| Bladder cancer is a severe disease which seriously threatens the human health all over the world.《Cancer facts & figures 2008》indicates that bladder cancer is the fourth most common malignancy in men, just after prostate cancer,lung & bronchus cancer,colon & rectum cancer and accounts for 7% of all malignancy in men. Bladder cancer is the eighth most common malignancy of men in China and the incidence is climbing up rapidly in some cities. Besides basic diagnosis based on symptom and physical exam, the cystoscopy is currently the gold standard for bladder cancer diagnosis. Meanwhile, radiological imaging is often preformed in conjunction with the cystoscopy for the evaluation of malignant invasion into adjacent structures. Radiological reading and interpretation process depends greatly on the experience of radiologists. Since the densities of image voxels inside carcinomatous tissue differ only subtly from those of the surroundings and the appearance of bladder cancer varies greatly, it is quite difficult to acquire some essential information such as the exact structure and degree of muscle invasion directly from image sequence, which is of great importance to accurate and early diagnosis and surgical treatment planning.The aim of this study was to explore characteristic texture features that could distinguish bladder cancer form normal wall tissue in MR images, which may help us determine the invasion depth of bladder cancer into wall muscleautomatically.The major work of this study includes:1,The collection of data and selection of parameter of MRI.Six consecutive patients with suspected bladder tumors and nine volunteers with fine bladders, collected from October 2008 to April 2009 in Tangdu Hospital, China, were included in this study. The age hasn't show significant difference between the two groups. All the six patients were finally confirmed as bladder cancer by postoperative pathological biopsy. All MR examinations were performed with a 3.0T scanner (MR-Signa EXCITE HD, GE). The scanning sequence included T2-FSE with scanning parameters of 2117.6/78.0 TR/TE, thickness of 2.1mm, 0.0sp (spacing of layers), and DFOV of 38cm 38cm.2,The selection of texture features between bladder carcinomatous tissue and bladder wall tissue.To reflect the texture characteristics more comprehensively, three categories of texture features were used, including: (1) features based on the probability distribution of gray-level, such as the mean, standard deviation, Skewness, Kurtosis, Third moment, Entropy, Uniformity, Smoothness of ROI intensities , (2) features extracted from co-occurrence matrices which demonstrate the spatial relationships among ROI pixels, such as Energy, Contrast, Relation, Energy, Local homogeneity,and (3) the auto-covariance coefficients derived from statistical analysis, such as Distance. 3,Calculation of the texture features and statistical analysisAll programs were performed with Matlab7.6. To alleviate the deviation caused by field inhomogeneity and other errors, texture features calculated form every slice of a dataset were averaged and set as the final value of the patient/volunteer correspondingly. Two-tailed independent-simple T test was used for evaluating the significance of the difference between the two groups on texture features extracted. The difference between groups was considered to be significant if the p-value was less than 0.05. All the analysis was performed by the SPSS12.0 software package.Conclusion:1,The Entropy,Smoothness,Uniformity and Relation (base on co-occurrence)are significantly difference between the patients and volunteers(p<0.05).2,The Contrast (base on co-occurrence matrices) and Distance (based on autocovariance coefficients) tend to be higher volunteers, but not reaching statistical significance.3,Among all the thirty-nine features, the most significant difference was obtained using Relation with the pixel distance varied from 1 to 3 pixels. But the results are not different when the pixel distance varied from 1 to 3 pixels. |
PDF Full Text Request