Deep Learning-Based Automatic Classification of Ischemic Stroke Subtype Using Diffusion-Weighted Images
- Author:
Wi-Sun RYU
1
;
Dawid SCHELLINGERHOUT
;
Hoyoun LEE
;
Keon-Joo LEE
;
Chi Kyung KIM
;
Beom Joon KIM
;
Jong-Won CHUNG
;
Jae-Sung LIM
;
Joon-Tae KIM
;
Dae-Hyun KIM
;
Jae-Kwan CHA
;
Leonard SUNWOO
;
Dongmin KIM
;
Sang-Il SUH
;
Oh Young BANG
;
Hee-Joon BAE
;
Dong-Eog KIM
Author Information
- Publication Type:Original Article
- From:Journal of Stroke 2024;26(2):300-311
- CountryRepublic of Korea
- Language:English
-
Abstract:
Background:and Purpose Accurate classification of ischemic stroke subtype is important for effective secondary prevention of stroke. We used diffusion-weighted image (DWI) and atrial fibrillation (AF) data to train a deep learning algorithm to classify stroke subtype.
Methods:Model development was done in 2,988 patients with ischemic stroke from three centers by using U-net for infarct segmentation and EfficientNetV2 for subtype classification. Experienced neurologists (n=5) determined subtypes for external test datasets, while establishing a consensus for clinical trial datasets. Automatically segmented infarcts were fed into the model (DWI-only algorithm). Subsequently, another model was trained, with AF included as a categorical variable (DWI+AF algorithm). These models were tested: (1) internally against the opinion of the labeling experts, (2) against fresh external DWI data, and (3) against clinical trial dataset.
Results:In the training-and-validation datasets, the mean (±standard deviation) age was 68.0±12.5 (61.1% male). In internal testing, compared with the experts, the DWI-only and the DWI+AF algorithms respectively achieved moderate (65.3%) and near-strong (79.1%) agreement. In external testing, both algorithms again showed good agreements (59.3%–60.7% and 73.7%–74.0%, respectively). In the clinical trial dataset, compared with the expert consensus, percentage agreements and Cohen’s kappa were respectively 58.1% and 0.34 for the DWI-only vs. 72.9% and 0.57 for the DWI+AF algorithms. The corresponding values between experts were comparable (76.0% and 0.61) to the DWI+AF algorithm.
Conclusion:Our model trained on a large dataset of DWI (both with or without AF information) was able to classify ischemic stroke subtypes comparable to a consensus of stroke experts.
