Show simple item record

dc.contributor.authorByth, Lachlan A
dc.date.accessioned2021-06-02T23:29:15Z
dc.date.available2021-06-02T23:29:15Z
dc.date.issued2020
dc.identifier.issn0004-8380
dc.identifier.doi10.1111/ajd.13405
dc.identifier.urihttp://hdl.handle.net/10072/404883
dc.description.abstractDear Editor, Machine learning promises to revolutionise medical disciplines that rely on visual pattern recognition.1 Deep learning algorithms based on convolutional neural networks can detect melanoma with sensitivity and specificity similar to expert dermatologists.2, 3 Such approaches are limited by the fact that perfect discrimination between benign and malignant lesions may not be achievable using visual information alone. Many lesions, particularly dysplastic naevi, require biopsy despite a low probability of melanoma in an individual lesion.4 Deep learning can be repurposed to provide decision support, assisting clinicians to approximate the biopsy decision-making of an expert.
dc.languageEnglish
dc.publisherWiley
dc.relation.ispartofpagefromE110
dc.relation.ispartofpagetoE112
dc.relation.ispartofissue1
dc.relation.ispartofjournalAustralasian Journal of Dermatology
dc.relation.ispartofvolume62
dc.subject.fieldofresearchClinical sciences
dc.subject.fieldofresearchcode3202
dc.subject.keywordsScience & Technology
dc.subject.keywordsLife Sciences & Biomedicine
dc.subject.keywordsDermatology
dc.titleGuiding biopsies of pigmented lesions with deep learning (Letter)
dc.typeJournal article
dc.type.descriptionC2 - Articles (Other)
dcterms.bibliographicCitationByth, LA, Guiding biopsies of pigmented lesions with deep learning (Letter), Australasian Journal of Dermatology, 2020, 62 (1), pp. E110-E112
dcterms.dateAccepted2020-06-20
dc.date.updated2021-06-02T23:27:30Z
gro.hasfulltextNo Full Text
gro.griffith.authorByth, Lachlan A.


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

  • Journal articles
    Contains articles published by Griffith authors in scholarly journals.

Show simple item record