Dataset Information

Smartphone-Based VO2max Measurement With Heart Snapshot in Clinical and Real-world Settings With a Diverse Population: Validation Study.

ABSTRACT:

Background

Maximal oxygen consumption (VO₂max) is one of the most predictive biometrics for cardiovascular health and overall mortality. However, VO₂max is rarely measured in large-scale research studies or routine clinical care because of the high cost, participant burden, and requirement for specialized equipment and staff.

Objective

To overcome the limitations of clinical VO₂max measurement, we aim to develop a digital VO₂max estimation protocol that can be self-administered remotely using only the sensors within a smartphone. We also aim to validate this measure within a broadly representative population across a spectrum of smartphone devices.

Methods

Two smartphone-based VO₂max estimation protocols were developed: a 12-minute run test (12-MRT) based on distance measured by GPS and a 3-minute step test (3-MST) based on heart rate recovery measured by a camera. In a 101-person cohort, balanced across age deciles and sex, participants completed a gold standard treadmill-based VO₂max measurement, two silver standard clinical protocols, and the smartphone-based 12-MRT and 3-MST protocols in the clinic and at home. In a separate 120-participant cohort, the video-based heart rate measurement underlying the 3-MST was measured for accuracy in individuals across the spectrum skin tones while using 8 different smartphones ranging in cost from US $99 to US $999.

Results

When compared with gold standard VO₂max testing, Lin concordance was p_c=0.66 for 12-MRT and p_c=0.61 for 3-MST. However, in remote settings, the 12-MRT was significantly less concordant with the gold standard (p_c=0.25) compared with the 3-MST (p_c=0.61), although both had high test-retest reliability (12-MRT intraclass correlation coefficient=0.88; 3-MST intraclass correlation coefficient=0.86). On the basis of the finding that 3-MST concordance was generalizable to remote settings whereas 12-MRT was not, the video-based heart rate measure within the 3-MST was selected for further investigation. Heart rate measurements in any of the combinations of the six Fitzpatrick skin tones and 8 smartphones resulted in a concordance of p_c≥0.81. Performance did not correlate with device cost, with all phones selling under US $200 performing better than p_c>0.92.

Conclusions

These findings demonstrate the importance of validating mobile health measures in the real world across a diverse cohort and spectrum of hardware. The 3-MST protocol, termed as heart snapshot, measured VO₂max with similar accuracy to supervised in-clinic tests such as the Tecumseh (p_c=0.94) protocol, while also generalizing to remote and unsupervised measurements. Heart snapshot measurements demonstrated fidelity across demographic variation in age and sex, across diverse skin pigmentation, and between various iOS and Android phone configurations. This software is freely available for all validation data and analysis code.

SUBMITTER: Webster DE

PROVIDER: S-EPMC8214186 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundUsing mobile communication technology as new personalized approach to treat mental disorders or to more generally improve quality of life is highly promising. Knowledge about intervention components that target key psychopathological processes in terms of transdiagnostic psychotherapy approaches is urgently needed. We explored the use of smartphone-based micro-interventions based on psychotherapeutic techniques, guided by short video-clips, to elicit mood changes.MethodAs part of a larger neurofeedback study, all subjects-after being randomly assigned to an experimental or control neurofeedback condition-underwent daily smartphone-based micro-interventions for 13 consecutive days. They were free to choose out of provided techniques, including viscerosensory attention, emotional imagery, facial expression, and contemplative repetition. Changes in mood were assessed in real world using the Multidimensional Mood State Questionnaire (scales: good-bad, GB; awake-tired, AT; and calm-nervous, CN).ResultsTwenty-seven men participated on at least 11 days and were thus included in the analyses. Altogether, they underwent 335, generally well-tolerated, micro-intervention sessions, with viscerosensory attention (178 sessions, 53.13%) and contemplative repetition (68 sessions, 20.30%) being the most frequently applied techniques. Mixed models indicated that subjects showed better mood [GB: b = 0.464, 95%confidence interval (CI) [0.068, 0.860], t (613.3) = 2.298, p = 0.022] and became more awake [AT: b = 0.514, 95%CI [0.103, 0.925], t (612.4) = 2.456, p = 0.014] and calmer [CN: b = 0.685, 95%CI [0.360, 1.010], t (612.3) = 4.137, p < 0.001] from pre- to post-micro-intervention. These mood improvements from pre- to post-micro-intervention were associated with changes in mood from the 1st day until the last day with regard to GB mood (r = 0.614, 95%CI [0.297, 0.809], p < 0.001), but not AT mood (r = 0.279, 95%CI [-0.122, 0.602], p = 0.167) and CN mood (r = 0.277, 95%CI [0.124, 0.601], p = 0.170).DiscussionOur findings provide evidence for the applicability of smartphone-based micro-interventions eliciting short-term mood changes, based on techniques used in psychotherapeutic approaches, such as mindfulness-based psychotherapy, transcendental meditation, and other contemplative therapies. The results encourage exploring these techniques' capability to improve mood in randomized controlled studies and patients. Smartphone-based micro-interventions are promising to modify mood in real-world settings, complementing other psychotherapeutic interventions, in line with the precision medicine approach. The here presented data were collected within a randomized trial, registered at ClinicalTrials.gov (Identifier: NCT01921088) https://clinicaltrials.gov/ct2/show/NCT01921088.

Project description:BackgroundWalking patterns can provide important indications of a person's health status and be beneficial in the early diagnosis of individuals with a potential walking disorder. For appropriate gait analysis, it is critical that natural functional walking characteristics are captured, rather than those experienced in artificial or observed settings. To better understand the extent to which setting influences gait patterns, and particularly whether observation plays a varying role on subjects of different ages, the current study investigates to what extent people walk differently in lab versus real-world environments and whether age dependencies exist.MethodsThe walking patterns of 20 young and 20 elderly healthy subjects were recorded with five wearable inertial measurement units (ZurichMOVE sensors) attached to both ankles, both wrists and the chest. An automated detection process based on dynamic time warping was developed to efficiently identify the relevant sequences. From the ZurichMOVE recordings, 15 spatio-temporal gait parameters were extracted, analyzed and compared between motion patterns captured in a controlled lab environment (10 m walking test) and the non-controlled ecologically valid real-world environment (72 h recording) in both groups.ResultsSeveral parameters (Cluster A) showed significant differences between the two environments for both groups, including an increased outward foot rotation, step width, number of steps per 180° turn, stance to swing ratio, and cycle time deviation in the real-world. A number of parameters (Cluster B) showed only significant differences between the two environments for elderly subjects, including a decreased gait velocity (p = 0.0072), decreased cadence (p = 0.0051) and increased cycle time (p = 0.0051) in real-world settings. Importantly, the real-world environment increased the differences in several parameters between the young and elderly groups.ConclusionElderly test subjects walked differently in controlled lab settings compared to their real-world environments, which indicates the need to better understand natural walking patterns under ecologically valid conditions before clinically relevant conclusions can be drawn on a subject's functional status. Moreover, the greater inter-group differences in real-world environments seem promising regarding the sensitive identification of subjects with indications of a walking disorder.

Project description:BackgroundPrevious cardiovascular risk prediction models in Japan have utilized prospective cohort studies with concise data. As the health information including health check-up records and administrative claims becomes digitalized and publicly available, application of large datasets based on such real-world data can achieve prediction accuracy and support social implementation of cardiovascular disease risk prediction models in preventive and clinical practice. In this study, classical regression and machine learning methods were explored to develop ischemic heart disease (IHD) and stroke prognostic models using real-world data.MethodsIQVIA Japan Claims Database was searched to include 691,160 individuals (predominantly corporate employees and their families working in secondary and tertiary industries) with at least one annual health check-up record during the identification period (April 2013-December 2018). The primary outcome of the study was the first recorded IHD or stroke event. Predictors were annual health check-up records at the index year-month, comprising demographic characteristics, laboratory tests, and questionnaire features. Four prediction models (Cox, Elnet-Cox, XGBoost, and Ensemble) were assessed in the present study to develop a cardiovascular disease risk prediction model for Japan.ResultsThe analysis cohort consisted of 572,971 invididuals. All prediction models showed similarly good performance. The Harrell's C-index was close to 0.9 for all IHD models, and above 0.7 for stroke models. In IHD models, age, sex, high-density lipoprotein, low-density lipoprotein, cholesterol, and systolic blood pressure had higher importance, while in stroke models systolic blood pressure and age had higher importance.ConclusionOur study analyzed classical regression and machine learning algorithms to develop cardiovascular disease risk prediction models for IHD and stroke in Japan that can be applied to practical use in a large population with predictive accuracy.

Project description:BackgroundStep counts are increasingly used in public health and clinical research to assess wellbeing, lifestyle, and health status. However, estimating step counts using commercial activity trackers has several limitations, including a lack of reproducibility, generalizability, and scalability. Smartphones are a potentially promising alternative, but their step-counting algorithms require robust validation that accounts for temporal sensor body location, individual gait characteristics, and heterogeneous health states.ObjectiveOur goal was to evaluate an open-source step-counting method for smartphones under various measurement conditions against step counts estimated from data collected simultaneously from different body locations ("internal" validation), manually ascertained ground truth ("manual" validation), and step counts from a commercial activity tracker (Fitbit Charge 2) in patients with advanced cancer ("wearable" validation).MethodsWe used eight independent datasets collected in controlled, semi-controlled, and free-living environments with different devices (primarily Android smartphones and wearable accelerometers) carried at typical body locations. Five datasets (N=103) were used for internal validation, two datasets (N=107) for manual validation, and one dataset (N=45) used for wearable validation. In each scenario, step counts were estimated using a previously published step-counting method for smartphones that uses raw sub-second level accelerometer data. We calculated mean bias and limits of agreement (LoA) between step count estimates and validation criteria using Bland-Altman analysis.ResultsIn the internal validation datasets, participants performed 751.7±581.2 (mean±SD) steps, and the mean bias was -7.2 steps (LoA -47.6, 33.3) or -0.5%. In the manual validation datasets, the ground truth step count was 367.4±359.4 steps while the mean bias was -0.4 steps (LoA -75.2, 74.3) or 0.1 %. In the wearable validation dataset, Fitbit devices indicated mean step counts of 1931.2±2338.4, while the calculated bias was equal to -67.1 steps (LoA -603.8, 469.7) or a difference of 0.3 %.ConclusionsThis study demonstrates that our open-source step counting method for smartphone data provides reliable step counts across sensor locations, measurement scenarios, and populations, including healthy adults and patients with cancer.