Deep Learning for Prediction of Obstructive Disease From Fast Myocardial Perfusion SPECT: A Multicenter Study.

Publication/Presentation Date

11-1-2018

Abstract

OBJECTIVES: The study evaluated the automatic prediction of obstructive disease from myocardial perfusion imaging (MPI) by deep learning as compared with total perfusion deficit (TPD).

BACKGROUND: Deep convolutional neural networks trained with a large multicenter population may provide improved prediction of per-patient and per-vessel coronary artery disease from single-photon emission computed tomography MPI.

METHODS: A total of 1,638 patients (67% men) without known coronary artery disease, undergoing stress

RESULTS: A total of 1,018 (62%) patients and 1,797 of 4,914 (37%) arteries had obstructive disease. Area under the receiver-operating characteristic curve for disease prediction by deep learning was higher than for TPD (per patient: 0.80 vs. 0.78; per vessel: 0.76 vs. 0.73: p < 0.01). With deep learning threshold set to the same specificity as TPD, per-patient sensitivity improved from 79.8% (TPD) to 82.3% (deep learning) (p < 0.05), and per-vessel sensitivity improved from 64.4% (TPD) to 69.8% (deep learning) (p < 0.01).

CONCLUSIONS: Deep learning has the potential to improve automatic interpretation of MPI as compared with current clinical methods.

Volume

11

Issue

11

First Page

1654

Last Page

1663

ISSN

1876-7591

Disciplines

Medicine and Health Sciences

PubMedID

29550305

Department(s)

Department of Medicine

Document Type

Article

Share

COinS