Ensemble approaches to classification and regression have attracted a great deal of interest in recent years. These methods can be shown both theoretically and empirically to outperform single predictors on a wide range of tasks. One of the elements required for accurate prediction when using an ensemble is recognised to be error "diversity". However, the exact meaning of this concept is not clear from the literature, particularly for classification tasks. In this paper we first review the varied attempts to provide a formal explanation of error diversity, including several heuristic and qualitative explanations in the literature. For completeness of discussion we include not only the classification literature but also some excerpts of the rather more mature regression literature, which we believe can still provide some insights. We proceed to survey the various techniques used for creating diverse ensembles, and categorise them, forming a preliminary taxonomy of diversity creation methods. As part of this taxonomy we introduce the idea of implicit and explicit diversity creation methods, and three dimensions along which these may be applied. Finally we propose some new directions that may prove fruitful in understanding classification error diversity.

No abstract

No abstract

Radiation-induced toxicity is a common adverse side effect of radiation therapy. Previous studies have demonstrated a lack of evidence to support common skincare advice for radiotherapy patients. The aim of the current study was to investigate the management of radiation-induced skin toxicity across Europe and the USA. Where previous surveys have focused on national practice or treatment of specific sites, the current study aimed to gain a broader representation of skincare practice. An anonymous online survey investigating various aspects of radiotherapy skincare management was distributed to departments across Europe and the USA (n = 181/737 responded i.e. 25%). The UK was excluded as a similar survey was carried out in 2011. The results highlight the lack of consistency in both the prevention and management of radiation-induced skin toxicity. Recommended products are often not based on evidence-based practice. Examples include the continued use of aqueous cream and gentian violet, as well as the recommendations on washing restrictions during treatment. To our knowledge, this is the most extensive survey to date on the current management of radiation-induced skin toxicity. This study highlights significant disparities between clinical practice and research-based evidence published in recent systematic reviews and guidelines. Ongoing large prospective randomised trials are urgently needed.

No abstract

Research from Allied Health Professionals (AHPs) is anecdotally known to lag behind that of other professions. The developing research landscape within other therapies and internationally led us to question how UK practice in therapeutic radiography was developing. The aim of the survey was to audit research capacity across therapy radiography in the UK. MethodAn electronic survey was sent to Radiotherapy Service Managers (RSM) and research leads in each of the radiotherapy centres in the UK. An adapted version of the Auditing Research Capacity tool (ARC © tool) was used as the basis of the questionnaire. ResultsA total of 45 RSM responded to the survey (67% response rate) and 30 Research radiographers (RR)(45% response rate). A total of 51 RR were in post equating to 40.3 whole time equivalents and averaging 1 RR per centre. Variation was evident in the commitment to the development of a research culture identified by practices such as linking research to the business planning cycle, inclusion of research in recruitment and advertising materials, or having a nominated therapeutic radiographer lead on research for the department. Over a third of responding centres did not have a research strategy and training for RRs was limited; specifically in areas such as writing funding bids, writing for publication and the research and governance process. ConclusionA number of short and long-term strategies are proposed that should enhance a positive research culture and improve research capacity for therapeutic radiography led research. These include utilisation of the existing infrastructure provided by the National Institute for Health Research, a lead or co-ordinator for research activity with a remit to motivate others. Development of links and networks, and the development of a research strategy linked to wider Trust research priorities. The research strategy should include mentoring or developing appropriate research skills for those engaged in research (including higher degree qualifications). RSMs should also encourage peerreviewed publications, and conference presentations from all staff to ensure research results are disseminated to the wider profession.

No abstract

No abstract