{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T04:38:52Z","timestamp":1776141532211,"version":"3.50.1"},"reference-count":43,"publisher":"Springer Science and Business Media LLC","issue":"7","license":[{"start":{"date-parts":[[2023,3,27]],"date-time":"2023-03-27T00:00:00Z","timestamp":1679875200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,3,27]],"date-time":"2023-03-27T00:00:00Z","timestamp":1679875200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000774","name":"The University of Newcastle","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100000774","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Process Lett"],"published-print":{"date-parts":[[2023,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Skin cancer can be fatal if it is found to be malignant. Modern diagnosis of skin cancer heavily relies on visual inspection through clinical screening, dermoscopy, or histopathological examinations. However, due to similarity among cancer types, it is usually challenging to identify the type of skin cancer, especially at its early stages. Deep learning techniques have been developed over the last few years and have achieved success in helping to improve the accuracy of diagnosis and classification. However, the latest deep learning algorithms still do not provide ideal classification accuracy. To further improve the performance of classification accuracy, this paper presents a novel method of classifying skin cancer in clinical skin images. The method consists of four blocks. First, class rebalancing is applied to the images of seven skin cancer types for better classification performance. Second, an image is preprocessed by being split into patches of the same size and then flattened into a series of tokens. Third, a transformer encoder is used to process the flattened patches. The transformer encoder consists of <jats:italic>N<\/jats:italic> identical layers with each layer containing two sublayers. Sublayer one is a multihead self-attention unit, and sublayer two is a fully connected feed-forward network unit. For each of the two sublayers, a normalization operation is applied to its input, and a residual connection of its input and its output is calculated. Finally, a classification block is implemented after the transformer encoder. The block consists of a flattened layer and a dense layer with batch normalization. Transfer learning is implemented to build the whole network, where the ImageNet dataset is used to pretrain the network and the HAM10000 dataset is used to fine-tune the network. Experiments have shown that the method has achieved a classification accuracy of 94.1%, outperforming the current state-of-the-art model IRv2 with soft attention on the same training and testing datasets. On the Edinburgh DERMOFIT dataset also, the method has better performance compared with baseline models.<\/jats:p>","DOI":"10.1007\/s11063-023-11204-5","type":"journal-article","created":{"date-parts":[[2023,3,27]],"date-time":"2023-03-27T18:02:47Z","timestamp":1679940167000},"page":"9335-9351","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":93,"title":["A Novel Vision Transformer Model for Skin Cancer Classification"],"prefix":"10.1007","volume":"55","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6747-6121","authenticated-orcid":false,"given":"Guang","family":"Yang","sequence":"first","affiliation":[]},{"given":"Suhuai","family":"Luo","sequence":"additional","affiliation":[]},{"given":"Peter","family":"Greer","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,3,27]]},"reference":[{"issue":"1","key":"11204_CR1","doi-asserted-by":"publisher","first-page":"10","DOI":"10.3322\/caac.20138","volume":"62","author":"RL Siegel","year":"2012","unstructured":"Siegel RL, Naishadham D, Jemal A (2012) Cancer statistics. CA 62(1):10\u201329. https:\/\/doi.org\/10.3322\/caac.20138","journal-title":"CA"},{"key":"11204_CR2","unstructured":"Australian Bureau of Statistics (2019) Causes of Death, Australia [Internet]. ABS, Canberra. Accessed 2022 Nov 1. https:\/\/www.abs.gov.au\/statistics\/health\/causes-death\/causes-death-australia\/2019."},{"key":"11204_CR3","unstructured":"Street W (2019) Cancer Facts & Figures. American Cancer Society, Atlanta, GA. http:\/\/cancerstatisticscenter.cancer.org. Accessed 2022 Nov 1."},{"issue":"6","key":"11204_CR4","first-page":"394","volume":"68","author":"F Bray","year":"2018","unstructured":"Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA 68(6):394\u2013424","journal-title":"CA"},{"issue":"1","key":"11204_CR5","first-page":"7","volume":"69","author":"RL Siegel","year":"2019","unstructured":"Siegel RL, Miller KD, Jemal A (2019) Cancer statistics. CA 69(1):7\u20133","journal-title":"CA"},{"issue":"3","key":"11204_CR6","first-page":"669","volume":"159","author":"ME Vestergaard","year":"2008","unstructured":"Vestergaard ME, Macaskill PH, Holt PE, Menzies SW (2008) Dermoscopy compared with naked eye examination for the diagnosis of primary melanoma: a meta-analysis of studies performed in a clinical setting. Br J Dermatol 159(3):669\u2013676","journal-title":"Br J Dermatol"},{"issue":"11","key":"11204_CR7","doi-asserted-by":"publisher","first-page":"1388","DOI":"10.1001\/archderm.141.11.1388","volume":"141","author":"SW Menzies","year":"2005","unstructured":"Menzies SW, Bischof L, Talbot H, Gutenev A, Avramidis M, Wong L, Lo SK, Mackellar G, Skladnev V, McCarthy W, Kelly J (2005) The performance of SolarScan: an automated dermoscopy image analysis instrument for the diagnosis of primary melanoma. Arch Dermatol 141(11):1388\u20131396","journal-title":"Arch Dermatol"},{"issue":"7639","key":"11204_CR8","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","volume":"542","author":"A Esteva","year":"2017","unstructured":"Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, Thrun S (2017) Dermatologist-level classification of skin cancer with deep neural networks. Nature 542(7639):115\u2013118","journal-title":"Nature"},{"key":"11204_CR9","doi-asserted-by":"crossref","unstructured":"Adeyinka AA, Viriri S (2018) Skin lesion images segmentation: a survey of the state-of-the-art. In: International conference on mining intelligence and knowledge exploration. Springer, Cham, pp. 321\u2013330","DOI":"10.1007\/978-3-030-05918-7_29"},{"key":"11204_CR10","first-page":"13","volume-title":"Soft Attention Improves Skin Cancer Classification Performance. InInterpretability of Machine Intelligence in Medical Image Computing, and Topological Data Analysis and Its Applications for Medical Data","author":"SK Datta","year":"2021","unstructured":"Datta SK, Shaikh MA, Srihari SN (2021) Soft Attention Improves Skin Cancer Classification Performance. InInterpretability of Machine Intelligence in Medical Image Computing, and Topological Data Analysis and Its Applications for Medical Data. Springer, Cham, pp 13\u201323"},{"key":"11204_CR11","unstructured":"Nadipineni H (2020) Method to classify skin lesions using dermoscopic images. arXiv preprint arXiv:2008.09418. 2020 Aug 21."},{"key":"11204_CR12","unstructured":"Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, Dehghani M, Minderer M, Heigold G, Gelly S, Uszkoreit J (2020) An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929. 2020 Oct 22"},{"issue":"1","key":"11204_CR13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/sdata.2018.161","volume":"5","author":"P Tschandl","year":"2018","unstructured":"Tschandl P, Rosendahl C, Kittler H (2018) The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci Data 5(1):1\u20139","journal-title":"Sci Data"},{"key":"11204_CR14","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser \u0141, Polosukhin I (2017) Attention is all you need. In: Advances in neural information processing systems. 2017, 30"},{"key":"11204_CR15","unstructured":"Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. 2018 Oct 11."},{"issue":"8","key":"11204_CR16","first-page":"9","volume":"1","author":"A Radford","year":"2019","unstructured":"Radford A, Wu J, Child R, Luan D, Amodei D, Sutskever I (2019) Language models are unsupervised multitask learners. OpenAI blog 1(8):9","journal-title":"OpenAI blog"},{"issue":"7","key":"11204_CR17","doi-asserted-by":"publisher","DOI":"10.1016\/j.mex.2020.100864","volume":"1","author":"N Gessert","year":"2020","unstructured":"Gessert N, Nielsen M, Shaikh M, Werner R, Schlaefer A (2020) Skin lesion classification using ensembles of multi-resolution EfficientNets with meta data. MethodsX 1(7):100864","journal-title":"MethodsX"},{"issue":"11","key":"11204_CR18","doi-asserted-by":"publisher","first-page":"3429","DOI":"10.1109\/TMI.2020.2995518","volume":"39","author":"Q Liu","year":"2020","unstructured":"Liu Q, Yu L, Luo L, Dou Q, Heng PA (2020) Semi-supervised medical image classification with relation-driven self-ensembling model. IEEE Trans Med Imaging 39(11):3429\u20133440","journal-title":"IEEE Trans Med Imaging"},{"key":"11204_CR19","doi-asserted-by":"crossref","unstructured":"Sun C, Shrivastava A, Singh S, Gupta A (2017) Revisiting unreasonable effectiveness of data in deep learning era. In: Proceedings of the IEEE international conference on computer vision 2017, pp 843\u2013852","DOI":"10.1109\/ICCV.2017.97"},{"key":"11204_CR20","doi-asserted-by":"crossref","unstructured":"Zhai X, Kolesnikov A, Houlsby N, Beyer L (2022) Scaling vision transformers. In: Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition 2022, pp 12104\u201312113","DOI":"10.1109\/CVPR52688.2022.01179"},{"key":"11204_CR21","unstructured":"Ba JL, Kiros JR, Hinton GE. Layer normalization. arXiv preprint arXiv:1607.06450. 2016 Jul 21."},{"key":"11204_CR22","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition 2016, pp 770\u2013778","DOI":"10.1109\/CVPR.2016.90"},{"issue":"11","key":"11204_CR23","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2019.14645","volume":"2","author":"N Tomita","year":"2019","unstructured":"Tomita N, Abdollahi B, Wei J, Ren B, Suriawinata A, Hassanpour S (2019) Attention-based deep neural networks for detection of cancerous and precancerous esophagus tissue on histopathological slides. JAMA Netw Open 2(11):e1914645","journal-title":"JAMA Netw Open"},{"key":"11204_CR24","unstructured":"Hendrycks D, Gimpel K. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415. 2016 Jun 27."},{"key":"11204_CR25","unstructured":"Melas-Kyriazi L. Do you even need attention? a stack of feed-forward layers does surprisingly well on imagenet. arXiv preprint arXiv:2105.02723. 2021 May 6."},{"issue":"34","key":"11204_CR26","first-page":"24261","volume":"6","author":"IO Tolstikhin","year":"2021","unstructured":"Tolstikhin IO, Houlsby N, Kolesnikov A, Beyer L, Zhai X, Unterthiner T, Yung J, Steiner A, Keysers D, Uszkoreit J, Lucic M (2021) Mlp-mixer: an all-mlp architecture for vision. Adv Neural Inf Process Syst 6(34):24261\u201324272","journal-title":"Adv Neural Inf Process Syst"},{"key":"11204_CR27","doi-asserted-by":"crossref","unstructured":"Touvron H, Bojanowski P, Caron M, Cord M, El-Nouby A, Grave E, Izacard G, Joulin A, Synnaeve G, Verbeek J, J\u00e9gou H (2022) Resmlp: feedforward networks for image classification with data-efficient training. IEEE Trans Pattern Anal Mach Intell","DOI":"10.1109\/TPAMI.2022.3206148"},{"key":"11204_CR28","doi-asserted-by":"crossref","unstructured":"Ballerini L, Fisher RB, Aldridge B, Rees J. A color and texture based hierarchical K-NN approach to the classification of non-melanoma skin lesions. In: Color medical image analysis 2013. Springer, Dordrecht, pp 63\u201386","DOI":"10.1007\/978-94-007-5389-1_4"},{"key":"11204_CR29","doi-asserted-by":"crossref","unstructured":"Fisher RB, Rees J, Bertrand A. Classification of ten skin lesion classes: Hierarchical knn versus deep net. In: Annual conference on medical image understanding and analysis 2019 Jul 24. Springer, Cham, pp 86\u201398","DOI":"10.1007\/978-3-030-39343-4_8"},{"issue":"149","key":"11204_CR30","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.105939","volume":"1","author":"C Xin","year":"2022","unstructured":"Xin C, Liu Z, Zhao K, Miao L, Ma Y, Zhu X, Zhou Q, Wang S, Li L, Yang F, Xu S (2022) An improved transformer network for skin cancer classification. Comput Biol Med 1(149):105939","journal-title":"Comput Biol Med"},{"issue":"77","key":"11204_CR31","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102357","volume":"1","author":"X He","year":"2022","unstructured":"He X, Tan EL, Bi H, Zhang X, Zhao S, Lei B (2022) Fully transformer network for skin lesion analysis. Med Image Anal 1(77):102357","journal-title":"Med Image Anal"},{"issue":"78","key":"11204_CR32","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.103997","volume":"1","author":"K Nakai","year":"2022","unstructured":"Nakai K, Chen YW, Han XH (2022) Enhanced deep bottleneck transformer model for skin lesion classification. Biomed Signal Process Control 1(78):103997","journal-title":"Biomed Signal Process Control"},{"issue":"1","key":"11204_CR33","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1109\/TLA.2016.7430097","volume":"14","author":"FE Alencar","year":"2016","unstructured":"Alencar FE, Lopes DC, Neto FM (2016) Development of a system classification of images dermoscopic for mobile devices. IEEE Latin Am Trans 14(1):325\u2013330","journal-title":"IEEE Latin Am Trans"},{"issue":"6","key":"11204_CR34","doi-asserted-by":"publisher","first-page":"448","DOI":"10.1049\/iet-ipr.2015.0385","volume":"10","author":"R Kasmi","year":"2016","unstructured":"Kasmi R, Mokrani K (2016) Classification of malignant melanoma and benign skin lesions: implementation of automatic ABCD rule. IET Image Proc 10(6):448\u2013455","journal-title":"IET Image Proc"},{"issue":"4","key":"11204_CR35","doi-asserted-by":"publisher","first-page":"1036","DOI":"10.1109\/TMI.2015.2506270","volume":"35","author":"A S\u00e1ez","year":"2015","unstructured":"S\u00e1ez A, S\u00e1nchez-Monedero J, Guti\u00e9rrez PA, Herv\u00e1s-Mart\u00ednez C (2015) Machine learning methods for binary and multiclass classification of melanoma thickness from dermoscopic images. IEEE Trans Med Imaging 35(4):1036\u20131045","journal-title":"IEEE Trans Med Imaging"},{"issue":"2","key":"11204_CR36","doi-asserted-by":"publisher","first-page":"615","DOI":"10.1109\/JBHI.2015.2390032","volume":"20","author":"Z Ma","year":"2015","unstructured":"Ma Z, Tavares JM (2015) A novel approach to segment skin lesions in dermoscopic images based on a deformable model. IEEE J Biomed Health Inform 20(2):615\u2013623","journal-title":"IEEE J Biomed Health Inform"},{"issue":"51","key":"11204_CR37","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/j.bspc.2019.02.013","volume":"1","author":"S Pathan","year":"2019","unstructured":"Pathan S, Prabhu KG, Siddalingaswamy PC (2019) Automated detection of melanocytes related pigmented skin lesions: a clinical framework. Biomed Signal Process Control 1(51):59\u201372","journal-title":"Biomed Signal Process Control"},{"issue":"6","key":"11204_CR38","doi-asserted-by":"publisher","first-page":"1675","DOI":"10.1109\/JBHI.2016.2637342","volume":"21","author":"P Kharazmi","year":"2016","unstructured":"Kharazmi P, AlJasser MI, Lui H, Wang ZJ, Lee TK (2016) Automated detection and segmentation of vascular structures of skin lesions seen in Dermoscopy, with an application to basal cell carcinoma classification. IEEE J Biomed Health Inform 21(6):1675\u20131684","journal-title":"IEEE J Biomed Health Inform"},{"issue":"140","key":"11204_CR39","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1016\/j.ijleo.2017.04.084","volume":"1","author":"F Dalila","year":"2017","unstructured":"Dalila F, Zohra A, Reda K, Hocine C (2017) Segmentation and classification of melanoma and benign skin lesions. Optik 1(140):749\u2013761","journal-title":"Optik"},{"issue":"40","key":"11204_CR40","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1016\/j.jvcir.2016.06.014","volume":"1","author":"N Noroozi","year":"2016","unstructured":"Noroozi N, Zakerolhosseini A (2016) Computer assisted diagnosis of basal cell carcinoma using Z-transform features. J Vis Commun Image Represent 1(40):128\u2013148","journal-title":"J Vis Commun Image Represent"},{"issue":"3","key":"11204_CR41","doi-asserted-by":"publisher","first-page":"456","DOI":"10.1016\/j.bbe.2018.03.005","volume":"38","author":"A Zakeri","year":"2018","unstructured":"Zakeri A, Hokmabadi A (2018) Improvement in the diagnosis of melanoma and dysplastic lesions by introducing ABCD-PDT features and a hybrid classifier. Biocybern Biomed Eng 38(3):456\u2013466","journal-title":"Biocybern Biomed Eng"},{"issue":"5","key":"11204_CR42","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/JTEHM.2017.2648797","volume":"16","author":"TY Satheesha","year":"2017","unstructured":"Satheesha TY, Satyanarayana D, Prasad MG, Dhruve KD (2017) Melanoma is skin deep: a 3D reconstruction technique for computerized dermoscopic skin lesion classification. IEEE J Transl Eng Health Med 16(5):1\u20137","journal-title":"IEEE J Transl Eng Health Med"},{"issue":"61","key":"11204_CR43","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/j.eswa.2016.05.017","volume":"1","author":"RB Oliveira","year":"2016","unstructured":"Oliveira RB, Marranghello N, Pereira AS, Tavares JM (2016) A computational approach for detecting pigmented skin lesions in macroscopic images. Expert Syst Appl 1(61):53\u201363","journal-title":"Expert Syst Appl"}],"container-title":["Neural Processing Letters"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11063-023-11204-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11063-023-11204-5\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11063-023-11204-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,11]],"date-time":"2023-11-11T17:17:01Z","timestamp":1699723021000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11063-023-11204-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,27]]},"references-count":43,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2023,12]]}},"alternative-id":["11204"],"URL":"https:\/\/doi.org\/10.1007\/s11063-023-11204-5","relation":{},"ISSN":["1370-4621","1573-773X"],"issn-type":[{"value":"1370-4621","type":"print"},{"value":"1573-773X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,27]]},"assertion":[{"value":"24 February 2023","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 March 2023","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"None declared.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"Not required.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}}]}}