Noninvasive Glucose Sensing in Dielectrically Equivalent Multilayer Skin Phantoms.

The interstitial fluid of the skin contains glucose levels comparable to those of blood. Noninvasive glucose sensing by microwaves has great potential to relieve diabetics from the burden of daily blood sampling, but improving the selectivity of this method remains a challenge. This study reports a dielectrically equivalent multilayer skin phantom and provides insight into the criteria for noninvasive glucose sensing by conducting dielectric analysis. The skin phantom was a hydrogel composed of gelatin, glucose, sodium chloride, and water covered by paraffin-impregnated paper. Investigations conducted on a wide range of component concentrations revealed characteristic relative permittivity and dielectric loss determined by the amount of electrolyte and solution that was independent of the amount of glucose. Since the microwave response due to glucose tends to be buried in noise, we developed a flowchart that first identifies the amounts of electrolytes and proteins, which are the major components other than glucose, and then quantifies the remaining glucose content. This noninvasive glucose sensing method would not be limited to the medical healthcare field; it could potentially be used in food manufacturing processes, livestock farming, and plant cultivation management.

Direct Measurement of Near-Wall Molecular Transport Rate in a Microchannel and Its Dependence on Diffusivity.

Solute transport in a narrow space is the most elemental process in chromatography and biological pattern formation. However, the observation of such transport has been quite difficult, and theoretical investigations have therefore preponderated. Here, using a space- and time-resolved surface plasmon resonance (SPR) method, we measured the nanoscale near-wall (next to the wall) transport rate in a narrow channel after a solution and its solvent had come into contact. By combining the SPR method with a capillary injection method, which enables two solution plugs to flow immediately after they have made contact, we were able to measure the solute concentration evolution at the channel wall. We tested three combinations of two plugs of solution-water-glucose, sodium chloride-water, and glucose-sodium chloride-and succeeded in measuring diffusion-coefficient-dependent changes in the concentration of solute flowing through a rectangular microchannel in less than 0.4 s. A numerical analysis of this system revealed the acceleration of the solute/solution boundary moving on the wall and its deceleration at the center of the channel cross section. The observed experimental transport rate agreed with the numerical result quantitatively. These results show that the solute transport followed a laminar flow with a no-slip model and that the molecules were transported in the order of their diffusivity. In the third combination, when the two solutions made contact and started flowing, the interdiffusion of the solutes resulted in temporal concentrations lower than either of the solutions before contact, which indicated that the contact between the two solutions quickly led to separation by the advection-diffusion processes. We found that such a concentration profile could actually be measured. Our techniques are simple and applicable to a wide range of molecules; the method opens the way to direct observation of the space-time near-wall solute transport process and can be used for the rapid determination of diffusivity.

Small Intestinal Lesions in Patients with Anemia on Hemodialysis: A Two-Center, Cross-Sectional Study.

AIM: The number of patients on chronic dialysis in Japan is increasing every year, and the average age of these patients is also increasing annually. Iron deficiency is an important cause of anemia in patients on hemodialysis (HD). However, it has not been clarified whether these patients might have small intestinal mucosal lesions causing iron deficiency anemia. METHODS: We conducted a cross-sectional study in -asymptomatic patients on HD between April 2014 and -December 2015. We performed small bowel capsule endoscopy (SBCE) and analyzed the relationship between small intestinal endoscopic findings and anemia. RESULTS: SBCE was successfully completed in 39 eligible patients. Univariate analysis revealed that there was a significant difference in blood hemoglobin levels between the morbid SBCE-finding group (median 7.7 g/dL; range 6.7-9.2 g/dL) and the non-morbid SBCE-finding group (median 10.65 g/dL; range 6.4-13.1 g/dL; p = 0.0006, Mann-Whitney U test). On multivariate analysis, the blood hemoglobin level was an independent predictor of morbid SBCE findings (p = 0.0033). The cutoff value of blood hemoglobin level for the morbid SBCE finding was determined as 9.2 g/dL using receiver operating characteristic curve analysis. CONCLUSIONS: Patients on HD with anemia are at a high risk of small intestinal lesions. Since the control of small intestinal lesion may improve the anemia, these outcomes are significant factors for managing patients on HD.

Thermally Degradable Water Diffusion Barrier Assembled by Gelatin and Beeswax toward Edible Electronics.

Making ingestible devices edible facilitates diagnosis and therapy inside the body without the risk of retention; however, food materials are generally soft, absorb water molecules, and are not suitable for electronic devices. Here, we fabricated an edible water diffusion barrier film made by gelatin-beeswax composites for the encapsulation of transient electronics. Hydrophobic beeswax and hydrophilic gelatin are inherently difficult to mix; therefore, we created an emulsion simply by raising the temperature high enough to melt the materials and vigorous stirring them. As they cool, the beeswax with a relatively high solidification temperature aggregates and forms microspheres, which increases the gelatin gel's viscoelasticity and immobilizes the emulsion structure in the film. The thermoresponsive gelatin imparts degradability to the barrier and its stickiness also enables transfer of metal patterned electronics. Furthermore, we designed an edible resonator on the film and demonstrated its operation in an abdominal phantom environment; the resonator was made to be degradable in a warm aqueous solution by optimizing the composition ratio of the gelatin and beeswax. Our findings provide insight into criteria for making transient electronics on hydrophilic substrates with hydrophobic water diffusion barriers. This proof-of-concept study expands the potential of operating edible electronics in aqueous environments in harmony with the human body and nature.

Enhancing Type 2 Diabetes Treatment Decisions With Interpretable Machine Learning Models for Predicting Hemoglobin A1c Changes: Machine Learning Model Development.

BACKGROUND: Type 2 diabetes (T2D) is a significant global health challenge. Physicians need to assess whether future glycemic control will be poor on the current trajectory of usual care and usual-care treatment intensifications so that they can consider taking extra treatment measures to prevent poor outcomes. Predicting poor glycemic control from trends in hemoglobin A1c (HbA1c) levels is difficult due to the influence of seasonal fluctuations and other factors. OBJECTIVE: We sought to develop a model that accurately predicts poor glycemic control among patients with T2D receiving usual care. METHODS: Our machine learning model predicts poor glycemic control (HbA1c≥8%) using the transformer architecture, incorporating an attention mechanism to process irregularly spaced HbA1c time series and quantify temporal relationships of past HbA1c levels at each time point. We assessed the model using HbA1c levels from 7787 patients with T2D seeing specialist physicians at the University of Tokyo Hospital. The training data include instances of poor glycemic control occurring during usual care with usual-care treatment intensifications. We compared prediction accuracy, assessed with the area under the receiver operating characteristic curve, the area under the precision-recall curve, and the accuracy rate, to that of LightGBM. RESULTS: The area under the receiver operating characteristic curve, the area under the precision-recall curve, and the accuracy rate (95% confidence limits) of the proposed model were 0.925 (95% CI 0.923-0.928), 0.864 (95% CI 0.852-0.875), and 0.864 (95% CI 0.86-0.869), respectively. The proposed model achieved high prediction accuracy comparable to or surpassing LightGBM's performance. The model prioritized the most recent HbA1c levels for predictions. Older HbA1c levels in patients with poor glycemic control were slightly more influential in predictions compared to patients with good glycemic control. CONCLUSIONS: The proposed model accurately predicts poor glycemic control for patients with T2D receiving usual care, including patients receiving usual-care treatment intensifications, allowing physicians to identify cases warranting extraordinary treatment intensifications. If used by a nonspecialist, the model's indication of likely future poor glycemic control may warrant a referral to a specialist. Future efforts could incorporate diverse and large-scale clinical data for improved accuracy.

Cooperative suction by vertical capillary array pump for controlling flow profiles of microfluidic sensor chips.

A passive pump consisting of integrated vertical capillaries has been developed for a microfluidic chip as an useful component with an excellent flow volume and flow rate. A fluidic chip built into a passive pump was used by connecting the bottoms of all the capillaries to a top surface consisting of a thin layer channel in the microfluidic chip where the thin layer channel depth was smaller than the capillary radius. As a result the vertical capillaries drew fluid cooperatively rather than independently, thus exerting the maximum suction efficiency at every instance. This meant that a flow rate was realized that exhibited little variation and without any external power or operation. A microfluidic chip built into this passive pump had the ability to achieve a quasi-steady rather than a rapidly decreasing flow rate, which is a universal flow characteristic in an ordinary capillary.

Data Processing of SPR Curve Data to Maximize the Extraction of Changes in Electrochemical SPR Measurements.

We developed a novel measuring and data-processing method for performing electrochemical surface plasmon resonance (EC-SPR) on sensor surfaces for which detecting a specific SPR angle is difficult, such as a polymer having a non-uniform thickness with coloration. SPR measurements are used in medicine and basic research as an analytical method capable of molecular detection without labeling. However, SPR is not good for detecting small molecules with small refractive index changes. The proposed EC-SPR, which combines SPR measurements with an electrochemical reaction, makes it possible to measure small molecules without increasing the number of measurement steps. A drawback of EC-SPR is that it is difficult to detect a specific SPR angle on electron mediators, and it was found that it may not be possible to capture all the features produced. The novel method we describe here is different from the conventional one in which a specific SPR angle is obtained from an SPR curve; rather, it processes the SPR curve itself and can efficiently aggregate the feature displacements in the SPR curves that are dispersed through multiple angles. As an application, we used our method to detect small concentrations of H2O2 (LOD 0.7 µM) and glutamate (LOD 5 µM).

Treatment Discontinuation Prediction in Patients With Diabetes Using a Ranking Model: Machine Learning Model Development.

BACKGROUND: Treatment discontinuation (TD) is one of the major prognostic issues in diabetes care, and several models have been proposed to predict a missed appointment that may lead to TD in patients with diabetes by using binary classification models for the early detection of TD and for providing intervention support for patients. However, as binary classification models output the probability of a missed appointment occurring within a predetermined period, they are limited in their ability to estimate the magnitude of TD risk in patients with inconsistent intervals between appointments, making it difficult to prioritize patients for whom intervention support should be provided. OBJECTIVE: This study aimed to develop a machine-learned prediction model that can output a TD risk score defined by the length of time until TD and prioritize patients for intervention according to their TD risk. METHODS: This model included patients with diagnostic codes indicative of diabetes at the University of Tokyo Hospital between September 3, 2012, and May 17, 2014. The model was internally validated with patients from the same hospital from May 18, 2014, to January 29, 2016. The data used in this study included 7551 patients who visited the hospital after January 1, 2004, and had diagnostic codes indicative of diabetes. In particular, data that were recorded in the electronic medical records between September 3, 2012, and January 29, 2016, were used. The main outcome was the TD of a patient, which was defined as missing a scheduled clinical appointment and having no hospital visits within 3 times the average number of days between the visits of the patient and within 60 days. The TD risk score was calculated by using the parameters derived from the machine-learned ranking model. The prediction capacity was evaluated by using test data with the C-index for the performance of ranking patients, area under the receiver operating characteristic curve, and area under the precision-recall curve for discrimination, in addition to a calibration plot. RESULTS: The means (95% confidence limits) of the C-index, area under the receiver operating characteristic curve, and area under the precision-recall curve for the TD risk score were 0.749 (0.655, 0.823), 0.758 (0.649, 0.857), and 0.713 (0.554, 0.841), respectively. The observed and predicted probabilities were correlated with the calibration plots. CONCLUSIONS: A TD risk score was developed for patients with diabetes by combining a machine-learned method with electronic medical records. The score calculation can be integrated into medical records to identify patients at high risk of TD, which would be useful in supporting diabetes care and preventing TD.

A reliable aptamer array prepared by repeating inkjet-spotting toward on-site measurement.

A preparation protocol is proposed for a reliable aptamer array utilizing an ink-jet spotter. We accumulated streptavidin and biotinylated-aptamer in this order on a biotinylated-polyethylene glycol-coated gold substrate to prepare an aptamer array. The aptamer array was prepared with an alternate spotting structure where each aptamer spot was placed between reference spots formed with blocking solution thus suppressing contamination from neighboring spots during the blocking and washing processes. Four aptamer spots were prepared in a small area of 1×4.8mm(2) with five reference spots made of blocking solution. We evaluated the thrombin binding ability of the spotted aptamer array using a multi-analysis surface plasmon resonance sensor. We prepared a disposable capillary-driven flow chip designed for on-site measurement (Miura et al., 2010) with our aptamer array and detected thrombin from phosphate-buffered saline at concentrations of 50ngmL(-1) and 1µgmL(-1) (equivalent to 1.35 and 27nM, respectively). A correlation was observed between the refractive index shift and thrombin concentration. This implies that our array preparation protocol meets the requirement for the preparation of a one-time-use chip for on-site measurement.

Machine-Learning-Based Prediction of a Missed Scheduled Clinical Appointment by Patients With Diabetes.

BACKGROUND: About 10% of patients with diabetes discontinue treatment, resulting in the progression of diabetes-related complications and reduced quality of life. OBJECTIVE: The objective was to predict a missed clinical appointment (MA), which can lead to discontinued treatment for diabetes patients. METHODS: A machine-learning algorithm was used to build a logistic regression model for MA predictions, with L2-norm regularization used to avoid over-fitting and 10-fold cross validation used to evaluate prediction performance. Data associated with patient MAs were extracted from electronic medical records and classified into two groups: one related to patients' clinical condition (X1) and the other related to previous findings (X2). The records used were those of the University of Tokyo Hospital, and they included the history of 16 026 clinical appointments scheduled by 879 patients whose initial clinical visit had been made after January 1, 2004, who had diagnostic codes indicating diabetes, and whose HbA1c had been tested within 3 months after their initial visit. Records between April 1, 2011, and June 30, 2014, were inspected for a history of MAs. RESULTS: The best predictor of MAs proved to be X1 + X2 (AUC = 0.958); precision and recall rates were, respectively, 0.757 and 0.659. Among all the appointment data, the day of the week when an appointment was made was most strongly associated with MA predictions (weight = 2.22). CONCLUSIONS: Our findings may provide information to help clinicians make timely interventions to avoid MAs.

On-chip graphene oxide aptasensor for multiple protein detection.

The versatility of an on-chip graphene oxide (GO) aptasensor was successfully confirmed by the detection of three different proteins, namely, thrombin (TB), prostate specific antigen (PSA), and hemagglutinin (HA), simply by changing the aptamers but with the sensor composition remaining the same. The results indicate that both DNA and RNA aptamers immobilized on the GO surface are sufficiently active to realize an on-chip aptasensor. Molecular selectivity and concentration dependence were investigated in relation to TB and PSA detection by using a dual, triple, and quintuple microchannel configuration. The multiple target detection of TB and PSA on a single chip was also demonstrated by using a 2×3 linear-array GO aptasensor. This work enables us to apply this sensor to the development of a multicomponent analysis system for a wide variety of targets by choosing appropriate aptamers.

Characteristics of refractory gastroesophageal reflux disease (GERD) symptoms -is switching proton pump inhibitors based on the patient's CYP2C19 genotype an effective management strategy?

OBJECTIVE: We investigated factors related to proton pump inhibitor (PPI) -refractory gastroesophageal reflux disease (GERD) symptoms, particularly with respect to acid, the CYP2C19 genotype and psychological aspects. METHODS: Patients with an Frequency Scale for the Symptoms of GERD (FSSG) score of ≥8 after the initial treatment were switched to therapy with rabeprazole at a dose of 20 mg once daily for eight weeks. We investigated the rate of improvement in PPI-refractory GERD symptoms, background factors, the Hospital Anxiety and Depression Scale (HADS) score and the CYP2C19 genotype. Patients Sixty patients endoscopically diagnosed with reflux esophagitis within the past six months who had received omeprazole at a dose of 20 mg once daily for eight weeks or longer were enrolled. RESULTS: In 71.6% of the patients, the FSSG score decreased to <8 after treatment with omeprazole at a dose of 20 mg once daily for ≥8 weeks, resulting in improvements in their GERD symptoms. Significant factors related to omeprazole-refractory GERD symptoms included a longer disease duration (p=0.0004) and higher HADS score (p=0.01). Among the omeprazole-refractory cases, only 23.5% of the patients showed symptom improvement after switching to rabeprazole. There were no significant differences in the average scores for FSSG (p=0.089) or HADS (p=0.182), before or after the drug change. A total of 92% of the rabeprazole poor responders were homo/hetero extensive metabolizers for the CYP2C19 genotype. CONCLUSION: Our findings suggest that switching the PPI from omeprazole (20 mg once daily) to rabeprazole (20 mg once daily) is not a significant effective therapeutic strategy for improving PPI-refractory GERD symptoms, taking into consideration possible psychometric factors and patients who require stronger acid suppression than that achieved with a double dose of PPIs for PPI-refractory GERD symptoms.

Differential measurement with a microfluidic device for the highly selective continuous measurement of histamine released from rat basophilic leukemia cells (RBL-2H3).

We fabricated a micro-fluidic device for the highly selective detection of the histamine released from rat basophilic leukemia (RBL) 2H3 cells. The device has two thin layer flow channels, each with one working electrode. One electrode was modified with Os-polyvinylpyridine based mediator containing horseradish peroxidase (Os-gel-HRP) and histamine oxidase (HAOx), the other was modified with Os-gel-HRP without any HAOx. We employed the device for differential measurement by using the HAOx modified electrode for detection and the unmodified electrode as a reference. The detection limit was greatly improved from 190 to 25 nM since the baseline noise level was suppressed. We used differential measurement to observe the histamine released from RBL-2H3 cells when stimulated with dinitrophenylated bovine serum albumin (DNP-BSA) as an antigen. We injected 5 microM of histamine solution into our device and it remained stable for more than 8 h.

Selective detection of L-glutamate using a microfluidic device integrated with an enzyme-modified pre-reactor and an electrochemical detector.

A microfluidic device integrated with a nanoliter volume enzyme pre-reactor and an enzyme-modified electrode was developed for the highly selective continuous measurement of glutamate (Glu). The device consists mainly of two glass plates. One plate incorporates an electrochemical cell that consists of working electrode (WE), reference electrode (RE) and counter electrode (CE). The WE is modified with a bilayer film of Os-polyvinylpyrridine-based mediator containing horseradish peroxidase (Os-gel-HRP). The WE was operated at -50 mV versus Ag. The other plate has a thin layer flow channel integrated with a pre-reactor. The reactor has a number of micropillars (20 microm in diameter, 20 microm high and separated from each other by a 20 microm gap) modified with ascorbate oxidase (AAOx) to eliminate L-ascorbic acid (AA). The enzymatic oxidation of AA is superior to that obtained with our previously reported pre-electrolysis type micro-reactor since electrochemically reversible transmitters such as catecholamines do not provide a cathodic current at the WE. In addition, the high operation potential of the pre-reactor causes unknown electroactive species, which also cause interference at the detection electrode. As a result, we were able to detect 1 microM Glu continuously at a low flow rate even when AA concentration was 100 microM.

Biocompatible glucose sensor prepared by modifying protein and vinylferrocene monomer composite membrane.

This paper proposes a very simple procedure for preparing a biocompatible sensor based on a protein (bovine serum albumin, BSA), enzyme and vinylferrocene (VF) composite membrane modified electrode. The membrane was prepared simply by first casting vinylferrocene and then coating it with BSA and glucose oxidase immobilised with glutaraldehyde. The sensor response was independent of dissolved oxygen concentration from 3 to 10 ppm and showed good stability for serum sample measurement, unlike the commonly used BSA/enzyme modified electrode. The sensor response was almost unchanged over the measurement time (>10 h) whereas the responses of a BSA and glucose oxidase modified platinum electrode and an osmium-polyvinylpyridine wired horseradish peroxidase modified electrode (Ohara et al., 1993) fell to 68% of their initial value in a serum sample containing 10mM glucose.

Continuous measurement of glutamate and hydrogen peroxide using a microfabricated biosensor for studying the neurotoxicity of tributyltin.

We first measured the effects of trace levels of an endocrine disruptor, tributyltin (TBT), on the secretion response from nerve cells using a microfabricated biosensor designed for the continuous measurement of L-glutamate and hydrogen peroxide. We observed higher and long-lasting glutamate and hydrogen peroxide concentrations from the cells when cultured rat cortical neurons were exposed to TBT. Glutamate and hydrogen peroxide release was induced even when we reduced the TBT concentration to 10 nM. This concentration is about two orders of magnitude lower than the concentration that induced apoptosis-like cell death. We also report on the effects of NMDA and non-NMDA receptor antagonists, which can help us to understand the mechanism of TBT neurotoxicity.

Molecular design for enhanced sensitivity of a FRET aptasensor built on the graphene oxide surface.

We designed a biomolecular probe for highly sensitive protein detection by modifying an aptamer with a DNA spacer. The spacer controls the distance between a fluorescence dye and a quencher, which is crucial for FRET-based sensors. We successfully demonstrated an improvement in the sensitivity of an on-chip graphene oxide aptasensor.

Graphene-modified interdigitated array electrode: fabrication, characterization, and electrochemical immunoassay application.

We have developed a new procedure for fabricating interdigitated array gold electrodes (Au-IDA) modified with reduced graphene oxide (rGO). In this procedure, we coated the gold surface of the micrometer order electrodes with graphene oxide (GO) prior to the reduction and the lift-off processes to avoid short-circuiting the pair of electrodes by conductive rGO flakes after the reduction. We then studied the basic electrochemical activity of the prepared electrodes, rGO/Au-IDA, mainly on p-aminophenol (pAP), because pAP is a good probe for an electrochemical immunoassay. The voltammograms showed that denser rGO provides better electrode reactivity for pAP. We confirmed that redox cycling between the anode and cathode at the rGO/Au-IDA was established, which yields more sensitive detection than with a single electrode. As one application of the electrochemical immunoassay using the rGO/Au-IDA, we demonstrated the quantitative detection of cortisol, a stress marker, at levels found in human saliva.

Selective detection of a catecholamine against electroactive interferents using an interdigitated heteroarray electrode consisting of a metal oxide electrode and a metal band electrode.

We developed an interdigitated array electrode (IDAE) consisting of a metal oxide electrode and a metal band heteroelectrode and employed it for the selective detection of catecholamines. We used an indium-tin oxide (ITO) film as the oxidation electrode of the IDAE because the ITO was able to suppress response currents from L-ascorbic acid (AA) and uric acid (UA), which are major electroactive interferents in biological fluids. However, the ITO film also suppresses the reduction of quinones including oxidized catecholamines. We developed a simple technique for fabricating our hetero IDAE, which also preserves the electrochemical properties of the ITO. When we compared hetero ITO-gold, homo ITO-ITO, and carbon-carbon IDAEs, we found that the hetero IDAE provided both high sensitivity and selectivity for DA detection. We achieved high selectivities for DA against AA and UA. The ratios of the response currents of AA and UA to DA were calculated as 6 and 5%, respectively.