Parsing HL7 aECG files and segmenting leads for interactive progressive-based interpretation of the 12-lead electrocardiogram

摘要

The 12-lead Electrocardiogram (ECG) is ubiquitously used as a diagnostic support tool to detect cardiovascular disease. Nevertheless, it is often difficult to read and incorrectly interpreted due to the significant cognitive load forced upon the interpreter. However, ECG interpretation errors can be reduced through utilizing interactive touch screen devices which facilitate a systematic approach to aid ECG interpretation. In view of this, a set of interactive questions and prompts to guide an interpreter through a typical ECG reporting process had been developed and coined `Interactive Progressive based Interpretation' (IPI). To realise the potential of this system a pathway for potential interpreters must be created. Often, an ECG is stored in XML format. Therefore, to allow practitioners to unobtrusively use the IPI system it must be capable of consuming this file format, and process its data. To achieve this, we have partnered with AMPS-LLC to create a model which can consume an HL 7-XML file, and converts it into specific segmented image files in a desired format (PDF, PNG or JPG). Once these images are segmented, they can then be placed automatically into the IPI system sequence. In conclusion, a pathway for a decision support model has been created to aid ECG interpretation. We hypothesize this could facilitate a diagnostic aid in ECG interpretation.