Articulated atlas for segmentation of the skeleton from head &amp; neck CT datasets

Steger, Sebastian; Kirschner, Matthias; Wesarg, Stefan

doi:10.1109/isbi.2012.6235790

Cited by 21 publications

(18 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…the vertebrae), simple image segmentation techniques like thresholding are not able to distinguish neighboring bones from each other. Therefore, we use the articulated atlas presented in [12]. It is capable of jointly segmenting the skull, mandible, the cervical vertebrae and the two upper thoracic vertebrae.…”

Section: Methodsmentioning

confidence: 99%

“…Once the articulated model converged, the segmentations S j for each bone item are extracted based on the probability maps and the CT intensities. Please refer to [12] for details.…”

Section: Methodsmentioning

confidence: 99%

“…One of the images must be a CT image -the fixed image -but the other one -the moving image -can be from a different modality for which rigid registration methods exist or an image of the same subject acquired at a different point in time. After an overall mutual information based rigid registration of the head, the articulated atlas presented in [12] is used to obtain a segmentation of the individual bones from the CT image. These bones are then successively rigidly registered with the moving image resulting in a rigid transformation for each of them.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated Skeleton Based Multi-modal Deformable Registration of Head&Neck Datasets

Steger

Wesarg

2012

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2012

Self Cite

View full text Add to dashboard Cite

Abstract. This paper presents a novel skeleton based method for the registration of head&neck datasets. Unlike existing approaches it is fully automated, spatial relation of the bones is considered during their registration and only one of the images must be a CT scan. An articulated atlas is used to jointly obtain a segmentation of the skull, the mandible and the vertebrae C1-Th2 from the CT image. These bones are then successively rigidly registered with the moving image, beginning at the skull, resulting in a rigid transformation for each of the bones. Linear combinations of those transformations describe the deformation in the soft tissue. The weights for the transformations are given by the solution of the Laplace equation. Optionally, the skin surface can be incorporated. The approach is evaluated on 20 CT/MRI pairs of head&neck datasets acquired in clinical routine. Visual inspection shows that the segmentation of the bones was successful in all cases and their successive alignment was successful in 19 cases. Based on manual segmentations of lymph nodes in both modalities, the registration accuracy in the soft tissue was assessed. The mean target registration error of the lymph node centroids was 5.33 ± 2.44 mm when the registration was solely based on the deformation of the skeleton and 5.00 ± 2.38 mm when the skin surface was additionally considered. The method's capture range is sufficient to cope with strongly deformed images and it can be modified to support other parts of the body. The overall registration process typically takes less than 2 minutes.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Once the articulated model converged, the segmentations S j for each bone item are extracted based on the probability maps and the CT intensities. Please refer to [12] for details.…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Skeleton Based Multi-modal Deformable Registration of Head&Neck Datasets

Steger

Wesarg

2012

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2012

Self Cite

View full text Add to dashboard Cite

show abstract

“…An articulated atlas is used to extract individual bones in the CT image ( Fig. 9) (Steger and Sakas 2012;Steger and Kirschner 2012). For each of the bones, a single rigid transformation is successively found based on the maximization of an image similarity metric.…”

Section: Image Fusionmentioning

confidence: 99%

Biomarkers in NeoMark European Project for Oral Cancers

Poli¹,

Copelli²,

Lanfranco³

et al. 2014

Biomarkers in Cancer

Self Cite

View full text Add to dashboard Cite

Oral cavity cancers are the seventh tumor by diffusion worldwide with more than 90 % being diagnosed as oral squamous cell carcinomas (OSCCs). According to the latest WHO statistics, OSCC accounts for 5 % of the cancer deaths worldwide, being the eighth more lethal cancer entity. Early identification of cancer relapses would have the potentiality to improve the disease control and the patient survival.NeoMark is a European co-funded research project (Seventh Framework Program, Information and Communication Technologies: EU-FP7-ICT-2007-2-22483-NeoMark) that has the objective to identify relevant biomarkers of OSCC recurrence. It integrates high-throughput gene expression analysis in tumor cells and IT-assisted imaging with traditional staging and follow-up protocols to improve the recurrence risk stratification and to obtain the earlier identification of locoregional relapses.The architecture of the project is based on the following key points:-Creation of a web application tool: a unified interface that helps the storage and management of all information -NeoMark database: the heterogeneous NeoMark data (demographics and risk factors; clinical, pathological, and immunohistochemical parameters; filtered and cleaned genomic and imaging data) are stored in a single database -the Integrated Health Record Repository (IHRR) -on a central NeoMark server. The server contains the marker definition functional environment (MDFE), a data analysis module. Based on the heterogeneous input data, it estimates the likelihood of a relapse and identifies OSCC risk factors. -Imaging biomarker extraction: several biomarkers are obtained from medical images such as CT and MRI scans (size, amount of necrosis from tumor and lymph nodes, etc.). To extract those features, a custom software tool -called the NeoMark Image Processing Tool -has specifically been developed. -Genomic data cleaning and filtering: extraction of genomic data and filtering of genes with low data quality and of those with high number of missing values.The NeoMark system was trained and initially validated in a multicenter pilot study (three European clinical centers involved: two in Italy and one in Spain) basing on 86 patients affected by OSCC with a minimum follow-up of 12 months. The clinicians recognized the usefulness of the disease bioprofile (or disease-specific profile) identified by NeoMark to evaluate the risk of disease reoccurrence of a patient at diagnosis, to stratify patients affected by OSCC at baseline according to the risk of recurrence, and to reserve a "tailored therapy" to each case. List of Abbreviations Key Facts• Oral squamous cell carcinoma (OSCC): it accounts for 5 % of the cancer deaths worldwide, being the eighth more lethal cancer entity.• Tumor relapse that afflicts 25-50 % of the surgically treated OSCC patients represents the main cause of failure and death.• Early identification of OSCC relapses can improve patient survival and quality of life.• The state of the art in OSCC patient classification is mainly based on TNM stag...

show abstract

“…On the other hand, those in which each voxel has a real value representing either the confidence or the probability of such a voxel of being part of the structure of interest. The first type, normally known as statistical atlas, is usually constructed by using techniques of Principal Component Analysis (PCA) using as input data the spatial coordinates of a set of relevant points (landmarks) chosen either manually or automatically . The second type, probabilistic atlases, utilizes techniques based on mathematical morphology and probabilistic models …”

Section: Introductionmentioning

confidence: 99%