Planning for healthcare services during the COVID-19 pandemic in the Southeast of England: a system dynamics modelling approach.

OBJECTIVES: To develop, test, validate and implement a system dynamics model to simulate the pandemic progress and the impact of various interventions on viral spread, healthcare utilisation and demand in secondary care. DESIGN: We adopted the system dynamics model incorporating susceptible, exposed, infection and recovery framework to simulate the progress of the pandemic and how the interventions for the COVID-19 response influence the outcomes with a focus on secondary care. SETTING: This study was carried out covering all the local health systems in Southeast of England with a catchment population of six million with a specific focus on Kent and Medway system. PARTICIPANTS: Six local health systems in Southeast of England using Kent and Medway as a case study. INTERVENTIONS: Short to medium 'what if' scenarios incorporating human behaviour, non-pharmaceutical interventions and medical interventions were tested using the model with regular and continuous feedback of the model results to the local health system leaders for monitoring, planning and rapid response as needed. MAIN OUTCOME MEASURES: Daily output from the model which included number infected in the population, hospital admissions needing COVID-19 care, occupied general beds, continuous positive airway pressure beds, intensive care beds, hospital discharge pathways and deaths. RESULTS: We successfully implemented a regional series of models based on the local population needs which were used in healthcare planning as part of the pandemic response. CONCLUSIONS: In this study, we have demonstrated the utility of system dynamics modelling incorporating local intelligence and collaborative working during the pandemic to respond rapidly and take decisions to protect the population. This led to strengthened cooperation among partners and ensured that the local population healthcare needs were met.

Are sustainable drainage systems (SuDS) effective at retaining dissolved trace elements?

Sustainable drainage systems (SuDS) are increasingly deployed to mitigate against increased trace element contaminant loads associated with urban and road runoff. However, there is a lack of research on their capabilities in removing these trace elements, particularly from the dissolved phase. Water samples were taken, following various rainfall events, from three different SuDS in Devon; one wetland pond adjacent to a busy dual carriageway, a new SuDS serving a housing estate and an established SuDS draining a mixed housing/light industrial area. A total of 15 elements were studied over the course of six rain events including the first flush of runoff. Removal rates varied within and between rain events as well as between types of SuDS. Although there was a general (modest) removal of dissolved elements within any given SuDS, this was not the case for all of the elements studied. Highest observed element concentrations entering the SuDS occurred at the onset of a rain event (first flush), the intensity of which, was related to the antecedent dry period. During high flows associated with intense rainfall, the SuDS could also act as a source of trace elements associated with fine particulates (e.g. lead) owing to resuspension of fine particulate material. Mature ponds with an abundance of macrophytes help retain solids and particulate metals, however poor maintenance leading to successional growth of shrubs and trees, reduces the efficiency of metal removal. This study highlighted the importance of long-term management planning to be included within any SuDs scheme.

Evaluation of the Use of Single- and Multi-Magnification Convolutional Neural Networks for the Determination and Quantitation of Lesions in Nonclinical Pathology Studies.

Digital pathology platforms with integrated artificial intelligence have the potential to increase the efficiency of the nonclinical pathologist's workflow through screening and prioritizing slides with lesions and highlighting areas with specific lesions for review. Herein, we describe the comparison of various single- and multi-magnification convolutional neural network (CNN) architectures to accelerate the detection of lesions in tissues. Different models were evaluated for defining performance characteristics and efficiency in accurately identifying lesions in 5 key rat organs (liver, kidney, heart, lung, and brain). Cohorts for liver and kidney were collected from TG-GATEs open-source repository, and heart, lung, and brain from internally selected R&D studies. Annotations were performed, and models were trained on each of the available lesion classes in the available organs. Various class-consolidation approaches were evaluated from generalized lesion detection to individual lesion detections. The relationship between the amount of annotated lesions and the precision/accuracy of model performance is elucidated. The utility of multi-magnification CNN implementations in specific tissue subtypes is also demonstrated. The use of these CNN-based models offers users the ability to apply generalized lesion detection to whole-slide images, with the potential to generate novel quantitative data that would not be possible with conventional image analysis techniques.

Using Deep Learning Artificial Intelligence Algorithms to Verify N-Nitroso-N-Methylurea and Urethane Positive Control Proliferative Changes in Tg-RasH2 Mouse Carcinogenicity Studies.

In Tg-rasH2 carcinogenicity mouse models, a positive control group is treated with a carcinogen such as urethane or N-nitroso-N-methylurea to test study validity based on the presence of the expected proliferative lesions in the transgenic mice. We hypothesized that artificial intelligence-based deep learning (DL) could provide decision support for the toxicologic pathologist by screening for the proliferative changes, verifying the expected pattern for the positive control groups. Whole slide images (WSIs) of the lungs, thymus, and stomach from positive control groups were used for supervised training of a convolutional neural network (CNN). A single pathologist annotated WSIs of normal and abnormal tissue regions for training the CNN-based supervised classifier using INHAND criteria. The algorithm was evaluated using a subset of tissue regions that were not used for training and then additional tissues were evaluated blindly by 2 independent pathologists. A binary output (proliferative classes present or not) from the pathologists was compared to that of the CNN classifier. The CNN model grouped proliferative lesion positive and negative animals at high concordance with the pathologists. This process simulated a workflow for review of these studies, whereby a DL algorithm could provide decision support for the pathologists in a nonclinical study.