Dr Daniel Chalk

About me:

I am an applied health services researcher focused on using modelling, simulation and data science techniques to help provide evidence to solve problems in health and social care settings, and inform decision making. My areas of expertise include the use of Discrete Event Simulation to model patient pathways, System Dynamics to model whole systems, Agent Based Simulation to model emergent population dynamics from individual-level behaviours and machine learning methods to train machines to replicate decision making.

I have a particular research interest in the application of AI-based Natural Language Processing techniques to health and social care data to automate the extraction of information from free text data. This includes identifying important named information (Named Entity Recognition) and predicting the sentiment of free text data (such as patient surveys) and the particular issues raised to help organisations better understand what is and isn’t working well in the delivery of their services (Aspect-Level Sentiment Analysis).

My prior projects have led to significant impacts for patients and services across the region, including significant reductions in referral to treatment time for muscle-invasive bladder cancer patients in Cornwall, and close work with Kernow CCG to improve the provision of urgent and emergency care provision across Cornwall.

In addition to my research work, I lead the development and delivery of the PenARC Health Service Modelling Associates (HSMA) Programme – an innovative and impactful programme in which health and social care staff are released from their usual role for a day a week for a year to undertake a modelling or data science project of importance for their organisation. We provide training and ongoing mentoring support for HSMAs, and previous HSMA projects have led to £multi-million investments in mental health services and emergency department facilities, as well as the establishment of new in-house Operational Research teams in NHS organisations across the region.

I am a firm advocate of the use of Free and Open Source software in developing my models and solutions, to ensure that the work I do can be freely shared, developed upon and critically appraised. I have significant programming experience, primarily in the use of the Python programming language, which I use for most of my work.