Source Codes

In order to facilitate the reproduction of our experiments, we are organizing the source codes in online repositories.

cvpr-skin-solo-ensemble

• This is the repository of the paper [Solo or ensemble? Choosing a CNN architecture for melanoma classification].
• The source codes can be download at: https://github.com/learningtitans/cvpr-skin-solo-ensemble

data-depth-design

• This is the repository of the paper [Data, Depth, and Design: Learning Reliable Models for Melanoma Screening].
• The source codes can be download at: https://github.com/learningtitans/data-depth-design

debiasing-skin

• This is the repository of the paper [Debiasing skin lesion datasets and models? Not so fast]
• The source codes can be download at: https://github.com/alceubissoto/debiasing-skin

deconstructing-bias-skin-lesion

• This is the repository of the paper [(De) Constructing Bias in Skin Lesion Datasets]
• The source codes can be download at: https://github.com/alceubissoto/deconstructing-bias-skin-lesion

gan-skin-lesion

• This is the repository of the paper [Skin lesion synthesis with generative adversarial networks]
• The source codes can be download at: https://github.com/alceubissoto/gan-skin-lesion6

isbi2017

• This is the repository of the paper [RECOD Titans at ISIC Challenge 2017].
  • The source codes can be download at: https://github.com/learningtitans/isbi2017-part1 (Part 1: Lesion Segmentation) and https://github.com/learningtitans/isbi2017-part3 (Part 3: Lesion Classification).

melanoma-transfer

• This is the repository of the paper [Knowledge Transfer for Melanoma Screening with Deep Learning].
  • The source code can be download at https://github.com/learningtitans/melanoma-transfer

melanoma-screening

• This is the repository of the paper [Towards Automated Melanoma Screening: Proper Computer Vision & Reliable Results].
• The source code can be download at https://github.com/learningtitans/melanoma-screening

skin-lesion-segmentation-agreement

• This is the repository of the papers [Handling Inter-Annotator Agreement for Automated Skin Lesion Segmentation] and [Less is More: Sample Selection and Label Conditioning Improve Skin Lesion Segmentation].
• The source codes can be download at: https://github.com/vribeiro1/skin-lesion-segmentation-agreement