A text-guided CLIP-style model with a frozen BERT text encoder and cross-entropy alignment classifies focal liver lesions from multi-phase CT slices with about 79 percent average accuracy, outperforming CLIP and MedCLIP on a small in-house dataset.
In: CVPR 2021, pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Vision-Language Model for Focal Liver Lesion Classification
A text-guided CLIP-style model with a frozen BERT text encoder and cross-entropy alignment classifies focal liver lesions from multi-phase CT slices with about 79 percent average accuracy, outperforming CLIP and MedCLIP on a small in-house dataset.