Automated brain tumor segmentation in magnetic resonance imaging based on sliding-window technique and symmetry analysis / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Brain tumor segmentation from magnetic resonance imaging (MRI) is an important step toward surgical planning, treatment planning, monitoring of therapy. However, manual tumor segmentation commonly used in clinic is time-consuming and challenging, and none of the existed automated methods are highly robust, reliable and efficient in clinic application. An accurate and automated tumor segmentation method has been developed for brain tumor segmentation that will provide reproducible and objective results close to manual segmentation results.</p><p><b>METHODS</b>Based on the symmetry of human brain, we employed sliding-window technique and correlation coefficient to locate the tumor position. At first, the image to be segmented was normalized, rotated, denoised, and bisected. Subsequently, through vertical and horizontal sliding-windows technique in turn, that is, two windows in the left and the right part of brain image moving simultaneously pixel by pixel in two parts of brain image, along with calculating of correlation coefficient of two windows, two windows with minimal correlation coefficient were obtained, and the window with bigger average gray value is the location of tumor and the pixel with biggest gray value is the locating point of tumor. At last, the segmentation threshold was decided by the average gray value of the pixels in the square with center at the locating point and 10 pixels of side length, and threshold segmentation and morphological operations were used to acquire the final tumor region.</p><p><b>RESULTS</b>The method was evaluated on 3D FSPGR brain MR images of 10 patients. As a result, the average ratio of correct location was 93.4% for 575 slices containing tumor, the average Dice similarity coefficient was 0.77 for one scan, and the average time spent on one scan was 40 seconds.</p><p><b>CONCLUSIONS</b>An fully automated, simple and efficient segmentation method for brain tumor is proposed and promising for future clinic use. Correlation coefficient is a new and effective feature for tumor location.</p>

RSF model optimization and its application to brain tumor segmentation in MRI / 生物医学工程学杂志

Magnetic resonance imaging (MRI) is usually obscure and non-uniform in gray, and the tumors inside are poorly circumscribed, hence the automatic tumor segmentation in MRI is very difficult. Region-scalable fitting (RSF) energy model is a new segmentation approach for some uneven grayscale images. However, the level set formulation (LSF) of RSF model is not suitable for the environment with different grey level distribution inside and outside the intial contour, and the complex intensity environment of MRI always makes it hard to get ideal segmentation results. Therefore, we improved the model by a new LSF and combined it with the mean shift method, which can be helpful for tumor segmentation and has better convergence and target direction. The proposed method has been utilized in a series of studies for real MRI images, and the results showed that it could realize fast, accurate and robust segmentations for brain tumors in MRI, which has great clinical significance.

Distribution and polymorphisms of host killing genes in the plasmids metagenome of Escherichia coli / 中华微生物学和免疫学杂志

Objective To investigate the genotypes of host killing genes and their single nucleotide polymorphisms (SNPs). Methods Three hundred and twenty strains of Escherichia coli that collected from the First Affiliated Hospital of Wenzhou Medical College were analyzed. The first sample ( E1 ) contains 160 strains isolated during the years from 2002 to 2003. The second sample (E2) contains 160 strains covering the years from 2008 to 2009. The plasmids of Escherichia coli were extracted by alkaline lysis method. Solexa/Illumina sequencing technology was used to sequence plasmids metagenome. Solexa Genome Analysis System and Soap programs were used to analyze gene distribution, SNPs and lineage-specific mutations. Results 11 077 768 reads were generated and 0. 045% of them can map to the reference sequences from El sample. Whereas 9 377 792 reads were generated and 0. 053% of which mapped to the reference from E2 sample. There are nine host killing genes identified in the two samples, of which hok gene is the most prevalent. A total of 29 SNP sites dispersed in five genes of the two samples. Approximately 33% of them were non-synonymous mutations. One position of A and G is the most prevalent polymorphism. Conclusion The known nine genotypes of host killing genes were all identified in plasmids of Escherichia coli in Wenzhou. hok gene showed the highest frequency. There were SNPs in five genotypes.

The processing of point clouds for brain deformation existing in image guided neurosurgery system / 生物医学工程学杂志

The finite element method (FEM) plays an important role in solving the brain deformation problem in the image guided neurosurgery system. The position of the brain cortex during the surgery provides the boundary condition for the FEM model. In this paper, the information of brain cortex is represented by the unstructured points and the boundary condition is achieved by the processing of unstructured points. The processing includes the mapping of texture, segmentation, simplification and denoising. The method of k-nearest clustering based on local surface properties is used to simplify and denoise the unstructured point clouds. The results of experiment prove the efficiency of point clouds processing.

Research of the key technique in virtual endoscopy and its clinical application / 生物医学工程学杂志

Studying the critical technique of virtual endoscopy (VE), we developed a VE system for clinical application. Computerized tomograph (CT) VE images built by the VE system were compared to those by fiberscopy and pathology. The results showed that the VE system could satisfy the demand of clinical application. The technique being applied to VE system is feasible.

Atherosclerotic calcification of coronary artery detected by electron beam CT: A new probation of calcific algorithm / 生物医学工程学杂志

Electron beam computed tomography (EBCT) can detect the atherosclerotic calcification of coronary artery qualitatively and quantitatively. It was also verified that the atherosclerotic calcification was directly related to the atherosclerotic extent and had a limited relation to the occurrence of coronary heart disease (CHD). So EBCT is one of the good non-invasive methods for predicting the risk of CHD. However, there are some problems in the calcification parameters (calcification area, calcification score) adopted by EBCT which have high variability and low reproducibility. As a result, these parameters have imperfection and need to be improved further. This research provides a new calcification parameter (calcification volume) which makes the use of three dimensional information of all calcific pixels in EBCT scanning images of coronary artery. After experiment in 11 human coronary artery specimens, it was testified that calcification volume had a lower variability than calcification area and calcification score in 25% percentile, median, 75% percentile, Mean, respectively. P value of t test in Mean variability is 0.027, and 0.058. These results suggest that calcification volume may be a new calcification parameter.

Three-dimensional reconstruction and display of intraocular and orbital tumors based on multi-modality image fusion theory / 第二军医大学学报

Objective: To study the application of CT and MRI fusion technique in the diagnosis and treatment of intraocular and orbital tumors. Methods: 2D-images of 13 patients with intraocular and orbital tumors were fused by special-point registration and Iterative Local Closest Point(ILCP) method; 3D-fusion images were reconstructed by Ray Tracing method. Results: A 3D-CT-MRI fusion images of intraocular and orbital tumors were reconstructed and displayed. The CT and MRI data of intraocular and orbital tumors were displayed on the same image as a comprehensive whole,which provided a stereogram of 3D-structure of the normal and abnormal orbital tissues. Anatomical structure of the orbit was clearly visualized by 3D-CT-MRI image. Conclusion: The multi-modality fusion technique can provide more accurate and comprehensive information for clinical diagnosis of intraocular and orbital tumors, which is helpful for doctors' planning of surgical operations,clinical education and doctor-patient communication.

A medical image fusion method based on fuzzy mathematics / 生物医学工程学杂志

In this paper, a method based on fuzzy mathematics to fuse multimodality medical images was presented. The improved FCM algorithm was adopted to segment images, and automatic fuzzy redistribution algorithm to define the subject degree. 16 different combinations of image tissues and 16 context relations, 256 models altogether, were considered. The result showed that the method had the great ability of anti-error and anti-segmentation interference and had the characteristics of robustness, quickness, and accuracy.

PRIMARY STUDY ON VIRTUAL ANATOMY OF THE THIGH PART ON COMPUTED TOMOGRAPHY DATA SOURCE / 解剖学报

Objective To reconstruct the three-dimensional(3D) image and to study on the virtual anatomy of the thigh part based on computed tomography(CT) data source. Methods Ten patients' contrast-enhanced light-speed CT images of the thigh part were chosen, “3D medical studio” was applied to read CT data, then to reconstruct and anatomize the thigh part. Results The reconstructed 3D images of the thigh part were realistically pseudocolored displayed, it was composed of four kinds of tissues(skin, muscle, skeleton and artery) and could be freely rotated and incised and restored for many times at any shape and depth on real time. Its' layers were distinct among tissues. Conclusion This research realizd partial function of virtual anatomy, it may play a definite role on regional anatomy and sectional image anatomy teaching and clinical radiology diagnosis.

Reconstruct three-dimensional medical images based on Internet / 生物医学工程学杂志

<p><b>UNLABELLED</b>This study was conducted to reconstruct three-dimensional medical images based on web and to achieve highly realistic display. Using the volume rendering techniques to reconstruct and display three-dimensional images in Java Applet program and utilizing signed Applets for solving security problem, We got two-dimensional images of human organs from ultrafast CT as sources and reconstructed the organs configuration of heart, coronary artery, head, cervical vertebrae, and pelvis. This reconstruction can be run in Web browser on different kinds of computers and for virtual surgical planning.</p><p><b>CONCLUSIONS</b>1. Three-dimensional medical image reconstruction in Web browser implemented by Java Applet is feasible, which will prompt clinical use of three-dimensional images. 2. The solid conformation of human organs, especially the anatomic structure of the coronary artery, can be displayed by using three-dimensional reconstruction techniques, which may offer more information to clinics.</p>

RECONSTRUCTION AND FUSION WITH MRI OF CAUDATE NUCLEUS, LENTIFORM NUCLEUS AND THALAMUS / 解剖学报

Objective To reconstruct caudate nucleus,lentiform nucleus and thalamus three-dimensionally,fuse these neural structures with (patient's) radiological images and provide anatomical data for diagnosis and treatment of neurosurgical diseases. Methods Chinese Virtual Human dataset was segmented using image-transparency method at first and each image was smoothed by eroding and dilating, then fused neural structures with MRI,the neural structures and brain in MRI were reconstructed and displayed using surface rendering and volume rendering lastly. Results The reconstructed neural structures were smooth,natural and realistic.Their shapes and positions was clearly displayed and could be rotated, observed in any direction.On the fused images,the neural structures and tumor could be seen at the same time.Conclusion The reconstructed structures and fused images have great reference value to teaching and learning and diagnosis and treatment of neurosurgical diseases.

THREE-DIMENSIONAL COMPUTERIZED RECONSTRUCTION OF HEART AND CORONARY ARTERY ON INTERNET / 解剖学报

Objective To reconstruct three-dimensional (3D) image of heart and coronary artery on internet and explore the reliability. Methods After the contrast medium was injected into coronary artery of 2 heart specimens, hearts were scanned by UFCT with enhancement volume. One patient's UFCT images of coronary artery angiography were also chosen.Then the primitive images were transposed format, removed noise, specified interesting areas. 3D software on server was applied to reconstruct heart and coronary artery. Results The reconstructed 3D images of the heart and coronary artery were realistically displayed, freely rotated and transected.Main parts and branches of coronary artery were similar to those in the images of the UFCT. Conclusion Heart and coronary artery can be reconstructed on internet,which offers a new approach for reconstructing organs and tissues of human being.;