Interactive computersied approach to guide human interpretation of the 12-lead electrocardiogram

Background: Cardiovascular disease (CVD) kills 17 million people per year. The 12-lead Electrocardiogram (ECG) is an important diagnostic support tool to detect cardiac disease. Up to 33% of ECGs are incorrectly interpreted and clinicians often impulsively provide a diagnosis based on their first impression. It is therefore imperative to optimise how physicians interpret the 12-lead ECG. The standard presentation for the 12-lead ECG was developed over 70 years ago and currently presents the reader with all 12 lead signals on paper. This presentation format delivers a significant cognitive load. We hypothesise that ECG interpretation errors will be reduced using an algorithmic web-based user interface to facilitate a systematic approach to ECG interpretation. Methods: Following a literature review and interviews with expert clinicians and teaching professionals, a set of interactive questions and prompts were developed to guide a reader through a series of sub-tasks when interpreting an ECG. This has been named ‘Interactive Progressive based Interpretation’ (IPI). This system runs on touch screen devices to recreate an algorithmic approach to aid ECG interpretation. The IPI system was implemented using emerging web technologies, i.e. HyperText Markup Language (HTML5), JavaScript and Cascading Style Sheets (CSS3). Using this model, the 12-lead ECG is segmented into five parts and presented over five web pages. Page one presents the rhythm strip, page two presents lead II, page three presents all limb leads, page four presents the precordial leads and the last page presents the full 12-lead ECG. Each page displays a prompt and a series of questions that require input before the reader can ‘progress’ to the next page.Results: When comparing preliminary results with meta-analysis from studies in ECG interpretation competency [1], using Chi-square test (p-value <0.01), the IPI system has improved interpretation accuracy from 40% to 57%. Conclusion: A new algorithmic model has been proposed to aid ECG interpretation where interpreters systematically and sequentially interpret the 12-lead ECG as a series of sub-tasks.