Skip to main content
. 2023 Feb 11;14:788. doi: 10.1038/s41467-023-36102-1

Fig. 5. Adaptability to pocket-sized US images.

Fig. 5

a The deep learning model trained on high-quality US images is adapted to low-quality pocket-size US images. b ROCs comparison of blind evaluation results of the junior radiologist. c ROCs comparison of blind evaluation results of the senior radiologist in diagnosing breast cancer. d Examples of typical case results. For lesions with different tumor depth and different benign and malignant types, the model can capture the elastic information effectively. Source data are provided as a Source Data file.