Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments
Abstract Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (VO2 m a x), or indirectly assessed using heart rate responses to standard exercise tests. However, such testing is costly and burdensome beca...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-12-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-022-00719-1 |
| _version_ | 1827775424181043200 |
|---|---|
| author | Dimitris Spathis Ignacio Perez-Pozuelo Tomas I. Gonzales Yu Wu Soren Brage Nicholas Wareham Cecilia Mascolo |
| author_facet | Dimitris Spathis Ignacio Perez-Pozuelo Tomas I. Gonzales Yu Wu Soren Brage Nicholas Wareham Cecilia Mascolo |
| author_sort | Dimitris Spathis |
| collection | DOAJ |
| description | Abstract Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (VO2 m a x), or indirectly assessed using heart rate responses to standard exercise tests. However, such testing is costly and burdensome because it requires specialized equipment such as treadmills and oxygen masks, limiting its utility. Modern wearables capture dynamic real-world data which could improve fitness prediction. In this work, we design algorithms and models that convert raw wearable sensor data into cardiorespiratory fitness estimates. We validate these estimates’ ability to capture fitness profiles in free-living conditions using the Fenland Study (N=11,059), along with its longitudinal cohort (N = 2675), and a third external cohort using the UK Biobank Validation Study (N = 181) who underwent maximal VO2 m a x testing, the gold standard measurement of fitness. Our results show that the combination of wearables and other biomarkers as inputs to neural networks yields a strong correlation to ground truth in a holdout sample (r = 0.82, 95CI 0.80–0.83), outperforming other approaches and models and detects fitness change over time (e.g., after 7 years). We also show how the model’s latent space can be used for fitness-aware patient subtyping paving the way to scalable interventions and personalized trial recruitment. These results demonstrate the value of wearables for fitness estimation that today can be measured only with laboratory tests. |
| first_indexed | 2024-03-11T13:49:41Z |
| format | Article |
| id | doaj.art-ea171e38e4ae4bd885b2e3340cd138f3 |
| institution | Directory Open Access Journal |
| issn | 2398-6352 |
| language | English |
| last_indexed | 2024-03-11T13:49:41Z |
| publishDate | 2022-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj.art-ea171e38e4ae4bd885b2e3340cd138f32023-11-02T09:23:10ZengNature Portfolionpj Digital Medicine2398-63522022-12-015111110.1038/s41746-022-00719-1Longitudinal cardio-respiratory fitness prediction through wearables in free-living environmentsDimitris Spathis0Ignacio Perez-Pozuelo1Tomas I. Gonzales2Yu Wu3Soren Brage4Nicholas Wareham5Cecilia Mascolo6Department of Computer Science and Technology, University of CambridgeMRC Epidemiology Unit, School of Clinical Medicine, University of CambridgeMRC Epidemiology Unit, School of Clinical Medicine, University of CambridgeDepartment of Computer Science and Technology, University of CambridgeMRC Epidemiology Unit, School of Clinical Medicine, University of CambridgeMRC Epidemiology Unit, School of Clinical Medicine, University of CambridgeDepartment of Computer Science and Technology, University of CambridgeAbstract Cardiorespiratory fitness is an established predictor of metabolic disease and mortality. Fitness is directly measured as maximal oxygen consumption (VO2 m a x), or indirectly assessed using heart rate responses to standard exercise tests. However, such testing is costly and burdensome because it requires specialized equipment such as treadmills and oxygen masks, limiting its utility. Modern wearables capture dynamic real-world data which could improve fitness prediction. In this work, we design algorithms and models that convert raw wearable sensor data into cardiorespiratory fitness estimates. We validate these estimates’ ability to capture fitness profiles in free-living conditions using the Fenland Study (N=11,059), along with its longitudinal cohort (N = 2675), and a third external cohort using the UK Biobank Validation Study (N = 181) who underwent maximal VO2 m a x testing, the gold standard measurement of fitness. Our results show that the combination of wearables and other biomarkers as inputs to neural networks yields a strong correlation to ground truth in a holdout sample (r = 0.82, 95CI 0.80–0.83), outperforming other approaches and models and detects fitness change over time (e.g., after 7 years). We also show how the model’s latent space can be used for fitness-aware patient subtyping paving the way to scalable interventions and personalized trial recruitment. These results demonstrate the value of wearables for fitness estimation that today can be measured only with laboratory tests.https://doi.org/10.1038/s41746-022-00719-1 |
| spellingShingle | Dimitris Spathis Ignacio Perez-Pozuelo Tomas I. Gonzales Yu Wu Soren Brage Nicholas Wareham Cecilia Mascolo Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments npj Digital Medicine |
| title | Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments |
| title_full | Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments |
| title_fullStr | Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments |
| title_full_unstemmed | Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments |
| title_short | Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments |
| title_sort | longitudinal cardio respiratory fitness prediction through wearables in free living environments |
| url | https://doi.org/10.1038/s41746-022-00719-1 |
| work_keys_str_mv | AT dimitrisspathis longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT ignacioperezpozuelo longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT tomasigonzales longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT yuwu longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT sorenbrage longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT nicholaswareham longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments AT ceciliamascolo longitudinalcardiorespiratoryfitnesspredictionthroughwearablesinfreelivingenvironments |