A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi-atlas registration, which uses the atlas prior information; the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label; the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86% on the left hippocampus and 87.4% on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.
Algorithms ; Hippocampus ; anatomy & histology ; Humans ; Image Processing, Computer-Assisted ; methods ; Magnetic Resonance Imaging ; Neuroimaging ; Software

We propose a multi-weighted probabilistic atlas to obtain accurate, robust, and reliable segmentation. The local similarity measure is used as the weight to compute the probabilistic atlas, and the distance field is used as the weight to incorporate the locality information of the atlas; the self-similarity is used as the weight to incorporate the local information of target image to refine the probabilistic atlas. Experimental results with brain MRI images showed that the proposed algorithm outperforms the common brain image segmentation methods and achieved a median Dice coefficient of 87.1% on the left hippocampus and 87.6% on the right.
Algorithms ; Brain ; anatomy & histology ; Humans ; Magnetic Resonance Imaging ; Neuroimaging

A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi- atlas registration, which uses the atlas prior information;the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label;the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86%on the left hippocampus and 87.4%on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.

Objective To establish a model for discrimination between benign and malignant gastrointestinal stromal tumors (GIST) by analyzing the texture features extracted from computed tomography (CT) images. Methods The CT datasets were collected from 110 patients with GIST (including 80 as the training cohort and 30 as the validation cohort). Feature set reduction was executed with the 0.632+ bootstrap method in the initial feature set followed by stepwise forward feature selection in the feature subset, and the classification model was generated by logistic regression. Results The 6-texture-feature-based classification model successfully discriminated between benign and malignant GIST in both the training and validation cohorts with AUCs of 0.93 and 0.91, sensitivity of 0.88 and 0.87, specificity of 0.85 and 0.86, and accuracy of 0.87 and 0.86 in the two cohorts, respectively. Conclusion This classification model established by radiomics analysis is capable of discrimination between benign and malignant GIST to provide assistance in preoperative diagnosis of GIST.

A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi- atlas registration, which uses the atlas prior information;the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label;the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86%on the left hippocampus and 87.4%on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.

Objective To establish a model for discrimination between benign and malignant gastrointestinal stromal tumors (GIST) by analyzing the texture features extracted from computed tomography (CT) images. Methods The CT datasets were collected from 110 patients with GIST (including 80 as the training cohort and 30 as the validation cohort). Feature set reduction was executed with the 0.632+ bootstrap method in the initial feature set followed by stepwise forward feature selection in the feature subset, and the classification model was generated by logistic regression. Results The 6-texture-feature-based classification model successfully discriminated between benign and malignant GIST in both the training and validation cohorts with AUCs of 0.93 and 0.91, sensitivity of 0.88 and 0.87, specificity of 0.85 and 0.86, and accuracy of 0.87 and 0.86 in the two cohorts, respectively. Conclusion This classification model established by radiomics analysis is capable of discrimination between benign and malignant GIST to provide assistance in preoperative diagnosis of GIST.

OBJECTIVETo propose a new method for automatic segmentation of manually determined knee articular cartilage into 9 subregions for T2 measurement.

METHODSThe middle line and normal line were automatically obtained based on the outline of articular cartilage manually drawn by experienced radiologists. The region of articular cartilage was then equidistantly divided into 3 layers along the direction of the normal line, and each layer was further equidistantly divided into 3 segments along the direction of the middle line. Finally the mean T2 value of each subregion was calculated. Bland-Altman analysis was used to evaluate the agreement between the proposed and manual subregion segmentation methods.

RESULTSThe 95% limits of agreement of manual and automatic methods ranged from -3.04 to 3.20 ms, demonstrating a narrow 95% limits of agreement (less than half of the minimum average). The coefficient of variation between the manual and proposed subregion methods was 4.04%.

CONCLUSIONThe proposed subregion segmentation method shows a good agreement with the manual segmentation method and minimizes potential subjectivity of the manual method.

Objective:To construct a diabetic foot classification prediction model based on radiomics features of fundus photographs.Methods:A total of 2 035 fundus photographs of patients with type 2 diabetes diagnosed at Nanfang Hospital between December 2011 and December 2018 were retrospectively collected [282 photographs from patients with diabetic foot(DF), and 1 753 from patients with diabetes mellitus(DM)]. All fundus photographs were randomly divided into a training set(1 424 photos) and a test set(611 photos) using a computer generated random number at 7∶3. After pre-processing the fundus photographs, a total of 4 128 texture features based on the gray matrix were extracted by the Radiomic toolkit, and 11 339 other features were extracted using the ToolboxDESC toolkit. The LASSO algorithm was used to select the 30 features most relevant to DF, and then the Bootstrap + 0.632 self-sampling method was used to further select the 7 best combinations. Logistic regression analysis was used to obtain the regression coefficients and establish the final diabetic foot classification prediction model. ROC curve was drawn, and AUC, sensitivity, specificity, and accuracy of the training and test sets were calculated to verify its prediction performance. Results:We screened 7 fundus radiomics markers for diabetic foot patients, and based on this established a DF/DM classification prediction model. The AUC, sensitivity, specificity, and accuracy of the model were 0.958 6, 0.984 0, 0.920 0, and 0.928 0 in the training set, and 0.927 1, 0.988 9, 0.881 0, and 0.896 9 in the test set, respectively.Conclusion:In this study, seven DF fundus markers were screened using radiomics technology. Based on this, a highly accurate and easy-to-use DF/DM classification model was constructed. This technology has the potential to increase the efficiency of DF screening programs.

OBJECTIVE: To compare the effectiveness and sensitivity of entropy and regional homogeneity (ReHo) for identifying irritable bowel syndrome (IBS) based on functional magnetic resonance imaging (fMRI). METHODS: Voxel-based approximate entropy (ApEn) was calculated based on findings of resting fMRI of 54 patients with IBS and 54 healthy control subjects. Feature selection was performed using independent sample -test, and support vector machine was then used to classify and identify different groups. The classification performance obtained from ApEn was compared with that from ReHo. RESULTS: Significant differences between the two groups were found in the left triangle part of inferior prefrontal gyrus, right angular gyrus of the inferior parietal lobule, left inferior temporal gyrus, left middle temporal gyrus, left lingual gyrus, bilateral middle occipital gyrus and bilateral superior occipital gyrus for ReHo ( < 0.05), and in the bilateral postcentral gyrus, right precentral gyrus, right inferior temporal gyrus, bilateral middle temporal gyrus and left superior occipital gyrus for ApEn ( < 0.05). ApEn consistently showed better performance than ReHo regardless of the variations in the number of features. The classification accuracy, specificity and sensitivity of ApEn were 93.5185%, 90.7407% and 96.2963%, respectively, as compared with 86.1111%, 85.1852% and 87.037% of ReHo. CONCLUSIONS Entropy analysis based on fMRI can be more sensitive and effective than ReHo for identification of IBS.
Brain ; diagnostic imaging ; Brain Mapping ; Case-Control Studies ; Entropy ; Humans ; Irritable Bowel Syndrome ; diagnostic imaging ; Magnetic Resonance Imaging

OBJECTIVE: To establish a fast adaptive active contour model based on local gray difference for parotid duct image segmentation. METHODS: On the basis of the LBF model, we added the mean difference of the local gray scale inside and outside the contour as the energy term of the driving evolution curve, and the local gray-scale variance difference was used to replace and as the control term of the energy parameter value. Two local similarity factors of different neighborhood sizes were introduced to correct the effects of image gray unevenness and boundary blur to improve the segmentation efficiency. RESULTS: During image segmentation, this algorithm allowed for adaptive adjustment of the evolution direction, velocity and the energy weight of the internal and external regions according to the difference of gray mean and variance between the internal and external regions. This algorithm was also capable of detecting the actual boundary in a complex gradient boundary region, thus enabling the evolution curve to approach the target boundary quickly and accurately. CONCLUSIONS The proposed algorithm is superior to the existing segmentation algorithms and allows fast and accurate segmentation of the parotid duct with well-preserved image details.
Algorithms ; Color ; Image Processing, Computer-Assisted ; Parotid Gland ; diagnostic imaging ; Salivary Ducts ; diagnostic imaging