Existing accelerometer-based human activity recognition (HAR) benchmark datasets that were recorded during free living suffer from non-fixed sensor placement, the usage of only one sensor, and unreliable annotations. We make two contributions in this work. First, we present the publicly available Human Activity Recognition Trondheim dataset (HARTH). Twenty-two participants were recorded for 90 to 120 min during their regular working hours using two three-axial accelerometers, attached to the thigh and lower back, and a chest-mounted camera. Experts annotated the data independently using the camera’s video signal and achieved high inter-rater agreement (Fleiss’ Kappa $=0.96$). They labeled twelve activities. The second contribution of this paper is the training of seven different baseline machine learning models for HAR on our dataset. We used a support vector machine, k-nearest neighbor, random forest, extreme gradient boost, convolutional neural network, bidirectional long short-term memory, and convolutional neural network with multi-resolution blocks. The support vector machine achieved the best results with an F1-score of $0.81$ (standard deviation: $\pm 0.18$), recall of $0.85\pm 0.13$, and precision of $0.79\pm 0.22$ in a leave-one-subject-out cross-validation. Our highly professional recordings and annotations provide a promising benchmark dataset for researchers to develop innovative machine learning approaches for precise HAR in free living.

Background: Aging is associated with reduced muscle mass and strength leading to impaired physical function. Resistance training programs incorporated into older adults’ real-life settings may have the potential to counteract these changes. We evaluated the effectiveness of 8 months resistance training using easily available, low cost equipment compared to physical activity counselling on physical function, muscle strength, and body composition in community-dwelling older adults receiving home care. Methods: This open label, two-armed, parallel group, cluster randomized trial recruited older adults above 70 years (median age 86.0 (Interquartile range 80–90) years) receiving home care. Participants were randomized at cluster level to the resistance training group (RTG) or the control group (CG). The RTG trained twice a week while the CG were informed about the national recommendations for physical activity and received a motivational talk every 6th week. Outcomes were assessed at participant level at baseline, after four, and 8 months and included tests of physical function (chair rise, 8 ft-up-and-go, preferred- and maximal gait speed, and stair climb), maximal strength, rate of force development, and body composition. Results: Twelve clusters were allocated to RTG (7 clusters, 60 participants) or CG (5 clusters, 44 participants). The number of participants analyzed was 56–64 (6–7 clusters) in RTG and 20–42 (5 clusters) in CG. After 8 months, multilevel linear mixed models showed that RTG improved in all tests of physical function and maximal leg strength (9–24%, p = 0.01–0.03) compared to CG. No effects were seen for rate of force development or body composition. Conclusion: This study show that resistance training using easily available, low cost equipment is more effective than physical activity counselling for improving physical function and maximal strength in community-dwelling older adults receiving home care. Trial registration: ISRCTN1067873