Brain Hematoma Marker Recognition Using Multitask Learning: SwinTransformer and Swin-Unet
This work addresses a domain-specific medical imaging problem for hematoma detection, with incremental improvements in handling covariate shifts.
The paper tackled brain hematoma marker recognition by proposing MTL-Swin-Unet, a multi-task learning method using transformers for classification and semantic segmentation, which outperformed other classifiers in F-value without covariate shift and in AUC with covariate shift.
This paper proposes a method MTL-Swin-Unet which is multi-task learning using transformers for classification and semantic segmentation. For spurious-correlation problems, this method allows us to enhance the image representation with two other image representations: representation obtained by semantic segmentation and representation obtained by image reconstruction. In our experiments, the proposed method outperformed in F-value measure than other classifiers when the test data included slices from the same patient (no covariate shift). Similarly, when the test data did not include slices from the same patient (covariate shift setting), the proposed method outperformed in AUC measure.