Evaluation of image features and search strategies for segmentation of bone structures in radiographs using Active Shape Models

Behiels, Gert; Maes, Frederik; Vandermeulen, Dirk; Suetens, Paul

doi:10.1016/s1361-8415(01)00051-2

Cited by 69 publications

(38 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The appearance model determines for each landmark its optimal new position, among candidate positions along the line perpendicular to the contour, on both sides. To reduce the effect of outliers, consistent displacement of neighboring landmarks is enforced by computing the global optimal path through the evaluated positions [6]. The shape approximation itself is also an iterative procedure.…”

Section: Segmentation Schemementioning

confidence: 99%

“…A non-linear gray value model is proposed which can deal with the highly variable boundary appearance of AAA and exploits information of different MR images. The shape parameters are more robustly estimated using dynamic programming regularization [6] and a weighted fit. To increase segmentation speed and robustness, a multi-resolution approach is used.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Segmentation of Abdominal Aortic Aneurysms in Multi-spectral MR Images

Bruijne

Ginneken

Bartels

et al. 2003

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. An automated method for segmenting the outer boundary of abdominal aortic aneurysms in MR images is presented. The method is based on the well known Active Shape Models (ASM), which fit a global landmark-based shape model on the basis of local boundary appearance models. The original threedimensional ASM scheme is modified to deal with multi-spectral image information and inconsistent boundary appearance in a principled way, with only a limited amount of training data. In addition, a framework for user interaction is proposed. If required, the obtained segmentation can be corrected in an interactive manner by indicating points on the desired boundary. The methods are evaluated in leave-one-out experiments on 21 datasets. A segmentation scheme combining gray level information from two or three MR sequences produces significantly better results than a single-scan model. Average volume errors with respect to the manual segmentation are 4.0%, in 19 out of 21 datasets. In the cases in which the obtained error is large, results can easily be improved using the interactive scheme.

show abstract

Section: Segmentation Schemementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automated Segmentation of Abdominal Aortic Aneurysms in Multi-spectral MR Images

Bruijne

Ginneken

Bartels

et al. 2003

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Behiels et al [1] have shown the suitability of statistical shape models for proximal femur segmentation. Recent work on automatically segmenting the femur in radiographs using statistical shape models includes [11,12].…”

Section: Introductionmentioning

confidence: 99%

Accurate Fully Automatic Femur Segmentation in Pelvic Radiographs Using Regression Voting

Lindner

Thiagarajah

Wilkinson

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Extraction of bone contours from radiographs plays an important role in disease diagnosis, pre-operative planning, and treatment analysis. We present a fully automatic method to accurately segment the proximal femur in anteroposterior pelvic radiographs. A number of candidate positions are produced by a global search with a detector. Each is then refined using a statistical shape model together with local detectors for each model point. Both global and local models use Random Forest regression to vote for the optimal positions, leading to robust and accurate results. The performance of the system is evaluated using a set of 519 images. We show that the fully automated system is able to achieve a mean point-to-curve error of less than 1mm for 98% of all 519 images. To the best of our knowledge, this is the most accurate automatic method for segmenting the proximal femur in radiographs yet reported.

show abstract

“…time, cost and exposure to unacceptable levels of ionizing radiation) preclude their widespread use, on the other hand 2D medical imaging modalities such as X-rays, ultrasound or fluoroscopy, or direct pointer digitization can be used to derive patient-specific information pre-or intraoperatively from the patient's anatomy for the purpose of 3D reconstruction. The acquired patient-specific data can be processed and used to deform a shape model to finally reconstruct the patient's anatomy  Predicting the shape of one bone from the observation of another from the same joint (Yang et al 2008)  Segmentation (Baldwin et al 2010, Behiels et al 2002, Heimann et al 2009, Tang and Ellis 2005, SSM allows encapsulation of prior anatomical knowledge for compensating low contrast and/or high levels of noise in the images, such models can achieve robust segmentation by constraining the possible shapes Taylor 2004, Cootes et al 1995)  Design of prosthesis and biomechanical finite element analysis (Bryan et al 2009)  Aiding in the detection of pathologies related to shape (e.g. cam impingement, osteoarthritis), anatomical differences related to sex and aging (Styner et al 2005).…”

Section: Statistical Shape Modelsmentioning

confidence: 99%

Patient-Specific Modelling in Orthopedics: From Image to Surgery

Gomes

Cauter

Beule

et al. 2012

Biomedical Imaging and Computational Modeling in Biomechanics

View full text Add to dashboard Cite

In orthopedic surgery, to decide upon intervention and how it can be optimized, surgeons usually rely on subjective analysis of medical images of the patient, obtained from computed tomography, magnetic resonance imaging, ultrasound or other techniques. Recent advancements in computational performance, image analysis and in silico modeling techniques have started to revolutionize clinical practice through the development of quantitative tools, including patient-specific models aiming at improving clinical diagnosis and surgical treatment. Anatomical and surgical landmarks as well as features extraction can be automated allowing for the creation of general or patient-specific models based on statistical shape models. Preoperative virtual planning and rapid prototyping tools allow the implementation of customized surgical solutions in real clinical environments. In the present chapter we discuss the applications of some of these techniques in orthopedics and present new computer-aided tools that can take us from image analysis to customized surgical treatment.

show abstract

Evaluation of image features and search strategies for segmentation of bone structures in radiographs using Active Shape Models

Cited by 69 publications

References 12 publications

Automated Segmentation of Abdominal Aortic Aneurysms in Multi-spectral MR Images

Automated Segmentation of Abdominal Aortic Aneurysms in Multi-spectral MR Images

Accurate Fully Automatic Femur Segmentation in Pelvic Radiographs Using Regression Voting

Patient-Specific Modelling in Orthopedics: From Image to Surgery

Contact Info

Product

Resources

About