Search

Since the emergence of COVID-19, deep learning models have been developed to identify COVID-19 from chest X-rays. With little to no direct access to hospital data, the AI community relies heavily on public data comprising numerous data sources. Model performance results have been exceptional when training and testing on open-source data, surpassing the reported capabilities of AI in pneumonia-detection prior to the COVID-19 outbreak. In this study impactful models are trained on a widely used open-source data and tested on an external test set and a hospital dataset, for the task of classifying chest X-rays into one of three classes: COVID-19, non-COVID pneumonia and no-pneumonia. Classification performance of the models investigated is evaluated through ROC curves, confusion matrices and standard classification metrics. Explainability modules are implemented to explore the image features most important to classification. Data analysis and model evalutions show that the popular open-source dataset COVIDx is not representative of the real clinical problem and that results from testing on this are inflated. Dependence on open-source data can leave models vulnerable to bias and confounding variables, requiring careful analysis to develop clinically useful/viable AI tools for COVID-19 detection in chest X-rays.

Most data collected by hospitals as a consequence of the delivery of routine care is not utilised for analytics or organisational intelligence. This project aims to develop tools to enhance the utilisation of routinely collected cancer data within hospitals across England. This was achieved by developing a web application using open source tools to provide health care professionals and hospital managers with easy to use, interactive analytics for cancer data. The application uses data items hospitals in England are mandated to collect as part of the Cancer Outcomes and Services Dataset (COSD), to provide clinical insight into survival outcomes, population distributions, service demands, waiting times, geographical case distributions and treatment information in real-time or near real-time. Development was guided by end user needs through the use of panels of clinical and non-clinical end users.

Most data collected by hospitals as a consequence of the delivery of routine care is not utilised for analytics or organisational intelligence. This project aims to develop tools to enhance the utilisation of routinely collected cancer data within hospitals across England. This was achieved by developing a web application using open source tools to provide health care professionals and hospital managers with easy to use, interactive analytics for cancer data. The application uses data items hospitals in England are mandated to collect as part of the Cancer Outcomes and Services Dataset (COSD), to provide clinical insight into survival outcomes, population distributions, service demands, waiting times, geographical case distributions and treatment information in real-time or near real-time. Development was guided by end user needs through the use of panels of clinical and non-clinical end users.

Since the emergence of COVID-19, deep learning models have been developed to identify COVID-19 from chest X-rays. With little to no direct access to hospital data, the AI community relies heavily on public data comprising numerous data sources. Model performance results have been exceptional when training and testing on open-source data, surpassing the reported capabilities of AI in pneumonia-detection prior to the COVID-19 outbreak. In this study impactful models are trained on a widely used open-source data and tested on an external test set and a hospital dataset, for the task of classifying chest X-rays into one of three classes: COVID-19, non-COVID pneumonia and no-pneumonia. Classification performance of the models investigated is evaluated through ROC curves, confusion matrices and standard classification metrics. Explainability modules are implemented to explore the image features most important to classification. Data analysis and model evalutions show that the popular open-source dataset COVIDx is not representative of the real clinical problem and that results from testing on this are inflated. Dependence on open-source data can leave models vulnerable to bias and confounding variables, requiring careful analysis to develop clinically useful/viable AI tools for COVID-19 detection in chest X-rays.

The main challenge in the pathway analysis of cancer treatments is the complexity of the process. Process mining is one of the approaches that can be used to visualize and analyze these complex pathways. In this study, our purpose was to use process mining to explore variations in the treatment pathways of endometrial cancer. We extracted patient data from a hospital information system, created the process model, and analyzed the variations of the 62-day pathway from a General Practitioner referral to the first treatment in the hospital. We also analyzed the variations based on three different criteria: the type of the first treatment, the age at diagnosis, and the year of diagnosis. This approach should be of interest to others dealing with complex medical and healthcare processes.

The main challenge in the pathway analysis of cancer treatments is the complexity of the process. Process mining is one of the approaches that can be used to visualize and analyze these complex pathways. In this study, our purpose was to use process mining to explore variations in the treatment pathways of endometrial cancer. We extracted patient data from a hospital information system, created the process model, and analyzed the variations of the 62-day pathway from a General Practitioner referral to the first treatment in the hospital. We also analyzed the variations based on three different criteria: the type of the first treatment, the age at diagnosis, and the year of diagnosis. This approach should be of interest to others dealing with complex medical and healthcare processes.

The area of process change over time is a particular concern in healthcare, where patterns of care emerge and evolve in response to individual patient needs. We propose a structured approach to analyse process change over time that is suitable for the complex domain of healthcare. Our approach applies a qualitative process comparison at three levels of abstraction: a holistic perspective (process model), a middle-level perspective (trace), and a fine-grained detail (activity). Our aim was to detect change points, localise and characterise the change, and unravel/understand the process evolution. We illustrate the approach using a case study of cancer pathways in Leeds where we found evidence of change points identified at multiple levels. In this paper, we extend our study by analysing the miners used in process discovery and providing a deeper analysis of the activity of investigation in trace and activity levels. In the experiment, we show that this qualitative approach provides a useful understanding of process change over time. Examining change at three levels provides confirmatory evidence of process change where perspectives agree, while contradictory evidence can lead to focused discussions with domain experts. This approach should be of interest to others dealing with processes that undergo complex change over time.

An understudied challenge within process mining is the area of process change over time. This is a particular concern in healthcare, where patterns of care emerge and evolve in response to individual patient needs and through complex interactions between people, process, technology and changing organisational structure. We propose a structured approach to analyse process change over time suitable for the complex domain of healthcare. Our approach applies a qualitative process comparison at three levels of abstraction: a holistic perspective summarizing patient pathways (process model level), a middle level perspective based on activity sequences for individuals (trace level), and a fine-grained detail focus on activities (activity level). Our aim is to identify points in time where a process changed (detection), to localise and characterise the change (localisation and characterisation), and to understand process evolution (unravelling). We illustrate the approach using a case study of cancer pathways in Leeds Cancer Centre where we found evidence of agreement in process change identified at the process model and activity levels, but not at the trace level. In the experiment we show that this qualitative approach provides a useful understanding of process change over time. Examining change at the three levels provides confirmatory evidence of process change where perspectives agree, while contradictory evidence can lead to focused discussions with domain experts. The approach should be of interest to others dealing with processes that undergo complex change over time.