Abstract
Applying deep learning models to MRI scans of acute stroke patients to extract features that are indicative of short-term outcome could assist a clinician’s treatment decisions. Deep learning models are usually accurate but are not easily interpretable. Here, we trained a convolutional neural network on ADC maps from hyperacute ischaemic stroke patients for prediction of short-term functional outcome and used an interpretability technique to highlight regions in the ADC maps that were most important in the prediction of a bad outcome. Although highly accurate, the model’s predictions were not based on aspects of the ADC maps related to stroke pathophysiology.
Original language | English |
---|---|
DOIs | |
Publication status | Published - 2021 |
Event | Intl. Soc. Mag. Reson. Med - Duration: 1 Jan 2021 → … |
Conference
Conference | Intl. Soc. Mag. Reson. Med |
---|---|
Period | 1/01/21 → … |
Keywords
- deep learning
- MRI scans
- acute stroke
- short-term outcome
- clinician’s treatment decisions
- convolutional neural network
- ADC maps
- hyperacute ischaemic stroke
- functional outcome
- interpretability technique
- stroke pathophysiology