Computer-Aided Diagnosis and Quantification of Cirrhotic Livers Based on Morphological Analysis and Machine Learning

Abstract

It is widely known that morphological changes of the liver and the spleen occur during the clinical course of chronic liver diseases. In this paper, we proposed a morphological analysis method based on statistical shape models (SSMs) of the liver and spleen for computer-aided diagnosis and quantification of the chronic liver. We constructed not only the liver SSM but also the spleen SSM and a joint SSM of the liver and the spleen for a morphologic analysis of the cirrhotic liver in CT images. The effective modes are selected based on both its accumulation contribution rate and its correlation with doctor's opinions (stage labels). We then learn a mapping function between the selected mode and the stage of chronic liver. The mapping function was used for diagnosis and staging of chronic liver diseases.

Sparse and Low-Rank Matrix Decomposition for Local Morphological Analysis to Diagnose Cirrhosis

“…Although the morphologic changes in the liver can be detected by computed tomography (CT), visual assessment is subjective and limited in its ability to detect small changes. In our previous work [8], we proposed a statistical shape model (SSM) that used principal component analysis (PCA) to analyze the morphology of the liver and then selected suitable features to be used to diagnose cirrhosis. Machine learning techniques such as SVM and support vector regression (SVR) are used for classification or for estimation of the stage.…”

Computer-Aided Diagnosis of Liver Cirrhosis Based on Multiple Statistical Shape Models

“…To improve classificat ion accuracy, we use a non-linear SVM [19] as a classifier instead of our previous simple linear classifier [13] and NN [15]. In our classification experiments, two classifiers, nonlinear SVM and our previous NN, and two feature selection methods, the proposed FDA-based mode selection and conventional ACR-based mode selection, are used.…”

“…In our previous study, we constructed a SSM of the liver and proved the potential application of SSMs for the classification of normal and cirrhosis livers by using a simp le linear classifier [12,13]. We also constructed multip le SSMs (i.e., liver SSM, spleen SSM and joint SSM of the liver and the spleen) for morphological analysis [14,15]; this study is based on the fact that the chronic liver diseases and liver cirrhosis will also cause significant morphological changes in the spleen [16].…”

“…As an improvement on our previous study, we propose to use fisher discriminant analysis to select effective modes as features for the classification of chronic liver d iseases. We also use a non-linear SVM for classification instead of the previously used simp le linear classifier [13] and nearest neighbour (NN) [15]. Our experiments have shown that both the mode selection method and the non-linear SVM classifier significantly improved classification accuracy.…”

Quantitative Detection of Cirrhosis: Towards the Development of Computer-Assisted Detection Method

“…Chen et al [11] recently published on a statistical shape model (SSM) to diagnose and quantify cirrhotic livers from CT scans based on morphological analysis and machine learning, using purely changes in liver and spleen shape, and reported classification approaches of 88 and 90 % for normal and abnormal livers, respectively. This approach required performing the segmentation of the liver and the spleen under the guidance of a physician in order to obtain accurate shapes.…”