Visual Computing for Industry, Biomedicine, and Art

Table 4 Results of performing ablation study of our proposed method trained on the BUSI [21] dataset and tested on the UDIAT [50] dataset

From: Multi-task approach based on combined CNN-transformer for efficient segmentation and classification of breast tumors in ultrasound images

Fold	Accuracy (%)	DC (%)	IoU (%)	Precision (%)	Recall (%)
Baseline	97.760 ± 0.002	81.440 ± 0.040	70.260 ± 0.010	89.460 ± 0.020	76.400 ± 0.030
Baseline + CAF	97.880 ± 0.000	81.520 ± 0.007	70.320 ± 0.010	90.320 ± 0.020	76.680 ± 0.020

Back to article page