Brain Tumor Segmentation, 두뇌 ct 영상 내 Tumor segmentation을 진행합니다.
알고리즘은 3D Unet, 2D Unet을 적용하여 진행하였습니다.
Challenge Site : BraTs20
- Data Format : nii.gz
- Image
- T1 : native
- T1CE : post_contrast
- T2 : T2-weighted
- Flair : T2 Fluid Attenuated Inversion Recovery
- Labels
- Label 0 : Unlabeled Volumn
- Label 1 : Necrotic and non-enhancing tumor core (NCR / NET)
- Label 2 : Peritumoral Edema (ED)
- Label 3 : missing ( No pixels in all the volumnes contain label 3)
- Label 4 : GD-enhancing tumor (ET)
- Performance
3D UNET : Mean IoU = 0.8143
2D UNET : Dice_coeff = 1.20 (Multiclass label into Singleclass)