We present a personalized approach for frequent fitness monitoring in road cycling solely relying on sensor data collected during bike rides and without the need for maximal effort tests. We use... Show moreWe present a personalized approach for frequent fitness monitoring in road cycling solely relying on sensor data collected during bike rides and without the need for maximal effort tests. We use competition and training data of three world-class cyclists of Team Jumbo-Visma to construct personalised heart rate models that relate the heart rate during exercise to the pedal power signal. Our model captures the non-trivial dependency between exertion and corresponding response of the heart rate, which we show can be effectively estimated by an exponential kernel. To construct the daily heart rate models that are required for day-to-day fitness estimation, we aggregate all sessions in the previous week and apply sampling. On average, the explained variance of our models is 0.86, which we demonstrate is more than twice as large as for models that ignore the temporal integration involved in the heart's response to exercise. We show that the fitness of a cyclist can be monitored by tracking developments of parameters of our heart rate models. In particular, we monitor the decay constant of the kernel involved, and also analytically determine virtual aerobic and anaerobic thresholds. We demonstrate that our findings for the virtual anaerobic threshold on average agree with the results of exercise tests. We believe this work is an important step forward in performance optimization by opening up avenues for switching to adaptive training programs that take into account the current physiological state of an athlete. Show less
Leeuw, A.W. de; Baar, R. van.; Knobbe, A.J.; Zwaard, S. van der 2022
In this study, we investigated the relationships between training load, perceived wellness and match performance in professional volleyball by applying the machine learning techniques XGBoost,... Show moreIn this study, we investigated the relationships between training load, perceived wellness and match performance in professional volleyball by applying the machine learning techniques XGBoost, random forest regression and subgroup discovery. Physical load data were obtained by manually logging all physical activities and using wearable sensors. Daily wellness of players was monitored using questionnaires. Match performance was derived from annotated actions by a video scout during matches. We identified conditions of predictor variables that related to attack and pass performance (p < 0.05). Better attack performance is related to heavy weights of lower-body strength training exercises in the preceding four weeks. However, worse attack performance is linked to large variations in weights of full-body strength training exercises, excessively heavy upper-body strength training, low jump heights and small variations in the number of high jumps in the four weeks prior to competition. Lower passing performance was associated with small variations in the number of high jumps in the preceding week and an excessive amount of high jumps performed, on average, in the two weeks prior to competition. Differences in findings with respect to passing and attack performance suggest that elite volleyball players can improve their performance if training schedules are adapted to the position of a player. Show less
This work introduces RADIUS, a framework for anomaly detection in sewer pipes using stereovision. The framework employs three-dimensional geometry reconstruction from stereo vision, followed by... Show moreThis work introduces RADIUS, a framework for anomaly detection in sewer pipes using stereovision. The framework employs three-dimensional geometry reconstruction from stereo vision, followed by statistical modeling of the geometry with a generic pipe model. The framework is designed to be compatible with existing workflows for sewer pipe defect detection, as well as to provide opportunities for machine learning implementations in the future. We test the framework on 48 image sets of 26 sewer pipes in different conditions collected in the lab. Of these 48 image sets, 5 could not be properly reconstructed in three dimensions due to insufficient stereo matching. The surface fitting and anomaly detection performed well: a human-graded defect severity score had a moderate, positive Pearson correlation of 0.65 with our calculated anomaly scores, making this a promising approach to automated defect detection in urban drainage. Show less
Through the quantification of physical activity energy expenditure (PAEE), health care monitoring has the potential to stimulate vital and healthy ageing, inducing behavioural changes in older... Show moreThrough the quantification of physical activity energy expenditure (PAEE), health care monitoring has the potential to stimulate vital and healthy ageing, inducing behavioural changes in older people and linking these to personal health gains. To be able to measure PAEE in a health care perspective, methods from wearable accelerometers have been developed, however, mainly targeted towards younger people. Since elderly subjects differ in energy requirements and range of physical activities, the current models may not be suitable for estimating PAEE among the elderly. Furthermore, currently available methods seem to be either simple but non-generalizable or require elaborate (manual) feature construction steps. Because past activities influence present PAEE, we propose a modeling approach known for its ability to model sequential data, the recurrent neural network (RNN). To train the RNN for an elderly population, we used the growing old together validation (GOTOV) dataset with 34 healthy participants of 60 years and older (mean 65 years old), performing 16 different activities. We used accelerometers placed on wrist and ankle, and measurements of energy counts by means of indirect calorimetry. After optimization, we propose an architecture consisting of an RNN with 3 GRU layers and a feedforward network combining both accelerometer and participant-level data. Our efforts included switching mean to standard deviation for down-sampling the input data and combining temporal and static data (person-specific details such as age, weight, BMI). The resulting architecture produces accurate PAEE estimations while decreasing training input and time by a factor of 10. Subsequently, compared to the state-of-the-art, it is capable to integrate longer activity data which lead to more accurate estimations of low intensity activities EE. It can thus be employed to investigate associations of PAEE with vitality parameters of older people related to metabolic and cognitive health and mental well-being. Show less
Leeuw, A.W. de; Zwaard, S. van der; Baar, R. van; Knobbe, A.J. 2021
A population group that is often overlooked in the recent revolution of self-tracking is the group of older people. This growing proportion of the general population is often faced with increasing... Show moreA population group that is often overlooked in the recent revolution of self-tracking is the group of older people. This growing proportion of the general population is often faced with increasing health issues and discomfort. In order to come up with lifestyle advice towards the elderly, we need the ability to quantify their lifestyle, before and after an intervention. This research focuses on the task of activity recognition (AR) from accelerometer data. With that aim, we collect a substantial labelled dataset of older individuals wearing multiple devices simultaneously and performing a strict protocol of 16 activities (the GOTOV dataset, 𝑁=28N=28). Using this dataset, we trained Random Forest AR models, under varying sensor set-ups and levels of activity description granularity. The model that combines ankle and wrist accelerometers (GENEActiv) produced the best results (accuracy >80%>80%) for 16-class classification. At the same time, when additional physiological information is used, the accuracy increased (>85%>85%). To further investigate the role of granularity in our predictions, we developed the LARA algorithm, which uses a hierarchical ontology that captures prior biological knowledge to increase or decrease the level of activity granularity (merge classes). As a result, a 12-class model in which the different paces of walking were merged showed a performance above 93%93%. Testing this 12-class model in labelled free-living pilot data, the mean balanced accuracy appeared to be reasonably high, while using the LARA algorithm, we show that a 7-class model (lying down, sitting, standing, household, walking, cycling, jumping) was optimal for accuracy and granularity. Finally, we demonstrate the use of the latter model in unlabelled free-living data from a larger lifestyle intervention study. In this paper, we make the validation data as well as the derived prediction models available to the community. Show less
In professional soccer, increasing amounts of data are collected that harness great potential when it comes to analysing tactical behaviour. Unlocking this potential is difficult as big data... Show moreIn professional soccer, increasing amounts of data are collected that harness great potential when it comes to analysing tactical behaviour. Unlocking this potential is difficult as big data challenges the data management and analytics methods commonly employed in sports. By joining forces with computer science, solutions to these challenges could be achieved, helping sports science to find new insights, as is happening in other scientific domains. We aim to bring multiple domains together in the context of analysing tactical behaviour in soccer using position tracking data. A systematic literature search for studies employing position tracking data to study tactical behaviour in soccer was conducted in seven electronic databases, resulting in 2338 identified studies and finally the inclusion of 73 papers. Each domain clearly contributes to the analysis of tactical behaviour, albeit in - sometimes radically - different ways. Accordingly, we present a multidisciplinary framework where each domain's contributions to feature construction, modelling and interpretation can be situated. We discuss a set of key challenges concerning the data analytics process, specifically feature construction, spatial and temporal aggregation. Moreover, we discuss how these challenges could be resolved through multidisciplinary collaboration, which is pivotal in unlocking the potential of position tracking data in sports analytics. Show less
Okai, J.; Paraschiakos, S.; Beekman, M.; Knobbe, A.J.; Rebelo de Sá, C. 2019
Human Activity Recognition (HAR) is a growing field of research in biomedical engineering and it has many potential applications in the treatment and prevention of several diseases. Due to the... Show moreHuman Activity Recognition (HAR) is a growing field of research in biomedical engineering and it has many potential applications in the treatment and prevention of several diseases. Due to the recent advancement in technology, devices that collect position and orientation measurements (e.g. accelerometers and gyroscopes) are becoming ubiquitous. These measurements can then be used to train machine learning models for HAR. In this research, we propose one recurrent neural network architecture and a data augmentation approach for building robust and accurate models for HAR. We compared models with Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) layers. The proposed data augmentation approach was used to make the models robust to the cases where one or more sensors are missing. In this empirical study, we could also understand some relations between the ideal locations of the sensors in the participants and the types of activities performed. The proposed approaches were tested in the GOTOv dataset from a study which involved 35 participants performing 16 sedentary, ambulatory and lifestyle activities in a semi-structured environment. The results presented, clearly show that the models are able to detect these activities in a robust way. Show less
This article focuses on the performance of runners in official races. Based on extensive public data from participants of races organized by the Boston Athletic Association, we demonstrate how... Show moreThis article focuses on the performance of runners in official races. Based on extensive public data from participants of races organized by the Boston Athletic Association, we demonstrate how different pacing profiles can affect the performance in a race. An athlete's pacing profile refers to the running speed at various stages of the race. We aim to provide practical, data-driven advice for professional as well as recreational runners. Our data collection covers 3 years of data made public by the race organizers, and primarily concerns the times at various intermediate points, giving an indication of the speed profile of the individual runner. We consider the 10 km, half marathon, and full marathon, leading to a data set of 120,472 race results. Although these data were not primarily recorded for scientific analysis, we demonstrate that valuable information can be gleaned from these substantial data about the right way to approach a running challenge. In this article, we focus on the role of race distance, gender, age, and the pacing profile. Since age is a crucial but complex determinant of performance, we first model the age effect in a gender- and distance-specific manner. We consider polynomials of high degree and use cross-validation to select models that are both accurate and of sufficient generalizability. After that, we perform clustering of the race profiles to identify the dominant pacing profiles that runners select. Finally, after having compensated for age influences, we apply a descriptive pattern mining approach to select reliable and informative aspects of pacing that most determine an optimal performance. The mining paradigm produces relatively simple and readable patterns, such that both professionals and amateurs can use the results to their benefit. Show less
In this paper, we investigate two variants of association rules for preference data, Label Ranking Association Rules and Pairwise Association Rules. Label Ranking Association Rules (LRAR) are the... Show moreIn this paper, we investigate two variants of association rules for preference data, Label Ranking Association Rules and Pairwise Association Rules. Label Ranking Association Rules (LRAR) are the equivalent of Class Association Rules (CAR) for the Label Ranking task. In CAR, the consequent is a single class, to which the example is expected to belong to. In LRAR, the consequent is a ranking of the labels. The generation of LRAR requires special support and confidence measures to assess the similarity of rankings. In this work, we carry out a sensitivity analysis of these similarity-based measures. We want to understand which datasets benefit more from such measures and which parameters have more influence in the accuracy of the model. Furthermore, we propose an alternative type of rules, the Pairwise Association Rules (PAR), which are defined as association rules with a set of pairwise preferences in the consequent. While PAR can be used both as descriptive and predictive models, they are essentially descriptive models. Experimental results show the potential of both approaches. Show less