An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.

A communication and information technology infrastructure for real time monitoring and management of type 1 diabetes patients

This paper is focused on the integration of state-of-the-art technologies in the fields of telecommunications, simulation algorithms, and data mining in order to develop a Type 1 diabetes patient's semi to fully-automated monitoring and management system. The main components of the system are a glucose measurement device, an insulin delivery system (insulin injection or insulin pumps), a mobile phone for the GPRS network, and a PDA or laptop for the Internet. In the medical environment, appropriate infrastructure for storage, analysis and visualizing of patients' data has been implemented to facilitate treatment design by health care experts.

Modeling Calcium Concentration and Biochemical Reactions

Almost all regions of the brain receive one or more neuromodulatory inputs, and disrupting these inputs produces deficits in neuronal function. Neuromodulators act through intracellular second messenger pathways to influence the electrical properties of neurons, integration of synaptic inputs, spatio-temporal firing dynamics of neuronal networks, and, ultimately, systems behavior. Second messengers pathways consist of series of bimolecular reactions, enzymatic reactions, and diffusion. Calcium is the second messenger molecule with the most effectors, and thus is highly regulated by buffers, pumps and intracellular stores. Computational modeling provides an innovative, yet practical method to evaluate the spatial extent, time course and interaction among second messenger pathways, and the interaction of second messengers with neuron electrical properties. These processes occur both in compartments where the number of molecules are large enough to describe reactions deterministically (e.g. cell body), and in compartments where the number of molecules is small enough that reactions occur stochastically (e.g. spines). – In this tutorial, I explain how to develop models of second messenger pathways and calcium dynamics. The first part of the tutorial explains the equations used to model bimolecular reactions, enzyme reactions, calcium release channels, calcium pumps and diffusion. The second part explains some of the GENESIS, Kinetikit and Chemesis objects that implement the appropriate equations. In depth explanation of calcium and second messenger models is provided by reviewing code, both in XPP, Chemesis and Kinetikit, that implements simple models of calcium dynamics and second messenger cascades.

Long-term systemic administration of human recombinant interleukin-1beta induces a dose-dependent fall in circulating parathyroid hormone in rats

The synergism/antagonism between interleukin (IL)-1beta and parathyroid hormone (PTH) has been the subject of in vitro and in vivo work, but a possible direct action of the cytokine on PTH release has not been reported. We have investigated the effect of a continuous infusion of human recombinant IL-1beta (rIL-1beta) on circulating PTH during a 14-day period in 7-week-old female rats. This time interval was chosen in order to exclude initial hypercalcemia and to enable data collection under steady-state conditions. Five groups of 20 animals each had miniosmotic pumps (Alzet 2002, 200 microl) implanted subcutaneously and primed to release either distilled water (controls) or 100, 500, 1,000 and 2, 000 ng/24 h of rIL-1beta. Blood was drawn on days 1 and 14 for PTH, corticosterone and Ca2+ determinations. Adequate biological activity of the infused rIL-1beta was supported by elevated rectal temperature records and significant elevations of plasma corticosterone on day 14. The 100-ng dose had no effect but 500-2, 000 ng rIL-1beta/24 h significantly reduced plasma PTH in a dose-dependent manner down to 54% of basal value (20.4 +/- 1.1 vs. 15.3 +/- 1.4 pg/ml for 500 ng, p < 0.005; 20.5 +/- 1.3 vs 12.3 +/- 1.1 for 1,000 ng, p < 0.001, and 19.5 +/- 2.0 vs. 10.6 +/- 1.1 pg/ml for 2,000 ng, p < 0.0008). Despite these findings, no differences in blood Ca2+ could be detected between treated animals and controls. The following conclusions can be inferred from the foregoing: Systemic administration of rIL-1beta to rats induced a dose-dependent fall in circulating PTH without altering calcemia, calling into question the biological relevance of the former finding. Although the recorded PTH depression may indeed not have been severe enough to cause hypocalcemia, it can be hypothesized that osteoclast activation by rIL-1beta would enhance bone mineral release into the pool compensating for depressed PTH activity.

Nutrient Transport in the Mammary Gland: Calcium, Trace Minerals and Water Soluble Vitamins.

Milk nutrients are secreted by epithelial cells in the alveoli of the mammary gland by several complex and highly coordinated systems. Many of these nutrients are transported from the blood to the milk via transcellular pathways that involve the concerted activity of transport proteins on the apical and basolateral membranes of mammary epithelial cells. In this review, we focus on transport mechanisms that contribute to the secretion of calcium, trace minerals and water soluble vitamins into milk with particular focus on the role of transporters of the SLC series as well as calcium transport proteins (ion channels and pumps). Numerous members of the SLC family are involved in the regulation of essential nutrients in the milk, such as the divalent metal transporter-1 (SLC11A2), ferroportin-1 (SLC40A1) and the copper transporter CTR1 (SLC31A1). A deeper understanding of the physiology and pathophysiology of these transporters will be of great value for drug discovery and treatment of breast diseases.

Proteasome inhibitor (MG132) rescues Nav1.5 protein content and the cardiac sodium current in dystrophin-deficient mdx (5cv) mice

The cardiac voltage-gated sodium channel, Nav1.5, plays a central role in cardiac excitability and impulse propagation and associates with the dystrophin multiprotein complex at the lateral membrane of cardiomyocytes. It was previously shown that Nav1.5 protein content and the sodium current (l Na) were both decreased in cardiomyocytes of dystrophin-deficient mdx (5cv) mice. In this study, wild-type and mdx (5cv) mice were treated for 7 days with the proteasome inhibitor MG132 (10 μg/Kg/24 h) using implanted osmotic mini pumps. MG132 rescued both the total amount of Nav1.5 protein and l Na but, unlike in previous studies, de novo expression of dystrophin was not observed in skeletal or cardiac muscle. This study suggests that the reduced expression of Nav1.5 in dystrophin-deficient cells is dependent on proteasomal degradation.

An Extended Hierarchical Task Analysis for Error Prediction in Medical Devices

This paper introduces an extended hierarchical task analysis (HTA) methodology devised to evaluate and compare user interfaces on volumetric infusion pumps. The pumps were studied along the dimensions of overall usability and propensity for generating human error. With HTA as our framework, we analyzed six pumps on a variety of common tasks using Norman’s Action theory. The introduced method of evaluation divides the problem space between the external world of the device interface and the user’s internal cognitive world, allowing for predictions of potential user errors at the human-device level. In this paper, one detailed analysis is provided as an example, comparing two different pumps on two separate tasks. The results demonstrate the inherent variation, often the cause of usage errors, found with infusion pumps being used in hospitals today. The reported methodology is a useful tool for evaluating human performance and predicting potential user errors with infusion pumps and other simple medical devices.

Evaluating a medical error taxonomy

Healthcare has been slow in using human factors principles to reduce medical errors. The Center for Devices and Radiological Health (CDRH) recognizes that a lack of attention to human factors during product development may lead to errors that have the potential for patient injury, or even death. In response to the need for reducing medication errors, the National Coordinating Council for Medication Errors Reporting and Prevention (NCC MERP) released the NCC MERP taxonomy that provides a standard language for reporting medication errors. This project maps the NCC MERP taxonomy of medication error to MedWatch medical errors involving infusion pumps. Of particular interest are human factors associated with medical device errors. The NCC MERP taxonomy of medication errors is limited in mapping information from MEDWATCH because of the focus on the medical device and the format of reporting.

Groundwater age distributions at a public drinking water supply well field derived from multiple age tracers (⁸⁵Kr, ³H/³He, and ³⁹Ar)

Groundwater age is a key aspect of production well vulnerability. Public drinking water supply wells typically have long screens and are expected to produce a mixture of groundwater ages. The groundwater age distributions of seven production wells of the Holten well field (Netherlands) were estimated from tritium-helium (3H/3He), krypton-85 (85Kr), and argon-39 (39Ar), using a new application of a discrete age distribution model and existing mathematical models, by minimizing the uncertainty-weighted squared differences of modeled and measured tracer concentrations. The observed tracer concentrations fitted well to a 4-bin discrete age distribution model or a dispersion model with a fraction of old groundwater. Our results show that more than 75 of the water pumped by four shallow production wells has a groundwater age of less than 20 years and these wells are very vulnerable to recent surface contamination. More than 50 of the water pumped by three deep production wells is older than 60 years. 3H/3He samples from short screened monitoring wells surrounding the well field constrained the age stratification in the aquifer. The discrepancy between the age stratification with depth and the groundwater age distribution of the production wells showed that the well field preferentially pumps from the shallow part of the aquifer. The discrete groundwater age distribution model appears to be a suitable approach in settings where the shape of the age distribution cannot be assumed to follow a simple mathematical model, such as a production well field where wells compete for capture area.

Reply to the editor

We appreciate the comments and concerns expressed by Arakawa and colleagues regarding our article, titled “Pulsatile control of rotary blood pumps: Does the modulation waveform matter?”1 Unfortunately, we have to disagree with Arakawa and colleagues. As is obvious from the title of our article, it investigates the effect of different waveforms on the heart–device interaction. In contrast to the authors' claim, this is the first article in the literature that uses basic waveforms (sine, triangle, saw tooth, and rectangular) with different phase shifts to examines their impact on left ventricular unloading. The previous publications2, 3 and 4 just varied the pump speed during systole and diastole, which was first reported by Bearnson and associates5 in 1996, and studied its effect on aortic pressure, coronary flow, and end-diastolic volume. We should mention that dp/dtmax is a load-sensitive parameter of contractility and not representative for the degree of unloading. Moreover, none of the aforementioned reports has studied mechanical unloading and in particular the stroke work of the left ventricle. Our method is unique because we do not just alternate between high and low speed but have accurate control of the waveform because of the direct drive system of Levitronix Technologies LLC (Waltham, Mass) and a custom-developed pump controller. Without referring, Arakawa and associates state “several previous studies have already reported the coronary flow diminishes as the left ventricular assist device support increases.” It should be noted that all the waveforms used in our study have 2000 rpm average value with 1000 rpm amplitude, which is not an excessive speed for the CentriMag rotary pump (Levitronix) to collapse the ventricle and diminish the coronary flow. We agree with Arakawa and coworkers that there is a need for a heart failure model to come to more relevant results with respect to clinical expectations. However, we have explored many existing models, including species and breeds that have a native proneness to cardiomyopathy, but all of them differ from the genetic presentation in humans. We certainly do not believe that the use of microembolization, in which the coronary circulation is impaired by the injection of microspheres, would form a good model from which to draw conclusions about coronary flow change under different loading conditions. A model would be needed in which either an infarct is created to mimic ischemic heart failure or the coronary circulation remains untouched to simulate, for instance, dilated cardiomyopathy. Furthermore, in discussion we clearly mention that “lack of heart failure is a major limitation of our study.” We also believe that unloading is not the only factor of the cardiac functional recovery, and an excessive unloading of the left ventricle might lead to cardiac tissue atrophy. Therefore, in our article we mention that control of the level of cardiac unloading by assist devices has been suggested as a mechanical tool to promote recovery, and more studies are required to find better strategies for the speed modulation of rotary pumps and to achieve an optimal heart load control to enhance myocardial recovery. Finally, there are many publications about pulsing rotary blood pumps and it was impossible to include them all. We preferred to reference some of the earlier basic works such as an original research by Bearnson and coworkers5 and another article published by our group,6 which is more relevant.

A robust reference signal generator for synchronized ventricular assist devices

Ventricular assist devices (VADs) are blood pumps that offer an option to support the circulation of patients with severe heart failure. Since a failing heart has a remaining pump function, its interaction with the VAD influences the hemodynamics. Ideally, the heart's action is taken into account for actuating the device such that the device is synchronized to the natural cardiac cycle. To realize this in practice, a reliable real-time algorithm for the automatic synchronization of the VAD to the heart rate is required. This paper defines the tasks such an algorithm needs to fulfill: the automatic detection of irregular heart beats and the feedback control of the phase shift between the systolic phases of the heart and the assist device. We demonstrate a possible solution to these problems and analyze its performance in two steps. First, the algorithm is tested using the MIT-BIH arrhythmia database. Second, the algorithm is implemented in a controller for a pulsatile and a continuous-flow VAD. These devices are connected to a hybrid mock circulation where three test scenarios are evaluated. The proposed algorithm ensures a reliable synchronization of the VAD to the heart cycle, while being insensitive to irregularities in the heart rate.

Diffusion tensor imaging of cocaine-treated rodents.

Studies in cocaine-dependent human subjects have shown differences in white matter on diffusion tensor imaging (DTI) compared with non-drug-using controls. It is not known whether the differences in fractional anisotropy (FA) seen on DTI in white matter regions of cocaine-dependent humans result from a pre-existing predilection for drug use or purely from cocaine abuse. To study the effect of cocaine on brain white matter, DTI was performed on 24 rats after continuous infusion of cocaine or saline for 4 weeks, followed by brain histology. Voxel-based morphometry analysis showed an 18% FA decrease in the splenium of the corpus callosum (CC) in cocaine-treated animals relative to saline controls. On histology, significant increase in neurofilament expression (125%) and decrease in myelin basic protein (40%) were observed in the same region in cocaine-treated animals. This study supports the hypothesis that chronic cocaine use alters white matter integrity in human CC. Unlike humans, where the FA in the genu differed between cocaine users and non-users, the splenium was affected in rats. These differences between rodent and human findings could be due to several factors that include differences in the brain structure and function between species and/or the dose, timing, and duration of cocaine administration.

Characterization of calcium signaling mechanisms in osteoblasts

Bone remodeling is controlled by the osteoclast, which resorbs bone, and the osteoblast, which synthesizes and secretes proteins that are eventually mineralized into bone. Ca$\sp{2+}$ homeostasis and signaling contribute to the function of nearly all cell types, and understanding both in the osteoblast is of importance given its secretory properties and interaction with osteoclasts. This study was undertaken to identify and investigate the physiology of the Ca$\sp{2+}$ signaling mechanisms present in osteoblasts. The Ca$\sp{2+}$ pumps, stores and channels present in osteoblasts were studied. RT-PCR cloning revealed that osteoblast-like cells express PMCA1b, an alternatively spliced transcript of the plasma membrane Ca$\sp{2+}$-ATPase. The PMCA1b isoform contains a consensus phosphorylation site for cAMP-dependent protein kinase A and a modified calmodulin binding domain. The regulation of osteoblast function by agents that act via cAMP-mediated pathways may involve alterations in the activity of the plasma membrane Ca$\sp{2+}$-ATPase.^ Calcium release from intracellular stores is a signaling mechanism used universally by cells responding to hormones and growth factors, and the compartmentalization and regulated release of calcium is cell-type specific. Fura-2 was employed to monitor intracellular Ca$\sp{2+}$. Thapsigargin and 2,5,-di-(tert-butyl)-1,4-benzohydroquinone (tBuHQ), two inhibitors of endoplasmic reticulum Ca$\sp{2+}$-ATPase activity, both emptied a single intracellular calcium pool which was released in response to either ATP or thrombin, identifying it as the inositol 1,4,5-trisphosphate-sensitive calcium store. The Ca$\sp{2+}$ storage system present in osteoblasts is typical of a non-excitable cell type, despite these cells sharing characteristics of excitable cells such as voltage-sensitive Ca$\sp{2+}$ channels (VSCCs).^ VSCCs are important cell surface regulators of membrane permeability to Ca$\sp{2+}$. In non-excitable cells VSCCs act as cellular transducers of stimulus-secretion coupling, activators of intracellular proteins, and in control of cell growth and differentiation. Functional VSCCs have been shown to exist in osteoblasts, however, no molecular cloning has been reported. To obtain information concerning the molecular identity of the osteoblastic VSCC, we used an RT-PCR regional amplification approach. Sequencing of the products indicated that osteoblasts express at least two isoforms of the L-type VSCC, $\alpha 1\sb{\rm C-a}$ and the $\alpha 1\sb{\rm C-d}$, which share regions of identity to the $\alpha \sb{\rm 1C}$ isoform first identified in cardiac myocytes. The ability of $1,25(\rm OH)\sb2D\sb3$ and structural analogs to modulate expression of Ca$\sp{2+}$ channel mRNA was then investigated. Cells were cultured for 48 hr in the presence of $1,25(\rm OH)\sb2D\sb3$ or vitamin D analogs, and the levels of mRNA encoding VSCC $\alpha \sb{\rm 1C}$ were quantitated using a competitive RT-PCR assay. It was found that $1,25(\rm OH)\sb2D\sb3$ and analog BT reduced steady state levels of $\alpha \sb{\rm 1C}$ mRNA. Conversely, analog AT did not alter steady state levels of Ca$\sp{2+}$ channel mRNA. Since it has been shown previously that analog BT, but not AT, binds and activates the nuclear vitamin D receptor, these findings suggest that the down regulation of channel mRNA involves the nuclear receptor for $1,25(\rm OH)\sb2D\sb3$. ^

The Effect of Parathyroid Hormone on Osseointegration in Insulin-Treated Diabetic Rats.

OBJECTIVES Uncontrolled diabetes mellitus is associated with impaired osseointegration. Diabetic individuals might benefit from bone anabolic therapies. Intermittent administration of 1-34 parathyroid hormone (PTH) stimulates bone formation in rodent models. However, this anabolic effect fails in diabetic rats. Whether the anabolic effect of PTH can be achieved in insulin-controlled diabetic rats has not been investigated yet. MATERIALS AND METHODS After diabetes induction with streptozotocin in 40 female Wistar rats, the animals were randomly divided into 4 groups: diabetes, diabetes plus PTH, insulin-treated diabetes, and insulin-treated diabetes plus PTH. After 1 week, miniscrews were inserted in the tibiae. Osmotic pumps with insulin or saline solution were implanted. Animals received 60 mg/kg PTH or saline solution. Histomorphometric analysis was performed. RESULTS In diabetic rats, no changes of medullary periimplant bone area or bone-to-implant contacts (BICs) were achieved with or without treatment with PTH. However, also animals treated with insulin failed to response significantly to PTH regarding bone area (7.4 ± 4.1% and 8.1 ± 4.1%) and BICs (33.7 ± 16.9% and 49.9 ± 11.9%). CONCLUSION These results demonstrate that the metabolic characteristics of the diabetic rats produced a condition unable to respond to PTH treatment, even when hyperglycemia was controlled with insulin.

Molecular control of cardiac sodium homeostasis in health and disease.

INTRODUCTION Cardiac myocytes utilize three high-capacity Na transport processes whose precise function can determine myocyte fate and the triggering of arrhythmias in pathological settings. We present recent results on the regulation of all three transporters that may be important for an understanding of cardiac function during ischemia/reperfusion episodes. METHODS AND RESULTS Refined ion selective electrode (ISE) techniques and giant patch methods were used to analyze the function of cardiac Na/K pumps, Na/Ca exchange (NCX1), and Na/H exchange (NHE1) in excised cardiac patches and intact myocytes. To consider results cohesively, simulations were developed that account for electroneutrality of the cytoplasm, ion homeostasis, water homeostasis (i.e., cell volume), and cytoplasmic pH. The Na/K pump determines the average life-time of Na ions (3-10 minutes) as well as K ions (>30 minutes) in the cytoplasm. The long time course of K homeostasis can determine the time course of myocyte volume changes after ion homeostasis is perturbed. In excised patches, cardiac Na/K pumps turn on slowly (-30 seconds) with millimolar ATP dependence, when activated for the first time. In steady state, however, pumps are fully active with <0.2 mM ATP and are nearly unaffected by high ADP (2 mM) and Pi (10 mM) concentrations as may occur in ischemia. NCX1s appear to operate with slippage that contributes to background Na influx and inward current in heart. Thus, myocyte Na levels may be regulated by the inactivation reactions of the exchanger which are both Na- and proton-dependent. NHE1 also undergo strong Na-dependent inactivation, whereby a brief rise of cytoplasmic Na can cause inactivation that persists for many minutes after cytoplasmic Na is removed. This mechanism is blocked by pertussis toxin, suggesting involvement of a Na-dependent G-protein. Given that maximal NCX1- and NHE1-mediated ion fluxes are much greater than maximal Na/K pump-mediated Na extrusion in myocytes, the Na-dependent inactivation mechanisms of NCX1 and NHE1 may be important determinants of cardiac Na homeostasis. CONCLUSIONS Na/K pumps appear to be optimized to continue operation when energy reserves are compromised. Both NCX1 and NHE1 activities are regulated by accumulation of cytoplasmic Na. These principles may importantly control cardiac cytoplasmic Na and promote myocyte survival during ischemia/reperfusion episodes by preventing Ca overload.