Visual Computing for Industry, Biomedicine, and Art

Table 7 Classification results reported by the proposed models trained on BUSI [21] dataset

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

Method	Accuracy (%)	Precision (%)	Recall (%)	F1-score (%)	Sensitivity (%)	Specificity (%)
ViT-B/16	86.00	86.11	86.02	85.93	86.45	85.26
ViT-B/32	83.61	83.87	83.60	83.56	86.66	78.55
ViT-L/32	85.06	84.40	84.53	84.40	86.87	82.10
MLP-Mixer-B/16	84.13	84.49	84.13	84.09	89.42	79.64
MLP-Mixer-L/16	85.46	85.73	85.46	85.47	87.52	82.10

Back to article page