Data mining framework for fatty liver disease classification in ultrasound: a hybrid feature extraction paradigm

PURPOSE: Fatty liver disease (FLD) is an increasing prevalent disease that can be reversed if detected early. Ultrasound is the safest and ubiquitous method for identifying FLD. Since expert sonographers are required to accurately interpret the liver ultrasound images, lack of the same will result in interobserver variability. For more objective interpretation, high accuracy, and quick second opinions, computer aided diagnostic (CAD) techniques may be exploited. The purpose of this work is to develop one such CAD technique for accurate classification of normal livers and abnormal livers affected by FLD. METHODS: In this paper, the authors present a CAD technique (called Symtosis) that uses a novel combination of significant features based on the texture, wavelet transform, and higher order spectra of the liver ultrasound images in various supervised learning-based classifiers in order to determine parameters that classify normal and FLD-affected abnormal livers. RESULTS: On evaluating the proposed technique on a database of 58 abnormal and 42 normal liver ultrasound images, the authors were able to achieve a high classification accuracy of 93.3% using the decision tree classifier. CONCLUSIONS: This high accuracy added to the completely automated classification procedure makes the authors' proposed technique highly suitable for clinical deployment and usage.

The usefulness of ultrasound in the classification of chronic liver disease

Chronic Liver Disease is a progressive, most of the time asymptomatic, and potentially fatal disease. In this paper, a semi-automatic procedure to stage this disease is proposed based on ultrasound liver images, clinical and laboratorial data. In the core of the algorithm two classifiers are used: a k nearest neighbor and a Support Vector Machine, with different kernels. The classifiers were trained with the proposed multi-modal feature set and the results obtained were compared with the laboratorial and clinical feature set. The results showed that using ultrasound based features, in association with laboratorial and clinical features, improve the classification accuracy. The support vector machine, polynomial kernel, outperformed the others classifiers in every class studied. For the Normal class we achieved 100% accuracy, for the chronic hepatitis with cirrhosis 73.08%, for compensated cirrhosis 59.26% and for decompensated cirrhosis 91.67%.

Chronic liver disease staging classification based on ultrasound, clinical and laboratorial data

In this work the identification and diagnosis of various stages of chronic liver disease is addressed. The classification results of a support vector machine, a decision tree and a k-nearest neighbor classifier are compared. Ultrasound image intensity and textural features are jointly used with clinical and laboratorial data in the staging process. The classifiers training is performed by using a population of 97 patients at six different stages of chronic liver disease and a leave-one-out cross-validation strategy. The best results are obtained using the support vector machine with a radial-basis kernel, with 73.20% of overall accuracy. The good performance of the method is a promising indicator that it can be used, in a non invasive way, to provide reliable information about the chronic liver disease staging.

Diffuse liver disease classification from ultrasound surface characterization, clinical and laboratorial data

In this work liver contour is semi-automatically segmented and quantified in order to help the identification and diagnosis of diffuse liver disease. The features extracted from the liver contour are jointly used with clinical and laboratorial data in the staging process. The classification results of a support vector machine, a Bayesian and a k-nearest neighbor classifier are compared. A population of 88 patients at five different stages of diffuse liver disease and a leave-one-out cross-validation strategy are used in the classification process. The best results are obtained using the k-nearest neighbor classifier, with an overall accuracy of 80.68%. The good performance of the proposed method shows a reliable indicator that can improve the information in the staging of diffuse liver disease.

Specific treatment of advanced Schistosomiasis liver disease in man: favourable results

One hundred eighty-four patients with hepatosplenic schistosomiasis mansoni from the northeast of Brazil were studied. All were treated with a single dose of Oxamniquine or Praziquantel, and were observed over 6 to 12 months. Special attention was given to the evolution of severe hepatopathy. Favourable results were obtained, particularly with the compensated hepatosplenic form. Hepatic function showed great improvement. Hepatomegaly and splenomegaly were significantly reduced in size, to a greater or lesser extent, in the great majority of patients. The implications of the results obtained are considered below.

Drug-induced autoimmune liver disease: A diagnostic dilemma of an increasingly reported disease.

The aetiology of autoimmune hepatitis (AIH) is uncertain but the disease can be triggered in susceptible patients by external factors such as viruses or drugs. AIH usually develops in individuals with a genetic background mainly consisting of some risk alleles of the major histocompatibility complex (HLA). Many drugs have been linked to AIH phenotypes, which sometimes persist after drug discontinuation, suggesting that they awaken latent autoimmunity. At least three clinical scenarios have been proposed that refers to drug- induced autoimmune liver disease (DIAILD): AIH with drug-induced liver injury (DILI); drug induced-AIH (DI-AIH); and immune mediated DILI (IM-DILI). In addition, there are instances showing mixed features of DI-AIH and IM-DILI, as well as DILI cases with positive autoantibodies. Histologically distinguishing DILI from AIH remains a challenge. Even more challenging is the differentiation of AIH from DI-AIH mainly relying in histological features; however, a detailed standardised histologic evaluation of large cohorts of AIH and DI-AIH patients would probably render more subtle features that could be of help in the differential diagnosis between both entities. Growing information on the relationship of drugs and AIH is being available, being drugs like statins and biologic agents more frequently involved in cases of DIAILD. In addition, there is some evidence on the fact that patients diagnosed with DIAILD may have had a previous episode of hepatotoxicity. Further collaborative studies in DIAILD will strengthen the knowledge and understanding of this intriguing and complex disorder which might represent different phenotypes across the spectrum of disease.

Coxiella burnetii as a possible cause of autoimmune liver disease: a case report.

INTRODUCTION: Q fever is a zoonotic infection that may cause severe hepatitis. Q-fever hepatitis has not yet been associated with autoimmune hepatitis and/or primary biliary cirrhosis. CASE PRESENTATION: We describe a 39-year-old man of Sri Lankan origin with chronic Q-fever hepatitis who developed autoantibodies compatible with autoimmune hepatitis/primary biliary cirrhosis overlap syndrome. Ursodeoxycholic acid in addition to antibiotic therapy markedly improved hepatic enzyme levels suggesting that autoimmunity, potentially triggered by the underlying infection, was involved in the pathogenesis of liver damage. CONCLUSION: We suggest that Coxiella burnetii might trigger autoimmune liver disease. Patients with Q-fever hepatitis who respond poorly to antibiotics should be investigated for serological evidence of autoimmune hepatitis, primary biliary cirrhosis or overlap syndrome, as these patients could benefit from adjunctive therapy with ursodeoxycholic acid. Conversely, C. burnetii serology might be necessary in patients with autoimmune liver disease in order to exclude underlying Coxiella infection.

Role of dietary carbohydrates and macronutrients in the pathogenesis of nonalcoholic fatty liver disease.

PURPOSE OF REVIEW: The prevalence of nonalcoholic fatty liver disease is increasing worldwide and there is strong evidence that dietary factors play a role in its pathogenesis. The present review aims to provide a better understanding of how carbohydrates and other macronutrients may affect the disease. RECENT FINDINGS: The effects of carbohydrates on the development of nonalcoholic fatty liver disease differ depending upon the carbohydrate type; high-glycemic index foods are related to increased hepatic fat in both rodents and humans. Similarly, simple carbohydrates, such as fructose, stimulate hepatic de-novo lipogenesis and decrease lipid oxidation, thus leading to increased fat deposition. The underlying mechanisms may involve the activation of transcription factors. Fat intake broadly leads to hepatic fat deposition in rodents but few data are available on humans. Both carbohydrates and fat trigger inflammatory factors, which are closely related to metabolic disorders and nonalcoholic fatty liver disease. Lifestyle interventions appear to be the most appropriate first-line treatment for nonalcoholic fatty liver disease. SUMMARY: There is strong evidence that the diet may affect the development of nonalcoholic fatty liver disease. Although simple carbohydrates are clearly shown to have deleterious effects in humans, the role of fat remains controversial. Further studies will be required to evaluate the effects of macronutrient composition on the development of nonalcoholic fatty liver disease.

Major Histocompatibility Class II Pathway Is Not Required for the Development of Nonalcoholic Fatty Liver Disease in Mice.

Single-nucleotide polymorphisms within major histocompatibility class II (MHC II) genes have been associated with an increased risk of drug-induced liver injury. However, it has never been addressed whether the MHC II pathway plays an important role in the development of nonalcoholic fatty liver disease, the most common form of liver disease. We used a mouse model that has a complete knockdown of genes in the MHC II pathway (MHCII(Δ/Δ)). Firstly we studied the effect of high-fat diet-induced hepatic inflammation in these mice. Secondly we studied the development of carbon-tetra-chloride- (CCl4-) induced hepatic cirrhosis. After the high-fat diet, both groups developed obesity and hepatic steatosis with a similar degree of hepatic inflammation, suggesting no impact of the knockdown of MHC II on high-fat diet-induced inflammation in mice. In the second study, we confirmed that the CCl4 injection significantly upregulated the MHC II genes in wild-type mice. The CCl4 treatment significantly induced genes related to the fibrosis formation in wild-type mice, whereas this was lower in MHCII(Δ/Δ) mice. The liver histology, however, showed no detectable difference between groups, suggesting that the MHC II pathway is not required for the development of hepatic fibrosis induced by CCl4.

Genome-wide association analysis identifies variants associated with nonalcoholic fatty liver disease that have distinct effects on metabolic traits.

Nonalcoholic fatty liver disease (NAFLD) clusters in families, but the only known common genetic variants influencing risk are near PNPLA3. We sought to identify additional genetic variants influencing NAFLD using genome-wide association (GWA) analysis of computed tomography (CT) measured hepatic steatosis, a non-invasive measure of NAFLD, in large population based samples. Using variance components methods, we show that CT hepatic steatosis is heritable (∼26%-27%) in family-based Amish, Family Heart, and Framingham Heart Studies (n = 880 to 3,070). By carrying out a fixed-effects meta-analysis of genome-wide association (GWA) results between CT hepatic steatosis and ∼2.4 million imputed or genotyped SNPs in 7,176 individuals from the Old Order Amish, Age, Gene/Environment Susceptibility-Reykjavik study (AGES), Family Heart, and Framingham Heart Studies, we identify variants associated at genome-wide significant levels (p<5×10(-8)) in or near PNPLA3, NCAN, and PPP1R3B. We genotype these and 42 other top CT hepatic steatosis-associated SNPs in 592 subjects with biopsy-proven NAFLD from the NASH Clinical Research Network (NASH CRN). In comparisons with 1,405 healthy controls from the Myocardial Genetics Consortium (MIGen), we observe significant associations with histologic NAFLD at variants in or near NCAN, GCKR, LYPLAL1, and PNPLA3, but not PPP1R3B. Variants at these five loci exhibit distinct patterns of association with serum lipids, as well as glycemic and anthropometric traits. We identify common genetic variants influencing CT-assessed steatosis and risk of NAFLD. Hepatic steatosis associated variants are not uniformly associated with NASH/fibrosis or result in abnormalities in serum lipids or glycemic and anthropometric traits, suggesting genetic heterogeneity in the pathways influencing these traits.

Lymphocyte populations in liver biopsy specimens from patients with chronic liver disease

The characterisation of lymphocytes from liver biopsies indicates that 'activated' T lymphocytes are present in the liver in alcohol induced hepatitis, chronic active hepatitis (HBS+ve and -ve), and in primary biliary cirrhosis but not in inactive cirrhosis, chronic persistent hepatitis, extrahepatic and drug induced cholestasis. A greater percentage of lymphocytes bear Fc-receptors in chronic active hepatitis than in alcohol induced hepatitis or cholestatic liver disease. The concentration of 'activated' T cells in the peripheral blood in all groups studied was within the normal range, suggesting that the 'activated' T cells found in the liver were reacting to either native or foreign antigens within the liver. The data on Fc-receptor bearing cells are consistent with the involvement of antibody assisted K cell mediated cytotoxicity in chronic active hepatitis.

Changes in biomarkers of liver disease during successful combination antiretroviral therapy in HIV-HCV-coinfected individuals.

BACKGROUND: We investigated changes in biomarkers of liver disease in HIV-HCV-coinfected individuals during successful combination antiretroviral therapy (cART) compared to changes in biomarker levels during untreated HIV infection and to HIV-monoinfected individuals. METHODS: Non-invasive biomarkers of liver disease (hyaluronic acid [HYA], aspartate aminotransferase-to-platelet ratio index [APRI], Fibrosis-4 [FIB-4] index and cytokeratin-18 [CK-18]) were correlated with liver histology in 49 HIV-HCV-coinfected patients. Changes in biomarkers over time were then assessed longitudinally in HIV-HCV-coinfected patients during successful cART (n=58), during untreated HIV-infection (n=59), and in HIV-monoinfected individuals (n=17). The median follow-up time was 3.4 years on cART. All analyses were conducted before starting HCV treatment. RESULTS: Non-invasive biomarkers of liver disease correlated significantly with the histological METAVIR stage (P<0.002 for all comparisons). The mean ±sd area under the receiver operating characteristic (AUROC) curve values for advanced fibrosis (≥F3 METAVIR) for HYA, APRI, FIB-4 and CK-18 were 0.86 ±0.05, 0.84 ±0.08, 0.80 ±0.09 and 0.81 ±0.07, respectively. HYA, APRI and CK-18 levels were higher in HIV-HCV-coinfected compared to HIV-monoinfected patients (P<0.01). In the first year on cART, APRI and FIB-4 scores decreased (-35% and -33%, respectively; P=0.1), mainly due to the reversion of HIV-induced thrombocytopaenia, whereas HYA and CK-18 levels remained unchanged. During long-term cART, there were only small changes (<5%) in median biomarker levels. Median biomarker levels changed <3% during untreated HIV-infection. Overall, 3 patients died from end-stage liver disease, and 10 from other causes. CONCLUSIONS: Biomarkers of liver disease highly correlated with fibrosis in HIV-HCV-coinfected individuals and did not change significantly during successful cART. These findings suggest a slower than expected liver disease progression in many HIV-HCV-coinfected individuals, at least during successful cART.

Endocrine causes of nonalcoholic fatty liver disease.

Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease in the industrialized world. The prevalence of NAFLD is increasing, becoming a substantial public health burden. NAFLD includes a broad spectrum of disorders, from simple conditions such as steatosis to severe manifestations such as fibrosis and cirrhosis. The relationship of NAFLD with metabolic alterations such as type 2 diabetes is well described and related to insulin resistance, with NAFLD being recognized as the hepatic manifestation of metabolic syndrome. However, NAFLD may also coincide with endocrine diseases such as polycystic ovary syndrome, hypothyroidism, growth hormone deficiency or hypercortisolism. It is therefore essential to remember, when discovering altered liver enzymes or hepatic steatosis on radiological exams, that endocrine diseases can cause NAFLD. Indeed, the overall prognosis of NAFLD may be modified by treatment of the underlying endocrine pathology. In this review, we will discuss endocrine diseases that can cause NALFD. Underlying pathophysiological mechanisms will be presented and specific treatments will be reviewed.

Metabolic syndrome and nonalcoholic fatty liver disease: Is insulin resistance the link?

Metabolic syndrome (MetS) is a disease composed of different risk factors such as obesity, type 2 diabetes or dyslipidemia. The prevalence of this syndrome is increasing worldwide in parallel with the rise in obesity. Nonalcoholic fatty liver disease (NAFLD) is now the most frequent chronic liver disease in western countries, affecting more than 30% of the general population. NAFLD encompasses a spectrum of liver manifestations ranging from simple steatosis to nonalcoholic steatohepatitis (NASH), fibrosis and cirrhosis, which may ultimately progress to hepatocellular carcinoma. There is accumulating evidence supporting an association between NAFLD and MetS. Indeed, NAFLD is recognized as the liver manifestation of MetS. Insulin resistance is increasingly recognized as a key factor linking MetS and NAFLD. Insulin resistance is associated with excessive fat accumulation in ectopic tissues, such as the liver, and increased circulating free fatty acids, which can further promote inflammation and endoplasmic reticulum stress. This in turn aggravates and maintains the insulin resistant state, constituting a vicious cycle. Importantly, evidence shows that most of the patients developing NAFLD present at least one of the MetS traits. This review will define MetS and NAFLD, provide an overview of the common pathophysiological mechanisms linking MetS and NAFLD, and give a perspective regarding treatment of these ever growing metabolic diseases.