Hae-Gil Hwang

Comparison of thresholding methods for breast tumor cell segmentation

Breast cancer was the most commonly occurring malignancy among women, so study of breast cancer are important. This paper is preprocessing of breast tumor research for segmentation method. There are many thresholding methods. Many thresholding methods developed but they have different result in each image. So we need automatic thresholding method because manual operating is tedious, time-consuming. This study compared result of 6 automatic thresholding method and 1 semi-supervised thresholding method in breast tumor image. Otsus method and Iterative selection are good result in breast tumor cell images. So we expected results using Otsus method and Iterative Selection to help for determining and diagnosing the breast tumor.

Three-dimensional Texture Analysis of Renal Cell Carcinoma Cell Nuclei for Computerized Automatic Grading

The extraction of important features in cancer cell image analysis is a key process in grading renal cell carcinoma. In this study, we analyzed the three-dimensional chromatin texture of cell nuclei based on digital image cytometry. Individual images of 2,423 cell nuclei were extracted from 80 renal cell carcinomas (RCCs) using confocal laser scanning microscopy (CLSM). First, we applied the 3D texture mapping method to render the volume of entire tissue sections. Then, we determined the chromatin texture quantitatively by calculating 3D gray level co-occurrence matrices and 3D run length matrices. Finally, to demonstrate the suitability of 3D texture features for classification, we performed a discriminant analysis. In addition, we conducted a principal component analysis to obtain optimized texture features. Automatic grading of cell nuclei using 3D texture features had an accuracy of 78.30%. Combining 3D textural and 3D morphological features improved the accuracy to 82.19%.

Classification of breast tissue images based on wavelet transform using discriminant analysis, neural network and SVM

In this paper, we described breast tissue image analyses using texture features from Haar wavelet transformed images to classify breast lesion of ductal organ Benign, DCIS and CA. The approach for creating a classifier is composed of 2 steps: feature extraction and classification. Therefore, in the feature extraction step, we extracted texture features from wavelet transformed images with 10/spl times/ magnification. In the classification step, we created three classifiers from each image of extracted features using statistical discriminant analysis, neural networks (back-propagation algorithm) and SVM (support vector machines). In this study, we conclude that the best classifier in histological sections of breast tissue in the texture features from second-level wavelet transformed images used in discriminant function.

Multi-resolution wavelet-transformed image analysis of histological sections of breast carcinomas

Multi-resolution images of histological sections of breast cancer tissue were analyzed using texture features of Haar- and Daubechies transform wavelets. Tissue samples analyzed were from ductal regions of the breast and included benign ductal hyperplasia, ductal carcinoma in situ (DCIS), and invasive ductal carcinoma (CA). To assess the correlation between computerized image analysis and visual analysis by a pathologist, we created a two-step classification system based on feature extraction and classification. In the feature extraction step, we extracted texture features from wavelet-transformed images at 10× magnification. In the classification step, we applied two types of classifiers to the extracted features, namely a statistics-based multivariate (discriminant) analysis and a neural network. Using features from second-level Haar transform wavelet images in combination with discriminant analysis, we obtained classification accuracies of 96.67 and 87.78% for the training and testing set (90 images each), respectively. We conclude that the best classifier of carcinomas in histological sections of breast tissue are the texture features from the second-level Haar transform wavelet images used in a discriminant function.

Design of the breast carcinoma cell bank system

We describe a cell bank construction which was designed by a content-based image retrieval system using SQL database and XML for breast carcinoma images. However, the conventional pathological images for storage, management and data sharing have been done by a manual handling. We attempted to find a solution for the occurred processing problems by constructing a computerized standard system and a large dataset for the breast carcinoma images. It is possible that the system could classify the images according to the categorized cancer, text retrieval- and content-based retrieval system color- and texture features were applied. The software was designed and developed by using Visual Basic and the database constructed by using SQL the extract the texture features from the images. That was then stored in XML for MPEG-7 based retrieval standard system.

Development of image processing software for quantitative analysis of bioluminescence image

Molecular imaging is leading an important role in the era of molecular medicine. Optical imaging, a rising star in the filed of molecular imaging, largely consists of fluorescent imaging and bioluminescent imaging. Molecular imaging analysis will allow us to observe the incipience and progression of the disease. In this paper, we suggest image processing methods and develop software for bioluminescence image analysis. And we calculate the photon count and analyze bioluminescence image in time sequence. . The mouse was transplanted a colon cancer which has high photon count value in day 3 and day 4. Bacterial luciferase reporter gene expression becomes activity in this time. This study has the importance of the development software for bioluminescence image processing and analysis. We expected this study lead the development of image technology.

Image Analysis of Ductal Proliferative Lesions of Breast Using Architectural Features

We propose a method to classify breast lesions of ducatal origin. The materials were tissue sections of the intraductal proliferative lesions of the breast: benign(DH:ductal hyperplasia), ductal carcinoma in situ(DCIS). The total 40 images from 70 samples of ducts were digitally captured from 15 cases of DCIS and 25 cases of DH diagnosed by pathologist. To assess the correlation between computerized images analysis and visual analysis by a pathologist, we extracted the total lumen area/gland area, to segment the gland(duct) area used a snake algorithm, to segment the lumen used multilevel Otsus method in the duct from 20x images for distinguishing DH and DCIS. In duct image, we extracted the five texture features (correlation, entropy, contrast, angular second moment, and inverse difference moment) using the co-occurrence matrix for a distribution pattern of cells and pleomorphism of the nucleus. In the present study, we obtained classification accuracy rates of 91.33%, the architectural features of breast ducts has been advanced as a useful features in the classification for distiguishing DH and DCIS. We expect that the proposed method in this paper could be used as a useful diagnostic tool to differentiate the intraductal proliferative lesions of the breast.

Cancer Cell Image Analysis and Visualization

We review computerized cancer cell image analysis and visualization research over the past 30 years. Image acquisition, feature extraction, classification, and visualization from two-dimensional to three-dimensional image algorithms are introduced with case studies of bladder, prostate, breast, and renal carcinomas.

Molecular imaging and its tissue analysis

For the 3D image reconstruction we used 32 laser source images, a flat image and 3D surface rendering that confused for 3D fluorescence molecular imaging (FMI). A minimum often tissue sections for Lewis lung carcinoma (LLC) and chemotherapy resistant Lewis lung carcinoma (CR-LLC) tumors were analyzed per tumor for quantification of the TUNEL-positive cells, cell-associated Cy5.5-Annexin and vessel-associated Alexa Fluor-Lectin.

Development of fitting model for visualization of left ventricle in gated SPECT image using least squares method

Cardiac disorder is one of the important diseases. Three-dimensional visualization and motility analysis of heart are necessaries to diagnosis of cardiac disorder. Because it is an important that diagnosis of cardiac disorder considers changing of shape with the passage of time. The purpose of this paper is to create a standard model of left ventricle myocardium. We restored the lost apex of heart in the process of image acquisition and interpolated. We found an algorithm for it using the least squares method. Furthermore we added some conditions the passed the points in the general least square method. It is an important point to determine the shape of left ventricle myocardium. We expect that the proposed algorithm should be helpful for a development of standard model.