ABSTRACT
Objective: To explore the effect of three-dimensional reconstruction of abdominal organ CT images based on improved moving cube algorithm. Methods: An adaptive improved marching cube algorithm based on the universal tree structure and the contour points method based on the regional growth method were proposed. Firstly, the medical images were segmented, and all the voxels intersecting with the threshold were marked after the seed points were selected. A general tree structure was created to insert intersecting voxels into sub-nodes and determine the vertex index method based on the general tree. Simplify the acquisition of equivalence information by moving equivalence points to merge coplanar triangles. Based on abdominal CT images of a volunteer, a three-dimensional kidney model was constructed by using traditional moving cube algorithm and improved moving cube algorithm, and the effects were compared. Results: Compared with traditional algorithm, the triangle facets generated with the improved moving cube algorithm were reduced by 39.20%, the efficiency of the algorithm was improved by 37.59%, the surface of the three-dimensional model was smooth and lifelike, and the local details were more accurate. Conclusion: Based on the improved moving cube algorithm, three-dimensional reconstruction of abdominal organs in CT images can be achieved quickly and accurately.
ABSTRACT
Objective To explore the feasibility of applying threshold and region-growing based algorithm in computed tomography (CT) images for the estimation of hepatic lesion volumes in patients with Tusanqi related hepatic sinusoidal obstruction syndrome (SOS).Methods From July 2012 to January 2015,at the First Affiliated Hospital of Anhui Medical University,20 patients who were diagnosed with SOS and had complete CT images were enrolled.Stereology and threshold and regiongrowing based algorithm were performed to estimate volumes of the low-density region in the liver,respectively,and then the measured volumes of hepatic lesion and operation time were compared.Paired samples t test and the Bland-Altman statistical test were performed for statistical analysis.Results The hepatic lesion volumes measured by stereology and threshold and region-growing based algorithm were (1.001±0.327) dm3 and (1.045±0.363) dm3,respectively,and the difference was not statistically significant (P＞0.05).The consistency between the two methods was good.The operation time of threshold and region-growing based algorithm was (597.55±52.86) s (minimum 500 s),which was less than that of stereology (1 251.60 ± 105.88) s (minimum 1 075 s),and the difference was statistically significant (t =32.808,P＜ 0.01).Conclusion There is no statistically significant difference in the measured hepatic lesion volume of patients with SOS between stereology and the threshold and regiongrowing based algorithm,but the operation time of threshold and region-growing based algorithm is shorter.
ABSTRACT
Objective To explore the feasibility of tubular model based segmentation method for cystic artery and three-dimensional (3D) reconstruction model of Calot's triangle.Methods A tubular model based 3D region growing algorithm was proposed for the segmentation of cystic arteries and its adjacent vessels from 13 patients' CT images in DICOM format.The data was transferred to 3D visualization workstation based on a set of CalotShow1.0 software for 3D reconstruction.Results The method could effectively segment cystic artery and obtain the 3D model of Calot's triangle.Conclusions The 3D reconstruction model based on tubular model related vessel segmentation method and CalotShow1.0 can accurately display the spatial positions and adjacent relationships of cystic artery and Calot's triangle.
ABSTRACT
Objective:The legal medical experts usuallycalculate the area ofinjured organs in bodyor other injured items bycountingthe grids in coordination paper.The experiences of the legal medical experts determine the result of this method.The result is subjective and has serious errors.So it is not conductive to the court and case detection.Methods:This paper proposes a method to calculate the injured area by hybrid of K-means algorithm and region growing algorithm.Results:The experiment shows this method is effective to calculate the area of irregular region.Conclusion:The result of this method is more accurate than the method of counting the grids in coordination paper.
ABSTRACT
A medical image segmentation is the primary issue in computer aided diagnosis. The traditional methods did not perform the image segmentation well because of varieties of image, inadequate informations, noises, uncertain images, and deficient image data. We Propose a new medical image segmentation by machine learning using background knowledge of segmentation pattern. The proposed algorithm is applied to real brain CT images. First, a region growing algorithm extracts the regions and statistical data. Also, shape informations about each regions are gathered. A supervisor makes a set of learning examples by selecting the regions which should be in one region. In the next step, some rules for merging regions are discovered from common properties of the examples. Also there will be verification procedure whether the pattern is the desired one. The procedure is achieved by machine learning technique from the patterns of positive or negative examples. The systems try to recognize the improved patterns in the next step, and make a knowledge base for the segmentation. From the experimental results of the proposed algorithm which is applied to various brain images, we obtain an adaptable knowledge base and a segmented image with proper regions of brain shape.