Article Text

Download PDFPDF
Journal update monthly top five
  1. Joanna Sutton-Klein1,
  2. William James Doherty1,
  3. Anisa Jabeen Nasir Jafar2,
  4. Gregory Yates1,
  5. Richard Body1,3,
  6. Simon David Carley1,4,
  7. Gabrielle Prager5
  1. 1 Emergency Department, Manchester University NHS Foundation Trust, Manchester, UK
  2. 2 Emergency Department, Royal Manchester Children's Hospital, Manchester, UK
  3. 3 Division of Cardiovascular Sciences, The University of Manchester, Manchester, UK
  4. 4 Postgraduate Medicine, Manchester Metropolitan University, Manchester, UK
  5. 5 Johns Hopkins Bloomberg School of Public Health Center for Teaching and Learning, Baltimore, Maryland, USA
  1. Correspondence to Dr Joanna Sutton-Klein, The University of Manchester, Manchester, UK; joanna.sutton-klein{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

This month’s update is by the Manchester University NHS Foundation Trust team. We used a multimodal search strategy, drawing on free open-access medical education resources and literature searches. We identified the five most interesting and relevant papers (decided by consensus) and highlight the main findings, key limitations and clinical bottom line for each paper.

The papers are ranked as:

  • Worth a peek—interesting, but not yet ready for prime time.

  • Head turner—new concepts.

  • Game changer—this paper could/should change practice.

Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction by Al-Zaiti et al

Topic: myocardial infarction

Outcome rating: head turner

Occlusion myocardial infarction (OMI) is a term increasingly preferred to ST-elevation myocardial infarction to describe the clinical presentation of acute and total coronary artery blockage. Most patients with OMI will have ST segment elevation on their initial ECG. However, a significant minority (24%–35%) present with other, less well-recognised ECG abnormalities, which are easily missed.1

The authors developed a machine learning model to improve ECG diagnosis of OMI using angiographic OMI as the reference standard. They trained the model on a derivation set of prehospital ECGs without obvious ST-elevation taken from 4026 patients with chest pain (5.2% OMI). The model was then tested on 3287 patients from separate clinical sites. On external validation, the model had an area under the receiver operating characteristic curve (AUROC) of 0.87 (95% CI 0.85 to 0.90), which was higher than a commercial ECG system (AUROC 0.75 (95% CI 0.71 to 0.79), p < 0.001) and practising clinicians (p<0.001). However, the negative predictive value for ‘ruling in’ OMI was 0.99 (95% CI 0.98 to 0.99), though this was higher than clinicians (0.97).

The model was derived using patients who were ethnically diverse, had undifferentiated chest pain and had high rates of existing ischaemic heart disease. The study is limited by its reliance on a single prehospital ECG to train the model. The use of serial ECGs …

View Full Text


  • Twitter @jo_may_sk, @richardbody

  • Contributors JS-K led the team and was responsible for submission. JS-K, WD, GY, AJ and GP reviewed papers. RB and SC provided senior academic input.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.