A Multi-Model Analysis of Non-Invasive Oxygen Saturation from Smartphone-Captured Fingertip Video

Finger-based optical sensing for blood glucose monitoring using smartphone photoplethysmography

Accurate and continuous blood glucose monitoring remains a critical challenge due to the invasive nature of existing measurement techniques in remote and clinical settings. To address these challenges, this work presents a smartphonebased optical photoplethysmography (PPG) framework for low-cost, portable, and reliable blood glucose monitoring. Fingertip PPG signals are acquired from the left index finger using a smartphone camera for a duration of 15 s at 30 frames per second. The raw optical signal is extracted from the red colour channel, which is sensitive to blood volume variations dominated by haemoglobin absorption. Subsequently, a bandpass filter in the range of 0.5-5 Hz is used to isolate cardiacrelated pulsatile components. Baseline drift and motion-induced artifacts are attenuated using cubic spline interpolation while preserving PPG waveform morphology relevant to physiological analysis. From the pre-processed signals, time-domain and morphology-based PPG features are extracted and combined with demographic information and reference glucose values to construct the input dataset for glucose-level classification. The proposed framework is tested and evaluated using data collected from 80 subjects. Smartphone-acquired PPG signals are evaluated against the existing reference measurement system. The regression-based classification framework using XGBoost outperformed alternative algorithms, achieving a classification accuracy of 77%, precision of 85%, and numerical blood glucose prediction with a minimum RMSE of 15. The resultant findings demonstrated that smartphone based optical PPG represents a scalable noninvasive alternative for blood glucose monitoring, highlighting its potential for accessible remote monitoring applications.

Design and Implementation of Low-Cost Non-Invasive Blood Pressure Measurement Device from Photoplethysmography with Signal Quality Assessment

Cardiovascular diseases remain a leading global health concern, necessitating accurate, non-invasive blood pressure monitoring solutions. This study presents a novel photoplethysmography (PPG)-based system that integrates three key innovations: (1) a real-time signal quality assessment (SQA) model employing a one-class SVM classifier to filter unreliable PPG segments, (2) a computationally efficient two-stage neural network (1D U-Net followed by 1D MultiResUNet) for arterial blood pressure (ABP) waveform estimation, and (3) a wearable hardware platform combining a MAX30102 optical sensor and ESP32 microcontroller for portable deployment. The SQA model uses five optimized features to achieve robust motion artifact rejection. The deep learning pipeline reconstructs ABP waveforms with mean absolute errors of 4.7 mmHg (systolic) and 4.3 mmHg (diastolic), complying with the Association for the Advancement of Medical Instrumentation (AAMI) standards. Validated on 40 subjects against reference sphygmomanometer measurements, the system demonstrates 92.5% (systolic) and 95% (diastolic) classification accuracy, with subsecond latency and low-power operation. By addressing critical challenges in motion robustness, computational efficiency, and clinical validation, this work advances the practicality of cuffless BP monitoring for telehealth and resource-limited settings.

Real-Time SpO2 Estimation from Chest Reflectance Photoplethysmography: Algorithm Development and Clinical Validation Against Arterial SaO2

Peripheral oxygen saturation (SpO2) estimation using photoplethysmography (PPG) is typically based on the pulsatile components of red and infrared (IR) PPG signals acquired from peripheral sites, such as the finger or earlobe. Although these sites provide strong PPG signals, they are less suitable for integrated monitoring with electrocardiography (ECG), including arrhythmia detection. Simultaneous acquisition of PPG and ECG from a chest-worn device may therefore enable continuous and integrated physiological monitoring. However, chest PPG is characterized by lower perfusion and greater susceptibility to respiratory and motion artifacts, making reliable extraction of pulsatile red and IR components challenging. In this study, we developed and clinically evaluated a real-time SpO2 estimation algorithm based on chest reflectance PPG. Green PPG, which provides a relatively distinct pulsatile waveform under low-perfusion conditions, was used for beat detection. A heart-rate-adaptive analysis window, normalized cross-correlation-based signal quality assessment, and least-squares estimation of the AC amplitude ratio were combined to calculate the red-to-IR ratio of ratios, from which SpO2 was estimated using a calibration equation. Algorithm performance was evaluated against reference arterial oxygen saturation (SaO2) measured using a blood gas analyzer. The proposed method achieved an accuracy root mean square of 2.91% relative to the reference SaO2. These results support the technical feasibility of chest reflectance PPG-based SpO2 estimation and its accuracy under controlled desaturation conditions.

Clinical Evaluation of a Multimodal On-Body Sensor Array

Seamless vital sign monitoring with fiber optic sensing technology: validating the sitting SMART MAT for community health in static and mobile conditions

Background Continuous vital sign monitoring is critical but often labor-intensive. Fiberoptic technologies, like the SMART MAT, offer non-invasive assessment. Adapting this for a seated posture addresses clinical needs by promoting safe patient mobilization and reducing complications associated with prolonged bed rest. Objectives To validate the accuracy of the sitting SMART MAT for respiratory rate (RR), heart rate (HR), oxygen saturation (SpO2), and systolic and diastolic blood pressure (SBP/DBP) against established clinical reference standards. Methods This cross-sectional study compared sitting SMART MAT readings with clinical reference measurements under stationary and mobile wheelchair conditions. Accuracy was assessed using Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Bland-Altman Limits of Agreement (LoA), repeated-measures ANOVA, and effect size. Results Data from 130 participants were analyzed. HR readings demonstrated high accuracy (MAE = 3.67–4.81 bpm) with tight LoA. RR error was minimal (MAE = 2.85–2.86 breaths/min) statically but increased during motion (MAE = 5.03–5.21 breaths/min). SpO2 remained highly stable (MAE = 1.08–1.21%), well within standard clinical tolerance. SBP showed solid central accuracy (MAE = 6.48–7.51 mmHg) but wider LoA occasionally exceeded the ±5 mmHg threshold, functioning best as a continuous trend monitor. DBP variance was negligible (MAE = 4.65–5.46 mmHg). Conclusion The current SMART MAT is a viable, integrated multimodal platform. HR, RR, and DBP are comparable to clinical reference standards, while SBP requires careful interpretation. By accommodating a functional sitting posture, the contactless system reduces immobility-related complications and demonstrates high feasibility for continuous, real-world monitoring.

A Review of Machine Learning Approaches for Non-Contact Oxygen Saturation (SpO 2 ) and Heart Rate Estimation

Non-contact cardiovascular measurement systems have gained widespread interest as an alternative to the conventional contact methods. Of all these methods, facial based techniques that determine heart rate and oxygen saturation (SpO 2 ) hold the greatest promise for real-time use in the healthcare and other sectors. These methods have benefited from the recent advances made in machine learning in terms of accuracy, robustness, and scalability. This review examines the most recent advances in facial-based HR and SpO 2 estimation techniques, focusing especially on the role of ML algorithms. Algorithms, data problems, and comparative performance of facial-based HR and SpO 2 estimation methods, together with the disadvantages associated with these algorithms and ethical questions involved, will be discussed in the article. Future trends in the development of hybrid systems and applications in wearable IoT devices will also be highlighted. The purpose of this review is to familiarize the reader with the recent advances in the field of non-contact HR and SpO 2 estimation systems.