Statistical learning theory for geospatial data. Case study: Aral sea

In recent years there has been an explosive growth in the development of adaptive and data driven methods. One of the efficient and data-driven approaches is based on statistical learning theory (Vapnik 1998). The theory is based on Structural Risk Minimisation (SRM) principle and has a solid statistical background. When applying SRM we are trying not only to reduce training error ? to fit the available data with a model, but also to reduce the complexity of the model and to reduce generalisation error. Many nonlinear learning procedures recently developed in neural networks and statistics can be understood and interpreted in terms of the structural risk minimisation inductive principle. A recent methodology based on SRM is called Support Vector Machines (SVM). At present SLT is still under intensive development and SVM find new areas of application (www.kernel-machines.org). SVM develop robust and non linear data models with excellent generalisation abilities that is very important both for monitoring and forecasting. SVM are extremely good when input space is high dimensional and training data set i not big enough to develop corresponding nonlinear model. Moreover, SVM use only support vectors to derive decision boundaries. It opens a way to sampling optimization, estimation of noise in data, quantification of data redundancy etc. Presentation of SVM for spatially distributed data is given in (Kanevski and Maignan 2004).

Concomitant chemoradiotherapy versus acceleration of radiotherapy with or without concomitant chemotherapy in locally advanced head and neck carcinoma (GORTEC 99-02): an open-label phase 3 randomised trial.

BACKGROUND: Concomitant chemoradiotherapy and accelerated radiotherapy independently improve outcomes for patients with locally advanced head and neck squamous-cell carcinoma (HNSCC). We aimed to assess the efficacy and safety of a combination of these approaches. METHODS: In our open-label phase 3 randomised trial, we enrolled patients with locally advanced, stage III and IV (non-metastatic) HNSCC and an Eastern Cooperative Oncology Group performance status of 0-2. We randomly allocated patients centrally with a computer program (with centre, T stage, N stage, and localisation as minimisation factors) in a 1:1:1 ratio to receive conventional chemoradiotherapy (70 Gy in 7 weeks plus three cycles of 4 days' concomitant carboplatin-fluorouracil), accelerated radiotherapy-chemotherapy (70 Gy in 6 weeks plus two cycles of 5 days' concomitant carboplatin-fluorouracil), or very accelerated radiotherapy alone (64·8 Gy [1·8 Gy twice daily] in 3·5 weeks). The primary endpoint, progression-free survival (PFS), was assessed in all enrolled patients. This trial is completed. The trial is registered with ClinicalTrials.gov, number NCT00828386. FINDINGS: Between Feb 29, 2000, and May 9, 2007, we randomly allocated 279 patients to receive conventional chemoradiotherapy, 280 to accelerated radiotherapy-chemotherapy, and 281 to very accelerated radiotherapy. Median follow-up was 5·2 years (IQR 4·9-6·2); rates of chemotherapy and radiotherapy compliance were good in all groups. Accelerated radiotherapy-chemotherapy offered no PFS benefit compared with conventional chemoradiotherapy (HR 1·02, 95% CI 0·84-1·23; p=0·88) or very accelerated radiotherapy (0·83, 0·69-1·01; p=0·060); conventional chemoradiotherapy improved PFS compared with very accelerated radiotherapy (0·82, 0·67-0·99; p=0·041). 3-year PFS was 37·6% (95% CI 32·1-43·4) after conventional chemoradiotherapy, 34·1% (28·7-39·8) after accelerated radiotherapy-chemotherapy, and 32·2% (27·0-37·9) after very accelerated radiotherapy. More patients in the very accelerated radiotherapy group had RTOG grade 3-4 acute mucosal toxicity (226 [84%] of 268 patients) compared with accelerated radiotherapy-chemotherapy (205 [76%] of 271 patients) or conventional chemoradiotherapy (180 [69%] of 262; p=0·0001). 158 (60%) of 265 patients in the conventional chemoradiotherapy group, 176 (64%) of 276 patients in the accelerated radiotherapy-chemotherapy group, and 190 (70%) of 272 patients in the very accelerated radiotherapy group were intubated with feeding tubes during treatment (p=0·045). INTERPRETATION: Chemotherapy has a substantial treatment effect given concomitantly with radiotherapy and acceleration of radiotherapy cannot compensate for the absence of chemotherapy. We noted the most favourable outcomes for conventional chemoradiotherapy, suggesting that acceleration of radiotherapy is probably not beneficial in concomitant chemoradiotherapy schedules. FUNDING: French Ministry of Health.

External irradiation with or without long-term androgen suppression for prostate cancer with high metastatic risk: 10-year results of an EORTC randomised study.

BACKGROUND: We did a randomised phase 3 trial assessing the benefit of addition of long-term androgen suppression with a luteinising-hormone-releasing hormone (LHRH) agonist to external irradiation in patients with prostate cancer with high metastatic risk. In this report, we present the 10-year results. METHODS: For this open-label randomised trial, eligible patients were younger than 80 years and had newly diagnosed histologically proven T1-2 prostatic adenocarcinoma with WHO histological grade 3 or T3-4 prostatic adenocarcinoma of any histological grade, and a WHO performance status of 0-2. Patients were randomly assigned (1:1) to receive radiotherapy alone or radiotherapy plus immediate androgen suppression. Treatment allocation was open label and used a minimisation algorithm with institution, clinical stage of the disease, results of pelvic-lymph-node dissection, and irradiation fields extension as minimisation factors. Patients were irradiated externally, once a day, 5 days a week, for 7 weeks to a total dose of 50 Gy to the whole pelvis, with an additional 20 Gy to the prostate and seminal vesicles. The LHRH agonist, goserelin acetate (3·6 mg subcutaneously every 4 weeks), was started on the first day of irradiation and continued for 3 years; cyproterone acetate (50 mg orally three times a day) was given for 1 month starting a week before the first goserelin injection. The primary endpoint was clinical disease-free survival. Analysis was by intention to treat. The trial is registered at ClinicalTrials.gov, number NCT00849082. FINDINGS: Between May 22, 1987, and Oct 31, 1995, 415 patients were randomly assigned to treatment groups and were included in the analysis (208 radiotherapy alone, 207 combined treatment). Median follow-up was 9·1 years (IQR 5·1-12·6). 10-year clinical disease-free survival was 22·7% (95% CI 16·3-29·7) in the radiotherapy-alone group and 47·7% (39·0-56·0) in the combined treatment group (hazard ratio [HR] 0·42, 95% CI 0·33-0·55, p<0·0001). 10-year overall survival was 39·8% (95% CI 31·9-47·5) in patients receiving radiotherapy alone and 58·1% (49·2-66·0) in those allocated combined treatment (HR 0·60, 95% CI 0·45-0·80, p=0·0004), and 10-year prostate-cancer mortality was 30·4% (95% CI 23·2-37·5) and 10·3% (5·1-15·4), respectively (HR 0·38, 95% CI 0·24-0·60, p<0·0001). No significant difference in cardiovascular mortality was noted between treatment groups both in patients who had cardiovascular problems at study entry (eight of 53 patients in the combined treatment group had a cardiovascular-related cause of death vs 11 of 63 in the radiotherapy group; p=0·60) and in those who did not (14 of 154 vs six of 145; p=0·25). Two fractures were reported in patients allocated combined treatment. INTERPRETATION: In patients with prostate cancer with high metastatic risk, immediate androgen suppression with an LHRH agonist given during and for 3 years after external irradiation improves 10-year disease-free and overall survival without increasing late cardiovascular toxicity.

Le rôle des employeurs dans la réinsertion professionnelle

Les évolutions récentes du marché du travail et les transformations de l'action de l'Etat social ont contribué à augmenter l'importance de la réinsertion professionnelle. Le renforcement de l'efficacité des mesures de réinsertion nécessite une prise en compte de tous les acteurs concernés, y compris des employeurs dont le rôle est souvent négligé. Ce travail étudie leur rôle d'un point de vue théorique et empirique, grâce à des entretiens réalisés avec des employeurs de quatre pays, y compris la Suisse. Les facteurs (critère d'embauche, rapports avec les services publics de l'emploi, perception des mesures ...) qui conditionnent la participation des employeurs au processus de réinsertion sont mis en avant et analysés. Le travail montre les attentes des employeurs en termes de stratégies de recherche d'emploi et de comportement des bénéficiaires. Les rapports distants entre les employeurs et les agences publiques de placement sont eux aussi analysés. Les résultats obtenus mettent en lumière la volonté des employeurs de participer à la réinsertion, mais soulignent également l'importance des mécanismes de minimisation des risques pour les employeurs. De plus, la force de la flexibilisation du marché est également démontrée. Finalement, cette meilleure connaissance des craintes et attentes des employeurs permet d'établir des recommandations visant à augmenter l'efficacité des mesures de réinsertion et à améliorer la compréhension mutuelle des différents acteurs de la réinsertion.

Carboplatin and weekly paclitaxel doublet chemotherapy compared with monotherapy in elderly patients with advanced non-small-cell lung cancer: IFCT-0501 randomised, phase 3 trial.

BACKGROUND: Platinum-based doublet chemotherapy is recommended to treat advanced non-small-cell lung cancer (NSCLC) in fit, non-elderly adults, but monotherapy is recommended for patients older than 70 years. We compared a carboplatin and paclitaxel doublet chemotherapy regimen with monotherapy in elderly patients with advanced NSCLC. METHODS: In this multicentre, open-label, phase 3, randomised trial we recruited patients aged 70-89 years with locally advanced or metastatic NSCLC and WHO performance status scores of 0-2. Patients received either four cycles (3 weeks on treatment, 1 week off treatment) of carboplatin (on day 1) plus paclitaxel (on days 1, 8, and 15) or five cycles (2 weeks on treatment, 1 week off treatment) of vinorelbine or gemcitabine monotherapy. Randomisation was done centrally with the minimisation method. The primary endpoint was overall survival, and analysis was done by intention to treat. This trial is registered, number NCT00298415. FINDINGS: 451 patients were enrolled. 226 were randomly assigned monotherapy and 225 doublet chemotherapy. Median age was 77 years and median follow-up was 30.3 months (range 8.6-45.2). Median overall survival was 10.3 months for doublet chemotherapy and 6.2 months for monotherapy (hazard ratio 0.64, 95% CI 0.52-0.78; p<0.0001); 1-year survival was 44.5% (95% CI 37.9-50.9) and 25.4% (19.9-31.3), respectively. Toxic effects were more frequent in the doublet chemotherapy group than in the monotherapy group (most frequent, decreased neutrophil count (108 [48.4%] vs 28 [12.4%]; asthenia 23 [10.3%] vs 13 [5.8%]). INTERPRETATION: Despite increased toxic effects, platinum-based doublet chemotherapy was associated with survival benefits compared with vinorelbine or gemcitabine monotherapy in elderly patients with NSCLC. We feel that the current treatment paradigm for these patients should be reconsidered. FUNDING: Intergroupe Francophone de Cancérologie Thoracique, Institut National du Cancer.

Concurrent or sequential adjuvant letrozole and radiotherapy after conservative surgery for early-stage breast cancer (CO-HO-RT): a phase 2 randomised trial.

BACKGROUND: Letrozole radiosensitises breast cancer cells in vitro. In clinical settings, no data exist for the combination of letrozole and radiotherapy. We assessed concurrent and sequential radiotherapy and letrozole in the adjuvant setting. METHODS: This phase 2 randomised trial was undertaken in two centres in France and one in Switzerland between Jan 12, 2005, and Feb 21, 2007. 150 postmenopausal women with early-stage breast cancer were randomly assigned after conserving surgery to either concurrent radiotherapy and letrozole (n=75) or sequential radiotherapy and letrozole (n=75). Randomisation was open label with a minimisation technique, stratified by investigational centres, chemotherapy (yes vs no), radiation boost (yes vs no), and value of radiation-induced lymphocyte apoptosis (< or = 16% vs >16%). Whole breast was irradiated to a total dose of 50 Gy in 25 fractions over 5 weeks. In the case of supraclavicular and internal mammary node irradiation, the dose was 44-50 Gy. Letrozole was administered orally once daily at a dose of 2.5 mg for 5 years (beginning 3 weeks pre-radiotherapy in the concomitant group, and 3 weeks post-radiotherapy in the sequential group). The primary endpoint was the occurrence of acute (during and within 6 weeks of radiotherapy) and late (within 2 years) radiation-induced grade 2 or worse toxic effects of the skin. Analyses were by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00208273. FINDINGS: All patients were analysed apart from one in the concurrent group who withdrew consent before any treatment. During radiotherapy and within the first 12 weeks after radiotherapy, 31 patients in the concurrent group and 31 in the sequential group had any grade 2 or worse skin-related toxicity. The most common skin-related adverse event was dermatitis: four patients in the concurrent group and six in the sequential group had grade 3 acute skin dermatitis during radiotherapy. At a median follow-up of 26 months (range 3-40), two patients in each group had grade 2 or worse late effects (both radiation-induced subcutaneous fibrosis). INTERPRETATION: Letrozole can be safely delivered shortly after surgery and concomitantly with radiotherapy. Long-term follow-up is needed to investigate cardiac side-effects and cancer-specific outcomes. FUNDING: Novartis Oncology France.

Positive time-frequency distributions based on joint marginal constraints

This correspondence studies the formulation of members ofthe Cohen-Posch class of positive time-frequency energy distributions.Minimization of cross-entropy measures with respect to different priorsand the case of no prior or maximum entropy were considered. It isconcluded that, in general, the information provided by the classicalmarginal constraints is very limited, and thus, the final distributionheavily depends on the prior distribution. To overcome this limitation,joint time and frequency marginals are derived based on a "directioninvariance" criterion on the time-frequency plane that are directly relatedto the fractional Fourier transform.

Incidència de les lesions en el futbol. Primera divisió. Temporada 1999-2000

El futbol constitueix al nostre país, igual com en d’altres Estats del nostre entorn, un fenomen que transcendeix allò merament esportiu per esdevenir un referent social que desperta passions i que té el seu consegüent reflex als mitjans de comunicació social. No és, doncs, estrany, que de forma diària la premsa esportiva d’àmbit nacional proporcioni informació constant sobre la matèria, incloent-hi allò que concerneix a les lesions que afecten els jugadors, car consideren que és un fet noticiable en la mesura en què això pugui influir en les expectatives dels diferents equips. La informació esmentada, adequadament recollida, sistematitzada i analitzada és susceptible de proporcionar valuoses dades sobre la incidència de les lesions en aquesta modalitat esportiva. Aquest és, precisament, l’objectiu del nostre estudi, sense una altra pretensió que la de proporcionar unes pautes a prendre en consideració per part dels professionals, perquè el seu esforç se centri en l’optimització del rendiment, la planificació de la temporada, l’establiment de programes de prevenció i, al capdavall, en la minimització del risc de lesions.

Incidencia lesional en el fútbol. Primera división. Temporada 1999-2000

El fútbol constituye en nuestro país, al igual que en otros Estados de nuestro entorno, un fenómeno que trasciende de lo meramente deportivo para convertirse en un referente social que despierta pasiones y que tiene su consiguiente reflejo en los medios de comunicación social. No es pues de extrañar que de forma diaria la prensa deportiva de ámbito nacional proporcione información constante sobre la materia, incluyendo lo concerniente a las lesiones que aquejan a los jugadores considerándolo hecho noticiable en la medida que ello pueda afectar a las expectativas de los diferentes equipos. Dicha información, debidamente recogida, sistematizada y analizada, es susceptible de proporcionar valiosos datos sobre la incidencia lesional en esta modalidad deportiva. Éste es precisamente el objetivo de nuestro estudio, sin otra pretensión que la de proporcionar unas pautas a tomar en consideración por los profesionales cuyo esfuerzo se centre en la optimización del rendimiento, en la planificación de la temporada, en el establecimiento de programas de prevención y, en definitiva, en la minimización del riesgo lesional.

Structured sparsity through reweighting and application to diffusion MRI

We consider the problem of multiple correlated sparse signals reconstruction and propose a new implementation of structured sparsity through a reweighting scheme. We present a particular application for diffusion Magnetic Resonance Imaging data and show how this procedure can be used for fibre orientation reconstruction in the white matter of the brain. In that framework, our structured sparsity prior can be used to exploit the fundamental coherence between fibre directions in neighbour voxels. Our method approaches the ℓ0 minimisation through a reweighted ℓ1-minimisation scheme. The weights are here defined in such a way to promote correlated sparsity between neighbour signals.

Local map update for large scale SLAM

A technique for simultaneous localisation and mapping (SLAM) for large scale scenarios is presented. This solution is based on the use of independent submaps of a limited size to map large areas. In addition, a global stochastic map, containing the links between adjacent submaps, is built. The information in both levels is corrected every time a loop is closed: local maps are updated with the information from overlapping maps, and the global stochastic map is optimised by means of constrained minimisation

Efficient estimation using the characteristic function : theory and applications with high frequency data

The attached file is created with Scientific Workplace Latex

Représentation et détection des images et des surfaces déformables

La représentation d'une surface, son lissage et son utilisation pour l'identification, la comparaison, la classification, et l'étude des variations de volume, de courbure ou de topologie sont omniprésentes dans l'aire de la numérisation. Parmi les méthodes mathématiques, nous avons retenu les transformations difféomorphiques d'un pattern de référence. Il y a un grand intérêt théorique et numérique à approcher un difféomorphisme arbitraire par des difféomorphismes engendrés par des champs de vitesses. Sur le plan théorique la question est : "est-ce que le sous-groupe de difféomorphismes engendrés par des champs de vitesses est dense dans le groupe plus large de Micheletti pour la métrique de Courant ?" Malgré quelques progrès réalisés ici, cette question demeure ouverte. Les pistes empruntées ont alors convergé vers le sous-groupe de Azencott et de Trouvé et sa métrique dans le cadre de l'imagerie. Elle correspond à une notion de géodésique entre deux difféomorphismes dans leur sous-groupe. L'optimisation est utilisée pour obtenir un système d'équations état adjoint caractérisant la solution optimale du problème d'identification à partir des observations. Cette approche est adaptée à l'identification de surfaces obtenues par un numériseur tel que, par exemple, le scan d'un visage. Ce problème est beaucoup plus difficile que celui d'imagerie. On doit alors introduire un système de référence courbe et une surface à facettes pour les calculs. On donne la formulation du problème d'identification et du calcul du changement de volume par rapport à un scan de référence.

La télémédecine en radiothérapie : développement d’un modèle et analyse des coûts

But : La radiothérapie (RT) est disponible seulement dans les grandes villes au Québec. Les patients atteints de cancer vivant en zone rurale doivent voyager pour obtenir ces soins. Toute proportion gardée, moins de ces patients accèdent à la RT. L’accessibilité serait améliorée en instaurant de petits centres de RT qui dépendraient de la télémédecine (téléRT). Cette étude tente (1) de décrire un modèle (population visée et technologie) réaliste de téléRT; (2) d’en estimer les coûts, comparativement à la situation actuelle où les patients voyagent (itineRT). Méthode : (1) À l’aide de données probantes, le modèle de téléRT a été développé selon des critères de : faisabilité, sécurité, absence de transfert des patients et minimisation du personnel. (2) Les coûts ont été estimés du point de vue du payeur unique en utilisant une méthode publiée qui tient compte des coûts en capitaux, de la main d’oeuvre et des frais généraux. Résultats : (1) Le modèle de téléRT proposé se limiterait aux traitements palliatifs à 250 patients par année. (2) Les coûts sont de 5918$/patient (95% I.C. 4985 à 7095$) pour téléRT comparativement à 4541$/patient (95%I.C. 4351 à 4739$) pour itineRT. Les coûts annuels de téléRT sont de 1,48 M$ (d.s. 0,6 M$), avec une augmentation des coûts nets de seulement 0,54 M$ (d.s. 0,26 M$) comparativement à itineRT. Si on modifiait certaines conditions, le service de téléRT pourrait s’étendre au traitement curatif du cancer de prostate et du sein, à coûts similaires à itineRT. Conclusion : Ce modèle de téléRT pourrait améliorer l’accessibilité et l’équité aux soins, à des coûts modestes.

Entourage d'un toxicomane. Profil clinique: besoins, interventions et impacts.

Rapport de stage présenté à la Faculté des Arts et des Sciences en vue de l'obtention du grade de Maîtrise ès sciences (M.Sc.) en criminologie.