Biomedical subjects

R Murray-Smith

Publications and source records attributed to R Murray-Smith.

3 recordsLinked to original sources

PubMed2005

European union system for the evaluation of substances: the second version.

This publication presents major changes in the assessment of the risks of chemicals to human health and the environment as implemented in the second version of the European Union System for the Evaluation of Substances, EUSES 2.0. EUSES is a harmonised quantitative risk assessment tool for chemicals. It is the PC-implementation of the technical guidelines developed within the framework of EU chemical legislation for industrial chemicals and biocides. As such, it is designed to support decision making by risk managers in government and industry and to assist scientific institutions in the risk assessment for these substances. The development of EUSES 2.0 is a co-ordinated project of the European Chemicals Bureau, EU Member States and the European chemical industry. Several model concepts, the technical background and the user interface of EUSES have been improved considerably. Major changes in the environmental assessment such as the implementation of emission scenario documents for industrial chemicals and biocides, the addition of the marine risk assessment, the enhancement of the regional model to include global scales, and improvements in the secondary poisoning and environmental effects modelling will be discussed. The update of the human risk assessment module in EUSES focuses on the risk characterisation for both threshold and non-threshold substances with, among others, the introduction of assessment factors. The performance of EUSES is illustrated in an example showing the human and environmental risk assessment of a sanitation disinfectant for private use.

Animals↗

PubMed2000

Monitoring, modelling and environmental exposure assessment of industrial chemicals in the aquatic environment.

Monitoring and laboratory data play integral roles alongside fate and exposure models in comprehensive risk assessments. The principle in the European Union Technical Guidance Documents for risk assessment is that measured data may take precedence over model results but only after they are judged to be of adequate reliability and to be representative of the particular environmental compartments to which they are applied. In practice, laboratory and field data are used to provide parameters for the models, while monitoring data are used to validate the models' predictions. Thus, comprehensive risk assessments require the integration of laboratory and monitoring data with the model predictions. However, this interplay is often overlooked. Discrepancies between the results of models and monitoring should be investigated in terms of the representativeness of both. Certainly, in the context of the EU risk assessment of existing chemicals, the specific requirements for monitoring data have not been adequately addressed. The resources required for environmental monitoring, both in terms of manpower and equipment, can be very significant. The design of monitoring programmes to optimise the use of resources and the use of models as a cost-effective alternative are increasing in importance. Generic considerations and criteria for the design of new monitoring programmes to generate representative quality data for the aquatic compartment are outlined and the criteria for the use of existing data are discussed. In particular, there is a need to improve the accessibility to data sets, to standardise the data sets, to promote communication and harmonisation of programmes and to incorporate the flexibility to change monitoring protocols to amend the chemicals under investigation in line with changing needs and priorities.

Environmental Exposure↗

PubMed1996

Side effects of normalising radial basis function networks.

Normalisation of the basis function activations in a Radial Basis Function (RBF) network is a common way of achieving the partition of unity often desired for modelling applications. It results in the basis functions covering the whole of the input space to the same degree. However, normalisation of the basis functions can lead to other effects which are sometimes less desirable for modelling applications. This paper describes some side effects of normalisation which fundamentally alter properties of the basis functions, e.g. the shape is no longer uniform, maxima of basis functions can be shifted from their centres, and the basis functions are no longer guaranteed to decrease monotonically as distance from their centre increases--in many cases basis functions can 'reactivate', i.e. re-appear far from the basis function centre. This paper examines how these phenomena occur, discusses their relevance for non-linear function approximation and examines the effect of normalisation on the network condition number and weights.

Least-Squares Analysis↗