Leveraging 3D convolutional neural network and 3D visible-near-infrared multimodal imaging for enhanced contactless oximetry.

Significance: Monitoring oxygen saturation ( SpO 2 ) is important in healthcare, especially for diagnosing and managing pulmonary diseases. Non-contact approaches broaden the potential applications of SpO 2 measurement by better hygiene, comfort, and capability for long-term monitoring. However, existing studies often encounter challenges such as lower signal-to-noise ratios and stringent environmental conditions. Aim: We aim to develop and validate a contactless SpO 2 measurement approach using 3D convolutional neural networks (3D CNN) and 3D visible-near-infrared (VIS-NIR) multimodal imaging, to offer a convenient, accurate, and robust alternative for SpO 2 monitoring. Approach: We propose an approach that utilizes a 3D VIS-NIR multimodal camera system to capture facial videos, in which SpO 2 is estimated through 3D CNN by simultaneously extracting spatial and temporal features. Our approach includes registration of multimodal images, tracking of the 3D region of interest, spatial and temporal preprocessing, and 3D CNN-based feature extraction and SpO 2 regression. Results: In a breath-holding experiment involving 23 healthy participants, we obtained multimodal video data with reference SpO 2 values ranging from 80% to 99% measured by pulse oximeter on the fingertip. The approach achieved a mean absolute error (MAE) of 2.31% and a Pearson correlation coefficient of 0.64 in the experiment, demonstrating good agreement with traditional pulse oximetry. The discrepancy of estimated SpO 2 values was within 3% of the reference SpO 2 for ∼ 80 % of all 1-s time points. Besides, in clinical trials involving patients with sleep apnea syndrome, our approach demonstrated robust performance, with an MAE of less than 2% in SpO 2 estimations compared to gold-standard polysomnography. Conclusions: The proposed approach offers a promising alternative for non-contact oxygen saturation measurement with good sensitivity to desaturation, showing potential for applications in clinical settings.

Ninety years of pulse oximetry: history, current status, and outlook.

Significance: This year, 2024, marks the 50th anniversary of the invention of pulse oximetry (PO), which was first presented by Takuo Aoyagi, an engineer from the Nihon Kohden Company, at the 13th Conference of the Japanese Society of Medical Electronics and Biological Engineering in Osaka in 1974. His discovery and the development of PO for the non-invasive measurement of peripheral arterial oxygenation represents one of the most significant chapters in the history of medical technology. It resulted from research and development efforts conducted by biochemists, engineers, physicists, physiologists, and physicians since the 1930s. Aim: The objective of this work was to provide a narrative review of the history, current status, and future prospects of PO. Approach: A comprehensive review of the literature on oximetry and PO was conducted. Results and Conclusions: Our historical review examines the development of oximetry in general and PO in particular, tracing the key stages of a long and fascinating story that has unfolded from the first half of the twentieth century to the present day-an exciting journey in which serendipity has intersected with the hard work of key pioneers. This work has been made possible by the contributions of numerous key pioneers, including Kurt Kramer, Karl Matthes, Glenn Millikan, Evgenii M. Kreps, Earl H. Wood, Robert F. Show, Scott A. Wilber, William New, and, above all, Takuo Aoyagi. PO has become an integral part of modern medical care and has proven to be an important tool for physiological monitoring. The COVID-19 pandemic not only highlighted the clinical utility of PO but also revealed some of the problems with the technology. Current research in biomedical optics should address these issues to make the technology even more reliable and accurate. We discuss the necessary innovations in PO and present our thoughts on what the next generation of PO might look like.

Vital parameter monitoring in harsh environment by the MedSENS in-ear multisensor device.

Accurate assessment of vital parameters is essential for diagnosis and triage of critically ill patients, but not always feasible in out-of-hospital settings due to the lack of suitable devices. We performed an extensive validation of a novel prototype in-ear device, which was proposed for the non-invasive, combined measurement of core body temperature (Tc), oxygen saturation (SpO2), and heart rate (HR) in harsh environments. A pilot study with randomized controlled design was conducted in the terraXcube environmental chamber. Participants were subsequently exposed to three 15 min test sessions at the controlled ambient temperatures of 20 °C, 5 °C, and - 10 °C, in randomized order. Vital parameters measured by the prototype were compared with Tc measurements from commercial esophageal (reference) and tympanic (comparator) probes and SpO2 and HR measurements from a finger pulse-oximeter (reference). Performance was assessed in terms of bias and Lin's correlation coefficient (CCC) with respect to the reference measurements and analyzed with linear mixed models. Twenty-three participants (12 men, mean (SD) age, 35 (9) years) completed the experimental protocol. The mean Tc bias of the prototype ranged between - 0.39 and - 0.80 °C at ambient temperatures of 20 °C and 5 °C, and it reached - 1.38 °C only after 15 min of exposure to - 10 °C. CCC values ranged between 0.07 and 0.25. SpO2 and HR monitoring was feasible, although malfunctioning was observed in one third of the tests. SpO2 and HR bias did not show any significant dependence on environmental conditions, with values ranging from - 1.71 to - 0.52% for SpO2 and 1.12 bpm to 5.30 bpm for HR. High CCC values between 0.81 and 0.97 were observed for HR in all environmental conditions. This novel prototype device for measuring vital parameters in cold environments demonstrated reliability of Tc measurements and feasibility of SpO2 and HR monitoring. Through non-invasive and accurate monitoring of vital parameters from the ear canal our prototype may offer support in triage and treatment of critically ill patients in harsh out-of-hospital conditions.

Innovative approaches in imaging photoplethysmography for remote blood oxygen monitoring.

Peripheral Capillary Oxygen Saturation (SpO2) has received increasing attention during the COVID-19 pandemic. Clinical investigations have demonstrated that individuals afflicted with COVID-19 exhibit notably reduced levels of SpO2 before the deterioration of their health status. To cost-effectively enable individuals to monitor their SpO2, this paper proposes a novel neural network model named "ITSCAN" based on Temporal Shift Module. Benefiting from the widespread use of smartphones, this model can assess an individual's SpO2 in real time, utilizing standard facial video footage, with a temporal granularity of seconds. The model is interweaved by two distinct branches: the motion branch, responsible for extracting spatiotemporal data features and the appearance branch, focusing on the correlation between feature channels and the location information of feature map using coordinate attention mechanisms. Accordingly, the SpO2 estimator generates the corresponding SpO2 value. This paper summarizes for the first time 5 loss functions commonly used in the SpO2 estimation model. Subsequently, a novel loss function has been contributed through the examination of various combinations and careful selection of hyperparameters. Comprehensive ablation experiments analyze the independent impact of each module on the overall model performance. Finally, the experimental results based on the public dataset (VIPL-HR) show that our model has obvious advantages in MAE (1.10%) and RMSE (1.19%) compared with related work, which implies more accuracy of the proposed method to contribute to public health.

Reliability of Continuous Noninvasive Hemoglobin Monitoring in Healthy Participants During En Route Care Training.

INTRODUCTION: Current standards for hemoglobin monitoring during air transports of U.S. combat wounded are invasive and intermittent. Fielded pulse co-oximeters can noninvasively measure total hemoglobin, but this parameter is not currently utilized. The primary objective of this study was to assess the percentage of vital sign measurements with successful capture of total noninvasive hemoglobin measurement using spectrophotometry-based technology for Hb (SpHb) measurements in healthy participants during training flights. Secondary objectives were to assess the feasibility of a novel electronic data capture mechanism from usual patient movement items and perform a pilot analysis of SpHb changes in healthy participants during transitions from ground to air transport. METHODS: We conducted a feasibility study enrolling healthy participants who had hemodynamic monitoring during usual U.S. Air Force Critical Care Air Transport (CCAT) flight training exercises from 2022 to 2023. Usual CCAT monitoring equipment and currently used Masimo Rainbow® pulse co-oximeters had the capability to measure SpHb. After each training exercise, the study team wirelessly downloaded case files from patient monitors utilizing the Battlefield Assisted Trauma Distributed Observation Kit (BATDOKTM) Case Downloader application. We then calculated point and precision estimates for the percentage of time for successful SpHb capture during the exercise and compared this to pulse oximetry (SpO2) capture. An a priori precision analysis for percentage of flight-time with successful SpHb data capture and descriptive statistics were performed. This study received Exempt Determination by the 59th Medical Wing IRB. RESULTS: We analyzed 26 records with mean monitoring durations of 94.5 [59.3-119.9] minutes during ground phases and 78.0 [59.9-106.5] minutes during flight phases. SpHb measures were successfully captured for 97.7% (n = 4,620) of possible ground measurements and 97.2% (n = 3,973) of possible in-flight measurements compared to 99.5% ground and 98.2% in-flight capture for SpO2. Mean intervals of missing SpHb data were 2 ± 5 minutes on the ground and 4 ± 6 minutes in-flight. Mean SpHb increased by 0.93 ± 0.96 g/dL during the ground phase, but had minimal changes during ascent, cruising altitude or descent. The BATDOKTM Case downloader completed transfer for all files. CONCLUSION: Masimo Rainbow® SpHb pulse co-oximeters reliably captured continuous, noninvasive hemoglobin measurements using usual CCAT patient movement items in healthy participants during both ground and flight training. The BATDOKTM Case Downloader successfully imported case files from CCAT patient monitors. Mean SpHb measures had a small increase during the ground phase of monitoring followed by minimal changes when transitioning to flight altitude.

Monte Carlo simulation of the effect of melanin concentration on light-tissue interactions for transmittance pulse oximetry measurement.

Significance: Questions about the accuracy of pulse oximeters in measuring arterial oxygen saturation ( SpO 2 ) in individuals with darker skin pigmentation have resurfaced since the COVID-19 pandemic. This requires investigation to improve patient safety, clinical decision making, and research. Aim: We aim to use computational modeling to identify the potential causes of inaccuracy in SpO 2 measurement in individuals with dark skin and suggest practical solutions to minimize bias. Approach: An in silico model of the human finger was developed to explore how changing melanin concentration and arterial oxygen saturation ( SaO 2 ) affect pulse oximeter calibration algorithms using the Monte Carlo (MC) technique. The model generates calibration curves for Fitzpatrick skin types I, IV, and VI and an SaO 2 range between 70% and 100% in transmittance mode. SpO 2 was derived by inputting the computed ratio of ratios for light and dark skin into a widely used calibration algorithm equation to calculate bias ( SpO 2 - SaO 2 ). These were validated against an experimental study to suggest the validity of the Monte Carlo model. Further work included applying different multiplication factors to adjust the moderate and dark skin calibration curves relative to light skin. Results: Moderate and dark skin calibration curve equations were different from light skin, suggesting that a single algorithm may not be suitable for all skin types due to the varying behavior of light in different epidermal melanin concentrations, especially at 660 nm. The ratio between the mean bias in White and Black subjects in the cohort study was 6.6 and 5.47 for light and dark skin, respectively, from the Monte Carlo model. A linear multiplication factor of 1.23 and exponential factor of 1.8 were applied to moderate and dark skin calibration curves, resulting in similar alignment. Conclusions: This study underpins the careful re-assessment of pulse oximeter designs to minimize bias in SpO 2 measurements across diverse populations.

Oxygen reserve index versus conventional peripheral oxygen saturation for prevention of hypoxaemia: A randomised controlled trial.

BACKGROUND: Hypoxaemia occurs frequently during paediatric laryngeal microsurgery. OBJECTIVE: The oxygen reserve index is a noninvasive and continuous parameter to assess PaO2 levels in the range of 100 to 200 mmHg. It ranges from 0 to 1.0. We investigated whether monitoring the oxygen reserve index can reduce the incidence of SpO2 90% or less. DESIGN: Randomised controlled trial. SETTING: A tertiary care paediatric hospital. PARTICIPANTS: Paediatric patients aged 18 years or less scheduled to undergo laryngeal microsurgery. INTERVENTION: The patients were randomly allocated to the oxygen reserve index or control groups, and stratified based on the presence of a tracheostomy tube. Rescue intervention was performed when the oxygen reserve index was 0.2 or less and the SpO2 was 94% or less in the oxygen reserve index and control groups, respectively. MAIN OUTCOME MEASURE: The primary outcome was the incidence of SpO2 90% or less during the surgery. RESULTS: Data from 88 patients were analysed. The incidence of SpO2 ≤ 90% did not differ between the oxygen reserve index and control groups [P = 0.114; 11/44, 25% vs. 18/44, 40.9%; relative risk: 1.27; and 95% confidence interval (CI): 0.94 to 1.72]. Among the 128 rescue interventions, SpO2 ≤ 90% event developed in 18 out of 75 events (24%) and 42 out of 53 events (79.2%) in the oxygen reserve index and control groups, respectively (P < 0.001; difference: 55.2%; and 95% CI 38.5 to 67.2%). The number of SpO2 ≤ 90% events per patient in the oxygen reserve index group (median 0, maximum 3) was less than that in the control group (median 0, maximum 8, P = 0.031). CONCLUSION: Additional monitoring of the oxygen reserve index, with a target value of greater than 0.2 during paediatric airway surgery, alongside peripheral oxygen saturation, did not reduce the incidence of SpO2 ≤ 90%.

Continuous Capnography for Early Detection of Respiratory Compromise During Gastroenterological Procedural Sedation and Analgesia.

Gastroenterology nurses working across a variety of clinical settings are responsible for periprocedural monitoring during moderate to deep procedural sedation and analgesia (PSA) to identify signs of respiratory compromise and intervene to prevent cardiorespiratory events. Pulse oximetry is the standard of care for respiratory monitoring, but it may delay or fail to detect abnormal ventilation during PSA. Continuous capnography, which measures end-tidal CO2 as a marker of alveolar ventilation, has been endorsed by a number of clinical guidelines. Large clinical trials have demonstrated that the addition of continuous capnography to pulse oximetry during PSA for various gastroenterological procedures reduces the incidence of hypoxemia, severe hypoxemia, and apnea. Studies have shown that the cost of adding continuous capnography is offset by the reduction in adverse events and hospital length of stay. In the postanesthesia care unit, continuous capnography is being evaluated for monitoring opioid-induced respiratory depression and to guide artificial airway removal. Studies are also examining the utility of continuous capnography to predict the risk of opioid-induced respiratory depression among patients receiving opioids for primary analgesia. Continuous capnography monitoring has become an essential tool to detect early signs of respiratory compromise in patients receiving PSA during gastroenterological procedures. When combined with pulse oximetry, it can help reduce cardiorespiratory adverse events, improve patient outcomes and safety, and reduce health care costs.

Reliability, clinical performance and trending ability of a pulse oximeter and pulse co-oximeter, in monitoring blood oxygenation, at two measurement sites, in immobilised white rhinoceros (Ceratotherium simum).

BACKGROUND: Monitoring blood oxygenation is essential in immobilised rhinoceros, which are susceptible to opioid-induced hypoxaemia. This study assessed the reliability, clinical performance and trending ability of the Nonin PalmSAT 2500 A pulse oximeter's and the Masimo Radical-7 pulse co-oximeter's dual-wavelength technology, with their probes placed at two measurement sites, the inner surface of the third-eyelid and the scarified ear pinna of immobilised white rhinoceroses. Eight white rhinoceros were immobilised with etorphine-based drug combinations and given butorphanol after 12 min, and oxygen after 40 min, of recumbency. The Nonin and Masimo devices, with dual-wavelength probes attached to the third-eyelid and ear recorded arterial peripheral oxygen-haemoglobin saturation (SpO2) at pre-determined time points, concurrently with measurements of arterial oxygen-haemoglobin saturation (SaO2), from drawn blood samples, by a benchtop AVOXimeter 4000 co-oximeter (reference method). Reliability of the Nonin and Masimo devices was evaluated using the Bland-Altman and the area root mean squares (ARMS) methods. Clinical performance of the devices was evaluated for their ability to accurately detect clinical hypoxemia using receiver operating characteristic (ROC) curves and measures of sensitivity, specificity, and positive and negative predictive values. Trending ability of the devices was assessed by calculating concordance rates from four-quadrant plots. RESULTS: Only the Nonin device with transflectance probe attached to the third-eyelid provided reliable SpO2 measurements across the 70 to 100% saturation range (bias - 1%, precision 4%, ARMS 4%). Nonin and Masimo devices with transflectance probes attached to the third-eyelid both had high clinical performance at detecting clinical hypoxaemia [area under the ROC curves (AUC): 0.93 and 0.90, respectively]. However, the Nonin and Masimo devices with transmission probes attached to the ear were unreliable and provided only moderate clinical performance. Both Nonin and Masimo devices, at both measurement sites, had concordance rates lower than the recommended threshold of ≥ 90%, indicating poor trending ability. CONCLUSIONS: The overall assessment of reliability, clinical performance and trending ability indicate that the Nonin device with transflectance probe attached to the third-eyelid is best suited for monitoring of blood oxygenation in immobilised rhinoceros. The immobilisation procedure may have affected cardiovascular function to an extent that it limited the devices' performance.

Performance and safety of the PRICO closed-loop oxygen saturation targeting system in neonates: pragmatic multicentre cross-over study (TarOx Study).

OBJECTIVE: This study aims to evaluate the performance of the fabian-Predictive-Intelligent-Control-of-Oxygenation (PRICO) system for automated control of the fraction of inspired oxygen (FiO2). DESIGN: Multicentre randomised cross-over study. SETTING: Five neonatal intensive care units experienced with automated control of FiO2 and the fabian ventilator. PATIENTS: 39 infants: median gestational age of 27 weeks (IQR: 26-30), postnatal age 7 days (IQR: 2-17), weight 1120 g (IQR: 915-1588), FiO2 0.32 (IQR: 0.22-0.43) receiving both non-invasive (27) and invasive (12) respiratory support. INTERVENTION: Randomised sequential 24-hour periods of automated and manual FiO2 control. MAIN OUTCOME MEASURES: Proportion (%) of time in normoxaemia (90%-95% with FiO2>0.21 and 90%-100% when FiO2=0.21) was the primary endpoint. Secondary endpoints were severe hypoxaemia (<80%) and severe hyperoxaemia (>98% with FiO2>0.21) and prevalence of episodes ≥60 s at these two SpO2 extremes. RESULTS: During automated control, subjects spent more time in normoxaemia (74%±22% vs 51%±22%, p<0.001) with less time above and below (<90% (9%±8% vs 12%±11%, p<0.001) and >95% with FiO2>0.21 (16%±19% vs 35%±24%) p<0.001). They spent less time in severe hyperoxaemia (1% (0%-3.5%) vs 5% (1%-10%), p<0.001) but exposure to severe hypoxaemia was low in both arms and not different. The differences in prolonged episodes of SpO2 were consistent with the times at extremes. CONCLUSIONS: This study demonstrates the ability of the PRICO automated oxygen control algorithm to improve the maintenance of SpO2 in normoxaemia and to avoid hyperoxaemia without increasing hypoxaemia.

The effect of pleural drainage on pulse oximetry in a post-operative thoracic surgery population.

BACKGROUND: Pleural effusions in post-operative thoracic surgery patients are common. Effusions can result in prolonged hospitalizations or readmissions, with prior studies suggesting mixed effects of pleural drainage on hypoxia. We aimed to define the impact of pleural drainage on pulse oximetry (SpO2) in post-thoracic surgery patients. METHODS: A retrospective study of post-operative thoracic surgery patients undergoing pleural drainage was performed. SpO2 and supplemental oxygen (FiO2) values were recorded at pre- and post-procedure. The primary outcome was difference in pre-procedural and post-procedural SpO2. RESULTS: We identified 95 patients with a mean age of 65 (SD - 13.8) years undergoing 122 pleural drainage procedures. Mean drainage volume was 619 (SD-423) mL and the majority of procedures (88.5 %) included a drainage of <1000 mL. SpO2 was associated with an increase from 94.0 % (SD-2.6) to 97.3 % (SD-2.0) at 24-h (p < 0.0001). FiO2 was associated with a decrease from 0.31 (SD-0.15) to 0.29 (SD-0.12) at 24-h (p = 0.0081). SpO2/FiO2 was associated with an increase from 344.5 (SD-99.0) to 371.9 (SD-94.7) at 24-h post-procedure (p < 0.0001). CONCLUSIONS: Pleural drainage within post-operative thoracic surgery patients offers statistically significant improvements in oxygen saturation by peripheral pulse oximetry and oxygen supplementation; however the clinical significance of these changes remains unclear. Pleural drainage itself may be requested for numerous reasons, including diagnostic (fevers, leukocytosis, etc.) or therapeutic (worsening dyspnea) evaluation. However, pleural drainage may offer minimal clinical impact on pulse oximetry in post-operative thoracic surgery patients.

Hypoxaemia and risk of death among children: rethinking oxygen saturation, risk-stratification, and the role of pulse oximetry in primary care.

Pulse oximeters are essential for assessing blood oxygen levels in emergency departments, operating theatres, and hospital wards. However, although the role of pulse oximeters in detecting hypoxaemia and guiding oxygen therapy is widely recognised, their role in primary care settings is less clear. In this Viewpoint, we argue that pulse oximeters have a crucial role in risk-stratification in both hospital and primary care or outpatient settings. Our reanalysis of hospital and primary care data from diverse low-income and middle-income settings shows elevated risk of death for children with moderate hypoxaemia (ie, peripheral oxygen saturations [SpO2] 90-93%) and severe hypoxaemia (ie, SpO2 <90%). We suggest that moderate hypoxaemia in the primary care setting should prompt careful clinical re-assessment, consideration of referral, and close follow-up. We provide practical guidance to better support front-line health-care workers to use pulse oximetry, including rethinking traditional binary SpO2 thresholds and promoting a more nuanced approach to identification and emergency treatment of the severely ill child.

Machine Learning-Based Critical Congenital Heart Disease Screening Using Dual-Site Pulse Oximetry Measurements.

BACKGROUND: Oxygen saturation (Spo2) screening has not led to earlier detection of critical congenital heart disease (CCHD). Adding pulse oximetry features (ie, perfusion data and radiofemoral pulse delay) may improve CCHD detection, especially coarctation of the aorta (CoA). We developed and tested a machine learning (ML) pulse oximetry algorithm to enhance CCHD detection. METHODS AND RESULTS: Six sites prospectively enrolled newborns with and without CCHD and recorded simultaneous pre- and postductal pulse oximetry. We focused on models at 1 versus 2 time points and with/without pulse delay for our ML algorithms. The sensitivity, specificity, and area under the receiver operating characteristic curve were compared between the Spo2-alone and ML algorithms. A total of 523 newborns were enrolled (no CHD, 317; CHD, 74; CCHD, 132, of whom 21 had isolated CoA). When applying the Spo2-alone algorithm to all patients, 26.2% of CCHD would be missed. We narrowed the sample to patients with both 2 time point measurements and pulse-delay data (no CHD, 65; CCHD, 14) to compare ML performance. Among these patients, sensitivity for CCHD detection increased with both the addition of pulse delay and a second time point. All ML models had 100% specificity. With a 2-time-points+pulse-delay model, CCHD sensitivity increased to 92.86% (P=0.25) compared with Spo2 alone (71.43%), and CoA increased to 66.67% (P=0.5) from 0. The area under the receiver operating characteristic curve for CCHD and CoA detection significantly improved (0.96 versus 0.83 for CCHD, 0.83 versus 0.48 for CoA; both P=0.03) using the 2-time-points+pulse-delay model compared with Spo2 alone. CONCLUSIONS: ML pulse oximetry that combines oxygenation, perfusion data, and pulse delay at 2 time points may improve detection of CCHD and CoA within 48 hours after birth. REGISTRATION: URL: https://www.clinicaltrials.gov/study/NCT04056104?term=NCT04056104&rank=1; Unique identifier: NCT04056104.

[Study on the application value of portable pulse oximeter in adult patients with obstructive sleep apnea].

Objective: To evaluate the application value of portable pulse oximeter in adult obstructive sleep apnea (OSA). Methods: This study prospectively enrolled adult patients who underwent polysomnography (PSG) due to snoring at the Respiratory and Sleep Medicine Department of Peking University People's Hospital from July 2022 to July 2023. During PSG monitoring, CS-WOxi was continuously used to monitor blood oxygen levels. The consistency between 3% oxygen desaturation index (ODI3) measured by portable pulse oximeter and ODI3 of polysomnography was evaluated using difference test, Pearson's correlation coefficient, and Bland-altman method. Receiver operating characteristic curve was used to determine the optimal threshold for diagnosing OSA. Results: A total of 184 subjects were included, including 121 males (65.8%) and 63 females (34.2%). The mean age was 46.0 (34.3, 59.0) years, body mass index was 26.0 (23.3, 29.6) kg/m², and the apnea-hypopnea index was 18.2 (5.8, 40.8) events/h. There was a significant difference between CS-ODI3 and PSG-ODI3 [17.1(6.2, 42.7) vs. 14.0(2.9, 32.6), P<0.001], and the Pearson correlation coefficient was 0.93 (P<0.001). There was a good correlation between CS-ODI3 and PSG-AHI (r=0.92, P<0.001). Bland-Altman consistency test showed that the average difference between the two was 0.7 events/h, and the 95% consistency limit was (-17.9, 19.3 events/h). When the CS-ODI3≥5 events/h was used to identify OSA, the sensitivity was 94.4%, the specificity was 80.0%, and the accuracy was 91.3%. When PSG-AHI≥5 events/h was used as the diagnostic criteria, the area under the receiver operating characteristic curve was 0.933. Conclusion: Portable pulse oximeter can monitor pulse oxygen saturation accurately and has good sensitivity and specificity for OSA high-risk patients, and is a reliable tool for OSA screening.

Distribution-informed and wavelength-flexible data-driven photoacoustic oximetry.

Significance: Photoacoustic imaging (PAI) promises to measure spatially resolved blood oxygen saturation but suffers from a lack of accurate and robust spectral unmixing methods to deliver on this promise. Accurate blood oxygenation estimation could have important clinical applications from cancer detection to quantifying inflammation. Aim: We address the inflexibility of existing data-driven methods for estimating blood oxygenation in PAI by introducing a recurrent neural network architecture. Approach: We created 25 simulated training dataset variations to assess neural network performance. We used a long short-term memory network to implement a wavelength-flexible network architecture and proposed the Jensen-Shannon divergence to predict the most suitable training dataset. Results: The network architecture can flexibly handle the input wavelengths and outperforms linear unmixing and the previously proposed learned spectral decoloring method. Small changes in the training data significantly affect the accuracy of our method, but we find that the Jensen-Shannon divergence correlates with the estimation error and is thus suitable for predicting the most appropriate training datasets for any given application. Conclusions: A flexible data-driven network architecture combined with the Jensen-Shannon divergence to predict the best training data set provides a promising direction that might enable robust data-driven photoacoustic oximetry for clinical use cases.

Machine learning-based prediction of cerebral oxygen saturation based on multi-modal cerebral oximetry data.

This study develops machine learning-based algorithms that facilitate accurate prediction of cerebral oxygen saturation using waveform data in the near-infrared range from a multi-modal oxygen saturation sensor. Data were obtained from 150,000 observations of a popular cerebral oximeter, Masimo O3™ regional oximetry (Co., United States) and a multi-modal cerebral oximeter, Votem (Inc., Korea). Among these observations, 112,500 (75%) and 37,500 (25%) were used for training and test sets, respectively. The dependent variable was the cerebral oxygen saturation value from the Masimo O3™ (0-100%). The independent variables were the time of measurement (0-300,000 ms) and the 16-bit decimal amplitudes values (infrared and red) from Votem (0-65,535). For the right part of the forehead, the root mean square error of the random forest (0.06) was much smaller than those of linear regression (1.22) and the artificial neural network with one, two or three hidden layers (2.58). The result was similar for the left part of forehead, that is, random forest (0.05) vs logistic regression (1.22) and the artificial neural network with one, two or three hidden layers (2.97). Machine learning aids in accurately predicting of cerebral oxygen saturation, employing the data from a multi-modal cerebral oximeter.

Effect of Skin Pigmentation and Finger Choice on Accuracy of Oxygen Saturation Measurement in an IoT-Based Pulse Oximeter.

Pulse oximeters are widely used in hospitals and homes for measurement of blood oxygen saturation level (SpO2) and heart rate (HR). Concern has been raised regarding a possible bias in obtaining pulse oximeter measurements from different fingertips and the potential effect of skin pigmentation (white, brown, and dark). In this study, we obtained 600 SpO2 measurements from 20 volunteers using three UK NHS-approved commercial pulse oximeters alongside our custom-developed sensor, and used the Munsell colour system (5YR and 7.5YR cards) to classify the participants' skin pigmentation into three distinct categories (white, brown, and dark). The statistical analysis using ANOVA post hoc tests (Bonferroni correction), a Bland-Altman plot, and a correlation test were then carried out to determine if there was clinical significance in measuring the SpO2 from different fingertips and to highlight if skin pigmentation affects the accuracy of SpO2 measurement. The results indicate that although the three commercial pulse oximeters had different means and standard deviations, these differences had no clinical significance.

[Evaluation of revascularization efficiency using near-infrared oximetry in patients with diabetic foot syndrome]. / Otsenka effektivnosti revaskulyarizatsii u patsientov s sindromom diabeticheskoi stopy metodom parainfrakrasnoi oksimetrii.

OBJECTIVE: To evaluate prognostic significance of tissue oximetry in healing of trophic defects in patients with diabetic foot syndrome (DFS) after endovascular revascularization. MATERIAL AND METHODS: In 42 patients with DFS, tissue oximetry was performed in angiosome projection with the FORE-SIGHT MC-2000 (CASMED) device before and the next day after revascularization of lower limb arteries. The 1st group included 30 patients with wound healing throughout 3 months, the 2nd group included 12 patients with amputation or no healing of trophic defects. RESULTS: Direct revascularization was more common in the 1st group (p=0.001). On the day after intervention, oxygen saturation (StO2) increased in all angiosomes in both groups (p<0.05). StO2 increment differed significantly between groups in all angiosomes except for point I (p<0.05). According to ROC analysis, StO2 increment by 4.5% in absolute values and 7.9% in percentage leads to trophic defect healing within 3 months (sensitivity and specificity were 76.7% and 66.7% for absolute values, 80% and 58.3% for percentage, respectively). CONCLUSION: Evaluation of StO2 in target angiosome may be valuable to predict trophic defect healing after endovascular surgery.

Association between prehospital ROX index with 30-day mortality among septic shock.

PURPOSE: Respiratory dysfunction is one of the most frequent symptoms observed during sepsis reflecting hypoxemia and/or acidosis that may be assessed by the ROX index (ratio of oxygen saturation by pulse oximetry/fraction of inspired oxygen to respiratory rate). This study aimed to describe the relationship between the prehospital ROX index and 30-day mortality rate among septic shock patients cared for in the prehospital setting by a mobile intensive care unit (MICU). METHODS: From May 2016 to December 2021, 530 septic shock patients cared for by a prehospital MICU were retrospectively analysed. Initial ROX index value was calculated at the first contact with MICU. A Cox regression analysis after propensity score matching was performed to assess the relationship between 30-day mortality rate and a ROX index ≤ 10. RESULTS: Pulmonary, digestive and urinary sepsis were suspected among 43%, 25% and 17% patients, respectively. The 30-day overall mortality reached 31%. Cox regression analysis showed a significant association between 30-day mortality and a ROX index ≤ 10: adjusted hazard ratio of 1.54 [1.08-2.31], p < 0.05. CONCLUSIONS: During the prehospital stage of septic shock patients cared for by a MICU, ROX index is significantly associated with 30-day mortality. A prehospital ROX ≤ 10 value is associated with a 1.5-fold 30-day mortality rate increase. Prospective studies are needed to confirm the ability of prehospital ROX to predict sepsis outcome since the prehospital setting.

Designing a Smartphone-Based Pulse Oximeter for Children in South Africa (Phefumla Project): Qualitative Analysis of Human-Centered Design Workshops With Health Care Workers.

Background: Pulse oximeters noninvasively measure blood oxygen levels, but these devices have rarely been designed for low-resource settings and are inconsistently available at outpatient clinics. Objective: The Phefumla project aims to develop and validate a pediatric smartphone-based pulse oximeter designed specifically for this context. We present the process of human-centered oximeter design with health care workers in South Africa. Methods: We purposively sampled 19 health care workers from 5 clinics in Khayelitsha, Cape Town. Using a human-centered design approach, we conducted participatory workshops with four activities with health care workers: (1) they received 3D-printed prototypes of potential oximeter designs to provide feedback; (2) we demonstrated on dolls how they would use the novel oximeter; (3) they used pile sorting to rank design features and suggest additional features they desired; and (4) they designed their preferred user interface using a whiteboard, marker, and magnetized features that could be repositioned. We audio recorded the workshops, photographed outputs, and took detailed field notes. Analysis involved iterative review of these data to describe preferences, identify key design updates, and provide modifications. Results: Participants expressed a positive sentiment toward the idea of a smartphone pulse oximeter and suggested that a pediatric device would address an important gap in outpatient care. Specifically, participants expressed a preference for the prototype that they felt enabled more diversity in the way it could be used. There was a strong tendency to prioritize pragmatic design features, such as robustness, which was largely dictated by health care worker context. They also added features that would allow the oximeter device to serve other clinical functions in addition to oxygen saturation measurement, such as temperature and respiratory rate measurements. Conclusions: Our end user-centered rapid participatory approach led to tangible design changes and prompted design discussions that the team had not previously considered. Overall, health care workers prioritized pragmatism for pediatric pulse oximeter device design.