Acute kidney injury (AKI) is now widely recognised as a serious health care issue, occurring in up to 25% of hospital in-patients, often with worsening of outcomes. There have been several reports of substandard care in AKI. This quality improvement (QI) programme aimed to improve AKI care and outcomes in a large teaching hospital.Areas of documented poor AKI care were identified and specific improvement activities implemented through sequential Plan-Do-Study-Act (PDSA) cycles. An electronic alert system (e-alert) for AKI was developed, a Priority Care Checklist (PCC) was tested with the aid of specialist nurses whilst targeted education activities were carried out and data on care processes and outcomes monitored.The e-alert had a sensitivity of 99% for the detection of new cases of AKI. Key aspects of the PCC saw significant improvements in their attainment: Detection of AKI within 24 hours from 53% to 100%, fluid assessment from 42% to 90%, drug review 48% to 95% and adherence to nine key aspects of care from 40% to 90%. There was a significant reduction in variability of delivered AKI care. AKI incidence reduced from 9% of all hospitalisations at baseline to 6.5% (28% reduction), AKI related length of stay reduced from 22.1 days to 17 days (23% reduction) and time to recovery (AKI days) 15.5 to 9.8 days (36% reduction). AKI related deaths also showed a trend towards reduction, from an average of 38 deaths to 34 (10.5%). The number of cases of hospital acquired AKI were reduced by 28% from 120 to 86 per month.This study demonstrates significant improvements related to a QI programme combining e-alerts, a checklist implemented by a nurse and education in improving key processes of care. This resulted in sustained improvement in key patient outcomes.

Concerns about inadequate patient hydration and suboptimal monitoring of fluid balance have been documented in recent reports. The Fluid Balance Improvement Project at Central Manchester University Hospitals NHS Foundation Trust was undertaken to identify risk factors influencing hydration and to implement a revised process to manage these risks, resulting in the development of a hydration pathway. This new approach to monitoring patient hydration, together with staff education and support, has resulted in improved compliance with fluid balance monitoring standards, as well as significant improvements in identifying patients at risk of dehydration, and an increase in patients with acute kidney injury commencing appropriate fluid balance monitoring.

We address the challenge of model transfer learning for a shape model matching (SMM) system. The goal is to adapt an existing SMM system to work effectively with new data without rebuilding the system from scratch. Recently, several SMM systems have been proposed that combine the outcome of a Random Forest (RF) regression step with shape constraints. These methods have been shown to lead to accurate and robust results when applied to the localisation of landmarks annotating skeletal structures in radiographs. However, as these methods contain a supervised learning component, their performance heavily depends on the data that was used to train the system, limiting their applicability to a new dataset with different properties. Here we show how to tune an existing SMM system by both updating the RFs with new samples and re-estimating the shape model. We demonstrate the effectiveness of tuning a cephalometric SMM system to replicate the annotation style of a new observer. Our results demonstrate that tuning an existing system leads to significant improvements in performance on new data, up to the extent of performing a well as a system that was fully rebuilt using samples from the new dataset. The proposed approach is fast and does not require access to the original training data.

No abstract

No abstract