Non-Intrusive Real Time Eye Tracking Using Facial Alignment for Assistive Technologies.

Most affordable eye tracking systems use either intrusive setup such as head-mounted cameras or use fixed cameras with infrared corneal reflections via illuminators. In the case of assistive technologies, using intrusive eye tracking systems can be a burden to wear for extended periods of time and infrared based solutions generally do not work in all environments, especially outside or inside if the sunlight reaches the space. Therefore, we propose an eye-tracking solution using state-of-the-art convolutional neural network face alignment algorithms that is both accurate and lightweight for assistive tasks such as selecting an object for use with assistive robotics arms. This solution uses a simple webcam for gaze and face position and pose estimation. We achieve a much faster computation time than the current state-of-the-art while maintaining comparable accuracy. This paves the way for accurate appearance-based gaze estimation even on mobile devices, giving an average error of around 4.5°on the MPIIGaze dataset [1] and state-of-the-art average errors of 3.9°and 3.3°on the UTMultiview [2] and GazeCapture [3], [4] datasets respectively, while achieving a decrease in computation time of up to 91%.

Using a cost function based on kinematics and electromyographic data to quantify muscle forces.

A reliable evaluation of muscle forces in the human body is highly desirable for several applications in both clinical and research contexts. Several models of muscle force distribution based on non-invasive measurements have been proposed since 1836 (Weber and Weber, 1836), amongst which Crowninshield's model (Crowninshield and Brand, 1981), which maximizes a cost-function representing the muscle fiber endurance, is the most popular. It is worth noting that Crowninshield's model is the most widely adopted notwithstanding its major limitations of physiological coherence. Forster et al. (2004) pointed out that "these (conventional) criteria however do not predict co-contraction adequately". Besides, electromyographic (EMG)-driven models have been proposed to assess individual muscle forces, which have not been broadly adopted due to their complexity and the need for a calibration before each test. In this context, a cost function based on kinematic and electromyographic data could provide the advantage of being physiologically more coherent with muscle activations compared to conventional cost-functions based on kinematics solely, and easier to use than the EMG-driven models. The objective of this study is to propose the first cost-function based on kinematics and electromyographic data to quantify muscle forces. When applying this new cost-function on a database of upper limb motions data of 17 subjects, healthy or with cerebral palsy, the muscle force prediction of the proposed model was 17.74% more coherent with the EMG pattern than the prediction of Crowninshield's model. And on average, these results were more consistent whether the subjects were healthy or with cerebral palsy. In conclusion, we propose this cost-function for the quantification of muscle forces.

Decomposition of three-dimensional ground-reaction forces under both feet during gait.

Three-dimensional ground reaction forces (3D-GRF) are essential for functional evaluation for rehabilitation. A platform path is required to obtain the 3D-GRF. The main shortcoming of these platform paths is that during double stance phases of gait, both feet can be placed on the same force platform causing the need for decomposing the 3D-GRF under each foot. Despite the high number of studies on force decomposition, there is still no method on the decomposition of 3D-GRF based on data from platforms. OBJECTIVE: This study aims to present an automatic method using parametric curve fitting modeling to increase the accuracy of decomposition of 3D-GRF during double stances under each foot. METHODS: The decomposition method was applied to the global 3D-GRF using 3rd order polynomial, sine, and sine-sigmoid functions. The computed 3D-GRF was compared to the 3D-GRF independently recorded by force platforms for each subject. RESULTS: The relative average error between the computed 3D-GRF and the recorded 3D-GRF were equal to 3.3±1.6%. In details for the vertical, antero-posterior, and medio-lateral GRF, these errors were 2.9±1.6%, 6.3±4.3%, and, 9.5±3.6%, respectively, for 30 subjects. CONCLUSION: The global error on the GRF is the best one in the literature. This method can be validated on various populations with musculoskeletal disorders.

Custom sizing of lower limb exoskeleton actuators using gait dynamic modelling of children with cerebral palsy.

The use of exoskeletons as an aid for people with musculoskeletal disorder is the subject to an increasing interest in the research community. These devices are expected to meet the specific needs of users, such as children with cerebral palsy (CP) who are considered a significant population in pediatric rehabilitation. Although these exoskeletons should be designed to ease the movement of people with physical shortcoming, their design is generally based on data obtained from healthy adults, which leads to oversized components that are inadequate to the targeted users. Consequently, the objective of this study is to custom-size the lower limb exoskeleton actuators based on dynamic modeling of the human body for children with CP on the basis of hip, knee, and ankle joint kinematics and dynamics of human body during gait. For this purpose, a multibody modeling of the human body of 3 typically developed children (TD) and 3 children with CP is used. The results show significant differences in gait patterns especially in knee and ankle with respectively 0.39 and -0.33 (Nm/kg) maximum torque differences between TD children and children with CP. This study provides the recommendations to support the design of actuators to normalize the movement of children with CP.

[Urinary calculi in a geriatric setting]. / Lithiase urinaire en milieu gériatrique.

The authors report a retrospective series of 174 patients with urolithiasis. They were all age between 65 and 88 years. This group represents one tenth of all patients treated for urinary stones in the Urology department of Sfax over the last decade. Neither the clinical symptoms nor radiological findings observed in this group differed from those other patients hospitalized for urolithiasis; on the other hand associated diseases related to aging appear to be well represented which alters the management, prognosis and outcome in this type of patients.