Advanced Kernel Methods For Remote Sensing Image Classification

Résumé Suite aux recentes avancées technologiques, les archives d'images digitales ont connu une croissance qualitative et quantitative sans précédent. Malgré les énormes possibilités qu'elles offrent, ces avancées posent de nouvelles questions quant au traitement des masses de données saisies. Cette question est à la base de cette Thèse: les problèmes de traitement d'information digitale à très haute résolution spatiale et/ou spectrale y sont considérés en recourant à des approches d'apprentissage statistique, les méthodes à noyau. Cette Thèse étudie des problèmes de classification d'images, c'est à dire de catégorisation de pixels en un nombre réduit de classes refletant les propriétés spectrales et contextuelles des objets qu'elles représentent. L'accent est mis sur l'efficience des algorithmes, ainsi que sur leur simplicité, de manière à augmenter leur potentiel d'implementation pour les utilisateurs. De plus, le défi de cette Thèse est de rester proche des problèmes concrets des utilisateurs d'images satellite sans pour autant perdre de vue l'intéret des méthodes proposées pour le milieu du machine learning dont elles sont issues. En ce sens, ce travail joue la carte de la transdisciplinarité en maintenant un lien fort entre les deux sciences dans tous les développements proposés. Quatre modèles sont proposés: le premier répond au problème de la haute dimensionalité et de la redondance des données par un modèle optimisant les performances en classification en s'adaptant aux particularités de l'image. Ceci est rendu possible par un système de ranking des variables (les bandes) qui est optimisé en même temps que le modèle de base: ce faisant, seules les variables importantes pour résoudre le problème sont utilisées par le classifieur. Le manque d'information étiquétée et l'incertitude quant à sa pertinence pour le problème sont à la source des deux modèles suivants, basés respectivement sur l'apprentissage actif et les méthodes semi-supervisées: le premier permet d'améliorer la qualité d'un ensemble d'entraînement par interaction directe entre l'utilisateur et la machine, alors que le deuxième utilise les pixels non étiquetés pour améliorer la description des données disponibles et la robustesse du modèle. Enfin, le dernier modèle proposé considère la question plus théorique de la structure entre les outputs: l'intègration de cette source d'information, jusqu'à présent jamais considérée en télédétection, ouvre des nouveaux défis de recherche. Advanced kernel methods for remote sensing image classification Devis Tuia Institut de Géomatique et d'Analyse du Risque September 2009 Abstract The technical developments in recent years have brought the quantity and quality of digital information to an unprecedented level, as enormous archives of satellite images are available to the users. However, even if these advances open more and more possibilities in the use of digital imagery, they also rise several problems of storage and treatment. The latter is considered in this Thesis: the processing of very high spatial and spectral resolution images is treated with approaches based on data-driven algorithms relying on kernel methods. In particular, the problem of image classification, i.e. the categorization of the image's pixels into a reduced number of classes reflecting spectral and contextual properties, is studied through the different models presented. The accent is put on algorithmic efficiency and the simplicity of the approaches proposed, to avoid too complex models that would not be used by users. The major challenge of the Thesis is to remain close to concrete remote sensing problems, without losing the methodological interest from the machine learning viewpoint: in this sense, this work aims at building a bridge between the machine learning and remote sensing communities and all the models proposed have been developed keeping in mind the need for such a synergy. Four models are proposed: first, an adaptive model learning the relevant image features has been proposed to solve the problem of high dimensionality and collinearity of the image features. This model provides automatically an accurate classifier and a ranking of the relevance of the single features. The scarcity and unreliability of labeled. information were the common root of the second and third models proposed: when confronted to such problems, the user can either construct the labeled set iteratively by direct interaction with the machine or use the unlabeled data to increase robustness and quality of the description of data. Both solutions have been explored resulting into two methodological contributions, based respectively on active learning and semisupervised learning. Finally, the more theoretical issue of structured outputs has been considered in the last model, which, by integrating outputs similarity into a model, opens new challenges and opportunities for remote sensing image processing.

Evaluation Findings: Rural Reentry Program Works, Data Download, Iowa Department of Corrections, April 2010

Iowa’s Division of Criminal and Juvenile Justice Planning (CJJP) recently completed an evaluation of the 2nd Judicial District’s Rural Prisoner Reentry Initiative (PRI), which provided reentry services to offenders both while in prison and after release.

Design, data management, and population baseline characteristics of the PERFORM magnetic resonance imaging project.

Quantitative information from magnetic resonance imaging (MRI) may substantiate clinical findings and provide additional insight into the mechanism of clinical interventions in therapeutic stroke trials. The PERFORM study is exploring the efficacy of terutroban versus aspirin for secondary prevention in patients with a history of ischemic stroke. We report on the design of an exploratory longitudinal MRI follow-up study that was performed in a subgroup of the PERFORM trial. An international multi-centre longitudinal follow-up MRI study was designed for different MR systems employing safety and efficacy readouts: new T2 lesions, new DWI lesions, whole brain volume change, hippocampal volume change, changes in tissue microstructure as depicted by mean diffusivity and fractional anisotropy, vessel patency on MR angiography, and the presence of and development of new microbleeds. A total of 1,056 patients (men and women ≥ 55 years) were included. The data analysis included 3D reformation, image registration of different contrasts, tissue segmentation, and automated lesion detection. This large international multi-centre study demonstrates how new MRI readouts can be used to provide key information on the evolution of cerebral tissue lesions and within the macrovasculature after atherothrombotic stroke in a large sample of patients.

Clinical characteristics and outcome of inpatients versus outpatients with venous thromboembolism: findings from the RIETE Registry.

BACKGROUND: Patients with venous thromboembolism (VTE) treated with anticoagulants are at risk of death from pulmonary embolism (PE) and/or bleeding. However, whether patients who develop VTE in hospital have a higher complication rate than those who develop VTE in an outpatient setting is unclear. PATIENTS AND METHODS: RIETE is an ongoing, prospective registry of consecutive patients with acute, objectively confirmed, symptomatic VTE. We compared the 3-month incidence of fatal PE and fatal bleeding in patients in whom the VTE had developed while in hospital for another medical condition (inpatients) with those who presented to the emergency ward because of VTE (outpatients). RESULTS: Up to April 2008, 22,133 patients with acute VTE were enrolled: 10,461 (47%) presented with PE, 11,672 with deep vein thrombosis. Overall, 6445 (29%) were inpatients. During the study period, those who developed VTE as inpatients had a significantly higher incidence of fatal PE (2.1% vs. 1.5%; odds ratio: 1.4; 95% CI: 1.1-1.7), overall death (7.0% vs. 5.4%; odds ratio: 1.3; 95% CI: 1.2-1.5), and major bleeding (2.9% vs. 2.1%; odds ratio: 1.4; 95% CI: 1.1-1.6) than outpatients. The incidence of fatal bleeding was not significantly increased (0.7% vs. 0.5%; odds ratio: 1.2; 95% CI: 0.9-1.8). In multivariable analysis, inpatient status was significantly associated with a higher risk for fatal PE (odds ratio: 1.3; 95% CI: 1.1-1.7). CONCLUSIONS: VTE occurring in hospitalized patients carries a significantly higher risk for death of PE than in outpatients, underscoring the importance of VTE prevention strategies in the hospital setting.

ZHARP: three-dimensional motion tracking from a single image plane.

Three-dimensional imaging and quantification of myocardial function are essential steps in the evaluation of cardiac disease. We propose a tagged magnetic resonance imaging methodology called zHARP that encodes and automatically tracks myocardial displacement in three dimensions. Unlike other motion encoding techniques, zHARP encodes both in-plane and through-plane motion in a single image plane without affecting the acquisition speed. Postprocessing unravels this encoding in order to directly track the 3-D displacement of every point within the image plane throughout an entire image sequence. Experimental results include a phantom validation experiment, which compares zHARP to phase contrast imaging, and an in vivo study of a normal human volunteer. Results demonstrate that the simultaneous extraction of in-plane and through-plane displacements from tagged images is feasible.

Pure sensory neuropathy in patients with primary Sjogren's syndrome: clinical, immunological and electromyographic findings

A pure sensory neuropathy caused by lymphocytic infiltration of the dorsal root ganglia has been reported in a few patients with Sjögren's syndrome. The clinical, immunological, and electromyographic findings of five patients with this type of neuropathy and primary Sjögren's syndrome were reviewed. Typical clinical indications were the presence of a chronic asymmetrical sensory deficit, initial disease in the hands with a predominant loss of the vibratory and joint position senses, and an association with Adie's pupil syndrome or trigeminal sensory neuropathy. The simultaneous impairment of the central and peripheral evoked cortical potentials suggested that there was a lesion of the neuronal cell body. The neuropathy preceded the diagnosis of Sjögren's syndrome in four patients. Four patients were positive for Ro antibodies, but systemic vasculitis or malignancy was not found after a mean follow up of six years. These findings indicate that in patients with a sensory neuropathy the diagnosis of Sjögren's syndrome has to be considered, even if the patient denies the presence of sicca symptoms, and that appropriate tests must be carried out.

Premature epiphyseal closure in an adolescent treated by retinoids for acne: An unusual cause of anterior knee pain

Introduction.- Retinoids are effective and widely used for the treatment of severe acne. Their use can be, however, associated with numerous side effects. For example, some rare cases of premature epiphyseal closure were reported.Observation.- A sixteen-year-old soccer player consulted for bilateral progressive anterior knee pain, since two months, evoking a femoro-patellar origin. After physiotherapy, the pain decreases on the right but remained on the left. The history taking brought out the use of isotretinoin for more than 6 months (0.5 mg/kg). Magnetic resonance imaging (MRI) findings showed an irregularity of the growth plate and an important metaphyso-epiphyseal oedema, more marked on the left. The diagnosis of retinoid induced premature ephysieal closure was retained. The treatment was stopped, with a resolution of symptoms within two months. The control MRI of the left knee present persisting small sequelar thumbprint-like growth plate lesion. Eighteen months later, neither limb-length discrepancy nor static disorder was noticed.Discussion.- Premature epiphyseal closure is a rare complication of retinoid treatment of acne. Retinoids induce an invasion of the growth plate by osteoclasts and a decrease in proteoglycans synthesis. The knee seems the most involved joint. The clinical presentation is aspecific, sometimes lightly symptomatic. A careful pharmacological history and an appropriate imaging are necessary. MRI is now the gold standard. It shows an irregularity of the growth plate with an oedema on both sides. In chronic phase, a thumbprint-like image may persist. The symptoms resolution arises in few weeks after the treatment interruption. A single case of static disorder was reported until now. The small size of the growth plate interruptions, insufficient to lead to a growth disorder if the medicament is stopped early enough, explains probably it. This complication being rare, a radiological follow-up of the young patients treated by retinoids is not proposed.