Non-invasive Blood Glucose Measurement Research Articles

Assessment of Optical Attenuation and Skin Thickness in Type 2 Diabetes Mellitus Patients Using Optical CoherenceTomography.

Diabetes management often involves invasive blood glucose monitoring, which can be uncomfortable for patients. Non-invasive techniques like multiple μ-spatially offset Raman spectroscopy (mμSORS) offer a promising alternative. To provide clinical data supporting mμSORS, we conducted a clinical trial with 198 participants to evaluate mμSORS for non-invasive blood glucose measurement. Using Optical Coherence Tomography, we studied skin thickness and optical attenuation in 172 diabetic and 26 healthy subjects. Results showed thicker stratum corneum and stratum spinosum (SS) in diabetics. Epidermal thickness increased with age and body mass index (BMI), decreased with skin brightness, and varied minimally with gender. Optical attenuation in SS was lower in diabetics, decreased with increasing a*, and was minimally affected by gender and BMI but increased with age in the upper dermis. These findings support mμSORS for accurate non-invasive glucose monitoring.

Evaluation and Validation on Sensitivity of Near-Infrared Diffuse Reflectance in Non-Invasive Human Blood Glucose Measurement.

In non-invasive blood glucose measurement, the sensitivity of glucose-induced optical signals within human tissue is a crucial reference point. This study evaluates the sensitivity of glucose-induced diffuse reflectance in the 1000-1700 nm range. A key factor in understanding this sensitivity is the rate at which the scattering coefficient changes due to glucose, as it is significantly higher than in non-living media and predominantly influences the diffuse light signal level when blood glucose levels change. The study measured and calculated the changes in the scattering coefficient at 1314 nm, a wavelength chosen for its minimal interference from glucose absorption and other bodily constituents. Based on the Mie scattering theory and the results at 1314 nm, the changes in the scattering coefficient within the 1000-1700 nm range were estimated. Subsequently, the sensitivity of the glucose signal across this range was determined through Monte Carlo (MC) simulations. The findings from 25 human trials indicate that the measured sensitivities at five other typical wavelengths within this band generally align with the sensitivities calculated using the aforementioned method. This research can guide the identification of blood glucose signals and the selection of wavelengths for non-invasive blood glucose measurements.

Flexible cellulose paper-based biosensor from inkjet printing for non-invasive glucose monitoring

Wearable glucose sensors have attracted significant attention for enabling non-invasive blood glucose measurement without discomfort and risk of infection. However, it is a challenge to simultaneously realize wearable adaptability, biodegradability, and excellent sensing performance. Herein, a cellulose paper-based non-invasive biosensor relying on reverse iontophoresis was designed to detect glucose in interstitial fluid (ISF), and two different enzyme immobilization strategies have been compared. The results showed inkjet-printed cellulose paper-based biosensor (IPB) performances better than the drop-coated cellulose paper-based biosensor (DPB). IPB has twice response current more than DPB in detection range (0–10 mM). In sensitivity, IPB is 1.170 μA/mM three times higher than the DPB (0.376 μA/mM). Besides, IPB's electron-transfer resistance (Rct) is 7.27 kΩ smaller than DPB (Rct = 10.51 kΩ) about 30 %. More importantly, IPB exhibited a good reproducibility (RSD, 4.82 %), which was much less than DPB (RSD, 18.35 %). Furthermore, the IPB realizes noninvasive continuous glucose monitoring over 6 h in volunteer experiments with great analytical performance comparable to commercial devices (Pearson correlation 0.732). Cellulose paper-based glucose sensors with inkjet printing provide non-invasive access to statistically significant diagnostic information, simple and cost-effective, which promotes the application of flexible, wearable, degradable bioelectrodes in continuous glucose monitoring at home, providing a concept for full integration in a compact and portable way in the future.

Healthcare and the Internet of Medical Things: Applications, Trends, Key Challenges, and Proposed Resolutions

In recent years, the Internet of medical things (IoMT) has become a significant technological advancement in the healthcare sector. This systematic review aims to identify and summarize the various applications, key challenges, and proposed technical solutions within this domain, based on a comprehensive analysis of the existing literature. This review highlights diverse applications of the IoMT, including mobile health (mHealth) applications, remote biomarker detection, hybrid RFID-IoT solutions for scrub distribution in operating rooms, IoT-based disease prediction using machine learning, and the efficient sharing of personal health records through searchable symmetric encryption, blockchain, and IPFS. Other notable applications include remote healthcare management systems, non-invasive real-time blood glucose measurement devices, distributed ledger technology (DLT) platforms, ultra-wideband (UWB) radar systems, IoT-based pulse oximeters, accident and emergency informatics (A&EI), and integrated wearable smart patches. The key challenges identified include privacy protection, sustainable power sources, sensor intelligence, human adaptation to sensors, data speed, device reliability, and storage efficiency. The proposed mitigations encompass network control, cryptography, edge-fog computing, and blockchain, alongside rigorous risk planning. The review also identifies trends and advancements in the IoMT architecture, remote monitoring innovations, the integration of machine learning and AI, and enhanced security measures. This review makes several novel contributions compared to the existing literature, including (1) a comprehensive categorization of IoMT applications, extending beyond the traditional use cases to include emerging technologies such as UWB radar systems and DLT platforms; (2) an in-depth analysis of the integration of machine learning and AI in IoMT, highlighting innovative approaches in disease prediction and remote monitoring; (3) a detailed examination of privacy and security measures, proposing advanced cryptographic solutions and blockchain implementations to enhance data protection; and (4) the identification of future research directions, providing a roadmap for addressing current limitations and advancing the scientific understanding of IoMT in healthcare. By addressing current limitations and suggesting future research directions, this work aims to advance scientific understanding of the IoMT in healthcare.

Comparison of Ensemble Method Performance in Classifying Blood Sugar Levels Output from Non-Invasive Device

Diabetes Mellitus (DM) is a persistent health issue in many countries and is a leading cause of heart disease, kidney failure, and blindness The International Diabetes Federation (IDF) estimated in 2019 that at least 463 million people worldwide aged 20-79 suffer from diabetes. This number is expected to rise to 578 million by 2030 and 700 million by 2045. Machine learning is a type of machine learning that is very helpful in various fields, including healthcare. In classification cases, ensemble methods classify by combining decisions from several other models, one way being through majority voting. Ensemble methods often produce more accurate classification or prediction results. Several ensemble methods include random forest, extra trees, rotation forest, and double random forest. The data used in this study is part of research on the development and clinical testing of a prototype non-invasive blood glucose monitoring device by the non-invasive biomarking team at IPB. The data includes both invasive and non-invasive blood glucose measurements collected in 2019. This study compares the performance of the random forest, extra trees, rotation forest, and double random forest models on blood glucose level data obtained from non-invasive devices. The research results show that the Rotation Forest algorithm is the best model, with the highest average accuracy compared to the other three algorithms, achieving an accuracy level of 0.7142857 (71.42%).

Multi-View Cross-Fusion Transformer Based on Kinetic Features for Non-Invasive Blood Glucose Measurement Using PPG Signal.

Noninvasive blood glucose (BG) measurement could significantly improve the prevention and management of diabetes. In this paper, we present a robust novel paradigm based on analyzing photoplethysmogram (PPG) signals. The method includes signal pre-processing optimization and a multi-view cross-fusion transformer (MvCFT) network for non-invasive BG assessment. Specifically, a multi-size weighted fitting (MSWF) time-domain filtering algorithm is proposed to optimally preserve the most authentic morphological features of the original signals. Meanwhile, the spatial position encoding-based kinetics features are reconstructed and embedded as prior knowledge to discern the implicit physiological patterns. In addition, a cross-view feature fusion (CVFF) module is designed to incorporate pairwise mutual information among different views to adequately capture the potential complementary features in physiological sequences. Finally, the subject- wise 5- fold cross-validation is performed on a clinical dataset of 260 subjects. The root mean square error (RMSE) and mean absolute error (MAE) of BG measurements are 1.129mmol/L and 0.659mmol/L, respectively, and the optimal Zone A in the Clark error grid, representing none clinical risk, is 87.89%. The results indicate that the proposed method has great potential for homecare applications.

Accuracy of a Non-Invasive Home Glucose Monitor for Measurement of Blood Glucose.

Patients with diabetes mellitus monitor their blood glucose at home with monitors that require a drop of blood or use a continuous glucose monitoring device that implants a small needle in the body. However, both cause discomfort to the patients which may inhibit them for regular blood glucose checks. Photoplethysmogram (PPG) sensing technology is an approach for non-invasive blood glucose measurement and PPG sensors can be used to predict hypoglycaemic episodes. InChcek is a PPG-based non-invasive glucose monitor. However, its accuracy has not been checked yet. Hence, this study aimed to evaluate the accuracy of InCheck, a non-invasive glucose monitor for the estimation of blood glucose. In a tertiary care hospital, patients who came for blood glucose estimation were tested for blood glucose non-invasively on the InCheck device and then by the laboratory method (glucose oxidase-peroxidase). These two readings were compared. We used International Organization for Standardization (ISO) 15197:2013 (95% of values should be within ± 15 mg/dL of reference reading if reference glucose <100 mg/dL or within ± 15% of reference reading if reference glucose ≥100 mg/dL and 99% of the values should be within zones A and B in consensus error grid), and Surveillance Error Grid for analyzing the accuracy. A total of 1223 samples were analyzed. There was a significant difference between the reference method glucose level (135 [Q1-Q3: 97 - 179] mg/dL) and monitor-measured glucose level (188.33 [Q1-Q3: 167.33-209.33] mg/dL) (P < 0.0001). A total of 18.5% of readings were following ISO 15197:2013 criteria and 67.25% of coordinates were within zone A and zone B of the consensus error grid. In the surveillance error grid analysis, about 29.4% of values were in the no-risk zone, 51.8% in slight risk, 18.6% in moderate risk, and 0.2% were in the severe risk zone. The accuracy of the InCheck device for the estimation of blood glucose by PPG signal is not following the recommended guidelines. Hence, further research is necessary for programming or redesigning the hardware and software for a better result from this optical sensor-based non-invasive home glucose monitor.

Implicit HbA1c Achieving 87% Accuracy within 90 Days in Non-Invasive Fasting Blood Glucose Measurements Using Photoplethysmography.

To reduce the error induced by overfitting or underfitting in predicting non-invasive fasting blood glucose (NIBG) levels using photoplethysmography (PPG) data alone, we previously demonstrated that incorporating HbA1c led to a notable 10% improvement in NIBG prediction accuracy (the ratio in zone A of Clarke's error grid). However, this enhancement came at the cost of requiring an additional HbA1c measurement, thus being unfriendly to users. In this study, the enhanced HbA1c NIBG deep learning model (blood glucose level predicted from PPG and HbA1c) was trained with 1494 measurements, and we replaced the HbA1c measurement (explicit HbA1c) with "implicit HbA1c" which is reversely derived from pretested PPG and finger-pricked blood glucose levels. The implicit HbA1c is then evaluated across intervals up to 90 days since the pretest, achieving an impressive 87% accuracy, while the remaining 13% falls near the CEG zone A boundary. The implicit HbA1c approach exhibits a remarkable 16% improvement over the explicit HbA1c method by covering personal correction items automatically. This improvement not only refines the accuracy of the model but also enhances the practicality of the previously proposed model that relied on an HbA1c input. The nonparametric Wilcoxon paired test conducted on the percentage error of implicit and explicit HbA1c prediction results reveals a substantial difference, with a p-value of 2.75 × 10-7.

Personalized diabetes monitoring platform leveraging IoMT and AI for non-invasive estimation

Non-invasive blood glucose estimation is an extensively researched area since current gold-standard invasive glucose monitoring methods present numerous inconveniences and challenges in terms of comfort and cost. We present the design, development, and validation of an Internet of Medical Things (IoMT) based wearable device for non-invasive and real-time measurement of blood glucose. This paper presents a diabetic health monitoring platform architecture that consists of (a) a user-worn photoplethysmography (PPG) device, (b) a smart analytics cloud that deploys models for blood glucose estimation, and (c) an end-to-end mobile/web application for monitoring diabetes patients. Blood glucose computation is achieved using a novel light-weight 1-dimensional input-reinforced deep neural network architecture, which we call as GlucoNet. This captures both long and short, temporal and spatial features from the PPG signal. The training and validation of the model were conducted on a dataset of 283 participants which demonstrated a mean absolute percentage error (MAPE) of 17.8% (± 12.8%) wherein 100% of predictions fall in the clinically acceptable zones A and B of the Clarke-error grid. The lightweight model is also deployed on edge devices for real-time and offline blood glucose measurement. We report a clinical outcome deployment study and insights from 20,000＋ glucose measurements obtained from another 600 patients. To our knowledge, this is the largest reported work employing a non-calibrated, non-invasive, demography, and time-of-food agnostic IoMT glucose monitoring system that does not require any feature engineering and is capable of running on edge devices.

Progress and Perspectives of Mid-Infrared Photoacoustic Spectroscopy for Non-Invasive Glucose Detection.

The prevalence of diabetes is rapidly increasing worldwide and can lead to a range of severe health complications that have the potential to be life-threatening. Patients need to monitor and control blood glucose levels as it has no cure. The development of non-invasive techniques for the measurement of blood glucose based on photoacoustic spectroscopy (PAS) has advanced tremendously in the last couple of years. Among them, PAS in the mid-infrared (MIR) region shows great promise as it shows the distinct fingerprint region for glucose. However, two problems are generally encountered when it is applied to monitor real samples for in vivo measurements in this MIR spectral range: (i) low penetration depth of MIR light into the human skin, and (ii) the effect of other interfering components in blood, which affects the selectivity of the detection system. This review paper systematically describes the basics of PAS in the MIR region, along with recent developments, technical challenges, and data analysis strategies, and proposes improvements for the detection sensitivity of glucose concentration in human bodies. It also highlights the recent trends of incorporating machine learning (ML) to enhance the detection sensitivity of the overall system. With further optimization of the experimental setup and incorporation of ML, this PAS in the MIR spectral region could be a viable solution for the non-invasive measurement of blood glucose in the near future.

Estimation of the Differential Pathlength Factor for Human Skin Using Monte Carlo Simulations.

Near-infrared technology is an emerging non-invasive technique utilized for various medical applications. Recently, there have been many attempts to utilize NIR technology for the continues monitoring of blood glucose levels through the skin. Different approaches and designs have been proposed for non-invasive blood glucose measurements. Light photons penetrating the skin can undergo multiple scattering events, and the actual optical pathlength becomes larger than the source-to-detector separation (optode spacing) in the reflection-mode configuration. Thus, the differential pathlength factor (DPF) must be incorporated into the modified Beer-Lambert law. The accurate estimation of the DPF values will lead to an accurate quantification of the physiological variations within the tissue. In this work, the aim was to systematically estimate the DPF for human skin for a range of source-to-detector separations and wavelengths. The Monte Carlo (MC) method was utilized to mimic the different layers of human skin with different optical properties and blood and water volume fractions. This work could help improve the accuracy of the near-infrared technique in the measurement of physiological variations within skin tissue.

Basic Study on Measurement of Return Loss and Smith Chart Change Using Microstrip Patch Antenna with Concentration Transition for Non-invasive Blood Glucose Measurement.

This paper describes a basic study on the measurement of return loss and change in Smith chart using a microstrip patch antenna (MPA) with concentration transition to perform non-invasive blood glucose measurements. To evaluate blood glucose level changes in the human body, the concentration of measurements was changed 10 times to an equivalent of 100mg/dL in a range of 0-1000mg/dL to reflect a concentration of 400mg/dL, which is the fatal level of diabetes. Five types of MPAs were fabricated that formed resonant frequencies in the 1, 2, 3, 4, and 5GHz bands, and were used in the experiment. Each MPA constituted a sharp narrow band characteristic and induced a large change in return loss. By measuring the return loss and Smith chart to evaluate the concentration change at resonant frequencies, the return loss was observed to change by an average of 0.058 dB for every 100mg/dL change, and the impedance magnitudes and phase angles analyzed through the Smith chart were observed to have a certain tendency, which confirmed that they were changed. This study shows that for performing non-invasive measurements of blood sugar level, measuring the change in return loss can provide a more stable and reliable measurement compared to the two methods that are simultaneously used to view the samples.

Near-Infrared Spectroscopy for Noninvasive Measurement of Blood Glucose: Problems, Progress, Tasks

Near-Infrared Spectroscopy for Noninvasive Measurement of Blood Glucose: Problems, Progress, Tasks

Deduction learning for precise noninvasive measurements of blood glucose with a dozen rounds of data for model training

Personalized modeling has long been anticipated to approach precise noninvasive blood glucose measurements, but challenged by limited data for training personal model and its unavoidable outlier predictions. To overcome these long-standing problems, we largely enhanced the training efficiency with the limited personal data by an innovative Deduction Learning (DL), instead of the conventional Induction Learning (IL). The domain theory of our deductive method, DL, made use of accumulated comparison of paired inputs leading to corrections to preceded measured blood glucose to construct our deep neural network architecture. DL method involves the use of paired adjacent rounds of finger pulsation Photoplethysmography signal recordings as the input to a convolutional-neural-network (CNN) based deep learning model. Our study reveals that CNN filters of DL model generated extra and non-uniform feature patterns than that of IL models, which suggests DL is superior to IL in terms of learning efficiency under limited training data. Among 30 diabetic patients as our recruited volunteers, DL model achieved 80% of test prediction in zone A of Clarke Error Grid (CEG) for model training with 12 rounds of data, which was 20% improvement over IL method. Furthermore, we developed an automatic screening algorithm to delete low confidence outlier predictions. With only a dozen rounds of training data, DL with automatic screening achieved a correlation coefficient ({R}_{P}) of 0.81, an accuracy score ({R}_{A}) of 93.5, a root mean squared error of 13.93 mg/dl, a mean absolute error of 12.07 mg/dl, and 100% predictions in zone A of CEG. The nonparametric Wilcoxon paired test on {R}_{A} for DL versus IL revealed near significant difference with p-value 0.06. These significant improvements indicate that a very simple and precise noninvasive measurement of blood glucose concentration is achievable.

Diffuse Imaging Approach for Universal Noninvasive Blood Glucose Measurements

We proposed a diffuse imaging approach for universal noninvasive blood glucose measurements based on visible light, which can predict the blood glucose concentration without personal calibration. The proposed approach used a CCD to obtain diffuse images from human index finger pulp. The denoising autoencoder algorithm adopted effectively extracted the scattering information highly related to blood glucose concentration from the diffuse images, and the gradient boosting regression algorithm enabled an accurate calculation of blood glucose concentration without prior personalized calibration.In vivoexperimental results showed that the proposed approach had a mean absolute error of 19.44 mg/dl, with all the predicted results observed within the clinically acceptable region (Region A: 78.9%) in the Clarke error grid analysis. Compared to other blood glucose concentration measurement methods of scattering coefficient, this new method does not require individual calibration, therefore it is easier to implement and popularize, which is critical for the noninvasive monitoring of blood glucose concentration.

90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c

Previously published photoplethysmography-(PPG) based non-invasive blood glucose (NIBG) measurements have not yet been validated over 500 subjects. As illustrated in this work, we increased the number subjects recruited to 2538 and found that the prediction accuracy (the ratio in zone A of Clarke’s error grid) reduced to undesirable 60.6%. We suspect the low prediction accuracy induced by larger sample size might arise from the physiological diversity of subjects, and one possibility is that the diversity might originate from medication. Therefore, we split the subjects into two cohorts for deep learning: with and without medication (1682 and 856 recruited subjects, respectively). In comparison, the cohort training for subjects without any medication had approximately 30% higher prediction accuracy over the cohort training for those with medication. Furthermore, by adding quarterly (every 3 months) measured glycohemoglobin (HbA1c), we were able to significantly boost the prediction accuracy by approximately 10%. For subjects without medication, the best performing model with quarterly measured HbA1c achieved 94.3% prediction accuracy, RMSE of 12.4 mg/dL, MAE of 8.9 mg/dL, and MAPE of 0.08, which demonstrates a very promising solution for NIBG prediction via deep learning. Regarding subjects with medication, a personalized model could be a viable means of further investigation.

Human body model with blood flow properties for non-invasive blood glucose measurement

In this paper, the authors propose a human body model with the final goal of developing non-invasive blood glucose measurement. Most wearable devices today include antennas and signal processing circuitry capable of non-invasive measurements. Authors plan to extend this circuitry for non-invasive blood glucose measurement with microwave technique. An optimum design of a proposed system needs a human body model. The authors have studied and compared various proposed methods in the published literature for non-invasive blood glucose measurement. Comparative study has concluded that the published human body models have dielectric-based tissue modelling and it is difficult to change the blood glucose concentration in it. Hence, a new human body model is derived and implemented in this work. The proposed model has provision to imitate the human blood flow and can vary blood glucose concentration for simulation purposes. Its accuracy has been defined and the proposed model was compared with standard dataset. It is established in simulation that with changes in blood glucose concentration the dielectric constant, electrical conductivity and other electrical parameters of blood also vary. Based on this, the parameters of the antenna mounted on body are affected. An antenna is fabricated to compare the simulation results with the actual test. It is found that blood glucose variation on extreme ends of the glucose concentration has return loss variation in range of 4–6 dB, while in 170 mg/dL–450 mg/dL range the variation is in between 1 and 3 dB values.

Features development in machine learning model for non-invasive blood-glucose measurement

Diabetes mellitus is the result of changes in the body caused by the decrease of insulin performance which is characterized by an increase in blood sugar level. Detection of blood sugar can be done with an invasive method or a non-invasive method. However, the non-invasive methods are considered better because they do not hurt the body and can check earlier, faster, and more accurately. A non-invasive blood glucose meter has been developed by a research team at IPB University. The output of the non-invasive tool is the intensity of the residual spectrum data, which will be related to the result of invasive measurement of blood glucose using some classification models, i.e. support vector machine and random forest. This research is aimed to compare the features development methods from the output of non-invasive tools and get the best features for modeling that can provide better predictions. The result of feature development shows that the best feature in the output of the non-invasive device is the trapezoidal area method at period because it has a higher accuracy value than the other four methods. The validation process shows that the random forest method has a higher accuracy value compared to the support vector machine.

Design of intelligent diabetes mellitus detection system using hybrid feature selection based XGBoost classifier

In this work, a non-invasive diabetes mellitus detection system is proposed based on the wristband photoplethysmography (PPG) signal and basic physiological parameters (PhyP) to enable easy detection of diabetes mellitus (DM). A dataset of 217 participants with diabetes, prediabetes and normal conditions is used to develop the system. The Mel frequency cepstral coefficients (MFCC) extracted from 5s PPG signal segments and the PhyP are used as input for the machine learning algorithms. The K-nearest neighbors, support vector machine, random forest and extreme gradient boost (XGBoost) classifiers are used for classification. In addition, a hybrid feature selection method (Hybrid FS) is proposed to reduce the size of the input data. The Hybrid FS-based XGBoost system achieves a high accuracy of 99.93 % for non-invasive diabetes detection with fewer features and less computational effort. The analysis suggests that the PPG signal from a wearable sensor is a good alternative for simple non-invasive blood glucose measurements in routine applications.

Towards non-invasive blood glucose measurement using machine learning: An all-purpose PPG system design

Diabetes, the result of excessive or uncontrolled glucose in the blood, is one of the leading causes of human mortality. Due to the unavailability of non-invasive glucose level checker until now, the most trustworthy day-to-day life glucose test for personal healthcare is the use of glucometer in which case painful finger pricking is an obvious part. However, researches have been done to prove the usage of pulse oximeter to measure the blood glucose level besides other physiological indicators such as heart rate, percentage of blood oxygen, etc. Here, as the first of two studies, we try to develop an all-purpose commercial prototype photoplethysmography (PPG) system to monitor necessary health indicator parameters in a non-invasive way. The developed fingertip PPG device consists of both transmissive and reflective type data acquisition system after illuminating the skin with red, green, and IR LEDs. Next, as the second study, special consideration is given to prove the efficiency of the device for measuring blood glucose level (BGL). To measure blood glucose from PPG signal, a few discriminative and related features are extracted from the obtained PPG signals. Machine learning algorithms are employed to predict the actual value of BGL from the extracted features. The proposed algorithm and system can predict the BGL level with a level of clinical accuracy. In the Clarke error grid plot, 96.15% and 3.85% of data are in the zone A and zone B, respectively, with 0% data in the critical zones.