• Journal of Biomedical Engineering Research
• /
• v.17 no.4
• /
• pp.433-442
• /
• 1996

Since one texture property(i.e coarseness, orientation, regularity, granularity) for ultrasound liver ages was not sufficient enough to classify the characteristics of livers, we used multi texture vectors tracted from ultrasound liver images and a statistical classifier. Multi texture vectors are selected among the feature vectors of the normal liver, fat liver and cirrhosis images which have a good separability in those ultrasound liver images. The statistical classifier uses multi texture vectors as input vectors and classifies ultrasound liver images for each multi texture vector by the Bayes decision rule. Then the decision of the liver disease is made by choosing the maximum value from the averages of a posteriori probability for each multi texture vector In our simulation, we obtained higtler correct ratio than that of other methods using single feature vector, for the test set the correct ratio is 94% in the normal liver, 84% in the fat liver and 86% in the cirrhosis liver.

• Journal of Biomedical Engineering Research
• /
• v.41 no.1
• /
• pp.48-54
• /
• 2020

In this paper, we deal with a liver fibrosis classification problem using ultrasound B-mode images. Commonly representative methods for classifying the stages of liver fibrosis include liver biopsy and diagnosis based on ultrasound images. The overall liver shape and the smoothness and roughness of speckle pattern represented in ultrasound images are used for determining the fibrosis stages. Although the ultrasound image based classification is used frequently as an alternative or complementary method of the invasive biopsy, it also has the limitations that liver fibrosis stage decision depends on the image quality and the doctor's experience. With the rapid development of deep learning algorithms, several studies using deep learning methods have been carried out for automated liver fibrosis classification and showed superior performance of high accuracy. The performance of those deep learning methods depends closely on the amount of datasets. We propose an enhanced U-net architecture to maximize the classification accuracy with limited small amount of image datasets. U-net is well known as a neural network for fast and precise segmentation of medical images. We design it newly for the purpose of classifying liver fibrosis stages. In order to assess the performance of the proposed architecture, numerical experiments are conducted on a total of 118 ultrasound B-mode images acquired from 78 patients with liver fibrosis symptoms of F0~F4 stages. The experimental results support that the performance of the proposed architecture is much better compared to the transfer learning using the pre-trained model of VGGNet.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.735-742
• /
• 2023

Non-alcoholic fatty liver disease is an independent risk factor for the development of cardiovascular disease, diabetes, hypertension, and kidney disease, and the clinical importance of non-alcoholic fatty liver disease has recently been increasing. In this study, we aim to extract feature values by applying GLCM, a texture analysis method, to ultrasound images of patients with non-alcoholic fatty liver disease. By applying an artificial neural network model using extracted feature values, we would like to classify the degree of fat deposition in non-alcoholic fatty liver into normal liver, mild fatty liver, moderate fatty liver, and severe fatty liver. As a result of applying the GLCM algorithm, the parameters Autocorrelation, Sum of squares, Sum average, and sum variance showed a tendency for the average value of the feature values to increase as it progressed from mild fatty liver to moderate fatty liver to severe fatty liver. The four parameters of Autocorrelation, Sum of squares, Sum average, and sum variance extracted by applying the GLCM algorithm to ultrasound images of non-alcoholic fatty liver disease were applied as inputs to the artificial neural network model. The classification accuracy was evaluated by applying the GLCM algorithm to the ultrasound images of non-alcoholic fatty liver disease and applying the extracted images to an artificial neural network, showing a high accuracy of 92.5%. Through these results, we would like to present the results of this study as basic data when conducting a texture analysis GLCM study on ultrasound images of patients with non-alcoholic fatty liver disease.

• Journal of the Korean Society of Radiology
• /
• v.16 no.7
• /
• pp.927-934
• /
• 2022

This study attempted to find out the degree of agreement between ultrasound image findings along with analysis of attenuation index and scatter distribution index values within tissues through quantitative measurement analysis using liver ultrasound images. From August 2022 to October 2022, liver ultrasound was performed on 45 patients who were suspected of having fatty liver and who received a prescription for liver ultrasound. As a result of the study, as a result of analyzing the agreement between the ultrasound image findings and the tissue attenuation index, the Kappa value was 0.82 (p<0.05), showing a very high agreement between the two examination methods. In addition, as a result of the agreement analysis between the ultrasound image findings and the scatter distribution index in the tissue, the Kappa value was 0.642 (p<0.05), showing high agreement between the two examination methods. At the time of fat liver prediction, the use of liver ultrasound findings and quantitative ultrasonography techniques, such as intra-tissue attenuation index and intra-tissue scatter distribution index, may be helpful in determining the degree of progression of fatty liver patients.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.9
• /
• pp.2206-2212
• /
• 2013

We propose a method for the classification of fatty liver by ultrasound imaging using Fuzzy Contrast Enhancement Technique and FCM. ROI images are extracted after removal of information data except ultrasound image of the liver and the kidney then image contrast is improved by Fuzzy Contrast Enhancement Algorithm. The images applied Fuzzy Contrast Enhancement Technique is applied average binarization then ROI images of liver and kidney parenchyma are extracted using Blob algorithm. Representative brightness is extracted in the liver and kidney images using the most frequent brightness level after classification of 10 brightness levels. We applied this method to ultrasound images and a radiologist confirmed the accuracy of diagnosis for fatty liver. This method would be a model for automatic method in the diagnosis of fatty liver.

• Journal of radiological science and technology
• /
• v.46 no.6
• /
• pp.475-484
• /
• 2023

This study aimed to compare filters for reducing speckle noise in ultrasound images using clinical liver images. We acquired the clinical liver ultrasound images, and noisy images were obtained by adding 0.01, 0.05, 0.10, and 0.50 intensity levels of speckle noise to the liver images. The Wiener filter, median modified Wiener filter, gamma filter, and Lee filter were designed for the noisy images by setting window sizes at 3×3, 5×5, and 7×7. The coefficient of variation (COV) and contrast to noise ratio (CNR) were calculated to evaluate noise reduction and various filters. Moreover, the filter with the highest image quality was selected and quantitatively compared to a noisy image. As a result, COV and CNR showed the noise improved result when the Lee filter was applied. Furthermore, the Lee filter image with a window size of 7×7 was noted to possess approximately a minimum of 1.28 to a maximum of 3.38 times better COV and a minimum of 2.18 to a maximum of 5.50 times better CNR than the noisy image. In conclusion, we confirmed that the Lee filter was effective in reducing speckle noise and proved that an appropriate window size needs to be set considering blurring.

• Journal of the Korean Society of Radiology
• /
• v.18 no.2
• /
• pp.135-144
• /
• 2024

Ultrasonography examination has limitations in quantifying hepatic fat quantification. Therefore, this study aimed to experimentally demonstrate whether changes in signal attenuation during ultrasound imaging can be quantified using simulated hepatic phantoms to assess hepatic fat content. Additionally, we aimed to evaluate the potential of ultrasound imaging for diagnosing hepatic fatty liver by analyzing the relationship between hepatic fat content and signal intensity in ultrasound images. In this study, we developed a total of five stimulated hepatic phantoms by homogeneously mixing water and oil. We confirmed the fat content of the phantoms using magnetic resonance imaging (MRI) and ultrasound imaging, and measured signal intensity according to distance in ultrasound images to analyze the correlation and mean comparison between fat content and signal intensity. We observed that as the fat content increased, the ultrasound penetration intensity decreased, confirming the potential for quantifying hepatic fat content using ultrasound. Additionally, the analysis of the correlation between the measured fat content using MRI and the signal intensity measured in ultrasound images showed a high correlation. Statistical analysis in our study confirmed that as the fat content increased, the slope representing signal during ultrasound imaging (US-GRE) decreased. In this study, it was statistically confirmed that the US-GRE value of ultrasound images gradually decreases as the fat content increases, and it is believed that US-GRE can serve as a biomarker expressing fatty liver content.

• Proceedings of the KOSOMBE Conference
• /
• v.1995 no.05
• /
• pp.37-41
• /
• 1995

In this paper, we proposed the 2-stage ultrasound liver image classifier which uses the fractal dimensions obtained from the original image and its 1/2 subsampled image, and the Normalized Fourier Power Spectrum. The fractal dimension based on Fractional Brownian Motion (FBM) is calculated from the variance of the same scale pixels instead of the mean of them. Since the actual ultrasound. liver images does not fully match the FBM, to get the fractal dimension, we use the scale vectors which satisfy the FBM model. In 2-stage classifier, we first classified normal and diffuse liver and then classified the fat liver and cirrhosis from the diffuse liver. For the test liver images. 70% of normal liver and 80% of fat liver and 90% of cirrhosis is classified classified with our 2-stage classifier.

• Journal of the Korean Society of Radiology
• /
• v.17 no.6
• /
• pp.911-917
• /
• 2023

The method of observing nodular changes on the liver surface using clinical ultrasonography is useful for diagnosing cirrhosis. However, the speckle noise that inevitably occurs in ultrasound images makes it difficult to identify changes in the liver surface and echo patterns, which has a negative impact on the diagnosis of cirrhosis. The purpose of this study is to model the median modified Wiener filter (MMWF), which can efficiently reduce noise in cirrhotic ultrasound images, and confirm its applicability. Ultrasound images were acquired using an ACR phantom and an actual cirrhotic patient, and the proposed MMWF algorithm and conventional noise reduction algorithm were applied to each image. Coefficient of variation (COV) and edge rise distance (ERD) were used as quantitative image quality evaluation factors for the acquired ultrasound images. We confirmed that the MMWF algorithm improved both COV and ERD values compared to the conventional noise reduction algorithm in both ACR phantom and real ultrasound images of cirrhotic patients. In conclusion, the proposed MMWF algorithm is expected to contribute to improving the diagnosis rate of cirrhosis patients by reducing the noise level and improving spatial resolution at the same time.

• Proceedings of the KOSOMBE Conference
• /
• v.1995 no.11
• /
• pp.177-182
• /
• 1995

Since one property(i.e. coarseness, orientation, regularity, granularity etc.) of ultrasound liver images was not sufficiently enough to classify the characteristics of livers, we used the multi-feature vectors from ultrasound images to diagnose the liver disease. The proposed classifier, which uses the multi-feature vectors and Bayes decision rule, performed well for the classification of normal, fat and cirrhosis liver. In our simulation, we used the Battacharyya distance and Hotelling Trace Criterion to select the best multi-feature vectors for the classifier and obtained less classification errors than other methods using single feature vector.