Malignancy Prediction and Lesion Identification from Clinical Dermatological Images.

Xia, Meng

Kheterpal, Meenal K

Wong, Samantha C

Park, Christine

Ratliff, William

Carin, Lawrence

Henao, Ricardo

2021-05-01T16:22:34Z

2021-05-01T16:22:34Z

2021

2021-05-01T16:22:32Z

We consider machine-learning-based malignancy prediction and lesion identification from clinical dermatological images, which can be indistinctly acquired via smartphone or dermoscopy capture. Additionally, we do not assume that images contain single lesions, thus the framework supports both focal or wide-field images. Specifically, we propose a two-stage approach in which we first identify all lesions present in the image regardless of sub-type or likelihood of malignancy, then it estimates their likelihood of malignancy, and through aggregation, it also generates an image-level likelihood of malignancy that can be used for high-level screening processes. Further, we consider augmenting the proposed approach with clinical covariates (from electronic health records) and publicly available data (the ISIC dataset). Comprehensive experiments validated on an independent test dataset demonstrate that i) the proposed approach outperforms alternative model architectures; ii) the model based on images outperforms a pure clinical model by a large margin, and the combination of images and clinical data does not significantly improves over the image-only model; and iii) the proposed framework offers comparable performance in terms of malignancy classification relative to three board certified dermatologists with different levels of experience.

https://hdl.handle.net/10161/22722

CoRR

cs.CV

cs.CV

cs.LG

Malignancy Prediction and Lesion Identification from Clinical Dermatological Images.

Journal article

Kheterpal, Meenal K|0000-0002-0460-6400

Henao, Ricardo|0000-0003-4980-845X

School of Medicine

Duke Cancer Institute

Dermatology

Duke

Institutes and Centers

Clinical Science Departments

abs/2104.02652