Neural network based glucose - insulin metabolism models for children with Type 1 diabetes

In this paper two models for the simulation of glucose-insulin metabolism of children with Type 1 diabetes are presented. The models are based on the combined use of Compartmental Models (CMs) and artificial Neural Networks (NNs). Data from children with Type 1 diabetes, stored in a database, have been used as input to the models. The data are taken from four children with Type 1 diabetes and contain information about glucose levels taken from continuous glucose monitoring system, insulin intake and food intake, along with corresponding time. The influences of taken insulin on plasma insulin concentration, as well as the effect of food intake on glucose input into the blood from the gut, are estimated from the CMs. The outputs of CMs, along with previous glucose measurements, are fed to a NN, which provides short-term prediction of glucose values. For comparative reasons two different NN architectures have been tested: a Feed-Forward NN (FFNN) trained with the back-propagation algorithm with adaptive learning rate and momentum, and a Recurrent NN (RNN), trained with the Real Time Recurrent Learning (RTRL) algorithm. The results indicate that the best prediction performance can be achieved by the use of RNN.

A real time simulation model of glucose-insulin metabolism for type 1 diabetes patients

In this paper, a simulation model of glucose-insulin metabolism for Type 1 diabetes patients is presented. The proposed system is based on the combination of Compartmental Models (CMs) and artificial Neural Networks (NNs). This model aims at the development of an accurate system, in order to assist Type 1 diabetes patients to handle their blood glucose profile and recognize dangerous metabolic states. Data from a Type 1 diabetes patient, stored in a database, have been used as input to the hybrid system. The data contain information about measured blood glucose levels, insulin intake, and description of food intake, along with the corresponding time. The data are passed to three separate CMs, which produce estimations about (i) the effect of Short Acting (SA) insulin intake on blood insulin concentration, (ii) the effect of Intermediate Acting (IA) insulin intake on blood insulin concentration, and (iii) the effect of carbohydrate intake on blood glucose absorption from the gut. The outputs of the three CMs are passed to a Recurrent NN (RNN) in order to predict subsequent blood glucose levels. The RNN is trained with the Real Time Recurrent Learning (RTRL) algorithm. The resulted blood glucose predictions are promising for the use of the proposed model for blood glucose level estimation for Type 1 diabetes patients.

The effect of an oral glucose load on sodium and water excretion after rapid intravenous infusion of 0.9% (w/v) saline

Previous studies have suggested that oral or intravenous glucose enhances salt and water retention following a saline load. To test this, we studied the effects of an oral glucose load on urinary sodium and water excretion and serum biochemistry in response to a 2l intravenous infusion of 0.9% saline in normal subjects.

Role of diuretics, blockers, and statins in increasing the risk of diabetes in patients with impaired glucose tolerance: reanalysis of data from the NAVIGATOR study

OBJECTIVE To examine the degree to which use of β blockers, statins, and diuretics in patients with impaired glucose tolerance and other cardiovascular risk factors is associated with new onset diabetes. DESIGN Reanalysis of data from the Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR) trial. SETTING NAVIGATOR trial. PARTICIPANTS Patients who at baseline (enrolment) were treatment naïve to β blockers (n=5640), diuretics (n=6346), statins (n=6146), and calcium channel blockers (n=6294). Use of calcium channel blocker was used as a metabolically neutral control. MAIN OUTCOME MEASURES Development of new onset diabetes diagnosed by standard plasma glucose level in all participants and confirmed with glucose tolerance testing within 12 weeks after the increased glucose value was recorded. The relation between each treatment and new onset diabetes was evaluated using marginal structural models for causal inference, to account for time dependent confounding in treatment assignment. RESULTS During the median five years of follow-up, β blockers were started in 915 (16.2%) patients, diuretics in 1316 (20.7%), statins in 1353 (22.0%), and calcium channel blockers in 1171 (18.6%). After adjusting for baseline characteristics and time varying confounders, diuretics and statins were both associated with an increased risk of new onset diabetes (hazard ratio 1.23, 95% confidence interval 1.06 to 1.44, and 1.32, 1.14 to 1.48, respectively), whereas β blockers and calcium channel blockers were not associated with new onset diabetes (1.10, 0.92 to 1.31, and 0.95, 0.79 to 1.13, respectively). CONCLUSIONS Among people with impaired glucose tolerance and other cardiovascular risk factors and with serial glucose measurements, diuretics and statins were associated with an increased risk of new onset diabetes, whereas the effect of β blockers was non-significant.

Variation in fat mobilization during early lactation differently affects feed intake, body condition, and lipid and glucose metabolism in high-yielding dairy cows.

Fat mobilization to meet energy requirements during early lactation is inevitable because of insufficient feed intake, but differs greatly among high-yielding dairy cows. Therefore, we studied milk production, feed intake, and body condition as well as metabolic and endocrine changes in high-yielding dairy cows to identify variable strategies in metabolic and endocrine adaptation to overcome postpartum metabolic load attributable to milk production. Cows used in this study varied in fat mobilization around calving, as classified by mean total liver fat concentrations (LFC) postpartum. German Holstein cows (n=27) were studied from dry off until d 63 postpartum in their third lactation. All cows were fed the same total mixed rations ad libitum during the dry period and lactation. Plasma concentrations of metabolites and hormones were measured in blood samples taken at d 56, 28, 15, and 5 before expected calving and at d 1 and once weekly up to d 63 postpartum. Liver biopsies were taken on d 56 and 15 before calving, and on d 1, 14, 28, and 49 postpartum to measure LFC and glycogen concentrations. Cows were grouped accordingly to mean total LFC on d 1, 14, and 28 in high, medium, and low fat-mobilizing cows. Mean LFC (±SEM) differed among groups and were 351±14, 250±10, and 159±9 mg/g of dry matter for high, medium, and low fat-mobilizing cows, respectively, whereas hepatic glycogen concentrations postpartum were the highest in low fat-mobilizing cows. Cows in the low group showed the highest dry matter intake and the least negative energy balance postpartum, but energy-corrected milk yield was similar among groups. The decrease in body weight postpartum was greatest in high fat-mobilizing cows, but the decrease in backfat thickness was greatest in medium fat-mobilizing cows. Plasma concentrations of nonesterified fatty acids and β-hydroxybutyrate were highest around calving in high fat-mobilizing cows. Plasma triglycerides were highest in the medium group and plasma cholesterol concentrations were lowest in the high group at calving. During early lactation, the decrease in plasma glucose concentrations was greatest in the high group, and plasma insulin concentrations postpartum were highest in the low group. The revised quantitative insulin sensitivity check index values decreased during the transition period and postpartum, and were highest in the medium group. Plasma cortisol concentrations during the transition period and postpartum period and plasma leptin concentrations were highest in the medium group. In conclusion, cows adapted differently to the metabolic load and used variable strategies for homeorhetic regulation of milk production. Differences in fat mobilization were part of these strategies and contributed to the individual adaptation of energy metabolism to milk production.

Supplementation of conjugated linoleic acid in dairy cows reduces endogenous glucose production during early lactation.

Trans-10,cis-12 conjugated linoleic acid (CLA) supplementation causes milk fat depression in dairy cows, but CLA effects on glucose metabolism are not clear. The objective of the study was to investigate glucose metabolism, especially endogenous glucose production (eGP) and glucose oxidation (GOx), as well as hepatic genes involved in endogenous glucose production in Holstein cows supplemented either with 50 g of rumen-protected CLA (9% trans-10,cis-12 and 10% cis-9,trans-11; CLA; n=10) or 50 g of control fat (24% C18:2; Ctrl; n=10) from wk 2 before parturition to wk 9 of lactation. Animal performance data were recorded and blood metabolites and hormones were taken weekly from 2 wk before to 12 wk after parturition. During wk 3 and 9 after parturition, glucose tolerance tests were performed and eGP and GOx were measured by [U-(13)C] glucose infusion. Liver biopsies were taken at the same time to measure total fat and glycogen concentrations and gene expression of pyruvate carboxylase, cytosolic phosphoenolpyruvate carboxykinase, glucose-6-phosphatase, and carnitine palmitoyl-transferase 1. Conjugated linoleic acid feeding reduced milk fat, but increased milk lactose output; milk yield was higher starting 5 wk after parturition in CLA-fed cows than in Ctrl-fed cows. Energy balance was more negative during CLA supplementation, and plasma concentrations of glucose were higher immediately after calving in CLA-fed cows. Conjugated linoleic acid supplementation did not affect insulin release during glucose tolerance tests, but reduced eGP in wk 3, and eGP and GOx increased with time after parturition. Hepatic gene expression of cytosolic phosphoenolpyruvate carboxykinase tended to be lower in CLA-fed cows than in Ctrl-fed cows. In spite of lower eGP in CLA-fed cows, lactose output and plasma glucose concentrations were greater in CLA-fed cows than in Ctrl-fed cows. This suggests a CLA-related glucose sparing effect most likely due to lower glucose utilization for milk fat synthesis and probably because of a more efficient whole-body energy utilization in CLA-fed cows.

Hepatic gene expression involved in glucose and lipid metabolism in transition cows: effects of fat mobilization during early lactation in relation to milk performance and metabolic changes.

Insufficient feed intake during early lactation results in elevated body fat mobilization to meet energy demands for milk production. Hepatic energy metabolism is involved by increasing endogenous glucose production and hepatic glucose output for milk synthesis and by adaptation of postcalving fuel oxidation. Given that cows differ in their degree of fat mobilization around parturition, indicated by variable total liver fat concentration (LFC), the study investigated the influence of peripartum fat mobilization on hepatic gene expression involved in gluconeogenesis, fatty acid oxidation, ketogenesis, and cholesterol synthesis, as well as transcriptional factors referring to energy metabolism. German Holstein cows were grouped according to mean total LFC on d 1, 14, and 28 after parturition as low [<200mg of total fat/g of dry matter (DM); n=10], medium (200-300 mg of total fat/g of DM; n=10), and high (>300 mg of total fat/g of DM; n=7), indicating fat mobilization during early lactation. Cows were fed total mixed rations ad libitum and held under equal conditions. Liver biopsies were taken at d 56 and 15 before and d 1, 14, 28, and 49 after parturition to measure mRNA abundances of pyruvate carboxylase (PC); phosphoenolpyruvate carboxykinase; glucose-6-phosphatase; propionyl-coenzyme A (CoA) carboxylase α; carnitine palmitoyl-transferase 1A (CPT1A); acyl-CoA synthetase, long chain 1 (ASCL1); acyl-CoA dehydrogenase, very long chain; 3-hydroxy-3-methylglutaryl-CoA synthase 1 and 2; sterol regulatory element-binding factor 1; and peroxisome proliferator-activated factor α. Total LFC postpartum differed greatly among cows, and the mRNA abundance of most enzymes and transcription factors changed with time during the experimental period. Abundance of PC mRNA increased at parturition to a greater extent in high- and medium-LFC groups than in the low-LFC group. Significant LFC × time interactions for ACSL1 and CPT1A during the experimental period indicated variable gene expression depending on LFC after parturition. Correlations between hepatic gene expression and performance data and plasma concentrations of metabolites and hormones showed time-specific relations during the transition period. Elevated body fat mobilization during early lactation affected gene expression involved in gluconeogenesis to a greater extent than gene expression involved in lipid metabolism, indicating the dependence of hepatic glucose metabolism on hepatic lipid status and fat mobilization during early lactation.

A specific expression profile of heat-shock proteins and glucose-regulated proteins is associated with response to neoadjuvant chemotherapy in oesophageal adenocarcinomas

BACKGROUND Oesophageal adenocarcinomas often show resistances to chemotherapy (CTX), therefore, it would be of high interest to better understand the mechanisms of resistance. We examined the expression of heat-shock proteins (HSPs) and glucose-regulated proteins (GRPs) in pretherapeutic biopsies of oesophageal adenocarcinomas to assess their potential role in CTX response. METHODS Ninety biopsies of locally advanced adenocarcinomas before platin/5-fluorouracil (FU)-based CTX were investigated by reverse phase protein arrays (RPPAs), immunohistochemistry (IHC) and quantitative RT-PCR. RESULTS CTX response strongly correlated with survival (P=0.001). Two groups of tumours with specific protein expression patterns were identified by RPPA: Group A was characterised by low expression of HSP90, HSP27 and p-HSP27((Ser15, Ser78, Ser82)) and high expression of GRP78, GRP94, HSP70 and HSP60; Group B exhibited the inverse pattern. Tumours of Group A were more likely to respond to CTX, resulting in histopathological tumour regression (P=0.041) and post-therapeutic down-categorisation from cT3 to ypT0-T2 (P=0.040). High HSP60 protein (IHC) and mRNA expression were also associated with tumour down-categorisation (P=0.016 and P=0.004). CONCLUSION Our findings may enhance the understanding of CTX response mechanisms, might be helpful to predict CTX response and might have translational relevance as they highlight the role of potentially targetable cellular stress proteins in the context of CTX response.

Genes in glucose metabolism and association with spina bifida.

The authors test single nucleotide polymorphisms (SNPs) in coding sequences of 12 candidate genes involved in glucose metabolism and obesity for associations with spina bifida. Genotyping was performed on 507 children with spina bifida and their parents plus anonymous control DNAs from Hispanic and Caucasian individuals. The transmission disequilibrium test was performed to test for genetic associations between transmission of alleles and spina bifida in the offspring (P < .05). A statistically significant association between Lys481 of HK1 (G allele), Arg109Lys of LEPR (G allele), and Pro196 of GLUT1 (A allele) was found ( P = .019, .039, and .040, respectively). Three SNPs on 3 genes involved with glucose metabolism and obesity may be associated with increased susceptibility to spina bifida.

Glucose transporter expression in eutopic endometrial tissue and ectopic endometriotic lesions

Endometriosis is an extremely prevalent disorder characterized by the growth of endometrial tissue at ectopic locations. Glycolysis is an energy-producing mechanism that occurs in almost all cells and requires an adequate uptake of glucose mediated by glucose transporter (GLUT) proteins. At present, however, very little is known about their expression in either the endometrium or the endometriotic lesions. The objective of this study was to examine the expression of SLC2A genes in the endometrium of women with and without endometriosis and in the matching ectopic tissue, and to confirm the presence of the GLUT proteins in ectopic lesions. There was a significantly higher expression of SLC2A3 and a significantly lower expression of SLC2A4 in women with endometriosis compared with those without. In women with endometriosis, the ectopic expression of SLC2A3, SLC2A4 and SLC2A5 was significantly higher than that observed in the matching eutopic tissue. GLUT1 protein expression was present in both epithelial and stromal cells and GLUT3 was confined to CD45-positive leukocytes. GLUT4 expression was strong in both ectopic epithelial and stromal cells and localized to the cellular membrane in epithelial cells. These results show that GLUT expression is altered between eutopic and ectopic tissue and between women with and without endometriosis, and that GLUT4 may represent a significant entry route for glucose into the endometriotic epithelial cells. The inducible nature of GLUT4 and its limited cellular expression may make GLUT4 an attractive target for non-hormone-based treatments of endometriosis.

Regulation of sodium/glucose cotransporter expression in LLC-PK$\sb1$ cells

Expression of the Na$\sp+$/glucose cotransporter (SGLT1), a differentiated function of the pig kidney epithelial cell line LLC-PK$\sb1$ derived from proximal tubule, was further investigated. The differentiation inducer hexamethylene bisacetamide (HMBA) and IBMX, an inhibitor of cAMP phosphodiesterase, each stimulated a significant increase in Na$\sp+$/glucose cotransport activity, levels of the 75 kD cotransporter subunit and steady-state levels of the SGLT1 message. The action of HMBA is associated with involvement of polyamines and protein kinase C, and is synergistic with cAMP. We provide evidence that cAMP-elevating agents increase Na$\sp+$/glucose cotransporter expression, at least in part, via a post-transcriptional mechanism. Two molecular species of SGLT1 mRNA (3.9 kb and 2.2 kb) are transcribed from the same gene in LLC-PK$\sb1$ cells and differ only in the length of the 3$\sp\prime$ untranslated region (3$\sp\prime$ UTR). cAMP elevation differentially stabilized the 3.9 kb SGLT1 transcript from degradation but not the 22 kb species. UV-cross-linking and label transfer experiments indicated that cyclic AMP elevation was associated with formation of a 48 kD protein complex with a specific domain within the 3$\sp\prime$ UTR of SGLT1 mRNA. The binding was competitively inhibited by poly (U) and other U-rich RNA species such as c-fos ARE, and modulated by a protein kinase A-mediated phosphorylation/dephosphorylation mechanism. The binding site was mapped to a 120-nucleotide 3$\sp\prime$ UTR sequence which contains a uridine-rich region (URE). Our study provides the first demonstration that renal SGLT1 is post-transcriptionally regulated by a phosphorylation/dephosphorylation mechanism, and provides a deeper insight into gene regulation of this physiologically important cotransporter. ^

Plasminogen activator inhibitor-1, monocyte chemoattractant protein-1, e-selectin and C-reactive protein levels in response to 4-week very-high-fructose or -glucose diets.

BACKGROUND/OBJECTIVES High intake of added sweeteners is considered to have a causal role in the pathogenesis of cardiometabolic disorders. Especially, high-fructose intake is regarded as potentially harmful to cardiometabolic health. It may cause not only weight gain but also low-grade inflammation, which represents an independent risk factor for developing type 2 diabetes and cardiovascular disease. In particular, fructose has been suggested to induce plasminogen activator inhibitor-1 (PAI-1) expression in the liver and to increase circulating inflammatory cytokines. We therefore aimed to investigate, whether high-fructose diet has an impact on PAI-1, monocyte chemoattractant protein-1 (MCP-1), e-selectin and C-reactive protein (CRP) concentrations in healthy humans. SUBJECTS/METHODS We studied 20 participants (12 males and 8 females) of the TUebingen FRuctose Or Glucose study. This is an exploratory, parallel, prospective, randomized, single-blinded, outpatient, hypercaloric, intervention study. The participants had a mean age of 30.9 ± 2.1 years and a mean body mass index of 26.0 ± 0.5 kg/m(2) and they received 150 g of either fructose or glucose per day for 4 weeks.Results:There were neither significant changes of PAI-1, MCP-1, e-selectin and CRP after fructose (n=10) and glucose (n=10) intervention nor treatment effects (all P>0.2). Moreover, we did not observe longitudinal associations of the inflammatory parameters with triglycerides, liver fat, visceral fat and body weight in the fructose group. CONCLUSIONS Temporary high-fructose intake does not seem to cause inflammation in apparently healthy people in this secondary analysis of a small feeding trial.

Predictive value of cerebrospinal fluid (CSF) lactate level versus CSF/blood glucose ratio for the diagnosis of bacterial meningitis following neurosurgery.

The value of cerebrospinal fluid (CSF) lactate level and CSF/blood glucose ratio for the identification of bacterial meningitis following neurosurgery was assessed in a retrospective study. During a 3-year period, 73 patients fulfilled the inclusion criteria and could be grouped by preset criteria in one of three categories: proven bacterial meningitis (n = 12), presumed bacterial meningitis (n = 14), and nonbacterial meningeal syndrome (n = 47). Of 73 patients analyzed, 45% were treated with antibiotics and 33% with steroids at the time of first lumbar puncture. CSF lactate values (cutoff, 4 mmol/L), in comparison with CSF/blood glucose ratios (cutoff, 0.4), were associated with higher sensitivity (0.88 vs. 0.77), specificity (0.98 vs. 0.87), and positive (0.96 vs. 0.77) and negative (0.94 vs. 0.87) predictive values. In conclusion, determination of the CSF lactate value is a quick, sensitive, and specific test to identify patients with bacterial meningitis after neurosurgery.

Personalized Tuning of a Reinforcement Learning Control Algorithm for Glucose Regulation

Artificial pancreas is in the forefront of research towards the automatic insulin infusion for patients with type 1 diabetes. Due to the high inter- and intra-variability of the diabetic population, the need for personalized approaches has been raised. This study presents an adaptive, patient-specific control strategy for glucose regulation based on reinforcement learning and more specifically on the Actor-Critic (AC) learning approach. The control algorithm provides daily updates of the basal rate and insulin-to-carbohydrate (IC) ratio in order to optimize glucose regulation. A method for the automatic and personalized initialization of the control algorithm is designed based on the estimation of the transfer entropy (TE) between insulin and glucose signals. The algorithm has been evaluated in silico in adults, adolescents and children for 10 days. Three scenarios of initialization to i) zero values, ii) random values and iii) TE-based values have been comparatively assessed. The results have shown that when the TE-based initialization is used, the algorithm achieves faster learning with 98%, 90% and 73% in the A+B zones of the Control Variability Grid Analysis for adults, adolescents and children respectively after five days compared to 95%, 78%, 41% for random initialization and 93%, 88%, 41% for zero initial values. Furthermore, in the case of children, the daily Low Blood Glucose Index reduces much faster when the TE-based tuning is applied. The results imply that automatic and personalized tuning based on TE reduces the learning period and improves the overall performance of the AC algorithm.