Non-Invasive In Vivo Estimation of HbA1c Using Monte Carlo Photon Propagation Simulation: Application of Tissue-Segmented 3D MRI Stacks of the Fingertip and Wrist for Wearable Systems.

The early diagnosis of diabetes mellitus in normal people or maintaining stable blood sugar concentrations in diabetic patients requires frequent monitoring of the blood sugar levels. However, regular monitoring of the sugar levels is problematic owing to the pain and inconvenience associated with pricking the fingertip or using minimally invasive patches. In this study, we devise a noninvasive method to estimate the percentage of the in vivo glycated hemoglobin (HbA1c) values from Monte Carlo photon propagation simulations, based on models of the wrist using 3D magnetic resonance (MR) image data. The MR image slices are first segmented for several different tissue types, and the proposed Monte Carlo photon propagation system with complex composite tissue support is then used to derive several models for the fingertip and wrist sections with different wavelengths of light sources and photodetector arrangements. The Pearson r values for the estimated percent HbA1c values are 0.94 and 0.96 for the fingertip transmission- and reflection-type measurements, respectively. This is found to be the best among the related studies. Furthermore, a single-detector multiple-source arrangement resulted in a Pearson r value of 0.97 for the wrist. The Bland-Altman bias values were found to be -0.003 ± 0.36, 0.01 ± 0.25, and 0.01 ± 0.21, for the two fingertip and wrist models, respectively, which conform to the standards of the current state-of-the-art invasive point-of-care devices. The implementation of these algorithms will be a suitable alternative to the invasive state-of-the-art methods.

A Comparative Analysis of Various Machine Learning Algorithms to Improve the Accuracy of HbA1c Estimation Using Wrist PPG Data.

Due to the inconvenience of drawing blood and the possibility of infection associated with invasive methods, research on non-invasive glycated hemoglobin (HbA1c) measurement methods is increasing. Utilizing wrist photoplethysmography (PPG) with machine learning to estimate HbA1c can be a promising method for non-invasive HbA1c monitoring in diabetic patients. This study aims to develop a HbA1c estimation system based on machine learning algorithms using PPG signals obtained from the wrist. We used a PPG based dataset of 22 subjects and algorithms such as extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), Categorical Boost (CatBoost) and random forest (RF) to estimate the HbA1c values. Note that the AC-to-DC ratios for three wavelengths were newly adopted as features in addition to the previously acquired 15 features from the PPG signal and a comparative analysis was performed between the performances of several algorithms. We showed that feature-importance-based selection can improve performance while reducing computational complexity. We also showed that AC-to-DC ratio (AC/DC) features play a dominant role in improving HbA1c estimation performance and, furthermore, a good performance can be obtained without the need for external features such as BMI and SpO2. These findings may help shape the future of wrist-based HbA1c estimation (e.g., via a wristwatch or wristband), which could increase the scope of noninvasive and effective monitoring techniques for diabetic patients.

Machine-Learning-Based Noninvasive In Vivo Estimation of HbA1c Using Photoplethysmography Signals.

Glycated hemoglobin (HbA1c) is an important factor in monitoring diabetes. Since the glycated hemoglobin value reflects the average blood glucose level over 3 months, it is not affected by exercise or food intake immediately prior to measurement. Thus, it is used as the most basic measure of evaluating blood-glucose control over a certain period and predicting the occurrence of long-term complications due to diabetes. However, as the existing measurement methods are invasive, there is a burden on the measurement subject who has to endure increased blood gathering and exposure to the risk of secondary infections. To overcome this problem, we propose a machine-learning-based noninvasive estimation method in this study using photoplethysmography (PPG) signals. First, the development of the device used to acquire the PPG signals is described in detail. Thereafter, discriminative and effective features are extracted from the acquired PPG signals using the device, and a machine-learning algorithm is used to estimate the glycated hemoglobin value from the extracted features. Finally, the performance of the proposed method is evaluated by comparison with existing model-based methods.

Cuffless Blood Pressure Estimation Based on Monte Carlo Simulation Using Photoplethysmography Signals.

Blood pressure measurements are one of the most routinely performed medical tests globally. Blood pressure is an important metric since it provides information that can be used to diagnose several vascular diseases. Conventional blood pressure measurement systems use cuff-based devices to measure the blood pressure, which may be uncomfortable and sometimes burdensome to the subjects. Therefore, in this study, we propose a cuffless blood pressure estimation model based on Monte Carlo simulation (MCS). We propose a heterogeneous finger model for the MCS at wavelengths of 905 nm and 940 nm. After recording the photon intensities from the MCS over a certain range of blood pressure values, the actual photoplethysmography (PPG) signals were used to estimate blood pressure. We used both publicly available and self-made datasets to evaluate the performance of the proposed model. In case of the publicly available dataset for transmission-type MCS, the mean absolute errors are 3.32 ± 6.03 mmHg for systolic blood pressure (SBP), 2.02 ± 2.64 mmHg for diastolic blood pressure (DBP), and 1.76 ± 2.8 mmHg for mean arterial pressure (MAP). The self-made dataset is used for both transmission- and reflection-type MCSs; its mean absolute errors are 2.54 ± 4.24 mmHg for SBP, 1.49 ± 2.82 mmHg for DBP, and 1.51 ± 2.41 mmHg for MAP in the transmission-type case as well as 3.35 ± 5.06 mmHg for SBP, 2.07 ± 2.83 mmHg for DBP, and 2.12 ± 2.83 mmHg for MAP in the reflection-type case. The estimated results of the SBP and DBP satisfy the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) standards and are within Grade A according to the British Hypertension Society (BHS) standards. These results show that the proposed model is efficient for estimating blood pressures using fingertip PPG signals.

Deep-Learning-Based Adaptive Symbol Decision for Visual MIMO System with Variable Channel Modeling.

A channel modeling method and deep-learning-based symbol decision method are proposed to improve the performance of a visual MIMO system for communication between a variable-color LED array and camera. Although image processing algorithms using color clustering are available to correct distorted color information in a channel, color-similarity-based approaches are limited by real-world distortions; to overcome such limitations, symbol decision is defined as a multiclass classification problem. Further, to learn a robust classifier against channel distortion, a deep neural network learning technique is applied to adaptively determine symbols from channel distortion. The network designed herein comprises the channel identification and symbol decision modules; the channel identification module extracts a channel identification vector for symbol determination from an input image using a two-dimensional deep convolutional neural network (CNN); the symbol decision module then generates a feature map by combining the channel identification vector and information on adjacent symbols to determine the symbol via learning correlations between adjacent symbols using a one-dimensional CNN. The two modules are connected together and learned simultaneously in an end-to-end manner. We also propose a new channel modeling method that intuitively reflects real-world distortion factors rather than the conventional additive white Gaussian noise channel to efficiently train deep-learning networks. Lastly, in the proposed channel distortion environment, the proposed method shows performance improvement by an average of about 41.8% (up to about 54.8%) compared to the existing Euclidean distance method, and about 6.3% (up to about 9.2%) on average compared to the SVM method.

Optical Measurement of Molar Absorption Coefficient of HbA1c: Comparison of Theoretical and Experimental Results.

Diabetes can cause dangerous complications if not diagnosed in a timely manner. The World Health Organization accepts glycated hemoglobin (HbA1c) as a measure of diagnosing diabetes as it provides significantly more information on the glycemic behavior from a single blood sample than the fasting blood sugar reading. The molar absorption coefficient of HbA1c is needed to quantify the amount of HbA1c present in a blood sample. In this study, we measured the molar absorption coefficient of HbA1c in the range of 450 nm to 700 nm using optical methods experimentally. We observed that the characteristic peaks of the molar absorption coefficient of HbA1c (at 545 nm and 579 nm for level 1, at 544 nm and 577 nm for level 2) are in close agreement with those reported in previous studies. The molar absorption coefficient values were also found to be close to those of earlier reports. The average molar absorption coefficient values of HbA1c were found to be 804,403.5 M−1cm−1 at 545 nm and 703,704.5 M−1cm−1 at 579 nm for level 1 as well as 503,352.4 M−1cm−1 at 544 nm and 476,344.6 M−1cm−1 at 577 nm for level 2. Our experiments focused on calculating the molar absorption coefficients of HbA1c in the visible wavelength region, and the proposed experimental method has an advantage of being able to easily obtain the molar absorption coefficient at any wavelength in the visible wavelength region. The results of this study are expected to help future investigations on noninvasive methods of estimating HbA1c levels.

Noninvasive In Vivo Estimation of Blood-Glucose Concentration by Monte Carlo Simulation.

Continuous monitoring of blood-glucose concentrations is essential for both diabetic and nondiabetic patients to plan a healthy lifestyle. Noninvasive in vivo blood-glucose measurements help reduce the pain of piercing human fingertips to collect blood. To facilitate noninvasive measurements, this work proposes a Monte Carlo photon simulation-based model to estimate blood-glucose concentration via photoplethysmography (PPG) on the fingertip. A heterogeneous finger model was exposed to light at 660 nm and 940 nm in the reflectance mode of PPG via Monte Carlo photon propagation. The bio-optical properties of the finger model were also deduced to design the photon simulation model for the finger layers. The intensities of the detected photons after simulation with the model were used to estimate the blood-glucose concentrations using a supervised machine-learning model, XGBoost. The XGBoost model was trained with synthetic data obtained from the Monte Carlo simulations and tested with both synthetic and real data (n = 35). For testing with synthetic data, the Pearson correlation coefficient (Pearson's r) of the model was found to be 0.91, and the coefficient of determination (R2) was found to be 0.83. On the other hand, for tests with real data, the Pearson's r of the model was 0.85, and R2 was 0.68. Error grid analysis and Bland-Altman analysis were also performed to confirm the accuracy. The results presented herein provide the necessary steps for noninvasive in vivo blood-glucose concentration estimation.

Time-Sharing-Based Synchronization and Performance Evaluation of Color-Independent Visual-MIMO Communication.

In the field of communication, synchronization is always an important issue. The communication between a light-emitting diode (LED) array (LEA) and a camera is known as visual multiple-input multiple-output (MIMO), for which the data transmitter and receiver must be synchronized for seamless communication. In visual-MIMO, LEDs generally have a faster data rate than the camera. Hence, we propose an effective time-sharing-based synchronization technique with its color-independent characteristics providing the key to overcome this synchronization problem in visual-MIMO communication. We also evaluated the performance of our synchronization technique by varying the distance between the LEA and camera. A graphical analysis is also presented to compare the symbol error rate (SER) at different distances.

LEA Detection and Tracking Method for Color-Independent Visual-MIMO.

Communication performance in the color-independent visual-multiple input multiple output (visual-MIMO) technique is deteriorated by light emitting array (LEA) detection and tracking errors in the received image because the image sensor included in the camera must be used as the receiver in the visual-MIMO system. In this paper, in order to improve detection reliability, we first set up the color-space-based region of interest (ROI) in which an LEA is likely to be placed, and then use the Harris corner detection method. Next, we use Kalman filtering for robust tracking by predicting the most probable location of the LEA when the relative position between the camera and the LEA varies. In the last step of our proposed method, the perspective projection is used to correct the distorted image, which can improve the symbol decision accuracy. Finally, through numerical simulation, we show the possibility of robust detection and tracking of the LEA, which results in a symbol error rate (SER) performance improvement.

Color-Space-Based Visual-MIMO for V2X Communication.

In this paper, we analyze the applicability of color-space-based, color-independent visual-MIMO for V2X. We aim to achieve a visual-MIMO scheme that can maintain the original color and brightness while performing seamless communication. We consider two scenarios of GCM based visual-MIMO for V2X. One is a multipath transmission using visual-MIMO networking and the other is multi-node V2X communication. In the scenario of multipath transmission, we analyze the channel capacity numerically and we illustrate the significance of networking information such as distance, reference color (symbol), and multiplexing-diversity mode transitions. In addition, in the V2X scenario of multiple access, we may achieve the simultaneous multiple access communication without node interferences by dividing the communication area using image processing. Finally, through numerical simulation, we show the superior SER performance of the visual-MIMO scheme compared with LED-PD communication and show the numerical result of the GCM based visual-MIMO channel capacity versus distance.

Noninvasive Estimation of Glycated Hemoglobin In-Vivo Based on Photon Diffusion Theory and Genetic Symbolic Regression Models.

The diagnosis and management of diabetes require frequent monitoring of blood sugar levels. Prolonged exposure of most of the monosaccharides in the bloodstream results in the glycation of hemoglobin. This glycated hemoglobin (HbA1c) based test plays an important role to avoid diabetic complications. However, noninvasive estimation of HbA1c is a very new, promising, and challenging topic in modern bioengineering scopes. The purpose of this study is to develop and verify mathematical models in order to quantify the glycated hemoglobin in-vivo percentage non-invasively. This research utilized photon diffusion theory to develop the finger models and genetic symbolic regression methods to solve the models to estimate the level of glycated hemoglobin in the blood. The validation of these models with human participants indicated a high degree of correlation (0.887 and 0.907 Pearson's r value), and high precision (2.56% and 2.96% coefficient of variation (%CV)) for transmission and reflection type noninvasive digital volume pulse-based signals. This research will be a breakthrough for the application of noninvasive HbA1c estimation.

Derivation and validation of gray-box models to estimate noninvasive in-vivo percentage glycated hemoglobin using digital volume pulse waveform.

Glycated hemoglobin and blood oxygenation are the two most important factors for monitoring a patient's average blood glucose and blood oxygen levels. Digital volume pulse acquisition is a convenient method, even for a person with no previous training or experience, can be utilized to estimate the two abovementioned physiological parameters. The physiological basis assumptions are utilized to develop two-finger models for estimating the percent glycated hemoglobin and blood oxygenation levels. The first model consists of a blood-vessel-only hypothesis, whereas the second model is based on a whole-finger model system. The two gray-box systems were validated on diabetic and nondiabetic patients. The mean absolute errors for the percent glycated hemoglobin (%HbA1c) and percent oxygen saturation (%SpO2) were 0.375 and 1.676 for the blood-vessel model and 0.271 and 1.395 for the whole-finger model, respectively. The repeatability analysis indicated that these models resulted in a mean percent coefficient of variation (%CV) of 2.08% and 1.74% for %HbA1c and 0.54% and 0.49% for %SpO2 in the respective models. Herein, both models exhibited similar performances (HbA1c estimation Pearson's R values were 0.92 and 0.96, respectively), despite the model assumptions differing greatly. The bias values in the Bland-Altman analysis for both models were - 0.03 ± 0.458 and - 0.063 ± 0.326 for HbA1c estimation, and 0.178 ± 2.002 and - 0.246 ± 1.69 for SpO2 estimation, respectively. Both models have a very high potential for use in real-world scenarios. The whole-finger model with a lower standard deviation in bias and higher Pearson's R value performs better in terms of higher precision and accuracy than the blood-vessel model.