Preliminary Study for Automated Recognition of Anatomical Structure from Torso CT images

Mesenteric vasculature-guided small bowel segmentation on 3-D CT

Due to its importance and possible applications in visualization, tumor detection and preoperative planning, automatic small bowel segmentation is essential for computer-aided diagnosis of small bowel pathology. However, segmenting the small bowel directly on computed tomography (CT) scans is very difficult because of the low image contrast on CT scans and high tortuosity of the small bowel and its close proximity to other abdominal organs. Motivated by the intensity characteristics of abdominal CT images, the anatomic relationship between the mesenteric vasculature and the small bowel, and potential usefulness of the mesenteric vasculature for establishing the path of the small bowel, we propose a novel mesenteric vasculature map-guided method for small bowel segmentation on high-resolution CT angiography scans. The major mesenteric arteries are first segmented using a vessel tracing method based on multi-linear subspace vessel model and Bayesian inference. Second, multi-view, multi-scale vesselness enhancement filters are used to segment small vessels, and vessels directly or indirectly connecting to the superior mesenteric artery are classified as mesenteric vessels. Third, a mesenteric vasculature map is built by linking vessel bifurcation points, and the small bowel is segmented by employing the mesenteric vessel map and fuzzy connectness. The method was evaluated on 11 abdominal CT scans of patients suspected of having carcinoid tumors with manually labeled reference standard. The result, 82.5% volume overlap accuracy compared with the reference standard, shows it is feasible to segment the small bowel on CT scans using the mesenteric vasculature as a roadmap.

Automated recognition of the psoas major muscles on X-ray CT images

The purpose of this study is to recognize the psoas major muscle on X-ray CT images. For this purpose, we propose a novel recognition method. The recognition process in this method involves three steps: the generation of a shape model for the psoas major muscle, recognition of anatomical points such as the origin and insertion, and the recognition of the psoas major muscles by the use of the shape model. We generated the shape model using 20 CT cases and tested the model for recognition in 20 other CT cases. The average Jaccard similarity coefficient (JSC) and reproducibility rate were 0.704 and 0.783, respectively. Experimental results indicate that our method was effective for a 2-D cross-sectional area (CSA) analysis.

Automated measurement of bone-mineral-density (BMD) values of vertebral bones based on X-ray torso CT images

Bone is one of the most important anatomical structures in humans and osteoporosis is one of the major public health concerns in the world. Osteoporosis is a main target disease of bone, which can be detected by medical image techniques. The purpose of this study is to develop a fully automated computer scheme to measure bone-mineral-density (BMD) values for vertebral trabecular bones. This scheme will aid osteoporosis diagnosis performed using computer tomography (CT) images. This scheme includes the following processing steps: segmentation of the bone region, recognition of the skeletal structures and measurement of the BMD value in vertebral trabecular bone of each vertebral body. The proposed scheme was applied to 20 X-ray torso CT cases to measure the BMD values for vertebral trabecular bones. The experimental results show that the mean and standard deviation of the difference between the BMD values measured by using the proposed method and those measured using a manual segmentation method were 6.93 mg/cm(3) and 6.82 mg/cm(3) respectively. The accuracy of the proposed scheme satisfied the requirement for a computer-aided system used in osteoporosis diagnosis.

Automated estimation of the upper surface of the diaphragm in 3-D CT images

This communication describes a fully automated method by which the position of the diaphragm surface can be estimated by deforming a thin-plate model to match the bottom surface of the lung in CT images. This method was applied to 338 X-ray CT scans, and its validity was proved by the experimental results.

Constructing a probabilistic model for automated liver region segmentation using non-contrast X-ray torso CT images

A probabilistic model was proposed in this research for fully-automated segmentation of liver region in non-contrast X-ray torso CT images. This probabilistic model was composed of two kinds of probability that show the location and density (CT number) of the liver in CT images. The probability of the liver on the spatial location was constructed from a number of CT scans in which the liver regions were pre-segmented manually as gold standards. The probability of the liver on density was estimated specifically using a Gaussian function. The proposed probabilistic model was used for automated liver segmentation from non-contrast CT images. 132 cases of the CT scans were used for the probabilistic model construction and then this model was applied to segment liver region based on a leave-one-out method. The performances of the probabilistic model were evaluated by comparing the segmented liver with the gold standard in each CT case. The validity and usefulness of the proposed model were proved.