Development of novel deep multimodal representation learning-based model for the differentiation of liver tumors on B-mode ultrasound images.

BACKGROUND AND AIM: Recently, multimodal representation learning for images and other information such as numbers or language has gained much attention. The aim of the current study was to analyze the diagnostic performance of deep multimodal representation model-based integration of tumor image, patient background, and blood biomarkers for the differentiation of liver tumors observed using B-mode ultrasonography (US). METHOD: First, we applied supervised learning with a convolutional neural network (CNN) to 972 liver nodules in the training and development sets to develop a predictive model using segmented B-mode tumor images. Additionally, we also applied a deep multimodal representation model to integrate information about patient background or blood biomarkers to B-mode images. We then investigated the performance of the models in an independent test set of 108 liver nodules. RESULTS: Using only the segmented B-mode images, the diagnostic accuracy and area under the curve (AUC) values were 68.52% and 0.721, respectively. As the information about patient background and blood biomarkers was integrated, the diagnostic performance increased in a stepwise manner. The diagnostic accuracy and AUC value of the multimodal DL model (which integrated B-mode tumor image, patient age, sex, aspartate aminotransferase, alanine aminotransferase, platelet count, and albumin data) reached 96.30% and 0.994, respectively. CONCLUSION: Integration of patient background and blood biomarkers in addition to US image using multimodal representation learning outperformed the CNN model using US images. We expect that the deep multimodal representation model could be a feasible and acceptable tool for the definitive diagnosis of liver tumors using B-mode US.

Identification of liver fibrosis using the hepatic vein waveform in patients with Fontan circulation.

AIM: Liver fibrosis caused by congestive hepatopathy has emerged as an important complication after Fontan procedure. We evaluated the utility of the hepatic vein (HV) waveform using Doppler ultrasound for identification of liver fibrosis in Fontan patients. METHODS: We investigated the HV waveforms in 41 Fontan patients and assessed correlations with clinical parameters, liver fibrosis markers, and hemodynamic data. RESULTS: Based on our preliminary analysis of 64 adult patients with chronic liver disease who underwent liver biopsy, we classified HV waveforms into five types with reference to the degree of flattening (from type 1, normal triphasic waveform; to type 5, a monophasic waveform indicating cirrhosis), and confirmed a significant correlation between waveform pattern and fibrosis stage. Notably, we detected HV waveforms in all of the Fontan patients and classified them into five types. The HV waveform pattern positively correlated with γ-glutamyl transferase and hyaluronic acid levels, and negatively correlated with albumin level and platelet count, but did not correlate with central venous pressure or brain natriuretic peptide level, suggesting that HV waveform could reflect pathophysiological changes in the liver without being affected by hepatic congestion. The highest area under the receiver operating characteristic curve of the HV waveform for detecting advanced liver fibrosis, as defined by ultrasonic findings and clinical features, was 0.829 (81.8% sensitivity, 73.3% specificity), which was higher than that of other non-invasive fibrosis markers. CONCLUSIONS: Hepatic vein waveforms change in accordance with liver fibrosis progression in Fontan patients, and can be a useful indicator of liver fibrosis after the Fontan procedure.

Liver stiffness measurements in chronic hepatitis C: Treatment evaluation and risk assessment.

BACKGROUND AND AIM: Liver stiffness (LS), measured by transient elastography, has been validated as a non-invasive surrogate for liver fibrosis. METHODS: We investigated the long-term predictive ability of LS for hepatocellular carcinoma (HCC) development and overall survival in 1146 patients with chronic hepatitis C by using LS value at enrollment. We also investigated chronological changes in LS based on antiviral therapy and its outcome in 752 patients. RESULTS: During the mean follow-up period of 6.6 years, 190 patients developed HCC. Cumulative HCC incidence rates at 5 years were clearly stratified as 1.7% in the ≤ 5 kPa, 3.3% in 5.1-10 kPa, 16.7% in 10.1-15 kPa, 24.4% in 15.1-20 kPa, 36.3% in 20.1-25 kPa, and 43.7% in > 25 kPa subgroups (P < 0.001). Overall survival was also stratified: 10-year survival rates were 99.3% in the ≤ 5 kPa, 95.4% in 5.1-10 kPa, 81.4% in 10.1-15 kPa, 79.5% in 15.1-20 kPa, 66.1% in 20.1-25 kPa, and 49.1% in > 25 kPa subgroups (P < 0.001). LS decreased at a rate of 8.1% per year in those who achieved sustained virological responses, but increased at 0.1% per year in those who could not achieve sustained virological response instead of antiviral therapy, and increased at 3.7% per year in those who did not undergo antiviral therapy. CONCLUSIONS: Liver stiffness measurements can be useful in the prediction of HCC development and overall survival and in the evaluation of chronological changes in liver fibrosis grade during and after antiviral therapy.

Potential associations between perihepatic lymph node enlargement and liver fibrosis, hepatocellular injury or hepatocarcinogenesis in chronic hepatitis B virus infection.

AIM: Although perihepatic lymph node enlargement (PLNE) is frequently observed in chronic liver disease, little is known about PLNE in chronic hepatitis B virus (HBV) infection. We aimed to evaluate this issue. METHODS: We originally enrolled a consecutive 502 patients with chronic HBV infection. Among them, 288 patients without history of interferon-based or nucleoside analog treatment and hepatocellular carcinoma (HCC) were primarily analyzed. RESULTS: PLNE was detected in 27 of 288 (9.4%) patients, which was fewer than that in chronic hepatitis C patients but more than that in subjects undertaking a general health examination as previously reported. The presence of PLNE was significantly associated with a higher probability of having an aspartate aminotransferase (AST) platelet ratio index of more than 1.5 (11.1% vs 1.5%, P = 0.01), a higher AST level (38.0 vs 26.8 U/L, P = 0.001), a higher alanine aminotransferase level (50.1 vs 28.0 U/L, P < 0.0001), and a lower platelet count (18.6 vs 20.6 × 10(4) /µL, P = 0.048) after adjustment for sex and age. However, in our original sample (n = 502), PLNE was observed in 1.4% of the patients with HCC and/or its history whereas 9.2% of the patients without HCC, and the proportion was significantly lower in patients with HCC and/or its history (P = 0.03). CONCLUSION: PLNE was associated with liver fibrosis and hepatocellular injury, but was negatively associated with HCC in chronic HBV infection.

Diagnostic performance of two-dimensional shear wave elastography and attenuation imaging for fibrosis and steatosis assessment in chronic liver disease.

PURPOSE: We investigated the diagnostic performance of two-dimensional shear wave elastography (2D-SWE) and attenuation imaging (ATI) in detecting fibrosis and steatosis in patients with chronic liver disease (CLD), comparing them with established methods. METHODS: In 190 patients with CLD, 2D-SWE and vibration-controlled transient elastography (VCTE) were used for liver stiffness measurement (LSM), and ATI and controlled attenuation parameter (CAP) were used for steatosis quantification. The correlations between these new and established methods were analyzed. RESULTS: Significant correlations were found between 2D-SWE and VCTE (r = 0.78, P < 0.001), and between ATI and CAP (r = 0.70, P < 0.001). Liver stiffness tended to be lower with 2D-SWE compared with that with VCTE, especially in cases with higher LSM, and ATI was less influenced by skin-capsular distance than CAP. Area under the receiver-operating characteristics curves (AUCs) and optimal cut-offs of 2D-SWE for diagnosing liver fibrosis stages F2, F3, and F4 were 0.73 (8.7 kPa), 0.79 (9.1 kPa), and 0.88 (11.6 kPa), respectively. The AUCs and optimal cut-offs of ATI for diagnosing hepatic steatosis grades S1, S2, and S3 were 0.91 (0.66 dB/cm/MHz), 0.80 (0.79 dB/cm/MHz), and 0.88 (0.86 dB/cm/MHz), respectively. A subgroup analysis of 86 patients with metabolic dysfunction-associated steatotic liver disease also demonstrated good performance for 2D-SWE and ATI. CONCLUSION: 2D-SWE and ATI performed comparably with conventional VCTE and CAP in evaluating CLD, offering reliable alternatives for diagnosing liver fibrosis and steatosis.

Perihepatic lymph node enlargement observed at a general health examination: A cross-sectional study.

AIM: Although perihepatic lymph node enlargement (PLNE) is known as a common finding in chronic liver disease, it can be found occasionally at a general health examination. We aimed to clarify the clinical significance of PLNE in general. METHODS: Between January 2008 and December 2011, 4234 subjects were enrolled, who underwent a general health examination at the University of Tokyo Hospital. RESULTS: PLNE was observed in 69 (1.6%) subjects, among whom 17 (0.4%) had liver disorders and 13 (0.3%) had malignancy, one of whom had both. No disorders were determined in the remaining 40 subjects (0.9%). Among 17 subjects with liver disorder-associated PLNE, anti-hepatitis C virus (HCV) antibody was determined in 11 and serum alanine aminotransferase levels were less than 40 U/L in eight. Among 13 subjects with malignancy-associated PLNE, para-aortic lymph nodes were also enlarged in eight. Among 40 subjects with PLNE of unknown etiology, 27 could be followed up for the mean period of 2.08 years, where no underlying disorders were newly determined with largely unaltered size of PLNE. CONCLUSION: The incidence of PLNE in the general population may vary with the prevalence of chronic liver disease, especially HCV infection. When PLNE is observed, liver disorders should be first surveyed including HCV infection even with normal serum alanine aminotransferase levels. PLNE with para-aortic lymph node enlargement may be suggestive of a malignant lesion. The incidence of PLNE of unknown etiology may be approximately 1% in the general population, which may be just followed up without further change.

Perihepatic lymph node enlargement is a negative predictor for sustained responses to pegylated interferon-α and ribavirin therapy for Japanese patients infected with hepatitis C virus genotype 1.

AIM: Although perihepatic lymph node enlargement (PLNE) is reportedly associated with the negative outcome of interferon therapy for chronic hepatitis C, there were limitations in that the results were obtained in patients with various genotypes, viral loads and treatment regimens. We aimed to precisely clarify the significance of PLNE in interferon therapy for chronic hepatitis C. METHODS: Between December 2004 and June 2005, 112 patients with hepatitis C virus (HCV) genotype 1 and HCV RNA of more than 100 KIU/mL were enrolled, who underwent pegylated interferon-α plus ribavirin therapy thereafter. PLNE was defined as a perihepatic lymph node of more than 1 cm in the longest axis by ultrasonography. RESULTS: The sustained virological response (SVR) rate was lower in patients with PLNE (4/22, 18.2%) than in those without (37/90, 41.1%; P = 0.045) and viral load decline was smaller in patients with PLNE than in those without (P = 0.028). The proportion of PLNE positive patients was the smallest in the SVR group (P = 0.033) among the patient groups divided by the treatment outcome. PLNE was retained as a negative predictor for SVR by multivariate logistic regression analysis (P = 0.012). Furthermore, PLNE was not significantly associated with the mutations at HCV core protein and at interferon sensitivity-determining region, or interleukin-28B polymorphism in 45 patients with HCV genotype 1, enrolled between December 2011 and March 2012. CONCLUSION: PLNE is a negative predictor for SVR in patients with HCV genotype 1 and HCV RNA of more than 100 KIU/mL treated with pegylated interferon-α plus ribavirin, independent of other known predictors for SVR.

A Novel Non-invasive Method for Predicting Liver Fibrosis by Quantifying the Hepatic Vein Waveform.

The hepatic vein (HV) waveform by Doppler ultrasound reflects the severity of liver fibrosis. We conducted a proof-of-concept study of a new method for quantifying the HV waveform. We calculated the coefficient of variation (CV) of the HV flow velocity and created a new index "q-HV" (quantified HV) and analyzed its performance for predicting histologic liver fibrosis in 114 patients with chronic liver disease. The CV of the HV flow velocity was well associated with flattening of the waveform and the q-HV significantly increased with the progression of liver fibrosis. The areas under the curve for the prediction of fibrosis stage were 0.732 for F2, 0.772 for F3 and 0.805 for F4. Combined q-HV and FIB-4 index (widely used liver fibrosis score) increased the diagnostic accuracy for liver fibrosis. The q-HV showed good accuracy for predicting liver fibrosis; thus, q-HV is feasible and acceptable as a non-invasive tool for predicting liver fibrosis.

Eradication of hepatitis C virus is associated with the attenuation of steatosis as evaluated using a controlled attenuation parameter.

Chronic hepatitis C virus (HCV) infection was shown to cause hepatic steatosis or suppression of serum lipid levels. However, little is known about the changes in hepatic steatosis following HCV eradication. We aimed to evaluate this issue using the controlled attenuation parameter (CAP), which was recently shown to provide a standardized non-invasive measure of hepatic steatosis. We enrolled 70 patients with chronic HCV infections and steatosis (CAP of over 248 dB/m) who had achieved a sustained viral response at 12 weeks after discontinuation of antiviral treatment using direct-acting antivirals (DAA). We then evaluated the state of hepatic steatosis before and after HCV eradication. We also investigated the changes in serum parameters such as cholesterol and glucose levels. The median value of CAP level decreased significantly after HCV eradication from 273 dB/m to 265 dB/m (P = 0.034). Also, LDL and HDL cholesterol levels increased significantly after HCV eradication (P = 0.002 and P = 0.027, respectively). In conclusion, a decrease in hepatic steatosis after HCV eradication with DAA was revealed in chronic hepatitis C patients with significant steatosis. Cancellation of the viral effect is a possible underlying cause of hepatic steatosis improvement and increase in HDL and LDL cholesterol levels.

Ultrasound evaluation of liver stiffness: accuracy of ultrasound imaging for the prediction of liver cirrhosis as evaluated using a liver stiffness measurement.

BACKGROUND AND AIMS: Because of the low penetration rate of transient elastography (TE) or its limitations in patients with obesity, narrow intercostal spaces, or ascites, the physical appearance of the liver as visualized using ultrasonography (US) is still thought to provide important information for the prediction of liver fibrosis. We examined the accuracy of various US signs when assessing the presence of liver cirrhosis, compared with TE. METHODS: We enrolled 189 patients who had undergone both conventional US and TE examinations. We then assessed the associations between US parameters of the liver (surface, edge, and parenchymal texture) or the US score (sum of each parameter score), and the presence of liver cirrhosis as determined based on a liver stiffness measurement (LSM) of >15. RESULTS: A significant increase in the LSM was observed according to the liver surface score (P < 0.001), liver edge score (P < 0.001), parenchymal texture score (P < 0.001), and US score (P < 0.001). The areas under the curves (AUROC) for the prediction of an LSM >15 for the liver surface, liver edge, parenchymal texture, and the US score were 0.859, 0.768, 0.837, and 0.902, respectively. The AUROC of the US score was higher than that of the APRI score (0.823) or the FIB-4 index (0.804). Using an optimal cut-off value of 3.5, the sensitivity and specificity of the US score were 0.815 and 0.858, respectively. CONCLUSIONS: The US score was clinically useful for the diagnosis of an LSM >15. The US score can be used as a substitute for TE data in patients with obesity, narrow intercostal spaces, or ascites or in hospitals where TE is unavailable.

Pancreatic cysts in general population on ultrasonography: Prevalence and development of risk score.

BACKGROUND: Pancreatic cysts are related to the presence of ductal adenocarcinomas elsewhere in the pancreas, and are also associated with an increased risk of pancreatic adenocarcinoma in the future. Most of the previous studies that investigated the prevalence of pancreatic cysts focused on patients within a hospital or out-patient setting, which may not be representative of the general population. We investigated the prevalence and predictive factors for the presence of pancreatic cysts within a large number of subjects via general health examination. METHODS: Between December 2007 and December 2013, a total of 5198 subjects were enrolled that underwent ultrasonography (US) on general health examination. We established a scoring system for predicting the presence of one or more pancreatic cysts using a split-sample method. RESULTS: Among the enrolled subjects, the prevalence of a pancreatic cyst was 3.5 %. In multivariate analysis, the prevalence was significantly increased with older age, female sex, and the presence of gall bladder adenomyomatosis (GB-ADM). Based on multivariate analysis in the training sample (n = 2,599), we established the scoring system consisting of age, sex, and the presence of GB-ADM to predict the presence of pancreatic cysts. This scoring system was validated in the testing sample (n = 2,599) and produced an area under the curve of 0.711. CONCLUSIONS: The prevalence of pancreatic cyst detected by US was 3.5 % in the general population, and increased with age, female sex, and the presence of GB-ADM. A new scoring system developed in the present study may help to identify better candidates for further examination when the pancreas is not visible by US.

Perihepatic lymph node enlargement is a negative predictor of liver cancer development in chronic hepatitis C patients.

BACKGROUND: Perihepatic lymph node enlargement (PLNE) is a common ultrasound finding in chronic hepatitis C patients. Although PLNE is considered to reflect the inflammatory response to hepatitis C virus (HCV), its clinical significance remains unclear. METHODS: Between December 2004 and June 2005, we enrolled 846 chronic hepatitis C patients in whom adequate ultrasound examinations had been performed. PLNE was defined as a perihepatic lymph node that was at least 1 cm in the longest axis by ultrasonography. We analyzed the clinical features of patients with PLNE and prospectively investigated the association between PLNE and hepatocellular carcinoma (HCC) development. RESULTS: We detected PLNE in 169 (20.0%) patients. Female sex, lower body mass index (BMI), and HCV serotype 1 were independently associated with the presence of PLNE. However, there were no significant differences in liver function tests, liver stiffness, and hepatitis C viral loads between patients with and without PLNE. During the follow-up period (mean 4.8 years), HCC developed in 121 patients. Unexpectedly, patients with PLNE revealed a significantly lower risk of HCC development than those without PLNE (p = 0.019, log rank test). Multivariate analysis revealed that the presence of PLNE was an independent negative predictor of HCC development (hazard ratio 0.551, p = 0.042). In addition, the sustained viral response rate in patients who received interferon (IFN) therapy was significantly lower in patients with PLNE than in patients without PLNE. CONCLUSIONS: Patients with PLNE had a lower risk of HCC development than those without PLNE. This study may provide new insights into daily clinical practice and the pathophysiology of HCV-induced hepatitis and hepatocarcinogenesis.

Autotaxin as a novel serum marker of liver fibrosis.

BACKGROUND: The clinical significance of autotaxin (ATX), a key enzyme for the production of the bioactive lysophospholipid lysophosphatidic acid remains unknown. Serum ATX enzymatic activity reportedly increases in parallel with liver fibrosis and exhibits a gender difference. METHODS: Serum ATX antigen level, measured easier than the activity, was evaluated as a marker of liver fibrosis in 2 cohorts of chronic liver disease caused by hepatitis C virus. RESULTS: In the first cohort, serum ATX level correlated significantly with liver fibrosis stage and was the best parameter for prediction of cirrhosis with an area under the receiver operating characteristic curve (AUROC) of 0.756 in male and 0.760 in female, when compared with serum hyaluronic acid and aminotransferase-to-platelet ratio index, an established marker of liver fibrosis. In another cohort, serum ATX level correlated significantly with liver stiffness, a novel reliable marker of liver fibrosis, being the second-best parameter in male (AUROC, 0.799) and in female (AUROC, 0.876) for prediction of significant fibrosis, and the best parameter in male (AUROC, 0.863) and the third-best parameter in female (AUROC, 0.872) for prediction of cirrhosis, both of which were judged by liver stiffness. CONCLUSIONS: Serum ATX level may be a novel marker of liver fibrosis.

Thrombocytopenia is more severe in patients with advanced chronic hepatitis C than B with the same grade of liver stiffness and splenomegaly.

BACKGROUND AND AIM: The mechanism responsible for thrombocytopenia in chronic liver diseases (CLD) is not yet fully understood. The prevalence of thrombocytopenia has been reported to be higher in patients with hepatitis C virus-related hepatocellular carcinoma (CLD-C) than in those with hepatitis B virus-related hepatocellular carcinoma (CDC-B). We have examined the potential difference in thrombocytopenia between patients with CLD-B and those with CLD-C in terms of liver fibrosis adjustment and splenomegaly. METHODS: The study cohort consisted of 102 patients with CLD-B and 143 patients with CLD-C were enrolled. Liver stiffness, which is reported to be well correlated with the degree of liver fibrosis, was measured by transient elastography. RESULTS: The analysis of covariance with liver stiffness as a covariate revealed that the platelet count was lower in CLD-C patients than in CLD-B patients. Following stratification for liver stiffness, thrombocytopenia was found to be more severe in CLD-C patients than CLD-B patients with advanced liver stiffness, whereas the degree of splenomegaly was not significantly different. The plasma thrombopoietin level was not different between CLD-B and CLD-C patients with advanced liver stiffness, and the immature platelet number was lower in CLD-C patients despite thrombocytopenia being more severe in these patients. CONCLUSIONS: CLD-C patients with advanced liver stiffness presented with more severe levels of thrombocytopenia than CLD-B patients even with the same grade of splenomegaly. Impaired platelet production rather than enhanced platelet destruction may underlie the mechanism responsible for thrombocytopenia in patients with CLD.

Ultrastructural localization of alpha-galactose-containing glycoconjugates in the rat vomeronasal organ.

Binding sites of Griffonia simplicifolia I-B4 isolectin (GS-I-B4), which recognizes terminal alpha-galactose residues of glycoconjugates, were examined in the juxtaluminal region of the rat vomeronasal sensory epithelium and its associated glands of the vomeronasal organ, using a lectin cytochemical technique. Lowicryl K4M-embedded ultra-thin sections, which were treated successively with biotinylated GS-I-B4 and streptavidin-conjugated 10 nm colloidal gold particles, were observed under a transmission electron microscope. Colloidal gold particles, which reflect the presence of terminal alpha-galactose-containing glycoconjugates, were present in vomeronasal receptor neurons in the sensory epithelium and secretory granules of acinar cells of associated glands of the epithelium. Quantitative analysis demonstrated that the density of colloidal gold particles associated with sensory cell microvilli that projected from dendritic endings of vomeronasal neurons was considerably higher than that of microvilli that projected from neighboring sustentacular cells. The same was true for the apical cytoplasms of these cells just below the microvilli. These results suggest that of the sensory microvilli and dendritic endings contained a much larger amount of the alpha-galactose-containing glycoconjugates, compared with those in sustentacular microvilli. Further, biochemical analyses demonstrated several vomeronasal organ-specific glycoproteins with terminal alpha-galactose.