REVIEW 1 cited by
A Multi-Level Deep Ensemble Model for Skin Lesion Classification in Dermoscopy Images
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
A Multi-Level Deep Ensemble Model for Skin Lesion Classification in Dermoscopy Images
read the original abstract
A multi-level deep ensemble (MLDE) model that can be trained in an 'end to end' manner is proposed for skin lesion classification in dermoscopy images. In this model, four pre-trained ResNet-50 networks are used to characterize the multiscale information of skin lesions and are combined by using an adaptive weighting scheme that can be learned during the error back propagation. The proposed MLDE model achieved an average AUC value of 86.5% on the ISIC-skin 2018 official validation dataset, which is substantially higher than the average AUC values achieved by each of four ResNet-50 networks.
Forward citations
Cited by 1 Pith paper
-
Privileged Lesion-Context Relational Distillation for Mask-Free Skin Lesion Classification
A mask-privileged teacher transfers relational lesion-context knowledge to an image-only student, improving skin-lesion classification on HAM10000 and ISIC 2018 without masks at inference.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.