Search

The adverse health effects of sedentary behavior and prolonged sedentary bouts are well-known. However, it is still unknown how physical activity can modify adverse health impacts related to sedentary behavior. The purpose of this study was to develop signal analysis methodology for sedentary behavior and physical activity classification from raw data of a hip-worn accelerometer and to investigate associations of patterns of sedentary behavior with lipid and glucose metabolism. A machine learning model was developed and validated using acceleration data, which included nine predefined and controlled typical daily activities ranging in intensity from sedentary to vigorous physical activity. Acceleration data was collected from 22 Finnish adults using a triaxial accelerometer attached to an elastic belt on a hip. The data were classified into five categories (lying down, sitting, and light, moderate, and vigorous physical activity). Thirty-six middle-aged Finnish adults wore an accelerometer for 14 days, and their sedentary behavior and sitting characteristics were determined. In addition, associations of sedentary behavior, sitting, and physical activity with glucose and lipid metabolism were investigated in the Northern Finland Birth Cohort 1966 46-year follow-up (n=5,832). Participants completed health and lifestyle questionnaires and attended clinical examinations and two weeks of sedentary behavior and physical activity measurements. Isotemporal substitution modeling was used for investigating time reallocations from sedentary to physical activities. The developed machine learning model provided acceptable accuracy for sedentary behavior and physical activity classifications. The method can be used for describing characteristics of sedentary behavior and sitting separately. Patterns of SB were more consistently associated with lipid metabolism than those of sitting. Associations between sedentary behavior and cardiometabolic health depended on moderate-t, Tiivistelmä Runsaan paikallaanolon ja pitkien paikallaanolojaksojen haitalliset vaikutukset terveyteen ovat hyvin tiedossa, mutta edelleen puuttuu tietoa, kuinka fyysinen aktiivisuus voi muokata paikallaanolon haitallisia vaikutuksia terveyteen. Työn tarkoituksena oli kehittää signaalinkäsittelymenetelmä paikallaanolon ja fyysisen aktiivisuuden luokittelemiseksi lantiolla pidettävän aktiivisuusmittarin raakakiihtyvyyksistä sekä tutkia paikallaanolon piirteiden yhteyksiä rasva- ja sokeriaineenvaihduntaan. Koneoppimismalli kehitettiin ja validoitiin käyttämällä kiihtyvyysdataa, joka sisälsi yhdeksän ennalta määriteltyä ja kontrolloitua tyypillistä arkiaktiviteettia, joiden intensiteetti vaihteli paikallaanolosta raskaaseen fyysiseen aktiivisuuteen. Kiihtyvyysdata kerättiin 22 suomalaisen aikuisen lantiolta elastiseen vyöhön kiinnitetyllä kolmiakselisella kiihtyvyysanturilla, jonka raakadata luokiteltiin viiteen eri luokkaan (makaaminen, istuminen sekä kevyt, keskiraskas ja raskas fyysinen aktiivisuus). Kolmekymmentäkuusi keski-ikäistä suomalaista käyttivät kiihtyvyysanturia 14 päivän ajan ja heidän paikallaanolonsa ja istumisensa piirteet määritettiin. Paikallaanolon, istumisen ja fyysisen aktiivisuuden yhteyksiä rasva- ja sokeriaineenvaihduntaan tutkittiin Pohjois-Suomen vuoden 1966 syntymäkohortin 46-vuotistutkimuksessa (n = 5832). Tutkittavat täyttivät terveys- ja elämäntapakyselyitä ja osallistuivat kliinisiin tutkimuksiin sekä kahden viikon mittaisiin paikallaanolon ja fyysisen aktiivisuuden mittauksiin. Isotemporaalisella korvausmallilla tutkittiin paikallaanoloajan korvaamista fyysisellä aktiivisuudella. Kehitetyllä koneoppimismenetelmällä voidaan riittävällä tarkkuudella luokitella paikallaanoloa ja fyysistä aktiivisuutta. Menetelmää voidaan käyttää paikallaanolon ja istumisen piirteiden kuvailussa erikseen. Paikallaanolon piirteet olivat selvemmin yhteydessä rasva-aineenvaihduntaan kuin istumisen piirteet. Paikallaanolon yhteydet rasva-aineenvaihduntaan oliva

Introduction : Physical inactivity, excessive total time spent in sedentary behavior (SB) and prolonged sedentary bouts have been proposed to be risk factors for chronic disease morbidity and mortality worldwide. However, which patterns and postures of SB have the most negative impacts on health outcomes is still unclear. This population-based study aimed to investigate the independent associations of the patterns of accelerometer-based overall SB and sitting with serum lipid biomarkers at different moderate- to vigorous-intensity physical activity (MVPA) levels. Methods: Physical activity and SB were measured in a birth cohort sample (N = 3272) at 46 yr using a triaxial hip-worn accelerometer in free-living conditions for 14 d. Raw acceleration data were classified into SB and PA using a machine learning–based model, and the bouts of overall SB and sitting were identified from the classified data. The participants also answered health-related questionnaires and participated in clinical examinations. Associations of overall SB (lying and sitting) and sitting patterns with serum lipid biomarkers were investigated using linear regression. Results: The overall SB patterns were more consistently associated with serum lipid biomarkers than the sitting patterns after adjustments. Among the participants with the least and the most MVPA, high total time spent in SB and SB bouts of 15–29.99 and ≥30 min were associated with impaired lipid metabolism. Among those with moderate amount of MVPA, higher time spent in SB and SB bouts of 15–29.99 min was unfavorably associated with serum lipid biomarkers. Conclusions: The associations between SB patterns and serum lipid biomarkers were dependent on MVPA level, which should be considered when planning evidence-based interventions to decrease SB in midlife.

Purpose: Machine-learning (ML) approaches have been repeatedly coupled with raw accelerometry to classify physical activity classes, but the features required to optimize their predictive performance are still unknown. Our aim was to identify appropriate combination of feature subsets and prediction algorithms for activity class prediction from hip-based raw acceleration data. Methods: The hip-based raw acceleration data collected from 27 participants was split into training (70 %) and validation (30 %) subsets. A total of 206 time- (TD) and frequencydomain (FD) features were extracted from 6-second non-overlapping windows of the signal. Feature selection was done using seven filter-based, two wrapper-based, and one embedded algorithm, and classification was performed with artificial neural network (ANN), support vector machine (SVM), and random forest (RF). For every combination between the feature selection method and the classifiers, the most appropriate feature subsets were found and used for model training within the training set. These models were then validated with the left-out validation set. Results: The appropriate number of features for the ANN, SVM, and RF ranged from 20 to 45. Overall, the accuracy of all the three classifiers was higher when trained with feature subsets generated using filter-based methods compared with when they were trained with wrapper-based methods (range: 78.1 %–88 % vs. 66 %–83.5 %). TD features that reflect how signals vary around the mean, how they differ with one another, and how much and how often they change were more frequently selected via the feature selection methods. Conclusions: A subset of TD features from raw accelerometry could be sufficient for ML-based activity classification if properly selected from different axes.

Purpose: To evaluate and enhance the generalization performance of machine learning physical activity intensity prediction models developed with raw acceleration data on populations monitored by different activity monitors. Method: Five datasets from four studies, each containing only hip- or wrist-based raw acceleration data (two hip- and three wrist-based) were extracted. The five datasets were then used to develop and validate artificial neural networks (ANN) in three setups to classify activity intensity categories (sedentary behavior, light, and moderate-to-vigorous). To examine generalizability, the ANN models were developed using within-dataset (leave-one-subject-out) cross-validation, and then cross-tested to other datasets with different accelerometers. To enhance the models’ generalizability, a combination of four of the five datasets was used for training and the fifth dataset for validation. Finally, all the five datasets were merged to develop a single model that is generalizable across the datasets (50% of the subjects from each dataset for training, the remaining for validation). Results: The datasets showed high performance in within-dataset cross-validation (accuracy 71.9–95.4%, Kappa K=0.63–0.94). The performance of the within-dataset validated models decreased when applied to datasets with different accelerometers (41.2–59.9%, K=0.21–0.48). The trained models on merged datasets consisting hip and wrist data predicted the left-out dataset with acceptable performance (65.9–83.7%, K=0.61–0.79). The model trained with all five datasets performed with acceptable performance across the datasets (80.4–90.7%, K=0.68–0.89). Conclusions: Integrating heterogeneous datasets in training sets seems a viable approach for enhancing the generalization performance of the models. Instead, within-dataset validation is not sufficient to understand the models’ performance on other populations with different accelerometers.

The purpose of this study was to describe the accelerometry-based characteristics of overall sedentary behavior (SB) and sitting among adults under free-living conditions. Thirty-six (mean age 47.6 years) volunteers carried a waist-worn accelerometer for ≥4 days with data ≥600 min/d during 14 consecutive days. A machine learning (ML) based method was used to classify the patterns of SB and sitting from raw 3D acceleration. The participants spent most (69.3%) of their waking time in SB, and a half (52.2%) of the SB was performed in a sitting posture. Men broke their overall sedentary time less often (4.1 vs. 6.1 bouts/h), but women sat more; however, women broke their sitting time as often as men (7.6 bouts/h). This study confirms that SB and sitting can be distinguished using ML methods, and more information about SB can be achieved when overall SB and sitting are analyzed separately in free-living conditions.