The ability to optimize power generation in sports is imperative, both for understanding and balancing training load correctly, and for optimizing competition performance. In this paper, we aim to estimate mechanical power output by employing a time-sequential information-based deep Long Short-Term Memory (LSTM) neural network from multiple inertial measurement units (IMUs). Thirteen athletes conducted roller ski skating trials on a treadmill with varying incline and speed. The acceleration and gyroscope data collected with the IMUs were run through statistical feature processing, before being used by the deep learning model to estimate power output. The model was thereafter used for prediction of power from test data using two approaches. First, a user-dependent case was explored, reaching a power estimation within 3.5% error. Second, a user-independent case was developed, reaching an error of 11.6% for the power estimation. Finally, the LSTM model was compared to two other machine learning models and was found to be superior. In conclusion, the user-dependent model allows for precise estimation of roller skiing power output after training the model on data from each athlete. The user-independent model provides less accurate estimation; however, the accuracy may be sufficient for providing valuable information for recreational skiers.

The pressure on the healthcare services is building up for several reasons. The ageing population trend, the increase in life-style related disease prevalence, as well as the increased treatment capabilities with associated general expectation all add pressure. The use of ambient healthcare technologies can alleviate the situation by enabling time and cost-efficient monitoring and follow-up of patients discharged from hospital care. We report on an ambulatory system developed for monitoring of physical rehabilitation patients. The system consists of a wearable multisensor monitoring device; a mobile phone with client application aggregating the data collected; a service-oriented-architecture based server solution; and a PC application facilitating patient follow-up by their health professional carers. The system has been tested and verified for accuracy in controlled environment trials on healthy volunteers, and also been usability tested by 5 congestive heart failure patients and their nurses. This investigation indicated that patients were able to use the system, and that nurses got an improved basis for patient follow-up.

No abstract

The cold and harsh climate in the High North represents a threat to safety and work performance. The aim of this study was to show that sensors integrated in clothing can provide information that can improve decision support for workers in cold climate without disturbing the user. Here, a wireless demonstrator consisting of a working jacket with integrated temperature, humidity and activity sensors has been developed. Preliminary results indicate that the demonstrator can provide easy accessible information about the thermal conditions at the site of the worker and local cooling effects of extremities. The demonstrator has the ability to distinguish between activity and rest, and enables implementation of more sophisticated sensor fusion algorithms to assess work load and pre-defined activities. This information can be used in an enhanced safety perspective as an improved tool to advice outdoor work control for workers in cold climate.

No abstract