According to the demographic evolution in industrialized countries, more and more elderly people will experience falls at home and will require emergency services. The main problem comes from fall-prone elderly living alone at home. To resolve this lack of safety, we propose a new method to detect falls at home, based on a multiple-cameras network for reconstructing the 3-D shape of people. Fall events are detected by analyzing the volume distribution along the vertical axis, and an alarm is triggered when the major part of this distribution is abnormally near the floor during a predefined period of time, which implies that a person has fallen on the floor. This method was validated with videos of a healthy subject who performed 24 realistic scenarios showing 22 fall events and 24 cofounding events (11 crouching position, 9 sitting position, and 4 lying on a sofa position) under several camera configurations, and achieved 99.7% sensitivity and specificity or better with four cameras or more. A real-time implementation using a graphic processing unit (GPU) reached 10 frames per second (fps) with 8 cameras, and 16 fps with 3 cameras.

Analyzing human poses with a Kinect is a promising method to evaluate potentials risks of musculoskeletal disorders at workstations. In ecological situations, complex 3D poses and constraints imposed by the environment make it difficult to obtain reliable kinematic information. Thus, being able to predict the potential accuracy of the measurement for such complex 3D poses and sensor placements is challenging in classical experimental setups. To tackle this problem, we propose a new evaluation method based on a virtual mannequin. In this study, we apply this method to the evaluation of joint positions (shoulder, elbow, and wrist), joint angles (shoulder and elbow), and the corresponding RULA (a popular ergonomics assessment grid) upper-limb score for a large set of poses and sensor placements. Thanks to this evaluation method, more than 500,000 configurations have been automatically tested, which would be almost impossible to evaluate with classical protocols. The results show that the kinematic information obtained by the Kinect software is generally accurate enough to fill in ergonomic assessment grids. However inaccuracy strongly increases for some specific poses and sensor positions. Using this evaluation method enabled us to report configurations that could lead to these high inaccuracies. As a supplementary material, we provide a software tool to help designers to evaluate the expected accuracy of this sensor for a set of upper-limb configurations. Results obtained with the virtual mannequin are in accordance with those obtained from a real subject for a limited set of poses and sensor placements.

Background: Accurate implant orientation reduces wear and increases stability in arthroplasty but is a technically demanding skill. Augmented reality (AR) headsets overlay digital information on top of the real world. We have developed an enhanced AR headset capable of tracking bony anatomy in relation to an implant, but it has not yet been assessed for its suitability as a training tool for implant orientation. Questions/purposes: (1) In the setting of simulated THA performed by novices, does an AR headset improve the accuracy of acetabular component positioning compared with hands-on training by an expert surgeon? (2) What are trainees’ perceptions of the AR headset in terms of realism of the task, acceptability of the technology, and its potential role for surgical training? Methods: Twenty-four study participants (medical students in their final year of school, who were applying to surgery residency programs, and who had no prior arthroplasty experience) participated in a randomized simulation trial using an AR headset and a simulated THA. Participants were randomized to two groups completing four once-weekly sessions of baseline assessment, training, and reassessment. One group trained using AR (with live holographic orientation feedback) and the other received one-on-one training from a hip arthroplasty surgeon. Demographics and baseline performance in orienting an acetabular implant to six patient-specific values on the phantom pelvis were collected before training and were comparable. The orientation error in degrees between the planned and achieved orientations was measured and was not different between groups with the numbers available (surgeon group mean error ± SD 16° ± 7° versus AR 14° ± 7°; p = 0.22). Participants trained by AR also completed a validated posttraining questionnaire evaluating their experiences. Results: During the four training sessions, participants using AR-guidance had smaller mean (± SD) errors in orientation than those receiving guidance from the surgeon: 1° ± 1° versus AR 6° ± 4°, p < 0.001. In the fourth session’s assessment, participants in both groups had improved (surgeon group mean improvement 6°, 95% CI, 4–8°; p < 0.001 versus AR group 9°, 95% CI 7–10°; p < 0.001). There was no difference between participants in the surgeon-trained and AR-trained group: mean difference 1.2°, 95% CI, -1.8 to 4.2°; p = 0.281. In posttraining evaluation, 11 of 12 participants would use the AR platform as a training tool for developing visuospatial skills and 10 of 12 for procedure-specific rehearsals. Most participants (11 of 12) stated that a combination of an expert trainer for learning and AR for unsupervised training would be preferred. Conclusions: A novel head-mounted AR platform tracked an implant in relation to bony anatomy to a clinically relevant level of accuracy during simulated THA. Learners were equally accurate, whether trained by AR or a surgeon. The platform enabled the use of real instruments and gave live feedback; AR was thus considered a feasible and valuable training tool as an adjunct to expert guidance in the operating room. Although there were no differences in accuracy between the groups trained using AR and those trained by an expert surgeon, we believe the tool may be useful in education because it demonstrates that some motor skills for arthroplasty may be learned in an unsupervised setting. Future studies will evaluate AR-training for arthroplasty skills other than cup orientation and its transfer validity to real surgery. Level of Evidence: Level I, therapeutic study.