Road accidents and victims provoked by drowsiness are a worldwide relevant social and economic problem. In EU, 25% of road accidents are related with fatigue. The relation between drowsiness and accidents is supported by scientific evidences that relate microsleep and other fatigue episodes with the loss of control of the vehicle. Nowadays there exist different technological approaches for drowsiness mitigating while driving, which try to reduce these accidents by detecting driver's physical condition and acting on the driver response. The techniques for mitigating accidents are based on algorithms of detection of physiological parameters that ensure drivers confidence and an appropriate use of the technology. The most frequently used non-invasive system are the cameras, which have been explored and used, mainly detecting eye movement and eyelid closure (PERCLOS). These systems have limitations due to artifacts and noise related to environmental and emotional conditions, which might lead to false positives. With the purpose of solving these shortcomings, the goal of the research is to develop a system capable of detecting the level of drowsiness based on the involuntary movements of the driver provoked by the respiration captured by means of cameras. In the current research, robustness in front of different types of users and circumstances has been explored. A system of reduced size automotive cameras of high dynamics is proposed to be used as breathing rate sensor. Images captured by these sensors will be processed to obtain the driver's chest/abdomen movement. These data will be analyzed in real time by a validated algorithm that interprets the movement and obtains the level of fatigue and drowsiness of the driver. A twofold “gold standard” is used to compare the robustness of the developed system. An experimental design has been developed in order to take into consideration different anthropometric characteristics, clothing types, user and vehicle movement and light conditions. The result will provide the boundary conditions of any system based on on-board cameras. These data will be used for building the algorithms to detect and interpret breathing patterns. The results show the capabilities of this approach and also permit to define the needs and requirements of the resulting technological developments

The presence of fatigue has been shown to modify running biomechanics. Overall in terms of gender, women are at lower risk than men for sustaining running-related injuries, although it depends on the factors taken into account. One possible reason for these differences in the injury rate and location might be the dissimilar running patterns between men and women. The purpose of this study was to determine the effect of fatigue and gender on the kinematic and ground reaction forces (GRF) parameters in recreational runners. Fifty-seven participants (28 males and 29 females) had kinematic and GRF variables measured while running at speed of 3.3 m s⁻¹ before and after a fatigue test protocol. The fatigue protocol included (1) a running Course-Navette test, (2) running up and down a flight of stairs for 5 min, and (3) performance of alternating jumps on a step (five sets of 1 minute each with 30 resting seconds between the sets). Fatigue decreased dorsiflexion (14.24 ± 4.98° in pre-fatigue and 12.65 ± 6.21° in fatigue condition, p < 0.05) at foot strike phase in females, and plantar flexion (−19.23 ± 4.12° in pre-fatigue and −18.26 ± 5.31° in fatigue condition, p < 0.05) at toe-off phase in males. These changes led to a decreased loading rate (88.14 ± 25.82 BW/s in pre-fatigue and 83.97 ± 18.83 BW/s in fatigue condition, p < 0.05) and the impact peak in females (1.95 ± 0.31 BW in pre-fatigue and 1.90 ± 0.31 BW in fatigue condition, p < 0.05), and higher peak propulsive forces in males (−0.26 ± 0.04 BW in pre-fatigue and −0.27 ± 0.05 BW in fatigue condition, p < 0.05) in the fatigue condition. It seems that better responses to impact under a fatigue condition are observed among women. Further studies should confirm whether these changes represent a strategy to optimize shock attenuation, prevent running injuries and improve running economy.

Falls in older people are a major health concern as the leading cause of disability and the second most common cause of accidental death. We developed a rapid fall risk assessment based on a combination of physical performance measurements made with an inertial sensor embedded in a smartphone. This study aimed to evaluate and validate the reliability and accuracy of an easy-to-use smartphone fall risk assessment by comparing it with the Physiological Profile Assessment (PPA) results. Sixty-five participants older than 55 performed a variation of the Timed Up and Go test using smartphone sensors. Balance and gait parameters were calculated, and their reliability was assessed by the (ICC) and compared with the PPAs. Since the PPA allows classification into six levels of fall risk, the data obtained from the smartphone assessment were categorised into six equivalent levels using different parametric and nonparametric classifier models with neural networks. The F1 score and geometric mean of each model were also calculated. All selected parameters showed ICCs around 0.9. The best classifier, in terms of accuracy, was the nonparametric mixed input data model with a 100% success rate in the classification category. In conclusion, fall risk can be reliably assessed using a simple, fast smartphone protocol that allows accurate fall risk classification among older people and can be a useful screening tool in clinical settings.

In biomechanics, movement is typically recorded by tracking the trajectories of anatomical landmarks previously marked using passive instrumentation, which entails several inconveniences. To overcome these disadvantages, researchers are exploring different markerless methods, such as pose estimation networks, to capture movement with equivalent accuracy to marker-based photogrammetry. However, pose estimation models usually only provide joint centers, which are incomplete data for calculating joint angles in all anatomical axes. Recently, marker augmentation models based on deep learning have emerged. These models transform pose estimation data into complete anatomical data. Building on this concept, this study presents three marker augmentation models of varying complexity that were compared to a photogrammetry system. The errors in anatomical landmark positions and the derived joint angles were calculated, and a statistical analysis of the errors was performed to identify the factors that most influence their magnitude. The proposed Transformer model improved upon the errors reported in the literature, yielding position errors of less than 1.5 cm for anatomical landmarks and 4.4 degrees for all seven movements evaluated. Anthropometric data did not influence the errors, while anatomical landmarks and movement influenced position errors, and model, rotation axis, and movement influenced joint angle errors.

In this paper, we present a dataset that takes 2D and 3D human pose keypoints estimated from images and relates them to the location of 3D anatomical landmarks. The dataset contains 51,051 poses obtained from 71 persons in A-Pose while performing 7 movements (walking, running, squatting, and four types of jumping). These poses were scanned to build a collection of 3D moving textured meshes with anatomical correspondence. Each mesh in that collection was used to obtain the 3D locations of 53 anatomical landmarks, and 48 images were created using virtual cameras with different perspectives. 2D pose keypoints from those images were obtained using the MediaPipe Human Pose Landmarker, and their corresponding 3D keypoints were calculated by linear triangulation. The dataset consists of a folder for each participant containing two Track Row Column (TRC) files and one JSON file for each movement sequence. One TRC file is used to store the 3D data of the triangulated 3D keypoints while the other contains the 3D anatomical landmarks. The JSON file is used to store the 2D keypoints and the calibration parameters of the virtual cameras. The anthropometric characteristics of the participants are annotated in a single CSV file. These data are intended to be used in developments that require the transformation of existing human pose solutions in computer vision into biomechanical applications or simulations. This dataset can also be used in other applications related to training neural networks for human motion analysis and studying their influence on anthropometric characteristics.

The new generation of videogame interfaces such as Microsoft’s Kinect opens the possibility of implementing exercise programs for physical training, and of evaluating and reducing the risks of elderly people falling. However, applications such as these might require measurements of joint kinematics that are more robust and accurate than the standard output given by the available middleware. This article presents a method based on particle filters for calculating joint angles from the positions of the anatomical points detected by PrimeSense’s NITE software. The application of this method to the measurement of lower limb kinematics reduced the error by one order of magnitude, to less than 10°, except for hip axial rotation, and it was advantageous over inverse kinematic analysis, in ensuring a robust and smooth solution without singularities, when the limbs are out-stretched and anatomical landmarks are aligned.