We propose FedScore, a privacy-preserving federated learning framework for scoring system generation across multiple sites to facilitate cross-institutional collaborations. The FedScore framework includes five modules: federated variable ranking, federated variable transformation, federated score derivation, federated model selection and federated model evaluation. To illustrate usage and assess FedScore's performance, we built a hypothetical global scoring system for mortality prediction within 30 days after a visit to an emergency department using 10 simulated sites divided from a tertiary hospital in Singapore. We employed a pre-existing score generator to construct 10 local scoring systems independently at each site and we also developed a scoring system using centralized data for comparison. We compared the acquired FedScore model's performance with that of other scoring models using the receiver operating characteristic (ROC) analysis. The FedScore model achieved an average area under the curve (AUC) value of 0.763 across all sites, with a standard deviation (SD) of 0.020. We also calculated the average AUC values and SDs for each local model, and the FedScore model showed promising accuracy and stability with a high average AUC value which was closest to the one of the pooled model and SD which was lower than that of most local models. This study demonstrates that FedScore is a privacy-preserving scoring system generator with potentially good generalizability.

A computer software is being rolled out in the NHS from today that can predict A&E admissions weeks in advance based on things like Covid rates and 111 calls. The AI technology will be used in over 100 hospitals with major A&E departments in England, nearly half of all NHS trusts. It was found to be able to make forecasts with'impressive' accuracy in a trial at nine trusts by looking at factors including local Covid and flu infection rates, traffic and 111 call data to model how many people will show up at A&E each day. The software also takes into consideration public holidays such as New Year's Eve, when emergency departments are more likely to fill up. There are plans to incorporate weather data in the future, with the cold associated with more falls and traffic accidents and hot temperatures linked to a rise in heart problems.

NHS Vale of York CCG has rolled out predictive intervention technology to identify patients at risk of unplanned hospital care. Health Navigator uses analytics and machine learning techniques to identify patients who may benefit from health coaching, particularly those with long-term health conditions. Delivered by registered clinicians, the service is designed to support patients with complex conditions and empower them to take control of their health, thus reducing A&E admissions and unplanned emergency care. The project has been commissioned by NHS Vale of York CCG and aims to address the NHS's increasing demand for urgent and emergency care services, as highlighted in figures released by NHS Digital recently which showed that emergency admissions have peaked nationally. Evidence from a local randomised control trial (RCT) at York Teaching Hospital showed a 36% reduction in A&E attendances for patients supported by health coaching.

We used longitudinal data from linked electronic health records of 4.6 million patients aged 18–100 years from 389 practices across England between 1985 to 2015. The population was divided into a derivation cohort (80%, 3.75 million patients from 300 general practices) and a validation cohort (20%, 0.88 million patients from 89 general practices) from geographically distinct regions with different risk levels. We first replicated a previously reported Cox proportional hazards (CPH) model for prediction of the risk of the first emergency admission up to 24 months after baseline. This reference model was then compared with 2 machine learning models, random forest (RF) and gradient boosting classifier (GBC). The initial set of predictors for all models included 43 variables, including patient demographics, lifestyle factors, laboratory tests, currently prescribed medications, selected morbidities, and previous emergency admissions.