Dataset Information

Smartphone camera oximetry in an induced hypoxemia study.

ABSTRACT: Hypoxemia, a medical condition that occurs when the blood is not carrying enough oxygen to adequately supply the tissues, is a leading indicator for dangerous complications of respiratory diseases like asthma, COPD, and COVID-19. While purpose-built pulse oximeters can provide accurate blood-oxygen saturation (SpO₂) readings that allow for diagnosis of hypoxemia, enabling this capability in unmodified smartphone cameras via a software update could give more people access to important information about their health. Towards this goal, we performed the first clinical development validation on a smartphone camera-based SpO₂ sensing system using a varied fraction of inspired oxygen (FiO₂) protocol, creating a clinically relevant validation dataset for solely smartphone-based contact PPG methods on a wider range of SpO₂ values (70-100%) than prior studies (85-100%). We built a deep learning model using this data to demonstrate an overall MAE = 5.00% SpO₂ while identifying positive cases of low SpO₂ < 90% with 81% sensitivity and 79% specificity. We also provide the data in open-source format, so that others may build on this work.

SUBMITTER: Hoffman JS

PROVIDER: S-EPMC9483471 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Smartphone camera oximetry in an induced hypoxemia study.

Hoffman Jason S JS Viswanath Varun K VK Tian Caiwei C Ding Xinyi X Thompson Matthew J MJ Larson Eric C EC Patel Shwetak N SN Wang Edward J EJ

NPJ digital medicine 20220919 1

Hypoxemia, a medical condition that occurs when the blood is not carrying enough oxygen to adequately supply the tissues, is a leading indicator for dangerous complications of respiratory diseases like asthma, COPD, and COVID-19. While purpose-built pulse oximeters can provide accurate blood-oxygen saturation (SpO<sub>2</sub>) readings that allow for diagnosis of hypoxemia, enabling this capability in unmodified smartphone cameras via a software update could give more people access to important ...[more]

PMID: 36123367

Similar Datasets

Project description:BackgroundPulse oximeters are used to measure [Formula: see text] and pulse rate. These devices are either standalone machines or integrated into physiologic monitoring systems. Some smartphones now have pulse oximetry capabilities. Because it is possible that some patients might utilize this technology, we sought to assess the accuracy and usability of smartphone pulse oximeters.MethodsThis was a prospective, observational study that involved noninvasive measurements of [Formula: see text] and heart rate with 3 devices: Masimo Radical-7, Kenek Edge with the Apple iPhone 6S, and the Samsung S8 smartphone. Ambulatory adult patients visiting our institution's pulmonary function lab for a 6-min walk test were eligible to participate in the study. Pretest and posttest results for each subject were obtained simultaneously using all 3 devices. All results were analyzed with the Spearman rho correlation test, and Bland-Altman plots were used to assess the agreement of measures between the devices.ResultsForty-seven subjects were enrolled in the study, with pulmonary hypertension (30%) and COPD (23%) being the 2 major diagnoses. The mean ± SD difference between the Masimo and Apple devices for pretest [Formula: see text] was 2.3 ± 2.4%, and the difference for posttest [Formula: see text] was 2.1 ± 3.9%. The mean difference between the Masimo and Samsung devices for pretest [Formula: see text] was 3.2 ± 2.8%, and the difference for posttest [Formula: see text] was 2.4 ± 3.5%. The number of subjects who were unable to obtain [Formula: see text] was higher with the Samsung device than with the Apple device in both pretest (14 of 47 vs 3 of 47) and posttest (17 of 47 vs 5 of 47). In contrast, the Masimo device was able to measure [Formula: see text] in all subjects.ConclusionsSmartphone pulse oximeters were unreliable compared to a hospital pulse oximeter. Further research is needed with evolving technology to better understand smartphone pulse oximetry. (ClinicalTrials.gov registration NCT03534271.).

Project description:Body composition is a key component of health in both individuals and populations, and excess adiposity is associated with an increased risk of developing chronic diseases. Body mass index (BMI) and other clinical or commercially available tools for quantifying body fat (BF) such as DXA, MRI, CT, and photonic scanners (3DPS) are often inaccurate, cost prohibitive, or cumbersome to use. The aim of the current study was to evaluate the performance of a novel automated computer vision method, visual body composition (VBC), that uses two-dimensional photographs captured via a conventional smartphone camera to estimate percentage total body fat (%BF). The VBC algorithm is based on a state-of-the-art convolutional neural network (CNN). The hypothesis is that VBC yields better accuracy than other consumer-grade fat measurements devices. 134 healthy adults ranging in age (21-76 years), sex (61.2% women), race (60.4% White; 23.9% Black), and body mass index (BMI, 18.5-51.6 kg/m2) were evaluated at two clinical sites (N = 64 at MGH, N = 70 at PBRC). Each participant had %BF measured with VBC, three consumer and two professional bioimpedance analysis (BIA) systems. The PBRC participants also had air displacement plethysmography (ADP) measured. %BF measured by dual-energy x-ray absorptiometry (DXA) was set as the reference against which all other %BF measurements were compared. To test our scientific hypothesis we run multiple, pair-wise Wilcoxon signed rank tests where we compare each competing measurement tool (VBC, BIA, …) with respect to the same ground-truth (DXA). Relative to DXA, VBC had the lowest mean absolute error and standard deviation (2.16 ± 1.54%) compared to all of the other evaluated methods (p < 0.05 for all comparisons). %BF measured by VBC also had good concordance with DXA (Lin's concordance correlation coefficient, CCC: all 0.96; women 0.93; men 0.94), whereas BMI had very poor concordance (CCC: all 0.45; women 0.40; men 0.74). Bland-Altman analysis of VBC revealed the tightest limits of agreement (LOA) and absence of significant bias relative to DXA (bias -0.42%, R2 = 0.03; p = 0.062; LOA -5.5% to +4.7%), whereas all other evaluated methods had significant (p < 0.01) bias and wider limits of agreement. Bias in Bland-Altman analyses is defined as the discordance between the y = 0 axis and the regressed line computed from the data in the plot. In this first validation study of a novel, accessible, and easy-to-use system, VBC body fat estimates were accurate and without significant bias compared to DXA as the reference; VBC performance exceeded those of all other BIA and ADP methods evaluated. The wide availability of smartphones suggests that the VBC method for evaluating %BF could play an important role in quantifying adiposity levels in a wide range of settings.Trial registration: ClinicalTrials.gov Identifier: NCT04854421.

Project description:ObjectivesTo assess disparities in hypoxemia detection by pulse oximetry across self-identified racial groups and associations with clinical outcomes.DesignObservational cohort study from May 5, 2018, to December 31, 2020.SettingThree academic medical centers in the United States.PatientsAdults greater than or equal to 18 years who self-identified as White, Black, Asian, or American Indian admitted to the ICU or undergoing surgery during inpatient hospitalization with simultaneous measurements of pulse oximetry-estimated oxygen saturation and arterial blood gas-derived oxygen saturation.InterventionsNone.Measurements and main resultsMultivariable models were employed to assess the relationships between race, occult hypoxemia (i.e., arterial blood gas-derived oxygen saturation < 88% despite pulse oximetry-estimated oxygen saturation ≥ 92%), and clinical outcomes of hospital mortality and hospital-free days. One-hundred twenty-eight-thousand two-hundred eighty-five paired pulse oximetry-estimated oxygen saturation-arterial blood gas-derived oxygen saturation measurements were included from 26,603 patients. Pulse oximetry-estimated oxygen saturation on average overestimated arterial blood gas-derived oxygen saturation by 1.57% (1.54-1.61%). Black, Asian, and American Indian patients were more likely to experience occult hypoxemia during hospitalization (estimated probability 6.2% [5.1-7.6%], 6.6% [4.9-8.8%], and 6.6% [4.4-10.0%], respectively) compared with White patients (3.6% [3.4-3.8%]). Black patients had increased odds of occult hypoxemia compared with White patients after adjustment (odds ratio, 1.65; 1.28-2.14; p < 0.001). Differences in occult hypoxemia between Asian and American Indian patients compared with White patients were not significant after adjustment (odds ratio, 1.53; 0.95-2.47; p = 0.077 and odds ratio, 1.31; 0.80-2.16; p = 0.288, respectively). Occult hypoxemia was associated with increased odds of mortality in surgical (odds ratio, 2.96; 1.20-7.28; p = 0.019) and ICU patients (1.36; 1.03-1.80; p = 0.033). Occult hypoxemia was associated with fewer hospital-free days in surgical (-2.5 d [-3.9 to -1.2 d]; p < 0.001) but not ICU patients (0.4 d [-0.7 to 1.4 d]; p = 0.500).ConclusionsOccult hypoxemia is more common in Black patients compared with White patients and is associated with increased mortality, suggesting potentially important outcome implications for undetected hypoxemia. It is imperative to validate pulse oximetry with expanded racial inclusion.

Dataset Information

Smartphone camera oximetry in an induced hypoxemia study.

Publications

Smartphone camera oximetry in an induced hypoxemia study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets