Show simple item record

dc.contributor.authorSaarela, Mirka
dc.contributor.authorGeogieva, Lilia
dc.date.accessioned2022-09-28T11:31:08Z
dc.date.available2022-09-28T11:31:08Z
dc.date.issued2022
dc.identifier.citationSaarela, M., & Geogieva, L. (2022). Robustness, Stability, and Fidelity of Explanations for a Deep Skin Cancer Classification Model. <i>Applied Sciences</i>, <i>12</i>(19), Article 9545. <a href="https://doi.org/10.3390/app12199545" target="_blank">https://doi.org/10.3390/app12199545</a>
dc.identifier.otherCONVID_156773385
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/83335
dc.description.abstractSkin cancer is one of the most prevalent of all cancers. Because of its being widespread and externally observable, there is a potential that machine learning models integrated into artificial intelligence systems will allow self-screening and automatic analysis in the future. Especially, the recent success of various deep machine learning models shows promise that, in the future, patients could self-analyse their external signs of skin cancer by uploading pictures of these signs to an artificial intelligence system, which runs such a deep learning model and returns the classification results. However, both patients and dermatologists, who might use such a system to aid their work, need to know why the system has made a particular decision. Recently, several explanation techniques for the deep learning algorithm’s decision-making process have been introduced. This study compares two popular local explanation techniques (integrated gradients and local model-agnostic explanations) for image data on top of a well-performing (80% accuracy) deep learning algorithm trained on the HAM10000 dataset, a large public collection of dermatoscopic images. Our results show that both methods have full local fidelity. However, the integrated gradients explanations perform better with regard to quantitative evaluation metrics (stability and robustness), while the model-agnostic method seem to provide more intuitive explanations. We conclude that there is still a long way before such automatic systems can be used reliably in practice.en
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.publisherMDPI AG
dc.relation.ispartofseriesApplied Sciences
dc.rightsCC BY 4.0
dc.subject.otherexplainable artificial intelligence
dc.subject.otherinterpretable machine learning
dc.subject.otherskin cancer
dc.subject.otherconvolutional neural network
dc.subject.otherdeep learning
dc.subject.otherintegrated gradients
dc.subject.otherlocal model-agnostic explanations
dc.titleRobustness, Stability, and Fidelity of Explanations for a Deep Skin Cancer Classification Model
dc.typearticle
dc.identifier.urnURN:NBN:fi:jyu-202209284691
dc.contributor.laitosInformaatioteknologian tiedekuntafi
dc.contributor.laitosFaculty of Information Technologyen
dc.contributor.oppiaineTietotekniikkafi
dc.contributor.oppiaineResurssiviisausyhteisöfi
dc.contributor.oppiaineKoulutusteknologia ja kognitiotiedefi
dc.contributor.oppiaineHuman and Machine based Intelligence in Learningfi
dc.contributor.oppiaineMathematical Information Technologyen
dc.contributor.oppiaineSchool of Resource Wisdomen
dc.contributor.oppiaineLearning and Cognitive Sciencesen
dc.contributor.oppiaineHuman and Machine based Intelligence in Learningen
dc.type.urihttp://purl.org/eprint/type/JournalArticle
dc.type.coarhttp://purl.org/coar/resource_type/c_2df8fbb1
dc.description.reviewstatuspeerReviewed
dc.relation.issn2076-3417
dc.relation.numberinseries19
dc.relation.volume12
dc.type.versionpublishedVersion
dc.rights.copyright© 2022 by the authors. Licensee MDPI, Basel, Switzerland
dc.rights.accesslevelopenAccessfi
dc.subject.ysosyväoppiminen
dc.subject.ysokoneoppiminen
dc.subject.ysoneuroverkot
dc.subject.ysodiagnostiikka
dc.subject.ysoihosyöpä
dc.subject.ysopäätöksentukijärjestelmät
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p39324
jyx.subject.urihttp://www.yso.fi/onto/yso/p21846
jyx.subject.urihttp://www.yso.fi/onto/yso/p7292
jyx.subject.urihttp://www.yso.fi/onto/yso/p416
jyx.subject.urihttp://www.yso.fi/onto/yso/p13613
jyx.subject.urihttp://www.yso.fi/onto/yso/p27803
dc.rights.urlhttps://creativecommons.org/licenses/by/4.0/
dc.relation.datasethttps://challenge.isic-archive.com/data/#2018
dc.relation.doi10.3390/app12199545
jyx.fundinginformationThe authors appreciate the HPC-Europa3 research visit programme, which is funded by the European Commission H2020—Research and Innovation programme (under grant agreement number 730897).
dc.type.okmA1


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

CC BY 4.0
Except where otherwise noted, this item's license is described as CC BY 4.0