Artificial intelligence – image interpretation platforms

Author(s): Dr Neelam Dugar

Hospital: Doncaster and Bassetlaw Teaching Hospitals NHS Foundation Trust

Reference: RAD Magazine, 45, 527, 21-22


In 1995, I worked as a junior doctor on the wards in medicine. I often performed an ECG on patients with acute chest pain. Interestingly, even back then the ECG graph was accompanied by a machine-generated clinical interpretation or ECG report, eg left bundle branch block, myocardial infarction etc. These interpretations helped the inexperienced junior doctors, non-specialist doctors and nurses.

As I progressed in my experience as a junior doctor doing medicine, my reliance on the computer-generated reports reduced. I started to see its limitations and inaccuracies. It was well-recognised by the medical community that the computer-generated ECG report could not be fully relied on.

I do not think a cardiologist could use the computer-generated ECG report in their medico-legal defence, for final decision making and prescribing drugs. The human interpretation of the ECG has remained the gold standard. Human interpretation takes into account the clinical history and patient symptoms/signs before decisions on patient management are made. The interesting article ‘Can we trust our ECG machines?’ published in the British Journal of Cardiology highlights this.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read more