Evaluation of a Continuous Blood Glucose Monitor: A Novel and Non-Invasive Wearable Using Bioimpedance Technology

Improvement of an Edge-IoT Architecture Driven by Artificial Intelligence for Smart-Health Chronic Disease Management

One of the health challenges in the 21st century is to rethink approaches to non-communicable disease prevention. A solution is a smart city that implements technology to make health smarter, enables healthcare access, and contributes to all residents' overall well-being. Thus, this paper proposes an architecture to deliver smart health. The architecture is anchored in the Internet of Things and edge computing, and it is driven by artificial intelligence to establish three foundational layers in smart care. Experimental results in a case study on glucose prediction noninvasively show that the architecture senses and acquires data that capture relevant characteristics. The study also establishes a baseline of twelve regression algorithms to assess the non-invasive glucose prediction performance regarding the mean squared error, root mean squared error, and r-squared score, and the catboost regressor outperforms the other models with 218.91 and 782.30 in MSE, 14.80 and 27.97 in RMSE, and 0.81 and 0.31 in R2, respectively, on training and test sets. Future research works involve extending the performance of the algorithms with new datasets, creating and optimizing embedded AI models, deploying edge-IoT with embedded AI for wearable devices, implementing an autonomous AI cloud engine, and implementing federated learning to deliver scalable smart health in a smart city context.

The effect of ionic redistributions on the microwave dielectric response of cytosol water upon glucose uptake

The sensitivity of cytosol water's microwave dielectric (MD) response to D-glucose uptake in Red Blood Cells (RBCs) allows the detailed study of cellular mechanisms as a function of controlled exposures to glucose and other related analytes like electrolytes. However, the underlying mechanism behind the sensitivity to glucose exposure remains a topic of debate. In this research, we utilize MDS within the frequency range of 0.5-40 GHz to explore how ionic redistributions within the cell impact the microwave dielectric characteristics associated with D-glucose uptake in RBC suspensions. Specifically, we compare glucose uptake in RBCs exposed to the physiological concentration of Ca2+ vs. Ca-free conditions. We also investigate the potential involvement of Na+/K+ redistribution in glucose-mediated dielectric response by studying RBCs treated with a specific Na+/K+ pump inhibitor, ouabain. We present some insights into the MD response of cytosol water when exposed to Ca2+ in the absence of D-glucose. The findings from this study confirm that ion-induced alterations in bound/bulk water balance do not affect the MD response of cytosol water during glucose uptake.

Prototype analysis of a low-power, small-scale wearable medical device

Wearable and portable devices are gaining significant popularity across consumer electronics as well as in medical and industrial fields. To ensure that these devices are both comfortable and appealing to users, they need to have low battery consumption and be compact in both size and weight. The EGluco project is focused on developing a wearable device for non-invasive blood glucose monitoring. This multi-sensor device incorporates electrical bioimpedance spectroscopy as one of its measurement techniques. One of the earlier versions of the device was deemed unsuitable as a wearable due to its large size and high power consumption. To make the device more suitable for wearability, the previous hardware was assessed, and a new design was proposed that simplified the system's power supply and reduced the operating voltage. This article presents two of these designs: an improved Howland current source with a supply voltage of 3.3 V, an output current of 250 μA, and the ability to conduct bioimpedance analysis up to 1 MHz using pulsed DIBS (Discrete Interval Binary Sequence) signals, and an instrumentation amplifier with the same supply voltage as the current source, a voltage gain of four, and a slew rate of 150 V/μs. By simplifying the power supply and implementing other changes, the device's size was reduced to a single 5 × 5 cm circuit board, compared to the previous configuration of four separate boards connected by cables.

Non-Invasive Glucose Sensing Technologies and Products: A Comprehensive Review for Researchers and Clinicians

Diabetes Mellitus incidence and its negative outcomes have dramatically increased worldwide and are expected to further increase in the future due to a combination of environmental and social factors. Several methods of measuring glucose concentration in various body compartments have been described in the literature over the years. Continuous advances in technology open the road to novel measuring methods and innovative measurement sites. The aim of this comprehensive review is to report all the methods and products for non-invasive glucose measurement described in the literature over the past five years that have been tested on both human subjects/samples and tissue models. A literature review was performed in the MDPI database, with 243 articles reviewed and 124 included in a narrative summary. Different comparisons of techniques focused on the mechanism of action, measurement site, and machine learning application, outlining the main advantages and disadvantages described/expected so far. This review represents a comprehensive guide for clinicians and industrial designers to sum the most recent results in non-invasive glucose sensing techniques' research and production to aid the progress in this promising field.

A comparison of continuous glucose monitors (CGMs) in diabetes management: A systematic literature review

BACKGROUND AND AIMS

As diabetes prevalence has continued to increase in the United States, as well as globally, utilization of disease management techniques has also improved. The evolution in disease management for diabetes has adapted greatly from the initial dipstix method. Continuous glucose monitors have grown in popularity with its introduction to the market. After introduction of CGMs as part of DM management, various advancements have been made to the current models to promote the usage CGMs to promote glycemic control. The main competitors in the CGM market is Medtronic, Dexcom, Freestyle, and Eversense.

METHODS

Information was primarily gathered by employing various PubMed scholarly articles for real-world examples in addition to data extraction from supplementary manuscripts. Articles were evaluated from over the past 20 years.

RESULTS

Clinically improvement of disease management of blood glucose levels, specifically with regards to mean absolute relative difference (MARD) was utilized to highlight effectiveness of continuous glucose monitors.

CONCLUSION

Of the four key continuous glucose monitors device on the market in the US, all have demonstrated to have similar beneficial qualities which can be utilized in both T1DM and T2DM patients. The best device for an individual would be based on their specific diabetes management goal (maintain TIR, decreasing TBR/TAR, decrease A1c).

Leveraging the future of diagnosis and management of diabetes: From old indexes to new technologies

BACKGROUND

Diabetes is a heterogeneous and multifactorial disease. However, glycemia and glycated hemoglobin have been the focus of diabetes diagnosis and management for the last decades. As diabetes management goes far beyond glucose control, it has become clear that assessment of other biochemical parameters gives a much wider view of the metabolic state of each individual, enabling a precision medicine approach.

METHODS

In this review, we summarize and discuss indexes that have been used in epidemiological studies and in the clinical practice.

RESULTS

Indexes of insulin secretion, sensitivity/resistance and metabolism have been developed and validated over the years to account also with insulin, C-peptide, triglycerides or even anthropometric measures. Nevertheless, each one has their own objective and consequently, advantages and disadvantages for specific cases. Thus, we discuss how new technologies, namely new sensors but also new softwares/applications, can improve the diagnosis and management of diabetes, both for healthcare professionals but also for caretakers and, importantly, to promote the empowerment of people living with diabetes.

CONCLUSIONS

In long-term, the solution for a better diabetes management would be a platform that allows to integrate all sorts of relevant information for the person with diabetes and for the healthcare practitioners, namely glucose, insulin and C-peptide or, in case of need, other parameters/indexes at home, sometimes more than once a day. This solution would allow a better and simpler disease management, more adequate therapeutics thereby improving patients' quality of life and reducing associated costs.

Sense and Learn: Recent Advances in Wearable Sensing and Machine Learning for Blood Glucose Monitoring and Trend-Detection

Diabetes mellitus is characterized by elevated blood glucose levels, however patients with diabetes may also develop hypoglycemia due to treatment. There is an increasing demand for non-invasive blood glucose monitoring and trends detection amongst people with diabetes and healthy individuals, especially athletes. Wearable devices and non-invasive sensors for blood glucose monitoring have witnessed considerable advances. This review is an update on recent contributions utilizing novel sensing technologies over the past five years which include electrocardiogram, electromagnetic, bioimpedance, photoplethysmography, and acceleration measures as well as bodily fluid glucose sensors to monitor glucose and trend detection. We also review methods that use machine learning algorithms to predict blood glucose trends, especially for high risk events such as hypoglycemia. Convolutional and recurrent neural networks, support vector machines, and decision trees are examples of such machine learning algorithms. Finally, we address the key limitations and challenges of these studies and provide recommendations for future work.