Fast Characteristic of Skin Lesions by Machine-Learning of Raman Spectrum

Fast Characteristic of Skin Lesions by Machine-Learning of Raman Spectrum | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Fast Characteristic of Skin Lesions by Machine-Learning of Raman Spectrum Hua Zhang, Danhua Wang, Limei Qu, Ying Xue, Xinli Li, Bei Li, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-154353/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The traditional diagnosis of skin lesions mainly relies on dermoscope and pathological biopsy, of which the former is non-objective and the latter is invasive and time-consuming. It is necessary to find an objective and non-invasive inspection method for the diagnosis of skin cancer which is the most common malignant tumor. Herein, we aimed to fast identify the skin cancers on ultrathin frozen fresh tissue sections by combining Raman spectroscopy detection and machine learning technology. Methods and material: 22 fresh frozen tissue sections including 3 squamous cell carcinomas, 11 basal cell carcinomas, 2 malignant melanomas, 3 seborrheic keratosis, and 3 melanocytic nevi, were included and performed Raman detection. To prevent the discrete Raman data distribution affecting the generalization ability of the learning model, a series of adaptive preprocessing algorithms were first applied to standardize the raw Raman data of five skin lesions. The processed Raman data were performed visualized cluster analysis by principal components analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE). And, using K-nearest Neighbor (KNN) and support vector machine (SVM) classifiers, two predictive models for diagnose were established and evaluated in the training set and test set by the confusion matrixes and receiver operating characteristic (ROC) curves. Results: The mean variance Raman spectrum graph of 5 skin lesion types were acquired after standardization procession and 4 peak positions with large differences were found. Through dimensionality reduction by PCA and t-SNE, the visual clustering results of Raman data showed heterogeneous intra-cluster homogeneity and inter-cluster dispersion. The test accuracies reached 94.56% and 98.94% in KNN and SVM classifiers respectively. The areas under the ROCs of the two classifiers, in the category dimension and the sample dimension, were all more than 0.99 which is close to the perfect classification effect. Conclusions: Raman spectroscopy is a competitive candidate for the fast and accurate diagnosis of skin lesions and the molecular information provided may be used in the pathological classification, predicting immunotherapy responsiveness and stratifying prognostic risk. Furthermore, the combination of Raman spectroscopy and machine learning methods showed great diagnostic capabilities with high accuracy is a promising tool for the diagnosis of skin lesions. Translational Medicine Internal Medicine skin tumor Raman spectroscopy machine learning molecular diagnosis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Full Text Tables Due to technical limitations, table 1 to 6 is only available as a download in the Supplemental Files section. Supplementary Files Table.doc Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-154353","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research","associatedPublications":[],"authors":[{"id":9419025,"identity":"bfe070fa-7e87-4276-9347-b006ff264585","order_by":0,"name":"Hua Zhang","email":"","orcid":"","institution":"Jilin University First Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hua","middleName":"","lastName":"Zhang","suffix":""},{"id":9419026,"identity":"748eab9b-46f4-4c7e-8ceb-d975285316de","order_by":1,"name":"Danhua Wang","email":"","orcid":"","institution":"Jilin University First Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Danhua","middleName":"","lastName":"Wang","suffix":""},{"id":9419027,"identity":"0905a359-f56e-465e-ba6e-7ed677350e90","order_by":2,"name":"Limei Qu","email":"","orcid":"","institution":"Jilin University First Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Limei","middleName":"","lastName":"Qu","suffix":""},{"id":9419028,"identity":"d7b82a2e-58e0-4286-89cb-9c1e83eb3210","order_by":3,"name":"Ying Xue","email":"","orcid":"","institution":"HOOKE Instruments Ltd.","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Xue","suffix":""},{"id":9419029,"identity":"60147e3f-741c-4c5f-888c-5739e755480c","order_by":4,"name":"Xinli Li","email":"","orcid":"","institution":"HOOKE Instruments Ltd.","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xinli","middleName":"","lastName":"Li","suffix":""},{"id":9419030,"identity":"9ac46fea-6a10-46fe-8726-29d1933875d6","order_by":5,"name":"Bei Li","email":"","orcid":"","institution":"Changchun Institute of Optics Fine Mechanics and Physics Chinese Academy of Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Bei","middleName":"","lastName":"Li","suffix":""},{"id":9419031,"identity":"fb8b3b4d-abe4-4dc1-8341-1ade3ef3f2a6","order_by":6,"name":"Bin Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAr0lEQVRIiWNgGAWjYBACPgYehgMMFWwyII4EUVrYwFrOsPGQpoWBsY2BJC28Bw8XzuPjMTjAfPA2D4NdHhFa+BIOz9zGBtTClmzNw5BcTIzDDA7zgrXwmEkD/ZXYQJyWOSAt/N9I0dIAtoWNWC1Av/AcY+ORPMxmbDnHIJmwFn4G3sOfeWqOyfEdb354402FHWEtDPIPQOQxBgZmEG1AUD0c1BCvdBSMglEwCkYeAAB67i6laVsSlAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-3179-6235","institution":"Jilin University First Hospital","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2021-01-25 18:14:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-154353/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-154353/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":5391144,"identity":"5756e646-9627-4171-a21d-1fe6f9c1a5bf","added_by":"auto","created_at":"2021-01-29 20:56:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":98078,"visible":true,"origin":"","legend":"Schematic of machine-learning of skin lesion Raman spectra for fast diagnosis. Ultrathin frozen fresh sections of skin lesions were acquired and prepared for Raman detection with a 785nm laser. Raman spectra were processed gradually and analyzed by machine learning methods. Diagnose information were put out at last.","description":"","filename":"Fig01.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/5d095dbc33806c7db1d16afd.png"},{"id":5391146,"identity":"2928e754-e6fb-4a30-8117-ef933a68fabb","added_by":"auto","created_at":"2021-01-29 20:56:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":577389,"visible":true,"origin":"","legend":"BCC Raman single spectrum standardization results. RI, Raman Intensity.","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/665ca6c847c4eccf18899954.png"},{"id":5391166,"identity":"3cefc1a2-8e7c-4943-a603-a54a78a14cf2","added_by":"auto","created_at":"2021-01-29 20:59:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":299727,"visible":true,"origin":"","legend":"The mean variance graph of Raman spectra of 5 skin lesions after preprocessing.","description":"","filename":"Fig03.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/048701b17bbb21a13f2b11dc.png"},{"id":5391039,"identity":"afab4c70-2d14-425c-8320-7777fb11b562","added_by":"auto","created_at":"2021-01-29 20:53:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":94046,"visible":true,"origin":"","legend":"Normalized Raman intensity for bands at 720 cm-1, 752 cm-1, 853 cm-1, and 1002 cm-1 of 5 skin lesions. RI, Raman Intensity. Data were presented as mean ± SD. * P \u003c 0.05; ** P \u003c 0.01; *** P \u003c 0.001. (Student’s t-test)","description":"","filename":"Fig04.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/5ce73df09aa0d5707d2c13e1.png"},{"id":5391148,"identity":"4add8226-0847-40e7-8ae7-2d894e98a273","added_by":"auto","created_at":"2021-01-29 20:56:24","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":275056,"visible":true,"origin":"","legend":"Visualized clustering results of 5 skin lesion types after dimensionality reduction in t-SNE (A) and PCA (B).","description":"","filename":"Fig05.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/7e0de30d39d504e8159e04a6.png"},{"id":5391145,"identity":"e279ee78-4fe3-4f72-9f8b-4e5656f826d6","added_by":"auto","created_at":"2021-01-29 20:56:24","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":249140,"visible":true,"origin":"","legend":"Confusion matrix and ROC curves of test results. (A, B) Confusion matrixes of 20% RS data test results in KNN (A) and SVM (B). (C, D) The ROC curves for the recognition probabilities of 5 skin lesion categories in KNN (C) and SVM (D).","description":"","filename":"Fig06.png","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/9ea0ecc62c75cdc1a8b5c3b5.png"},{"id":13580257,"identity":"ffd7f657-ae8f-4b98-bc66-db1763b178ba","added_by":"auto","created_at":"2021-09-17 04:22:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":640468,"visible":true,"origin":"","legend":"","description":"","filename":"20210110RSskinlesions2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1_covered.pdf"},{"id":5391257,"identity":"2d5b1aed-ae29-4305-accf-8b6c6520ed61","added_by":"auto","created_at":"2021-01-29 21:02:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":472691,"visible":true,"origin":"","legend":"","description":"","filename":"20210110RSskinlesions2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1_stamped.pdf"},{"id":5391036,"identity":"73011109-e9ae-4da0-a188-77bb3988804c","added_by":"auto","created_at":"2021-01-29 20:53:24","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":103936,"visible":true,"origin":"","legend":"","description":"","filename":"Table.doc","url":"https://assets-eu.researchsquare.com/files/rs-154353/v1/c00436dc9770c894e16f0422.doc"}],"financialInterests":"","formattedTitle":"Fast Characteristic of Skin Lesions by Machine-Learning of Raman Spectrum","fulltext":[{"header":"Full Text","content":"\u003cp\u003eThis preprint is available for \u003ca href='/article/rs-154353/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e.\u003c/p\u003e"},{"header":"Tables","content":"\u003cp\u003eDue to technical limitations, table 1 to 6 is only available as a download in the Supplemental Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"skin tumor, Raman spectroscopy, machine learning, molecular diagnosis ","lastPublishedDoi":"10.21203/rs.3.rs-154353/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-154353/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eThe traditional diagnosis of skin lesions mainly relies on dermoscope and pathological biopsy, of which the former is non-objective and the latter is invasive and time-consuming. It is necessary to find an objective and non-invasive inspection method for the diagnosis of skin cancer which is the most common malignant tumor. Herein, we aimed to fast identify the skin cancers on ultrathin frozen fresh tissue sections by combining Raman spectroscopy detection and machine learning technology.\u003cstrong\u003e \u003c/strong\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eMethods and material: \u003c/strong\u003e22 fresh frozen tissue sections including 3 squamous cell carcinomas, 11 basal cell carcinomas, 2 malignant melanomas, 3 seborrheic keratosis, and 3 melanocytic nevi, were included and performed Raman detection. To prevent the discrete Raman data distribution affecting the generalization ability of the learning model, a series of adaptive preprocessing algorithms were first applied to standardize the raw Raman data of five skin lesions. The processed Raman data were performed visualized cluster analysis by principal components analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE). And, using K-nearest Neighbor (KNN) and support vector machine (SVM) classifiers, two predictive models for diagnose were established and evaluated in the training set and test set by the confusion matrixes and receiver operating characteristic (ROC) curves.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eThe mean variance Raman spectrum graph of 5 skin lesion types were acquired after standardization procession and 4 peak positions with large differences were found. Through dimensionality reduction by PCA and t-SNE, the visual clustering results of Raman data showed heterogeneous intra-cluster homogeneity and inter-cluster dispersion. The test accuracies reached 94.56% and 98.94% in KNN and SVM classifiers respectively. The areas under the ROCs of the two classifiers, in the category dimension and the sample dimension, were all more than 0.99 which is close to the perfect classification effect. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eRaman spectroscopy is a competitive candidate for the fast and accurate diagnosis of skin lesions and the molecular information provided may be used in the pathological classification, predicting immunotherapy responsiveness and stratifying prognostic risk. Furthermore, the combination of Raman spectroscopy and machine learning methods showed great diagnostic capabilities with high accuracy is a promising tool for the diagnosis of skin lesions.\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"Fast Characteristic of Skin Lesions by Machine-Learning of Raman Spectrum","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-01-29 20:53:22","doi":"10.21203/rs.3.rs-154353/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7d901091-2417-4deb-b67d-992832fd4774","owner":[],"postedDate":"January 29th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":2131232,"name":"Translational Medicine"},{"id":2131233,"name":"Internal Medicine"}],"tags":[],"updatedAt":"2021-01-29T20:53:24+00:00","versionOfRecord":[],"versionCreatedAt":"2021-01-29 20:53:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-154353","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-154353","identity":"rs-154353","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}