No abstract

Computational models of the heart are now being used to assess the effectiveness and feasibility of interventions through in-silico clinical trials (ISCTs). As the adoption and acceptance of ISCTs increases, best practices for reporting the methodology and analysing the results will emerge. 

Focusing in the area of cardiology, we aim to evaluate the types of ISCTs, their analysis methods and their reporting standards. To this end, we conducted a systematic review of cardiac ISCTs over the period of January 1st 2012 to January 1st 2022, following the preferred reporting items for systematic reviews and meta-analysis (PRISMA). We considered cardiac ISCTs of human patient cohorts, and excluded studies of single individuals and those in which models were used to guide a procedure without comparing against a control group. 


We identified 36 publications that described cardiac ISCTs, with most of the studies coming from the US and the UK. In 75% of the studies, a validation step was performed, although the specific type of validation varied between the studies. ANSYS FLUENT was the most commonly used software in 19% of ISCTs. The specific software used was not reported in 14% of the studies.

Unlike clinical trials, we found a lack of consistent reporting of patient demographics, with 28% of the studies not reporting them. Uncertainty quantification was limited, with sensitivity analysis performed in only 19% of the studies. In 97% of the ISCTs, no link was provided to provide easy access to the data or models used in the study. There was no consistent naming of study types with a wide range of studies that could potentially be considered ISCTs. 

There is a clear need for community agreement on minimal reporting standards on patient demographics, accepted standards for ISCT cohort quality control, uncertainty quantification, and increased model and data sharing.

Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): The study was funded by a University of Edinburgh Wellcome Trust iTPA award. The authors acknowledge the support of the British Heart Foundation Centre for Research Excellence Award III (RE/18/5/34216). SEW is supported by the British Heart Foundation (FS/20/26/34952). Background Artificial intelligence-enhanced electrocardiogram (AI-ECG) analysis offers the potential to identify patterns unrecognisable to human interpreters and broaden the ECG’s utility. However, current algorithms rely on waveform signals derived from digital ECGs for input data, and these cannot be readily obtained from paper-based ECGs. This potentially presents a barrier to adoption as numerous workplaces continue to use paper-based ECGs. The views of stakeholders on the current use of paper-based ECGs and the potential future application of AI-ECG analysis are unknown. Purpose To explore stakeholders’ views about current and future ECG use. To determine the perceived utility of AI-analysis of paper-based ECGs. Methods A web-based survey was designed using Qualtrics and distributed to a variety of healthcare professionals from numerous locations across the United Kingdom (UK). The survey consisted of 12 questions about participants’ perceptions relating to current and future paper-based ECG use and the perceived advantages and disadvantages of AI-ECG. Results In total, 43 healthcare professionals from 15 health provider organisations in the National Health Service (NHS) completed the survey. Paper-based ECGs were in use in 86% (37/43) of the respondents’ workplaces and 61% (26/43) felt that it would be useful if AI-based algorithms could analyse paper-based ECGs in addition to digital ECGs (Figure 1). Views on future prevalence of paper-based ECGs were split with 47% (20/43) responding that it is likely or extremely likely paper-based ECGs will still be in use in the next 5 years in the NHS. Perceived advantages of AI-based analysis included the potential to improve clinical decision making (51%, (22/43)) and optimisation of healthcare professionals’ work (leaving more time for clinical patient management) (47%, (20/43)) (Figure 2A). The inability to explain how algorithms determine results (56%, (24/43)), a lack of clarity over the accountability for the results (44%, (19/43)), and a reduction in learning opportunities (44%, (19/43)) were identified as potential issues associated with use of AI-ECG (Figure 2B). Conclusions Whilst AI-ECG offers potential to improve clinical care, there is currently a gap between research and the integration of AI-ECG into real-world practice. Paper-based ECGs remain prevalent within the NHS, and the current requirement for algorithms to receive signal data presents a barrier to current and future AI-ECG implementation. There is currently an unmet clinical need to develop algorithms capable of interpreting paper-based ECGs. AI-ECG analysis of paper-based ECGs could enable a wider range of healthcare professionals to capitalise on any benefits offered by AI-ECG.