Biometal muscle to restore atrial transport function in a permanent atrial fibrillation animal model: a potential tool in the treatment of end-stage heart failure.

BACKGROUND: Half of the patients with end-stage heart failure suffer from persistent atrial fibrillation (AF). Atrial kick (AK) accounts for 10-15% of the ejection fraction. A device restoring AK should significantly improve cardiac output (CO) and possibly delay ventricular assist device (VAD) implantation. This study has been designed to assess the mechanical effects of a motorless pump on the right chambers of the heart in an animal model. METHODS: Atripump is a dome-shaped biometal actuator electrically driven by a pacemaker-like control unit. In eight sheep, the device was sutured onto the right atrium (RA). AF was simulated with rapid atrial pacing. RA ejection fraction (EF) was assessed with intracardiac ultrasound (ICUS) in baseline, AF and assisted-AF status. In two animals, the pump was left in place for 4 weeks and then explanted. Histology examination was carried out. The mean values for single measurement per animal with +/-SD were analysed. RESULTS: The contraction rate of the device was 60 per min. RA EF was 41% in baseline, 7% in AF and 21% in assisted-AF conditions. CO was 7+/-0.5 l min(-1) in baseline, 6.2+/-0.5 l min(-1) in AF and 6.7+/-0.5 l min(-1) in assisted-AF status (p<0.01). Histology of the atrium in the chronic group showed chronic tissue inflammation and no sign of tissue necrosis. CONCLUSIONS: The artificial muscle restores the AK and improves CO. In patients with end-stage cardiac failure and permanent AF, if implanted on both sides, it would improve CO and possibly delay or even avoid complex surgical treatment such as VAD implantation.

Active Learning of Very-High Resolution Optical Imagery with SVM: Entropy vs Margin Sampling

An active learning method is proposed for the semi-automatic selection of training sets in remote sensing image classification. The method adds iteratively to the current training set the unlabeled pixels for which the prediction of an ensemble of classifiers based on bagged training sets show maximum entropy. This way, the algorithm selects the pixels that are the most uncertain and that will improve the model if added in the training set. The user is asked to label such pixels at each iteration. Experiments using support vector machines (SVM) on an 8 classes QuickBird image show the excellent performances of the methods, that equals accuracies of both a model trained with ten times more pixels and a model whose training set has been built using a state-of-the-art SVM specific active learning method

Rôle de l'exercice dans l'approche multidisciplinaire de la lombalgie chronique

RÉSUMÉ L'exercice est utilisé dans le traitement de la lombalgie depuis plus de cent ans. La recherche dans ce domaine a commencé au milieu du XXème siècle puis s'est développée exponentiellement jusqu'à nos jours. La première étude de cette thèse a eu pour but de passer en revue cette abondante littérature scientifique. Il en est ressorti que l'exercice est un moyen efficace de prévention primaire et secondaire de la lombalgie. En tant que modalité de traitement, l'exercice permet de diminuer l'incapacité et la douleur et d'améliorer la condition physique et le statut professionnel des patients lombalgiques subaigus et chroniques. Parmi les caractéristiques de l'exercice, la supervision est essentielle. Des investigations ultérieures sont nécessaires afin d'identifier des sous-groupes de patients répondant favorablement à d'autres caractéristiques de l'exercice. L'exercice est souvent utilisé dans l'optique de maintenir les résultats obtenus à la suite d'un traitement, bien que peu d'études s'y soient penchées. La deuxième partie de cette thèse a eu pour objectifs d'évaluer l'efficacité d'un programme d'exercice (PE) suivi par des patients lombalgiques chroniques ayant complété une restauration fonctionnelle multidisciplinaire (RFM), en comparaison avec le suivi classique (SC) consistant simplement à encourager les patients à adopter un quotidien aussi actif que possible par la suite. Les résultats ont montré que les améliorations obtenues au terme de RFM étaient maintenues par les deux groupes à un an de suivi. Bien qu'aucune différence n'ait été obtenue entre les deux groupes, seul le groupe PE améliorait significativement l'incapacité et l'endurance isométrique des muscles du tronc. Une analyse économique a ensuite été réalisée afin d'évaluer la rentabilité de PE. L'évaluation de la qualité de vie des patients au terme de RFM et à un an de suivi permettait d'estimer les années de vie ajustées par leur qualité (QALYs) gagnées par chaque groupe. Les coûts directs (visites chez le médecin, spécialiste, physio, autres) et indirects (jours d'absence au travail) étaient estimés avant RFM et à un an de suivi à l'aide d'un agenda. Aucune différence significative n'était obtenue entre les groupes. Une mince différence de QALYs en faveur de PE ne se traduisait néanmoins pas en bénéfices mesurables. La recherche future devrait s'attacher à identifier un ou des sous-groupe(s) de patients pour lesquels SC ne permet pas de maintenir à long terme les améliorations obtenues au terme de RFM, et pour lesquels l'efficacité thérapeutique et la rentabilité économique de PE pourraient être accrues. ABSTRACT Exercise is used to treat low back pain for over a hundred years. Research in this area began in the mid-twentieth century and then grew exponentially until nowadays. The first study of this thesis was aimed to review this abundant scientific literature. It showed that exercise is effective in the primary and secondary prevention of low back pain. As a modality of treatment, exercise can reduce disability and pain and improve physical fitness and professional status of patients with subacute and chronic low back pain. Among different exercise characteristics, supervision is essential. Further investigations are needed to identify subgroups of patients responding positively to other characteristics of exercise. Exercise is often used as a post-treatment modality in order to maintain results over time, although only a few studies addressed this issue directly. The purpose of the second part of this thesis was to evaluate the effectiveness of an exercise program (EP) for patients with chronic low back pain who completed a functional multidisciplinary rehabilitation (FMR), compared to the routine follow-up (RF) which simply consisted of encouraging patients to adopt an active daily life thereafter. The results showed that improvements obtained at the end of FMR were maintained by both groups at one year follow-up. Although no difference was obtained between both groups, only the EP group significantly improved disability and isometric endurance of trunk muscles. An economic analysis was then carried out to assess the cost-effectiveness of EP. Based on the evaluation of patients' quality of life after FMR and at one year follow-up, an estimation of adjusted life years for their quality (QALYs) gained by each group was done. Direct costs (physician, specialist, physiotherapist, other therapists visits) and indirect costs (days off work) were measured before FMR and at one year follow-up using a cost diary. No significant difference was obtained between both groups. A slight difference in QALYs in favour of EP did yet not translate into measurable benefits. Future research should focus on identifying subgroups of patients for which RF is insufficient to reach long-term improvements after FMR, and for which the therapeutic effectiveness and cost-effectiveness of EP could be increased.