A Lightweight Network for Contextual and Morphological Awareness for Hepatic Vein Segmentation.

Accurate segmentation of the hepatic vein can improve the precision of liver disease diagnosis and treatment. Since the hepatic venous system is a small target and sparsely distributed, with various and diverse morphology, data labeling is difficult. Therefore, automatic hepatic vein segmentation is extremely challenging. We propose a lightweight contextual and morphological awareness network and design a novel morphology aware module based on attention mechanism and a 3D reconstruction module. The morphology aware module can obtain the slice similarity awareness mapping, which can enhance the continuous area of the hepatic veins in two adjacent slices through attention weighting. The 3D reconstruction module connects the 2D encoder and the 3D decoder to obtain the learning ability of 3D context with a very small amount of parameters. Compared with other SOTA methods, using the proposed method demonstrates an enhancement in the dice coefficient with few parameters on the two datasets. A small number of parameters can reduce hardware requirements and potentially have stronger generalization, which is an advantage in clinical deployment.

A Deep Learning Model for Automatic Segmentation of Intraparenchymal and Intraventricular Hemorrhage for Catheter Puncture Path Planning.

Intracerebral hemorrhage is the subtype of stroke with the highest mortality rate, especially when it also causes secondary intraventricular hemorrhage. The optimal surgical option for intracerebral hemorrhage remains one of the most controversial areas of neurosurgery. We aim to develop a deep learning model for the automatic segmentation of intraparenchymal and intraventricular hemorrhage for clinical catheter puncture path planning. First, we develop a 3D U-Net embedded with a multi-scale boundary aware module and a consistency loss for segmenting two types of hematoma in computed tomography images. The multi-scale boundary aware module can improve the model's ability to understand the two types of hematoma boundaries. The consistency loss can reduce the probability of classifying a pixel into two categories at the same time. Since different hematoma volumes and locations have different treatments. We also measure hematoma volume, estimate centroid deviation, and compare with clinical methods. Finally, we plan the puncture path and conduct clinical validation. We collected a total of 351 cases, and the test set contained 103 cases. For intraparenchymal hematomas, the accuracy can reach 96 % when the proposed method is applied for path planning. For intraventricular hematomas, the proposed model's segmentation efficiency and centroid prediction are superior to other comparable models. Experimental results and clinical practice show that the proposed model has potential for clinical application. In addition, our proposed method has no complicated modules and improves efficiency, with generalization ability.

A review of deep learning-based multiple-lesion recognition from medical images: classification, detection and segmentation.

Deep learning-based methods have become the dominant methodology in medical image processing with the advancement of deep learning in natural image classification, detection, and segmentation. Deep learning-based approaches have proven to be quite effective in single lesion recognition and segmentation. Multiple-lesion recognition is more difficult than single-lesion recognition due to the little variation between lesions or the too wide range of lesions involved. Several studies have recently explored deep learning-based algorithms to solve the multiple-lesion recognition challenge. This paper includes an in-depth overview and analysis of deep learning-based methods for multiple-lesion recognition developed in recent years, including multiple-lesion recognition in diverse body areas and recognition of whole-body multiple diseases. We discuss the challenges that still persist in the multiple-lesion recognition tasks by critically assessing these efforts. Finally, we outline existing problems and potential future research areas, with the hope that this review will help researchers in developing future approaches that will drive additional advances.

MRLA-Net: A tumor segmentation network embedded with a multiple receptive-field lesion attention module in PET-CT images.

The tumor image segmentation is an important basis for doctors to diagnose and formulate treatment planning. PET-CT is an extremely important technology for recognizing the systemic situation of diseases due to the complementary advantages of their respective modal information. However, current PET-CT tumor segmentation methods generally focus on the fusion of PET and CT features. The fusion of features will weaken the characteristics of the modality itself. Therefore, enhancing the modal features of the lesions can obtain optimized feature sets, which is extremely necessary to improve the segmentation results. This paper proposed an attention module that integrates the PET-CT diagnostic visual field and the modality characteristics of the lesion, that is, the multiple receptive-field lesion attention module. This paper made full use of the spatial domain, frequency domain, and channel attention, and proposed a large receptive-field lesion localization module and a small receptive-field lesion enhancement module, which together constitute the multiple receptive-field lesion attention module. In addition, a network embedded with a multiple receptive-field lesion attention module has been proposed for tumor segmentation. This paper conducted experiments on a private liver tumor dataset as well as two publicly available datasets, the soft tissue sarcoma dataset, and the head and neck tumor segmentation dataset. The experimental results showed that the proposed method achieves excellent performance on multiple datasets, and has a significant improvement compared with DenseUNet, and the tumor segmentation results on the above three PET/CT datasets were improved by 7.25%, 6.5%, 5.29% in Dice per case. Compared with the latest PET-CT liver tumor segmentation research, the proposed method improves by 8.32%.

SDA-UNet: a hepatic vein segmentation network based on the spatial distribution and density awareness of blood vessels.

Objective.Hepatic vein segmentation is a fundamental task for liver diagnosis and surgical navigation planning. Unlike other organs, the liver is the only organ with two sets of venous systems. Meanwhile, the segmentation target distribution in the hepatic vein scene is extremely unbalanced. The hepatic veins occupy a small area in abdominal CT slices. The morphology of each person's hepatic vein is different, which also makes segmentation difficult. The purpose of this study is to develop an automated hepatic vein segmentation model that guides clinical diagnosis.Approach.We introduce the 3D spatial distribution and density awareness (SDA) of hepatic veins and propose an automatic segmentation network based on 3D U-Net which includes a multi-axial squeeze and excitation module (MASE) and a distribution correction module (DCM). The MASE restrict the activation area to the area with hepatic veins. The DCM improves the awareness of the sparse spatial distribution of the hepatic veins. To obtain global axial information and spatial information at the same time, we study the effect of different training strategies on hepatic vein segmentation. Our method was evaluated by a public dataset and a private dataset. The Dice coefficient achieves 71.37% and 69.58%, improving 3.60% and 3.30% compared to the other SOTA models, respectively. Furthermore, metrics based on distance and volume also show the superiority of our method.Significance.The proposed method greatly reduced false positive areas and improved the segmentation performance of the hepatic vein in CT images. It will assist doctors in making accurate diagnoses and surgical navigation planning.

The effectiveness of zinc supplementation in men with isolated hypogonadotropic hypogonadism.

A multicenter, open-label, randomized, controlled superiority trial with 18 months of follow-up was conducted to investigate whether oral zinc supplementation could further promote spermatogenesis in males with isolated hypogonadotropic hypogonadism (IHH) receiving sequential purified urinary follicular-stimulating hormone/human chorionic gonadotropin (uFSH/hCG) replacement. Sixty-seven Chinese male IHH patients were recruited from the Departments of Endocrinology in eight tertiary hospitals and randomly allocated into the sequential uFSH/hCG group (Group A, n = 34) or the sequential uFSH plus zinc supplementation group (Group B, n = 33). In Group A, patients received sequential uFSH (75 U, three times a week every other 3 months) and hCG (2000 U, twice a week) treatments. In Group B, patients received oral zinc supplementation (40 mg day-1 ) in addition to the sequential uFSH/hCG treatment given to patients in Group A. The primary outcome was the proportion of patients with a sperm concentration ≥1.0 × 106 ml-1 during the 18 months. The comparison of efficacy between Groups A and B was analyzed. Nineteen of 34 (55.9%) patients receiving sequential uFSH/hCG and 20 of 33 (60.6%) patients receiving sequential uFSH/hCG plus zinc supplementation achieved sperm concentrations ≥1.0 × 106 ml-1 by intention to treat analyses. No differences between Group A and Group B were observed as far as the efficacy of inducing spermatogenesis (P = 0.69). We concluded that the sequential uFSH/hCG plus zinc supplementation regimen had a similar efficacy to the sequential uFSH/hCG treatment alone. The additional improvement of 40 mg day-1 oral zinc supplementation on spermatogenesis and masculinization in male IHH patients is very subtle.

Comparative efficacy of anti-diabetic agents on nonalcoholic fatty liver disease in patients with type 2 diabetes mellitus: a systematic review and meta-analysis of randomized and non-randomized studies.

BACKGROUND: Nonalcoholic fatty liver disease (NAFLD) has a high prevalence in patients with type 2 diabetes mellitus (T2DM). In this study, we sought to provide a comprehensive assessment regarding the effects of anti-diabetic agents on NAFLD in patients with T2DM. METHODS: MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials were searched for randomized controlled trials (RCTs) with different anti-diabetic agents in T2DM. Observational trials were also recruited to expand our population. Hepatic fat content and liver histology were evaluated as primary outcomes. Pooled estimates were calculated using a fixed effect model. RESULTS: One thousand one hundred ninety-six participants in 19 RCTs and 14 non-randomized studies were included. Evidence from RCTs and observational studies suggested that greater hepatic fat content reduction and improved liver histology were seen in thiazolidinediones for 12-72 weeks; glucagon-like peptide-1 receptor agonists had beneficial effects on hepatic fat content after 26-50 weeks intervention, and insulin/metformin combination with 3-7 months improved hepatic fat content. Initiating metformin or dapagliflozin showed no benefit on hepatic fat content or liver histology in 16-48 weeks. Besides, nateglinide for 18 months was reported in a small sample-size RCT to improve hepatic fat content and liver histology. Sitagliptin therapy of 1 year also provided benefit on nonalcoholic steatohepatitis score in an observational study. CONCLUSIONS: For T2DM with NAFLD, administrating thiazolidinediones and glucagon-like peptide-1 receptor agonists seems to provide more identified advances in attenuating hepatic fat content. Further RCTs are warranted to assess the efficacy of various hypoglycemic agents on clinical outcomes associated with NAFLD in T2DM. Copyright © 2015 John Wiley & Sons, Ltd.

Effect of exenatide, insulin and pioglitazone on bone metabolism in patients with newly diagnosed type 2 diabetes.

AIM: Preclinical studies suggested that insulin, incretin and thiazolidinediones had effect on regulation of bone metabolism. But clinical evidence is limited. We assessed the effects of these antihyperglycemic agents on bone metabolism in patients with newly diagnosed type 2 diabetes. METHODS: The present study was a two-center, randomized, parallel-group clinical trial. Sixty-two newly diagnosed and drug-naïve patients with type 2 diabetes were randomized to exenatide (EXE, n = 20), mixed protamine zinc recombinant human insulin lispro injection (25R; INS, n = 21) or pioglitazone (PIO, n = 21) group for a 24-week treatment. Glycosylated hemoglobin A1c (HbA1c), body weight, body mineral density (BMD) and fasting serum concentration of bone turnover markers including osteocalcin (OC), C-telopeptide of type I collagen (CTX) and tartrate-resistant alkaline phosphatase 5b (TRAcP5b) were assessed at baseline and week 24. RESULTS: Baseline characteristics were similar among groups. At week 24, HbA1c improved in all patients (EXE:-2.4 ± 0.3 %, INS:-2.4 ± 0.3 %, PIO:-2.0 ± 0.2 %; p > 0.05 among groups). Patients treated with exenatide lost body weight remarkably (-4.7 ± 0.8 kg). In spite of the amelioration of glucose control, no significant improvement of OC, CTX or TRAcP5b was observed at week 24 (EXE: OC -0.619 ± 0.728 ng/ml, CTX 0.147 ± 0.046 ng/ml, TRAcP5b 0.302 ± 0.149 U/L;INS: OC 0.637 ± 0.787 ng/ml, CTX -0.012 ± 0.074 ng/ml, TRAcP5b 0.124 ± 0.395 U/L; PIO: OC -0.150 ± 0.691 ng/ml, CTX 0.073 ± 0.094 ng/ml, TRAcP5b 0.586 ± 0.183 U/L; p > 0.05), as well as BMD measurement, regardless of the treatments. CONCLUSIONS: Twenty-four-week treatment with exenatide, insulin and pioglitazone improved glucose control in patients with newly diagnosed type 2 diabetes, but had no impact on bone turnover markers or BMD.

Sequential Versus Continual Purified Urinary FSH/hCG in Men With Idiopathic Hypogonadotropic Hypogonadism.

CONTEXT: Gonadotropin therapy using a human chorionic gonadotropin (hCG) and FSH preparation is an effective regimen in inducing masculinization and spermatogenesis in men with idiopathic hypogonadotropic hypogonadism (IHH). However, the high cost of medication and frequent injections affect compliance. OBJECTIVE: The aim of this study was to determine the efficacy of sequential use of highly purified urinary FSH (uFSH)/hCG in men with IHH. DESIGN AND SETTING: A randomized, open-label, prospective, controlled noninferiority trial with an 18-month follow-up was conducted in 9 tertiary hospitals. PATIENTS AND INTERVENTION: A total of 67 Chinese men with IHH were randomly allocated into group A receiving continual uFSH (75 U, 3 times a week) and hCG (2000 U, twice a week) injection and group B receiving sequential uFSH (75 U, 3 times a week every other 3 months) and hCG (2000 U, twice a week) injection. MAIN OUTCOME MEASURE: The primary outcome was the proportion of subjects with a sperm concentration of ≥ 1.0 × 10(6)/mL during the 18 months. The efficacy between groups A and B was compared for noninferiority. RESULTS: Of the patients, 17/33 (51.5%) receiving continual uFSH/hCG and 19/34 (55.9%) receiving sequential uFSH/hCG achieved sperm concentrations of ≥ 1.0 × 10(6)/mL. The efficacy in the sequential uFSH/hCG group was not inferior to that in the continual uFSH/hCG group (noninferiority, P = .008) by intention-to-treat analysis. CONCLUSIONS: The efficacy of the sequential uFSH/hCG regimen is not inferior to that of the continual uFSH/hCG regimen in inducing spermatogenesis and masculinization of patients with IHH.

Effects of exenatide, insulin, and pioglitazone on liver fat content and body fat distributions in drug-naive subjects with type 2 diabetes.

Ectopic accumulation of lipids in nonadipose tissues plays a primary role in the pathogenesis of type 2 diabetes mellitus (T2DM). This study was to examine the effects of exenatide, insulin, and pioglitazone on liver fat content and body fat distributions in T2DM. Thirty-three drug-naive T2DM patients (age 52.7 ± 1.7 years, HbA1c 8.7 ± 0.2 %, body mass index 24.5 ± 0.5 kg/m(2)) were randomized into exenatide, insulin, or pioglitazone for 6 months. Intrahepatic fat (IHF), visceral fat (VF), and subcutaneous fat (SF) were measured using proton nuclear magnetic resonance spectroscopy. Plasma tumor necrosis factor α (TNFα) and adiponectin were assayed by ELISA. HbA1c declined significantly in all three groups. Body weight, waist, and serum triglycerides decreased with exenatide. After interventions, IHF significantly reduced with three treatments (exenatide Δ = -68 %, insulin Δ = -58 %, pioglitazone Δ = -49 %). Exenatide reduced VF (Δ = -36 %) and SF (Δ = -13 %), and pioglitazone decreased VF (Δ = -30 %) with no impact on SF, whereas insulin had no impact on VF or SF. Levels of TNFα (exenatide/insulin/pioglitazone) decreased, and levels of adiponectin (exenatide/pioglitazone) increased. Analysis showed that ΔIHF correlated with ΔHbA1c and Δweight. Besides, ΔIHF correlated with Δtriglycerides and ΔTNFα, but the correlations fell short of significance after BMI adjustment. By linear regression analysis, ΔHbA1c alone explained 41.5 % of the variance of ΔIHF, and ΔHbA1c + Δweight explained 57.6 % of the variance. Liver fat content can be significantly reduced irrespective of using exenatide, insulin, and pioglitazone. Early glycaemic control plays an important role in slowing progression of fatty liver in T2DM.

Correlations between A1c, fasting glucose, 2h postload glucose, and ß-cell function in the Chinese population.

This study was aimed to assess the associations of hemoglobin A1c (HbA1c), fasting plasma glucose (FPG), and 2h postload plasma glucose (2hPG) with ß-cell function in the Chinese population. A total of 913 subjects underwent 75-g oral glucose tolerance test (OGTT) and HbA1c testing. According to OGTT, isolated impaired fasting glucose (i-IFG) was defined as 5.6 mmol/l ≤ FPG < 7.0 mmol/l and 2hPG < 7.8 mmol/l; isolated impaired glucose tolerance (i-IGT) was defined as FPG < 5.6 mmol/l and 7.8 mmol/l ≤ 2hPG < 11.1 mmol/l. HbA1c 5.7-6.4 % was used to identify subjects with prediabetes. Insulin release was calculated by basal homeostasis model assessment of insulin secretion (HOMA-ß), early-phase InsAUC30/GluAUC30, and total-phase InsAUC120/GluAUC120. ß-cell function relative to insulin sensitivity was expressed as disposition index (DI). All indices of insulin sensitivity and ß-cell function gradually decreased with increasing HbA1c, FPG, and 2hPG (all p < 0.01). ß-cell function decreased precipitously when HbA1c exceeded 5.5 %. Compared with HbA1c, FPG showed stronger correlations with HOMA-ß, InsAUC30/GluAUC30, InsAUC120/GluAUC120, DI30, and DI120 (all p < 0.05), and 2hPG was more closely related to DI30 and DI120 (all p < 0.01). Moreover, FPG was more strongly related to HOMA-ß and InsAUC30/GluAUC30 than 2hPG (all p < 0.05). The combination of i-IFG and HbA1c 5.7-6.4 % showed the greatest reduction in DI30 and DI120 compared with HbA1c 5.7-6.4 % alone, i-IGT, or i-IFG (p < 0.05). In conclusion, HbA1c could be used as a marker to identify subjects with impaired ß-cell function, but OGTT performs better than HbA1c. The combination of HbA1c and FPG is a simple and sensitive method to evaluate ß-cell function.

Intensive insulin therapy increases sex hormone-binding globulin in newly diagnosed type 2 diabetic patients.

OBJECTIVE: Many studies have shown that low sex hormone-binding globulin (SHBG) is associated with insulin resistance, but only few studies have examined how serum SHBG is regulated by insulin in humans. This interventional study aimed to investigate the effect of insulin therapy (IT) on serum SHBG levels in newly diagnosed type 2 diabetic patients. METHODS: A total of 80 newly diagnosed type 2 diabetic subjects were enrolled and randomly grouped into a 2-week intensive IT with/without metformin. Serum SHBG, total testosterone, glucose, liver enzymes, lipids, insulin, and C-peptide levels were measured before and after IT. RESULTS: Before IT, serum SHBG levels were negatively correlated with BMI, waist circumference (WC), alanine aminotransferase (ALT), gamma-glutamyl transpeptidase (γ-GT), triglyceride (TG), fasting insulin, and C-peptide, and homeostatic model assessment of insulin resistance (HOMA-IR), and positively with HDL-C (all P for trend <0.05), after adjustment for age and sex. IT increased serum SHBG levels from 26.5±14.5 to 33.2±15.0 nmol/l (P<0.001), increased by 25.2% (95% CI, 20.3 to 30.9%, P<0.001). In a multiple linear regression model adjusting for age, sex, BMI, and WC, the decreases in ΔALT (standardized regression coefficient ß=-0.374, P=0.012) and ΔTG (ß=-0.380, P=0.020) were independent contributors to the increase in ΔSHBG. CONCLUSIONS: IT increases serum SHBG likely through improving insulin resistance and liver function.

Decreased beta cell function and insulin sensitivity contributed to increasing fasting glucose in Chinese.

To evaluate the role of insulin resistance and beta cell function to increasing fasting plasma glucose (FPG), 1,272 Chinese subjects (18-80 years of age) were divided into normal glucose tolerance (NGT), impaired fasting glucose (IFG), impaired glucose tolerance (IGT), combined glucose intolerance (CGI), and type 2 diabetes mellitus (T2DM) according to oral glucose tolerance test. Insulin sensitivity was measured by Matsuda index (ISI(M)) and homeostasis model assessment of insulin resistance (1/HOMA-IR); ß-cell function adjusted by insulin sensitivity was assessed from disposition index (DI) at basal DI(0) (homeostasis model assessment of ß-cell function (HOMA-B) × [1/HOMA-IR]), early-phase DI(30) (the ratio of total insulin AUC and total glucose AUC during 0-30 min of the OGTT (InsAUC(30)/GluAUC(30)) × ISI(M)) and total DI(120) (the ratio of total insulin AUC and total glucose AUC during 0-120 min of the OGTT (InsAUC(120)/GluAUC(120)) × ISI(M)). Compared with NGT, in IFG, ISI(M) (-23%), DI(0) (-38%), DI(30) (-30%), and DI(120) (-31%) were decreased significantly. As the FPG increased across categories classified by FPG levels from NGT → IFG → T2DM with 2 h PG < 7.8 mmol/l, ISI(M), DI(0), DI(30) and DI(120) showed decline beginning from normal range of FPG, compared with the reference category of FPG < 4.0 mmol/l. Correlation analysis showed that ISI(M) and DI were correlated inversely with FPG concentration (r = -0.242 for ISI(M), r = -0.933 for DI(0), r = -0.806 for DI(30), r = -0.817 for DI(120); P < 0.001). Both the impairment of beta cell function and insulin sensitivity started at the low point of FPG within the normoglycemic range and contributed to the deterioration of fasting glucose.

Short-term intensive therapy in newly diagnosed type 2 diabetes partially restores both insulin sensitivity and ß-cell function in subjects with long-term remission.

OBJECTIVE: To examine the effect of intensive glycemic control therapy (IT) on insulin sensitivity and ß-cell function in newly diagnosed type 2 diabetic patients compared with subjects with normal glucose tolerance (NGT) and those with impaired glucose tolerance (IGT). RESEARCH DESIGN AND METHODS: Forty-eight newly diagnosed type 2 diabetic patients were randomly assigned to IT for 2 weeks and followed up for 1 year. Intravenous glucose tolerance tests were conducted in NGT, IGT, and diabetic subjects. Blood glucose and insulin were measured before and after IT and at the 1-year follow-up. RESULTS: IT lowered the homeostasis model assessment (HOMA) for insulin resistance (IR) significantly, from 3.12 ± 1.4 (mean ± SD) to 1.72 ± 0.8, a level comparable to the IGT (1.96 ± 1.1) and NGT (1.37 ± 0.6) subjects in the remission group; however, no HOMA-IR improvement was observed in nonremission subjects. HOMA-ß in the remission group was improved (mean, interquartile range) from 18.4 (8.3-28.5) to 44.6 (32.1-69.1) and acute insulin response of insulin (AIRins) from 1.50 ± 0.22 to 1.83 ± 0.19 µIU/mL after IT, but was still significantly lower than those in NGT individuals (HOMA-ß: 86.4 [56.7-185.2], P < 0.01; AIRins: 2.54 ± 0.39 µIU/mL, P < 0.01). After IT and at 1 year, the hyperbolic relationship between HOMA-ß and HOMA sensitivity of remission subjects shifted close to that of IGT subjects. CONCLUSIONS: IT in newly diagnosed type 2 diabetes not only partially restored ß-cell function but also greatly restored insulin sensitivity. Compared with IGT and NGT subjects, ß-cell function was less restored than insulin sensitivity after IT in the remission subjects.

Anti-inflammatory effects of simvastatin on adipokines in type 2 diabetic patients with carotid atherosclerosis.

OBJECTIVE: Statins are extensively used for lowering LDL-cholesterol and reducing cardiovascular events. Recent studies have shown that statins have beneficial anti-inflammatory effects. We aimed to determine whether and how adipokines are regulated during statin treatment in type 2 diabetic patients. METHOD: In this study,we investigated the changes of CRP and inflammation-related adipokines (SAA,IL-6,TNFalpha and adiponectin) in 23 type 2 diabetic patients with atherosclerosis who received statin therapy, and 20 diabetic patients with atherosclerosis and 14 diabetic patients without atherosclerosis who did not receive statin therapy for a period of three months. RESULTS: By the end of the simvastatin treatment (40 mg, daily), LDL-cholesterol was decreased by 16.7% and HDL-cholesterol was increased by 31.9%. SAA, CRP, TNFalpha and IL-6 levels were decreased by 31.8%, 66.2%, 53.9% and 14%, respectively and adiponectin was increased by 59.6%, compared with the baseline levels. Interestingly, the decrease of SAA was positively correlated with that of LDL-cholesterol but negatively with HDL-cholesterol during statin treatment. Among the adipokines, the decrease of SAA was positively correlated with TNFalpha (r = 0.50, p = 0.016). CONCLUSION: The results suggest that adipokines may be differentially regulated and independent of cholesterol changes and that adipokines may be a mediator, and the adipose tissue may be a target of statins' anti-inflammatory effect.