Risk Modelling Framework for Emergency Hospital Readmission, Using Hospital Episode Statistics Inpatient Data

The objective of this study was to develop, test and benchmark a framework and a predictive risk model for hospital emergency readmission within 12 months. We performed the development using routinely collected Hospital Episode Statistics data covering inpatient hospital admissions in England. Three different timeframes were used for training, testing and benchmarking: 1999 to 2004, 2000 to 2005 and 2004 to 2009 financial years. Each timeframe includes 20% of all inpatients admitted within the trigger year. The comparisons were made using positive predictive value, sensitivity and specificity for different risk cut-offs, risk bands and top risk segments, together with the receiver operating characteristic curve. The constructed Bayes Point Machine using this feature selection framework produces a risk probability for each admitted patient, and it was validated for different timeframes, sub-populations and cut-off points. At risk cut-off of 50%, the positive predictive value was 69.3% to 73.7%, the specificity was 88.0% to 88.9% and sensitivity was 44.5% to 46.3% across different timeframes. Also, the area under the receiver operating characteristic curve was 73.0% to 74.3%. The developed framework and model performed considerably better than existing modelling approaches with high precision and moderate sensitivity.

An empirical study of the variability in the composition of British freight trains

As part of the broader sustainability and economic efficiency agenda, European transport policy places considerable emphasis on improving rail’s competitiveness to increase its share of the freight market. Much attention is devoted to infrastructure characteristics which determine the number of freight trains which can operate and influence the operating characteristics of these trains. However, little attention has been devoted to the composition of the freight trains themselves, with scant published data relating to the practicalities of this important component of system utilisation and its impacts on rail freight viability and sustainability. This paper develops a better understanding of the extent to which freight train composition varies, through a large-scale empirical study of the composition of British freight trains. The investigation is based on a survey of almost 3,000 individual freight trains, with analysis at four levels of disaggregation, from the commodity groupings used in official statistics down to individual services. This provides considerable insight into rail freight operations with particular relevance to the efficiency of utilisation of trains using the available network paths. The results demonstrate the limitations of generalising about freight train formations since, within certain commodity groupings, considerable variability was identified even at fairly high levels of disaggregation.

Performance of an adaptive homodyne receiver in the presence of multipath, Rayleigh-fading and time-varying quadrature errors

In this paper, we carry out a detailed performance analysis of the blind source separation based I/Q corrector operating at the baseband. Performance of the digital I/Q corrector is evaluated not only under time-varying phase and gain errors but also in the presence of multipath and Rayleigh fading channels. Performance under low-SNR and different modulation formats and constellation sizes is also evaluated. What is more, BER improvement after correction is illustrated. The results indicate that the adaptive algorithm offers adequate performance for most communication applications hence, reducing the matching requirements of the analog front-end enabling higher levels of integration.