A Novel Focal Tversky loss function with improved Attention U-Net for lesion segmentation

18 Oct 2018 Nabila Abraham Naimul Mefraz Khan

We propose a generalized focal loss function based on the Tversky index to address the issue of data imbalance in medical image segmentation. Compared to the commonly used Dice loss, our loss function achieves a better trade off between precision and recall when training on small structures such as lesions... (read more)

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Lesion Segmentation BUS 2017 Dataset B Attn U-Net + Multi-Input + FTL Dice Score 0.804 # 1
Lesion Segmentation BUS 2017 Dataset B U-Net + FTL Dice Score 0.669 # 3
Lesion Segmentation BUS 2017 Dataset B Attn U-Net + DL Dice Score 0.615 # 4
Lesion Segmentation ISIC 2018 U-Net + FTL Dice Score 0.829 # 5
Lesion Segmentation ISIC 2018 Attn U-Net + Multi-Input + FTL Dice Score 0.856 # 3
Lesion Segmentation ISIC 2018 Attn U-Net + DL Dice Score 0.806 # 6

Methods used in the Paper


METHOD TYPE
Concatenated Skip Connection
Skip Connections
ReLU
Activation Functions
Max Pooling
Pooling Operations
Convolution
Convolutions
U-Net
Semantic Segmentation Models
Focal Loss
Loss Functions