Deep learning based melanoma diagnosis using dermoscopic images

Conor Wall, Fraser Young, Li Zhang, Emma-Jane Phillips, Richard Jiang, Yonghong Yu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The most common malignancies in the world are skin cancers, with melanomas being the most lethal. The emergence of Convolutional Neural Networks (CNNs) has provided a highly compelling method for medical diagnosis. This research therefore conducts transfer learning with grid search based hyper-parameter fine-tuning using six state-of-the-art CNN models for the classification of benign nevus and malignant melanomas, with the models then being exported, implemented, and tested on a proof-of-concept Android application. Evaluated using Dermofit Image Library and PH2 skin lesion data sets, the empirical results indicate that the ResNeXt50 model achieves the highest accuracy rate with fast execution time, and a relatively small model size. It compares favourably with other related methods for melanoma diagnosis reported in the literature.
Original languageEnglish
Title of host publicationDevelopments of Artificial Intelligence Technologies in Computation and Robotics
Subtitle of host publicationProceedings of the 14th International FLINS Conference on Robotics and Artificial Intelligence (FLINS 2020)
EditorsZhong Li , Chunrong Yuan, Jie Lu, Etienne E. Kerre
Place of PublicationSingapore
PublisherWorld Scientific
Pages907-914
Number of pages8
Volume12
ISBN (Electronic)9789811223341, 9789811223334
ISBN (Print)9789811223327
DOIs
Publication statusPublished - Oct 2020
EventThe 14th International FLINS Conference on Robotics and Artificial Intelligence (FLINS 2020) - FernUniversitä t in Hagen/TH Köln, Cologne, Germany
Duration: 18 Aug 202021 Aug 2020
https://www.hrm-bildung.de/flins2020/

Publication series

NameWorld Scientific Proceedings Series on Computer Engineering and Information Science
Volume12
ISSN (Print)1793-7868

Conference

ConferenceThe 14th International FLINS Conference on Robotics and Artificial Intelligence (FLINS 2020)
Country/TerritoryGermany
CityCologne
Period18/08/2021/08/20
Internet address

Fingerprint

Dive into the research topics of 'Deep learning based melanoma diagnosis using dermoscopic images'. Together they form a unique fingerprint.

Cite this