Entity

Time filter

Source Type

Nanjing, China

Shi Y.,Nanjing University | Gao Y.,Nanjing University | Yang Y.,Nanjing University | Zhang Y.,Bayi hospital | Wang D.,Bayi hospital
IEEE Transactions on Biomedical Engineering | Year: 2013

Lung needle biopsy image classification is a critical task for computer-aided lung cancer diagnosis. In this study, a novel method, multimodal sparse representation-based classification (mSRC), is proposed for classifying lung needle biopsy images. In the data acquisition procedure of our method, the cell nuclei are automatically segmented from the images captured by needle biopsy specimens. Then, features of three modalities (shape, color, and texture) are extracted from the segmented cell nuclei. After this procedure, mSRC goes through a training phase and a testing phase. In the training phase, three discriminative subdictionaries corresponding to the shape, color, and texture information are jointly learned by a genetic algorithm guided multimodal dictionary learning approach. The dictionary learning aims to select the topmost discriminative samples and encourage large disagreement among different subdictionaries. In the testing phase, when a new image comes, a hierarchical fusion strategy is applied, which first predicts the labels of the cell nuclei by fusing three modalities, then predicts the label of the image by majority voting. Our method is evaluated on a real image set of 4372 cell nuclei regions segmented from 271 images. These cell nuclei regions can be divided into five classes: four cancerous classes (corresponding to four types of lung cancer) plus one normal class (no cancer). The results demonstrate that the multimodal information is important for lung needle biopsy image classification. Moreover, compared to several state-of-the-art methods (LapRLS, MCMI-AB, mcSVM, ESRC, KSRC), the proposed mSRC can achieve significant improvement (mean accuracy of 88.1\%$, precision of 85.2\%$, recall of 92.8\%$, etc.), especially for classifying different cancerous types. © 1964-2012 IEEE. Source


Shi Y.,Nanjing University | Gao Y.,Nanjing University | Wang R.,Massey University | Zhang Y.,Bayi hospital | Wang D.,Bayi hospital
Applied Intelligence | Year: 2013

Previous computer-aided lung cancer image classification methods are all cost-blind, which assume that the misdiagnosis (categorizing a cancerous image as a normal one or categorizing a normal image as a cancerous one) costs are equal. In addition, previous methods usually require experienced pathologists to label a large amount of images as training samples. To this end, a novel transductive cost-sensitive method is proposed for lung cancer image classification on needle biopsies specimens, which only requires the pathologist to label a small amount of images. The proposed method analyzes lung cancer images in the following procedures: (i) an image capturing procedure to capture images from the needle biopsies specimens; (ii) a preprocessing procedure to segment the individual cells from the captured images; (iii) a feature extraction procedure to extract features (i.e. shape, color, texture and statistical information) from the obtained individual cells; (iv) a codebook learning procedure to learn a codebook on the extracted features by adopting k-means clustering, which aims to represent each image as a histogram over different codewords; (v) an image classification procedure to predict labels for testing images using the proposed multi-class cost-sensitive Laplacian regularized least squares (mCLRLS). We evaluate the proposed method on a real-image set provided by Bayi Hospital, which contains 271 images including normal ones and four types of cancerous ones (squamous carcinoma, adenocarcinoma, small cell cancer and nuclear atypia). The experimental results demonstrate that the proposed method achieves a lower cancer-misdiagnosis rate and lower total misdiagnosis costs comparing with previous methods, which includes the supervised learning approach (kNN, mcSVM and MCMI-AdaBoost), semi-supervised learning approach (LapRLS) and cost-sensitive approach (CS-SVM). Meanwhile, the experiments also disclose that both transductive and cost-sensitive settings are useful when only a small amount of training images are available. © 2012 Springer Science+Business Media, LLC. Source


Guo B.,Bayi hospital | Wang Y.,General Hospital of Coal Industry Ministry of China | Hui Y.,PLA Fourth Military Medical University | Yang X.,No. 4 Hospital of Xian | Fan Q.,Bayi hospital
Molecular Vision | Year: 2010

Purpose: To evaluate the effects of an anti-rat vascular endothelial growth factor antibody (ARVA) and bevacizumab (Avastin) on rat retinal Müller glial cells (RMGCs) in vivo and in vitro. Methods: Rat RMGCs were identified and cultivated, and were then treated with bevacizumab (0.1, 0.25, and 1 mg/ml), ARVA (0.1, 0.5, and 1 μg/ml), or 1 mg/ml of rat immunoglobulin G (IgG) for 12, 24, 48, and 72 h. The numbers of viable RMGCs were determined using a trypan blue dye exclusion assay and a methyl thiazolyl tetrazolium colorimetric assay. In the in vivo study, the rats received intravitreal injections of 5 μl bevacizumab (3.75 mg/ml), ARVA (15 μg/ml), and rat IgG (1 mg/ml). The electroretinogram was recorded. Seven days after the injections, histopathologic changes and glial fibrillary acidic protein expression of RMGCs in the retina were analyzed by immunohistochemistry with hematoxylineosin and fluorescent staining. Results: After exposure to bevacizumab at various concentrations for various periods of time, the stained cell numbers and optical density values of mitochondrial dehydrogenase activity of RMGCs had no significant differences (p>0.05) from those of the control group and IgG medium. In the stained cells, ARVA demonstrated a dose-dependent increase. Compared with those treated for 12 and 24 h, the increase of stained cells treated with 0.5 and 1 μg/ml ARVA at 48 and 72 h was very significant (p<0.01). The optical densities of RMGCs exposed to 0.5 and 1 μg/ml of ARVA at 48 and 72 h were significantly lower than cells exposed to a fresh culture medium (p<0.01). The histology of both treated and control eyes after intravitreal injection was similar and showed no anatomic signs of toxicity. There were no obvious glial fibrillary acidic protein upregulations of RMGCs in all groups. The scotopic electroretinogram responses to flashes of light in the control and treated eyes had similar b-wave amplitudes. Conclusions: Intravitreal bevacizumab and ARVA had no short-term, direct retinal toxicity in rats. Bevacizumab exerts no inhibition on rat RMGCs, while ARVA at higher doses (over 0.5 μg/ml) may be harmful to the growth of RMGCs © 2010 Molecular Vision. Source

Discover hidden collaborations