Advanced Modelling and Simulation (HBI128)

Module Objective

Modelling and simulation is the process of creating and analysing a digital prototype of a physical or biological model to predict its performance in the real world. In clinical practice, simulation and modelling is used to understand and predict outcomes that may be unethical, inefficient, or impractical to undertake by experimentation.

By the end of this module the Clinical Scientist in HSST will be able to analyse, synthesise and apply their knowledge and understanding of current advanced modelling and simulation techniques used in clinical practice. They will be able to identify the need for new approaches by critically reviewing the evidence base. They will be expected to lead the development, implementation and review of new software and algorithms that underpin modelling and simulation. Fundamental to this module is the ability to work within established clinical guidelines and with full ethical considerations, especially with regard to the validation of the model/simulation.

The Clinical Scientist in HSST will also be expected to consistently demonstrate the attitudes and behaviours necessary for the role of a CCS.

By the end of this module the Clinical Scientist in HSST will be able to analyse, synthesise, evaluate and critically apply their expert knowledge of the purpose, principles, application and limitations of modelling and simulation in clinical practice. They should have a level of basic familiarity with a range of modelling techniques, for example:

  • Monte Carlo technique for simulating radiotherapy treatment, diagnostic image formation or radiation protection.
  • Simulation of radiofrequency emission and heat absorption produced by magnetic resonance (MR) scanners or mobile phones.
  • Computational fluid dynamics simulation of blood flow.
  • Finite element analysis of stress in biomechanics.
  • Solution of coupled sets of differential equations, for example in physiologic or pharmacokinetic compartmental models.
  • Agent-based models of disease propagation.
  • Training simulators for surgical procedures such as arthroscopy
  • Scenario-based simulation of consequences of discrete choices, for example tools used to teach management of emergency department arrivals.
  • Mathematical network models, e.g. for infection propagation and mitigation.
  • Cellular automata, e.g. for radiobiology modelling.

The Clinical Scientist in HSST would be expected to have covered a minimum of six of the above or equivalent modelling techniques to a level that would allow them to judge whether the technique would be appropriate for solving a particular problem.

Many issues are common to all modelling techniques and by in-depth study and use of one technique, the Clinical Scientist in HSST should develop a critical understanding of:

  • The advantages and disadvantages of modelling compared with experimental measurements.
  • How to select a modelling paradigm based on comparative analysis of the published literature.
  • The mathematics underlying the modelling technique.
  • How to specify the model using, for example, computer aided-design file of equipment geometry, segmentation of computed tomography (CT) scan anatomy, or experimental measurement of physiological parameters.
  • How the accuracy of results is limited by the accuracy of the input parameters, the fidelity of geometrical representation and the scale of model discretisation.
  • How to estimate the magnitude of errors through techniques such as sensitivity analysis.
  • How high sensitivity to initial conditions may mean that it is impossible to obtain meaningful answers to some types of question.
  • How to estimate the computational resources – such as CPU time, memory and disc storage – required to obtain a solution to the desired accuracy.
  • When interpreting model results:
    • how to produce appropriate visualisations of the raw data as an aid to understanding, for communication purposes and for hypothesis generation;
    • how to reduce the raw data to meaningful summary statistics of physical quantities of interest;
    • how to interpret the clinical significance of the results.
  • How to validate the model against, for example, experiments, other models, or mathematically tractable idealised cases.

By the end of this module the Clinical Scientist in HSST will be able to demonstrate a critical understanding of advanced modelling and simulation techniques in their area of clinical practice. They will apply their knowledge in their area of clinical practice, performing and mastering a range of technical and clinical skills while considering the impact on confidentiality of patient data, and will be able to:

  • Assess the demand for and specify a modelling and simulation solution to address a clinical need with users, clinical colleagues and other relevant stakeholders.
  • Critically analyse the research and evidence to support the development plan so that a new development that results in adaptation to practice can be made in a quality-assured, safe, timely and cost-effective manner.
  • Critically evaluate measures to identify, actively manage and mitigate risk to patients and staff when developing modelling and simulation solutions.
  • Develop a model/simulation of a complex process such as those listed in the knowledge section.
  • Import, adapt and utilise experimental and/or epidemiological data in designing and refining a model.
  • Justify the model and its usage.
  • Interpret the results produced by the model/simulation.
  • Report on how the outcome of the model/simulation may be applied to improve healthcare.
  • Comply with quality assurance and governance procedures to ensure software 
  • developments are introduced and critically evaluated to identify, actively manage and mitigate risk to patients.
  • Provide consultant-level clinical scientific advice with respect to the current and future uses of modelling and simulation, including interpretation of the clinical impact and contribution to the service, their implications for patient care and management, and recommendations for additional or more complex solutions.
  • Collaborate with colleagues across organisational boundaries, e.g. university and industry, to develop, promote and participate in a multiprofessional approach to the development of modelling and simulation solutions that underpin high-quality patient care and management, ensuring compliance with data-sharing agreements, policy and legislation.
  • Reflect on the challenges of applying research to practice in relation to these areas of practice and suggest improvements, building on a critique of available evidence.

Definitions: A model is a program that has been developed to copy the way a system works in real life. It uses mathematical formulas and calculations to predict what is likely to happen based on data recorded about what actually did happen in the past. Computer simulations use computer models to also predict how a system will behave given a set of conditions. Again, they are created through mathematical formulas. The difference between a model and a simulation is that a simulation often uses something physical to mimic the system (e.g. haptic feedback or virtual reality).