Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
To be, or not to be Melanoma: Convolutional neural networks in skin lesion classification
KTH, School of Technology and Health (STH), Medical Engineering.
2016 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Machine learning methods provide an opportunity to improve the classification of skin lesions and the early diagnosis of melanoma by providing decision support for general practitioners. So far most studies have been looking at the creation of features that best indicate melanoma. Representation learning methods such as neural networks have outperformed hand-crafted features in many areas. This work aims to evaluate the performance of convolutional neural networks in relation to earlier machine learning algorithms and expert diagnosis. In this work, convolutional neural networks were trained on datasets of dermoscopy images using weights initialized from a random distribution, a network trained on the ImageNet dataset and a network trained on Dermnet, a skin disease atlas.  The ensemble sum prediction of the networks achieved an accuracy of 89.3% with a sensitivity of 77.1% and a specificity of 93.0% when based on the weights learned from the ImageNet dataset and the Dermnet skin disease atlas and trained on non-polarized light dermoscopy images.  The results from the different networks trained on little or no prior data confirms the idea that certain features are transferable between different data. Similar classification accuracies to that of the highest scoring network are achieved by expert dermatologists and slightly higher results are achieved by referenced hand-crafted classifiers.  The trained networks are found to be comparable to practicing dermatologists and state-of-the-art machine learning methods in binary classification accuracy, benign – melanoma, with only little pre-processing and tuning. 

Place, publisher, year, edition, pages
2016. , 80 p.
Series
TRITA-STH, 2016:84
Keyword [en]
Clinical Decision Support, Convolutional Neural Networks, Deep Learning, Dermoscopy, Machine Learning, Melanoma
National Category
Medical Engineering
Identifiers
URN: urn:nbn:se:kth:diva-190000OAI: oai:DiVA.org:kth-190000DiVA: diva2:950147
External cooperation
Gnosco
Subject / course
Medical Engineering
Educational program
Master of Science in Engineering - Medical Engineering
Examiners
Available from: 2016-08-01 Created: 2016-07-27 Last updated: 2016-08-01Bibliographically approved

Open Access in DiVA

ANDREAS_NYLUND_MASTER_THESIS(4543 kB)4123 downloads
File information
File name FULLTEXT01.pdfFile size 4543 kBChecksum SHA-512
8857a57a9252c8947166c0a842145b8d3b1dcc2ba6594fd6d223172630cd75decfe5ce8152d5a77db3fb8daaac405be39b673c85afa81084c9d3ebddb7d54836
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Nylund, Andreas
By organisation
Medical Engineering
Medical Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 4123 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 410 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf