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.

Child sun protection in New Zealand: parental view and societal responsibilities

Issue addressed: Cutaneous melanoma is a significant health problem in New Zealand. Excessive sun exposure in early life increases subsequent risk. This study investigated parental opinions, understanding and practices concerning the sun protection of young children. The study aimed to identify areas where improvements in sun protection may be most needed. Methods: Parents were recruited through licensed childcare centres and kindergartens in Dunedin to take part in semi-structured focus groups. Feedback was obtained from participants in response to summary reports based on audiotapes. Results: Parents noted increased social acceptability of sun protective behaviours and child sunburn was now unacceptable. Past media campaigns were well recalled. The 'time to burn' used in media weather reports was easier to understand than the Ultra Violet Index (UVI), about which more information was wanted. Protective messages were expected to be straightforward, consistent and readily and regularly available. Local radio may provide the most timely, relevant information. There was a perceived lack of authoritative information about sunscreens and sunglasses and a shortage of acceptable protective clothing. Fuller information on sunscreen containers and greater use of UV Protection Factor (UPF) ratings for clothing and Eye Protection Factor (EPF) for sunglasses would assist. The use of shade and rescheduling of activities were scarcely mentioned. Conclusions: Parents were aware of the need for child sun protection but lacked confidence about how best to achieve this. Future health promotion programs should emphasise how optimal protection can be achieved more than why sun protection is needed. Programs should include a repertoire of strategies targeted towards individuals through the education of children and caregivers. They should also aim at achieving modifications in physical and social environments, including appropriate product design and promotion. So what?: The development of a balanced, comprehensive program with environmental components that reinforce protective behaviours has the potential to sustain sun protection among the largest number of children in the longer term.

Clustering and hot spot detection in socio-economic spatio-temporal data

Distribution of socio-economic features in urban space is an important source of information for land and transportation planning. The metropolization phenomenon has changed the distribution of types of professions in space and has given birth to different spatial patterns that the urban planner must know in order to plan a sustainable city. Such distributions can be discovered by statistical and learning algorithms through different methods. In this paper, an unsupervised classification method and a cluster detection method are discussed and applied to analyze the socio-economic structure of Switzerland. The unsupervised classification method, based on Ward's classification and self-organized maps, is used to classify the municipalities of the country and allows to reduce a highly-dimensional input information to interpret the socio-economic landscape. The cluster detection method, the spatial scan statistics, is used in a more specific manner in order to detect hot spots of certain types of service activities. The method is applied to the distribution services in the agglomeration of Lausanne. Results show the emergence of new centralities and can be analyzed in both transportation and social terms.

A community survey of sun exposure, sunburn and sun protection.

AIM: To describe outdoor activities, sun protection behaviours and the experience of sunburn in a sample of New Zealanders during summer weekends of 1994. METHODS: 1243 respondents to a telephone survey provided information regarding their outdoor activities for the 5 hour period around midday of the previous Saturday and Sunday. The sample was drawn from those aged 15 to 65 years in the five centres of Auckland, Hamilton, Wellington, Christchurch and Dunedin. Respondents provided information on sun exposure, sunburn, sun protection and beliefs about tanning, as well as background demographic information, skin type and previous experience of sunburn. RESULTS: 12% of the sample (or 17% of all those outdoors) reported being sunburned on the preceding weekend, and those sunburned tended to be men, and to be under age 35 years. The face, neck and limbs were the areas most frequently reported as burned. Sporting activities and beach or water activities were associated with the highest number of episodes of burning. Overall 38% of those outside reported wearing a hat and 32% reported the use of a sunscreen. Positive attitudes to tanning were quite common and probably present the main target for change in the community. CONCLUSION: On any sunny weekend in summer about three-quarters of adult New Zealanders will be out in the sun for relatively long periods of time, and many will get sunburned. The reduction of such harmful sun exposures remains an important public health goal.

Ultraviolet damage to the eye revisited: eye-sun protection factor (E-SPF®), a new ultraviolet protection label for eyewear.

Ultraviolet (UV) radiation potentially damages the skin, the immune system, and structures of the eye. A useful UV sun protection for the skin has been established. Since a remarkable body of evidence shows an association between UV radiation and damage to structures of the eye, eye protection is important, but a reliable and practical tool to assess and compare the UV-protective properties of lenses has been lacking. Among the general lay public, misconceptions on eye-sun protection have been identified. For example, sun protection is mainly ascribed to sunglasses, but less so to clear lenses. Skin malignancies in the periorbital region are frequent, but usual topical skin protection does not include the lids. Recent research utilized exact dosimetry and demonstrated relevant differences in UV burden to the eye and skin at a given ambient irradiation. Chronic UV effects on the cornea and lens are cumulative, so effective UV protection of the eyes is important for all age groups and should be used systematically. Protection of children's eyes is especially important, because UV transmittance is higher at a very young age, allowing higher levels of UV radiation to reach the crystalline lens and even the retina. Sunglasses as well as clear lenses (plano and prescription) effectively reduce transmittance of UV radiation. However, an important share of the UV burden to the eye is explained by back reflection of radiation from lenses to the eye. UV radiation incident from an angle of 135°-150° behind a lens wearer is reflected from the back side of lenses. The usual antireflective coatings considerably increase reflection of UV radiation. To provide reliable labeling of the protective potential of lenses, an eye-sun protection factor (E-SPF®) has been developed. It integrates UV transmission as well as UV reflectance of lenses. The E-SPF® compares well with established skin-sun protection factors and provides clear messages to eye health care providers and to lay consumers.

New microfluidic-based sampling procedure for overcoming the hematocrit problem associated with dried blood spot analysis.

Hematocrit (Hct) is one of the most critical issues associated with the bioanalytical methods used for dried blood spot (DBS) sample analysis. Because Hct determines the viscosity of blood, it may affect the spreading of blood onto the filter paper. Hence, accurate quantitative data can only be obtained if the size of the paper filter extracted contains a fixed blood volume. We describe for the first time a microfluidic-based sampling procedure to enable accurate blood volume collection on commercially available DBS cards. The system allows the collection of a controlled volume of blood (e.g., 5 or 10 μL) within several seconds. Reproducibility of the sampling volume was examined in vivo on capillary blood by quantifying caffeine and paraxanthine on 5 different extracted DBS spots at two different time points and in vitro with a test compound, Mavoglurant, on 10 different spots at two Hct levels. Entire spots were extracted. In addition, the accuracy and precision (n = 3) data for the Mavoglurant quantitation in blood with Hct levels between 26% and 62% were evaluated. The interspot precision data were below 9.0%, which was equivalent to that of a manually spotted volume with a pipet. No Hct effect was observed in the quantitative results obtained for Hct levels from 26% to 62%. These data indicate that our microfluidic-based sampling procedure is accurate and precise and that the analysis of Mavoglurant is not affected by the Hct values. This provides a simple procedure for DBS sampling with a fixed volume of capillary blood, which could eliminate the recurrent Hct issue linked to DBS sample analysis.