CardiSort: a convolutional neural network for cross vendor automated sorting of cardiac MR images.

ABSTRACT:

Objectives

To develop an image-based automatic deep learning method to classify cardiac MR images by sequence type and imaging plane for improved clinical post-processing efficiency.

Methods

Multivendor cardiac MRI studies were retrospectively collected from 4 centres and 3 vendors. A two-head convolutional neural network ('CardiSort') was trained to classify 35 sequences by imaging sequence (n = 17) and plane (n = 10). Single vendor training (SVT) on single-centre images (n = 234 patients) and multivendor training (MVT) with multicentre images (n = 434 patients, 3 centres) were performed. Model accuracy and F1 scores on a hold-out test set were calculated, with ground truth labels by an expert radiologist. External validation of MVT (MVT_external) was performed on

SUBMITTER: Lim RP 

PROVIDER: S-EPMC9381634 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature