Journal
BMC MEDICAL IMAGING
Volume 22, Issue 1, Pages -Publisher
BMC
DOI: 10.1186/s12880-022-00763-z
Keywords
Artificial intelligence; Emergency radiology; Pulmonary embolism; Deep learning; Automated detection
Funding
- Turku University Hospital
- Paulo Foundation
Ask authors/readers for more resources
This study aimed to develop and evaluate a deep neural network model for automated detection of pulmonary embolism using weakly labelled training data. Two versions of the model, one pre-trained with chest X-rays and the other with natural images, performed well on both stack- and slice-based levels. The model pre-trained with chest X-rays showed slightly better specificity and sensitivity compared to the model pre-trained with natural images.
Background The aim of this study was to develop and evaluate a deep neural network model in the automated detection of pulmonary embolism (PE) from computed tomography pulmonary angiograms (CTPAs) using only weakly labelled training data. Methods We developed a deep neural network model consisting of two parts: a convolutional neural network architecture called InceptionResNet V2 and a long-short term memory network to process whole CTPA stacks as sequences of slices. Two versions of the model were created using either chest X-rays (Model A) or natural images (Model B) as pre-training data. We retrospectively collected 600 CTPAs to use in training and validation and 200 CTPAs to use in testing. CTPAs were annotated only with binary labels on both stack- and slice-based levels. Performance of the models was evaluated with ROC and precision-recall curves, specificity, sensitivity, accuracy, as well as positive and negative predictive values. Results Both models performed well on both stack- and slice-based levels. On the stack-based level, Model A reached specificity and sensitivity of 93.5% and 86.6%, respectively, outperforming Model B slightly (specificity 90.7% and sensitivity 83.5%). However, the difference between their ROC AUC scores was not statistically significant (0.94 vs 0.91, p = 0.07). Conclusions We show that a deep learning model trained with a relatively small, weakly annotated dataset can achieve excellent performance results in detecting PE from CTPAs.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available