Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Safety of human immunisation with a live-attenuated Mycobacterium tuberculosis vaccine: a randomised, double-blind, controlled phase I trial.

BACKGROUND: Tuberculosis remains one of the world's deadliest transmissible diseases despite widespread use of the BCG vaccine. MTBVAC is a new live tuberculosis vaccine based on genetically attenuated Mycobacterium tuberculosis that expresses most antigens present in human isolates of M tuberculosis. We aimed to compare the safety of MTBVAC with BCG in healthy adult volunteers. METHODS: We did this single-centre, randomised, double-blind, controlled phase 1 study at the Centre Hospitalier Universitaire Vaudois (CHUV; Lausanne, Switzerland). Volunteers were eligible for inclusion if they were aged 18-45 years, clinically healthy, HIV-negative and tuberculosis-negative, and had no history of active tuberculosis, chemoprophylaxis for tuberculosis, or BCG vaccination. Volunteers fulfilling the inclusion criteria were randomly assigned to three cohorts in a dose-escalation manner. Randomisation was done centrally by the CHUV Pharmacy and treatments were masked from the study team and volunteers. As participants were recruited within each cohort, they were randomly assigned 3:1 to receive MTBVAC or BCG. Of the participants allocated MTBVAC, those in the first cohort received 5 × 10(3) colony forming units (CFU) MTBVAC, those in the second cohort received 5 × 10(4) CFU MTBVAC, and those in the third cohort received 5 × 10(5) CFU MTBVAC. In all cohorts, participants assigned to receive BCG were given 5 × 10(5) CFU BCG. Each participant received a single intradermal injection of their assigned vaccine in 0·1 mL sterile water in their non-dominant arm. The primary outcome was safety in all vaccinated participants. Secondary outcomes included whole blood cell-mediated immune response to live MTBVAC and BCG, and interferon γ release assays (IGRA) of peripheral blood mononuclear cells. This trial is registered with ClinicalTrials.gov, number NCT02013245. FINDINGS: Between Jan 23, 2013, and Nov 6, 2013, we enrolled 36 volunteers into three cohorts, each of which consisted of nine participants who received MTBVAC and three who received BCG. 34 volunteers completed the trial. The safety of vaccination with MTBVAC at all doses was similar to that of BCG, and vaccination did not induce any serious adverse events. All individuals were IGRA negative at the end of follow-up (day 210). After whole blood stimulation with live MTBVAC or BCG, MTBVAC was at least as immunogenic as BCG. At the same dose as BCG (5×10(5) CFU), although no statistical significance could be achieved, there were more responders in the MTBVAC group than in the BCG group, with a greater frequency of polyfunctional CD4+ central memory T cells. INTERPRETATION: To our knowledge, MTBVAC is the first live-attenuated M tuberculosis vaccine to reach clinical assessment, showing similar safety to BCG. MTBVAC seemed to be at least as immunogenic as BCG, but the study was not powered to investigate this outcome. Further plans to use more immunogenicity endpoints in a larger number of volunteers (adults and adolescents) are underway, with the aim to thoroughly characterise and potentially distinguish immunogenicity between MTBVAC and BCG in tuberculosis-endemic countries. Combined with an excellent safety profile, these data support advanced clinical development in high-burden tuberculosis endemic countries. FUNDING: Biofabri and Bill & Melinda Gates Foundation through the TuBerculosis Vaccine Initiative (TBVI).