Détection de chute à l'aide d'une caméra de profondeur

Les chutes chez les personnes âgées représentent un problème important de santé publique. Des études montrent qu’environ 30 % des personnes âgées de 65 ans et plus chutent chaque année au Canada, entraînant des conséquences néfastes sur les plans individuel, familiale et sociale. Face à une telle situation la vidéosurveillance est une solution efficace assurant la sécurité de ces personnes. À ce jour de nombreux systèmes d’assistance de services à la personne existent. Ces dispositifs permettent à la personne âgée de vivre chez elle tout en assurant sa sécurité par le port d'un capteur. Cependant le port du capteur en permanence par le sujet est peu confortable et contraignant. C'est pourquoi la recherche s’est récemment intéressée à l’utilisation de caméras au lieu de capteurs portables. Le but de ce projet est de démontrer que l'utilisation d'un dispositif de vidéosurveillance peut contribuer à la réduction de ce fléau. Dans ce document nous présentons une approche de détection automatique de chute, basée sur une méthode de suivi 3D du sujet en utilisant une caméra de profondeur (Kinect de Microsoft) positionnée à la verticale du sol. Ce suivi est réalisé en utilisant la silhouette extraite en temps réel avec une approche robuste d’extraction de fond 3D basée sur la variation de profondeur des pixels dans la scène. Cette méthode se fondera sur une initialisation par une capture de la scène sans aucun sujet. Une fois la silhouette extraite, les 10% de la silhouette correspondant à la zone la plus haute de la silhouette (la plus proche de l'objectif de la Kinect) sera analysée en temps réel selon la vitesse et la position de son centre de gravité. Ces critères permettront donc après analyse de détecter la chute, puis d'émettre un signal (courrier ou texto) vers l'individu ou à l’autorité en charge de la personne âgée. Cette méthode a été validée à l’aide de plusieurs vidéos de chutes simulées par un cascadeur. La position de la caméra et son information de profondeur réduisent de façon considérable les risques de fausses alarmes de chute. Positionnée verticalement au sol, la caméra permet donc d'analyser la scène et surtout de procéder au suivi de la silhouette sans occultation majeure, qui conduisent dans certains cas à des fausses alertes. En outre les différents critères de détection de chute, sont des caractéristiques fiables pour différencier la chute d'une personne, d'un accroupissement ou d'une position assise. Néanmoins l'angle de vue de la caméra demeure un problème car il n'est pas assez grand pour couvrir une surface conséquente. Une solution à ce dilemme serait de fixer une lentille sur l'objectif de la Kinect permettant l’élargissement de la zone surveillée.

Analyse cinématique 3D de la coordination des articulations de l'épaule par l'entremise du rythme scapulo-huméral

L'épaule est souvent affectée par des troubles musculo-squelettiques. Toutefois, leur évaluation est limitée à des mesures qualitatives qui nuisent à la spécificité et justesse du diagnostic. L'analyse de mouvement tridimensionnel pourrait complémenter le traitement conventionnel à l'aide de mesures quantitatives fonctionnelles. L'interaction entre les articulations de l'épaule est estimée par le rythme scapulo-huméral, mais la variabilité prononcée qu'il affiche nuit à son utilisation clinique. Ainsi, l'objectif général de cette thèse était de réduire la variabilité de la mesure du rythme scapulo-huméral. L'effet de la méthode de calcul du rythme scapulo-huméral et des conditions d'exécution du mouvement (rotation axiale du bras, charge, vitesse, activité musculaire) ont été testées. La cinématique des articulations de l'épaule a été calculé par chaîne cinématique et filtre de Kalman étendu sur des sujets sains avec un système optoélectronique. La méthode usuelle de calcul du rythme scapulo-huméral extrait les angles d'élévation gléno-humérale et de rotation latérale scapulo-thoracique. Puisque ces angles ne sont pas co-planaires au thorax, leur somme ne correspond pas à l'angle d'élévation du bras. Une nouvelle approche de contribution articulaire incluant toutes les rotations de chaque articulation est proposée et comparée à la méthode usuelle. La méthode usuelle surestimait systématiquement la contribution gléno-humérale par rapport à la méthode proposée. Ce nouveau calcul du rythme scapulo-huméral permet une évaluation fonctionnelle dynamique de l'épaule et réduit la variabilité inter-sujets. La comparaison d'exercices de réadaptation du supra-épineux contrastant la rotation axiale du bras a été réalisée, ainsi que l'effet d'ajouter une charge externe. L'exercice «full-can» augmentait le rythme scapulo-huméral et la contribution gléno-humérale ce qui concorde avec la fonction du supra-épineux. Au contraire, l'exercice «empty-can» augmentait la contribution scapulo-thoracique ce qui est associé à une compensation pour éviter la contribution gléno-humérale. L'utilisation de charge externe lors de la réadaptation du supra-épineux semble justifiée par un rythme scapulo-huméral similaire et une élévation gléno-humérale supérieure. Le mouvement de l'épaule est souvent mesuré ou évalué en condition statique ou dynamique et passive ou active. Cependant, l'effet de ces conditions sur la coordination articulaire demeure incertain. La comparaison des ces conditions révélait des différences significatives qui montrent l'importance de considérer les conditions de mouvement pour l'acquisition ou la comparaison des données.

Effets de l’utilisation d’un chien d’assistance sur les efforts aux membres supérieurs lors de la montée d’une pente en fauteuil roulant chez les individus ayant une lésion médullaire

Problématique. L'utilisation d'un chien d'assistance à la mobilité (CAM) représente une option novatrice pour préserver l’intégrité des membres supérieurs (MSs) chez les utilisateurs de fauteuil roulant manuel (FRM). Aucune étude biomécanique n’a quantifié les effets du CAM sur les efforts aux MSs lors de la montée d’un plan incliné. Objectif. Cette étude quasi-expérimentale vise à comparer les efforts aux MSs lors de la montée d’un plan incliné avec et sans l’assistance d’un CAM. Méthodologie. Dix participants avec une lésion de la moelle épinière (LME) qui utilisent un FRM et possèdent un CAM ont monté un plan incliné à trois reprises avec et sans l’assistance du CAM. Les forces appliquées sur les cerceaux, mesurées avec des roues instrumentées, et les mouvements du FRM et des MSs, enregistrés avec un système d'analyse du mouvement, ont permis de mesurer les efforts mécaniques aux MSs. Simultanément, l'activité électromyographique (EMG) des muscles grand pectoral, deltoïde antérieur, biceps et triceps a été enregistrée et normalisée avec sa valeur maximale pour mesurer les efforts musculaires aux MSs. Résultats. En général, le CAM réduit significativement les valeurs moyennes et maximales de la force totale appliquée aux cerceaux et de sa composante tangentielle, des moments nets de flexion, de rotation interne et d’adduction aux épaules et des taux d’utilisation musculaire du deltoïde antérieur, du biceps et du triceps. Conclusion. L’assistance d’un CAM réduit les efforts aux MSs lors de la montée d’un plan incliné chez les utilisateurs d’un FRM ayant une LME.

Impacto de las cirugías múltiples, en un solo tiempo quirúrgico, sobre el patrón de marcha en niños con parálisis cerebral

Introducción: La Parálisis Cerebral (PC) es la enfermedad neurológica más incapacitante en niños, su historia natural tiende al deterioro motor y funcional. Con este estudio se busca establecer sí las cirugías múltiples de miembros inferiores, en un tiempo quirúrgico, mantienen el nivel motor y funcional. Material y Método: Estudio analítico de cohortes. Se compara un grupo de pacientes sometidos a cirugías múltiples contra un grupo de pacientes no operados, en el Instituto de Ortopedia Infantil Roosevelt. Se evaluaron los pacientes con dos Laboratorios para el Análisis del Movimiento (LAM) y se midieron los desenlaces mediante el cambio en la puntuación del perfil de marcha (GPS) y el nivel funcional motor grueso (GMFCS). Resultados: 109 pacientes cumplieron con los criterios de selección, 67 pacientes fueron sometidos a cirugía y 42 pacientes no. Los pacientes operados mejoraron el GPS promedio (diferencia -1,94; p=0,002) comparado con los pacientes no operados (diferencia 1,74; p=0,001), indicando una mejoría significativa de la cinemática de la marcha. En un modelo de regresión logística predictivo, el paciente que es operado tiene una probabilidad del 78% de mantener su patrón de marcha, mientras que sí no se opera su probabilidad disminuye al 37%. El nivel funcional motor GMFCS no mostró cambios significativos entre los grupos. Discusión: Las cirugías múltiples de miembros inferiores mantienen de manera significativa el patrón de marcha en pacientes con PC. Se destaca el seguimiento de los pacientes mediante el LAM y se sugiere el uso del GPS para valorar resultados en este tipo de pacientes.

View-based approaches to spatial representation in human vision

In an immersive virtual environment, observers fail to notice the expansion of a room around them and consequently make gross errors when comparing the size of objects. This result is difficult to explain if the visual system continuously generates a 3-D model of the scene based on known baseline information from interocular separation or proprioception as the observer walks. An alternative is that observers use view-based methods to guide their actions and to represent the spatial layout of the scene. In this case, they may have an expectation of the images they will receive but be insensitive to the rate at which images arrive as they walk. We describe the way in which the eye movement strategy of animals simplifies motion processing if their goal is to move towards a desired image and discuss dorsal and ventral stream processing of moving images in that context. Although many questions about view-based approaches to scene representation remain unanswered, the solutions are likely to be highly relevant to understanding biological 3-D vision.

Early identification of stimulant treatment responders, partial responders and non-responders using objective measures in children and adolescents with hyperkinetic disorder

Background: The aim of this study was to evaluate stimulant medication response following a single dose of methylphenidate (MPH) in children and young people with hyperkinetic disorder using infrared motion analysis combined with a continuous performance task (QbTest system) as objective measures. The hypothesis was put forward that a moderate testdose of stimulant medication could determine a robust treatment response, partial response and non-response in relation to activity, attention and impulse control measures. Methods: The study included 44 children and young people between the ages of 7-18 years with a diagnosis of hyperkinetic disorder (F90 & F90.1). A single dose-protocol incorporated the time course effects of both immediate release MPH and extended release MPH (Concerta XL, Equasym XL) to determine comparable peak efficacy periods post intake. Results: A robust treatment response with objective measures reverting to the population mean was found in 37 participants (84%). Three participants (7%) demonstrated a partial response to MPH and four participants (9%) were determined as non-responders due to deteriorating activity measures together with no improvements in attention and impulse control measures. Conclusion: Objective measures provide early into prescribing the opportunity to measure treatment response and monitor adverse reactions to stimulant medication. Most treatment responders demonstrated an effective response to MPH on a moderate testdose facilitating a swift and more optimal titration process.

First-principle data-driven models for assessment of motor disorders in Parkinson’s disease

Parkinson’s disease (PD) is an increasing neurological disorder in an aging society. The motor and non-motor symptoms of PD advance with the disease progression and occur in varying frequency and duration. In order to affirm the full extent of a patient’s condition, repeated assessments are necessary to adjust medical prescription. In clinical studies, symptoms are assessed using the unified Parkinson’s disease rating scale (UPDRS). On one hand, the subjective rating using UPDRS relies on clinical expertise. On the other hand, it requires the physical presence of patients in clinics which implies high logistical costs. Another limitation of clinical assessment is that the observation in hospital may not accurately represent a patient’s situation at home. For such reasons, the practical frequency of tracking PD symptoms may under-represent the true time scale of PD fluctuations and may result in an overall inaccurate assessment. Current technologies for at-home PD treatment are based on data-driven approaches for which the interpretation and reproduction of results are problematic.  The overall objective of this thesis is to develop and evaluate unobtrusive computer methods for enabling remote monitoring of patients with PD. It investigates first-principle data-driven model based novel signal and image processing techniques for extraction of clinically useful information from audio recordings of speech (in texts read aloud) and video recordings of gait and finger-tapping motor examinations. The aim is to map between PD symptoms severities estimated using novel computer methods and the clinical ratings based on UPDRS part-III (motor examination). A web-based test battery system consisting of self-assessment of symptoms and motor function tests was previously constructed for a touch screen mobile device. A comprehensive speech framework has been developed for this device to analyze text-dependent running speech by: (1) extracting novel signal features that are able to represent PD deficits in each individual component of the speech system, (2) mapping between clinical ratings and feature estimates of speech symptom severity, and (3) classifying between UPDRS part-III severity levels using speech features and statistical machine learning tools. A novel speech processing method called cepstral separation difference showed stronger ability to classify between speech symptom severities as compared to existing features of PD speech. In the case of finger tapping, the recorded videos of rapid finger tapping examination were processed using a novel computer-vision (CV) algorithm that extracts symptom information from video-based tapping signals using motion analysis of the index-finger which incorporates a face detection module for signal calibration. This algorithm was able to discriminate between UPDRS part III severity levels of finger tapping with high classification rates. Further analysis was performed on novel CV based gait features constructed using a standard human model to discriminate between a healthy gait and a Parkinsonian gait. The findings of this study suggest that the symptom severity levels in PD can be discriminated with high accuracies by involving a combination of first-principle (features) and data-driven (classification) approaches. The processing of audio and video recordings on one hand allows remote monitoring of speech, gait and finger-tapping examinations by the clinical staff. On the other hand, the first-principles approach eases the understanding of symptom estimates for clinicians. We have demonstrated that the selected features of speech, gait and finger tapping were able to discriminate between symptom severity levels, as well as, between healthy controls and PD patients with high classification rates. The findings support suitability of these methods to be used as decision support tools in the context of PD assessment.

A Computer Vision Framework For Finger-Tapping Evaluation In Parkinson's Disease

Objective: To define and evaluate a Computer-Vision (CV) method for scoring Paced Finger-Tapping (PFT) in Parkinson's disease (PD) using quantitative motion analysis of index-fingers and to compare the obtained scores to the UPDRS (Unified Parkinson's Disease Rating Scale) finger-taps (FT). Background: The naked-eye evaluation of PFT in clinical practice results in coarse resolution to determine PD status. Besides, sensor mechanisms for PFT evaluation may cause patients discomfort. In order to avoid cost and effort of applying wearable sensors, a CV system for non-invasive PFT evaluation is introduced. Methods: A database of 221 PFT videos from 6 PD patients was processed. The subjects were instructed to position their hands above their shoulders besides the face and tap the index-finger against the thumb consistently with speed. They were facing towards a pivoted camera during recording. The videos were rated by two clinicians between symptom levels 0-to-3 using UPDRS-FT. The CV method incorporates a motion analyzer and a face detector. The method detects the face of testee in each video-frame. The frame is split into two images from face-rectangle center. Two regions of interest are located in each image to detect index-finger motion of left and right hands respectively. The tracking of opening and closing phases of dominant hand index-finger produces a tapping time-series. This time-series is normalized by the face height. The normalization calibrates the amplitude in tapping signal which is affected by the varying distance between camera and subject (farther the camera, lesser the amplitude). A total of 15 features were classified using K-nearest neighbor (KNN) classifier to characterize the symptoms levels in UPDRS-FT. The target ratings provided by the raters were averaged. Results: A 10-fold cross validation in KNN classified 221 videos between 3 symptom levels with 75% accuracy. An area under the receiver operating characteristic curves of 82.6% supports feasibility of the obtained features to replicate clinical assessments. Conclusions: The system is able to track index-finger motion to estimate tapping symptoms in PD. It has certain advantages compared to other technologies (e.g. magnetic sensors, accelerometers etc.) for PFT evaluation to improve and automate the ratings

Target tracking with range and bearing measurements via robust linear filtering

In this paper we provide a robust version of a linear Kalman filter for target tracking with nonlinear range and bearing measurements. Moreover, we prove that the state estimation error is bounded in a probabilistic sense. We compare our approach with the current state of the art in converted radar measurement based linear filtering.

Ageing effects on knee and ankle joint angles at key events and phases of the gait cycle.

The objective of this research was to determine whether joint angles at critical gait events and during major energy generation/absorption phases of the gait cycle would reliably discriminate age-related degeneration during unobstructed walking. The gaits of 24 healthy adults (12 young and 12 elderly) were analysed using the PEAK Motus motion analysis system. The elderly participants showed significantly greater single (60.3% versus 62.3%, p < 0.01) and double ( p < 0.05) support times, reduced knee flexion (47.7° versus 43.0°, p < 0.05) and ankle plantarflexion (16.8° compared to 3.3°, p = 0.053) at toe off, reduced knee flexion during push-off and reduced ankle dorsiflexion (16.8° compared to 22.0°, p < 0.05) during the swing phase. The plantarflexing ankle joint motion during the stance to swing phase transition (A₂) for the young group (31.3°) was about twice ( p < 0.05) that of the elderly (16.9°). Reduced knee extension range of motion suggests that the elderly favoured a flexed-knee gait to assist in weight acceptance. Reduced dorsiflexion by the elderly in the swing phase implies greater risk of toe contact with obstacles. Overall, the results suggest that joint angle measures at critical events/phases in the gait cycle provide a useful indication of age-related degeneration in the control of lower limb trajectories during unobstructed walking.

Influence of posture on the range of axial rotation and coupled lateral flexion of the thoracic spine

Objective
This study examines the influence of posture on the range of axial rotation of the thorax and the range and direction of the coupled lateral flexion.

Methods
The ranges of mid thoracic axial rotation and coupled lateral flexion were measured in 52 asymptomatic subjects (aged 18-43 years) using an optical motion analysis system. To examine the influence of posture on primary and coupled motion, we initiated axial rotation from a neutral sitting posture and from end-range thoracic flexion and extension.

Results
There was a significant decrease in the range of thoracic rotation in flexion compared with the neutral and extended postures (P < .001). The mean range of coupled lateral flexion was 8.9% of the axial rotation range in the neutral posture and increased to 14.3% and 23.2% in the extended and flexed postures, respectively. Patterns of coupled motion varied between subjects, but an ipsilateral pattern was more common in the flexed posture, whereas a contralateral pattern was more common in the neutral and extended postures.

Conclusions
The ranges and patterns of coupled motion of the thorax appear to be strongly influenced by the posture from which the movement is initiated. This has important implications in relation to the interpretation of clinical tests of thoracic motion and in consideration of mechanisms of development of thoracic pain disorders.

Postural neck pain : an investigation of habitual sitting posture, perception of 'good' posture and cervicothoracic kinaesthesia

Impairments of cervico-cephalic kinaesthesia and habitual forward head posture have been considered important in the aetiology of postural neck pain, yet these factors have not been specifically examined in a homogeneous clinical population. The objective of this study was to compare the habitual sitting posture (HSP), perception of good posture and postural repositioning error (PRE) of the cervico-thoracic (CT) spine in individuals with postural neck pain, with a matched group of asymptomatic subjects. Twenty-one subjects with postural neck pain and 22 asymptomatic control subjects were recruited into the study. An optical motion analysis system was used to measure the HSP and perceived ‘good’ sitting posture. PRE was measured over six trials where the subject attempted to replicate their self-selected ‘good’ posture. There was no difference between the groups in the HSP but significant differences were identified in the perception of ‘good’ posture. Posture repositioning error was higher for the head posture variables than for CT and shoulder girdle variables in both groups. However, there was no significant difference in posture repositioning error between groups for any of the posture measures. The findings suggest that individuals with postural neck pain may have a different perception of ‘good’ posture, but no significant difference in HSP or kinaesthetic sensibility compared with matched asymptomatic subjects.

Human action recognition based on 3D SIFT and LDA model

How to recognize human action from videos captured by modern cameras efficiently and effectively is a challenge in real applications. Traditional methods which need professional analysts are facing a bottleneck because of their shortcomings. To cope with the disadvantage, methods based on computer vision techniques, without or with only a few human interventions, have been proposed to analyse human actions in videos automatically. This paper provides a method combining the three dimensional Scale Invariant Feature Transform (SIFT) detector and the Latent Dirichlet Allocation (LDA) model for human motion analysis. To represent videos effectively and robustly, we extract the 3D SIFT descriptor around each interest point, which is sampled densely from 3D Space-time video volumes. After obtaining the representation of each video frame, the LDA model is adopted to discover the underlying structure-the categorization of human actions in the collection of videos. Public available standard datasets are used to test our method. The concluding part discusses the research challenges and future directions.

Robust recognition and segmentation of human actions using HMMs with missing observations

This paper describes the integration of missing observation data with hidden Markov models to create a framework that is able to segment and classify individual actions from a stream of human motion using an incomplete 3D human pose estimation. Based on this framework, a model is trained to automatically segment and classify an activity sequence into its constituent subactions during inferencing. This is achieved by introducing action labels into the observation vector and setting these labels as missing data during inferencing, thus forcing the system to infer the probability of each action label. Additionally, missing data provides recognition-level support for occlusions and imperfect silhouette segmentation, permitting the use of a fast (real-time) pose estimation that delegates the burden of handling undetected limbs onto the action recognition system. Findings show that the use of missing data to segment activities is an accurate and elegant approach. Furthermore, action recognition can be accurate even when almost half of the pose feature data is missing due to occlusions, since not all of the pose data is important all of the time.

Early jump-out corner detectors

Two corner detectors are presented, one of which works by testing similarity of image patches along the contour direction to detect curves in the image contour, and the other of which uses direct estimation image curvature along the contour direction. The operators are fast, robust to noise, and self-thresholding. An interpretation of the Kitchen-Rosenfeld corner operator is presented which shows that this operator can also be viewed as the second derivative of the image function along the edge direction.