training image, n t points ( note : n t was empirically chosen as twenty) were sampled in its normal direction and also in the inverse direction of to obtain total (2 n t + 1) sampled points.…”
Section: Methodsmentioning
confidence: 99%
“…By averaging intensities of the ten points sampled on both sides of each (2 n t + 1) point, intensities at these (2 n t + 1) points of each landmark were represented by a (2 n t + 1) × 1 vector derived from the(2 n t + 1) × 10 array . Given a landmark i in a training radiograph j , its training appearance model g ij ' was then calculated as the derivative of intensity values in its (2 n t + 1) × 1 vector . Thus g ij ' had dimensions of (2 n t + 1) × 1, and each element of g ij ' needed to be normalized with …”
Section: Methodsmentioning
confidence: 99%
“…Attempts to develop an automatic solution for the proximal femur segmentation have been presented previously . Despite numerous methods developed and successful results achieved by previous work for delineating the proximal femur, a fast and robust segmentation of the complete pelvis in AP radiographs with sufficient accuracy has not yet been fully explored due to more complicated anatomical structures in the pelvic region than those in the femoral area.…”
Section: Introductionmentioning
confidence: 99%
“…In this article, we extended methods presented in to the complete‐pelvis delineation for image‐free navigation of THA, particularly for the postoperative determination of acetabular cup orientations in image‐free THA. We built our complete‐pelvis segmentation framework by using not only the statistical model‐based approach , but also anatomy‐specific information processing techniques as well as an unconstrained optimization method to cope with the complex pelvic structures in AP radiographs.…”
“…We divided our framework for the complete‐pelvis segmentation into two separate modules as shown in Figure : left hemi‐pelvis segmentation and right hemi‐pelvis segmentation. By methods presented in , contours of the left hemi‐pelvis were segmented in the first module and then taken as the input to the second module to initialize the right hemi‐pelvis segmentation. All details including the clinical study will be described in the following sections of this article.…”
Section: Methodsmentioning
confidence: 99%
“…The SAM of the left hemi‐pelvis was constructed as follows . For a landmark (depicted as the red dot in Figure (a)) in a training image, n t points ( note : n t was empirically chosen as twenty) were sampled in its normal direction and also in the inverse direction of to obtain total (2 n t + 1) sampled points.…”
Section: Methodsmentioning
confidence: 99%
“…By averaging intensities of the ten points sampled on both sides of each (2 n t + 1) point, intensities at these (2 n t + 1) points of each landmark were represented by a (2 n t + 1) × 1 vector derived from the(2 n t + 1) × 10 array . Given a landmark i in a training radiograph j , its training appearance model g ij ' was then calculated as the derivative of intensity values in its (2 n t + 1) × 1 vector . Thus g ij ' had dimensions of (2 n t + 1) × 1, and each element of g ij ' needed to be normalized with …”
Section: Methodsmentioning
confidence: 99%
“…Attempts to develop an automatic solution for the proximal femur segmentation have been presented previously . Despite numerous methods developed and successful results achieved by previous work for delineating the proximal femur, a fast and robust segmentation of the complete pelvis in AP radiographs with sufficient accuracy has not yet been fully explored due to more complicated anatomical structures in the pelvic region than those in the femoral area.…”
Section: Introductionmentioning
confidence: 99%
“…In this article, we extended methods presented in to the complete‐pelvis delineation for image‐free navigation of THA, particularly for the postoperative determination of acetabular cup orientations in image‐free THA. We built our complete‐pelvis segmentation framework by using not only the statistical model‐based approach , but also anatomy‐specific information processing techniques as well as an unconstrained optimization method to cope with the complex pelvic structures in AP radiographs.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.