Deep Learning Classification of Ischemic Stroke Territory on Diffusion-Weighted MRI: Added Value of Augmenting the Input with Image Transformations.
Authors
Affiliations (10)
Affiliations (10)
- Department of Radiology, Behçet Uz Children's Hospital, Izmir, Turkey. [email protected].
- Department of Biomedical Technologies, The Graduate School of Natural And Applied Sciences, Dokuz Eylül Universtiy, Izmir, Turkey. [email protected].
- Department of Biomedical Technologies, The Graduate School of Natural And Applied Sciences, Dokuz Eylül Universtiy, Izmir, Turkey.
- Department of Electrical and Electronics Engineering, Dokuz Eylül University, Adatepe Mahallesi Doğuş Caddesi, 35160, Buca, İzmir, Turkey.
- Izmir Health Technologies Development and Accelerator (BioIzmir), Dokuz Eylül University, Izmir, Turkey.
- Ataturk Training and Research Hospital, Department of Radiology, Izmir Katip Celebi University, Basin Sitesi, 35360, Izmir, Turkey.
- Department of Radiology, Behçet Uz Children's Hospital, Izmir, Turkey.
- Department of Radiology, Kahramanmaraş Sütçü İmam University Hospital, Kahramanmaraş, Turkey.
- Department of Radiology, Hatay Training and Research Hospital, Güzelburç, Hatay, Turkey.
- Department of Radiology, Dokuz Eylül University, 15 Temmuz Sağlık Sanat Yerleşkesi/İnciraltı, 35340, İzmir, Turkey.
Abstract
Our primary aim with this study was to build a patient-level classifier for stroke territory in DWI using AI to facilitate fast triage of stroke to a dedicated stroke center. A retrospective collection of DWI images of 271 and 122 consecutive acute ischemic stroke patients from two centers was carried out. Pretrained MobileNetV2 and EfficientNetB0 architectures were used to classify territorial subtypes as middle cerebral artery, posterior circulation, or watershed infarcts along with normal slices. Various input combinations using edge maps, thresholding, and hard attention versions were explored. The effect of augmenting the three-channel inputs of pre-trained models on classification performance was analyzed. ROC analyses and confusion matrix-derived performance metrics of the models were reported. Of the 271 patients included in this study, 151 (55.7%) were male and 120 (44.3%) were female. One hundred twenty-nine patients had MCA (47.6%), 65 patients had posterior circulation (24%), and 77 patients had watershed (28.0%) infarcts for center 1. Of the 122 patients from center 2, 78 (64%) were male and 44 (34%) were female. Fifty-two patients (43%) had MCA, 51 patients had posterior circulation (42%), and 19 (15%) patients had watershed infarcts. The Mobile-Crop model had the best performance with 0.95 accuracy and a 0.91 mean f1 score for slice-wise classification and 0.88 accuracy on external test sets, along with a 0.92 mean AUC. In conclusion, modified pre-trained models may be augmented with the transformation of images to provide a more accurate classification of affected territory by stroke in DWI.