The effect of confounding data features on a deep learning algorithm to predict complete coronary occlusion in a retrospective observational setting.

Autor:	Brisk R; Cardiovascular Research Unit, Craigavon Hospital, 68 Lurgan Road, Portadown BT63 5QQ, UK.; School of Computer Science, Ulster University, Shore Road, Jordanstown BT37 0QB, UK., Bond R; School of Computer Science, Ulster University, Shore Road, Jordanstown BT37 0QB, UK., Finlay D; Nanotechnology and Integrated Bioengineering Centre, Ulster University, Jordanstown, UK., McLaughlin J; Nanotechnology and Integrated Bioengineering Centre, Ulster University, Jordanstown, UK., Piadlo A; Cardiovascular Research Unit, Craigavon Hospital, 68 Lurgan Road, Portadown BT63 5QQ, UK., Leslie SJ; Cardiac Unit, Raigmore Hospital, Inverness IV32 3UJ, UK.; Division of Biomedical Sciences, University of the Highlands and Islands Institute of Health Research and Innovation, Old Perth Road, IV2 3JH, Inverness, UK., Gossman DE; Tufts University School of Medicine, 145 Harrison Avenue, Boston, MA 02111, USA.; Department of Cardiology, St Elizabeth Medical Centre, 736 Cambridge Street, Boston, MA 02135, USA., Menown IB; Cardiovascular Research Unit, Craigavon Hospital, 68 Lurgan Road, Portadown BT63 5QQ, UK.; Queens University, School of Medicine, Dentistry and Biomedical Sciences, University Road, Belfast, BT7 1NN, UK., McEneaney DJ; Cardiovascular Research Unit, Craigavon Hospital, 68 Lurgan Road, Portadown BT63 5QQ, UK.; Centre for Advanced Cardiovascular Research, Ulster University, Jordanstown, UK., Warren S; Cardiology Division, Department of Medicine, Anne Arundel Medical Center, Annapolis, MD, USA.
Jazyk:	angličtina
Zdroj:	European heart journal. Digital health [Eur Heart J Digit Health] 2021 Feb 20; Vol. 2 (1), pp. 127-134. Date of Electronic Publication: 2021 Feb 20 (Print Publication: 2021).
DOI:	10.1093/ehjdh/ztab002
Abstrakt:	Aims: Deep learning (DL) has emerged in recent years as an effective technique in automated ECG analysis. Methods and Results: A retrospective, observational study was designed to assess the feasibility of detecting induced coronary artery occlusion in human subjects earlier than experienced cardiologists using a DL algorithm. A deep convolutional neural network was trained using data from the STAFF III database. The task was to classify ECG samples as showing acute coronary artery occlusion, or no occlusion. Occluded samples were recorded after 60 s of balloon occlusion of a single coronary artery. For the first iteration of the experiment, non-occluded samples were taken from ECGs recorded in a restroom prior to entering theatres. For the second iteration of the experiment, non-occluded samples were taken in the theatre prior to balloon inflation. Results were obtained using a cross-validation approach. In the first iteration of the experiment, the DL model achieved an F1 score of 0.814, which was higher than any of three reviewing cardiologists or STEMI criteria. In the second iteration of the experiment, the DL model achieved an F1 score of 0.533, which is akin to the performance of a random chance classifier. Conclusion: The dataset was too small for the second model to achieve meaningful performance, despite the use of transfer learning. However, 'data leakage' during the first iteration of the experiment led to falsely high results. This study highlights the risk of DL models leveraging data leaks to produce spurious results. (© The Author(s) 2021. Published by Oxford University Press on behalf of the European Society of Cardiology.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu