Machine Learning Microplastic Characterisation Surpasses Human Performance and Uncovers Labelling Errors in Public FTIR Data

Machine Learning Microplastic Characterisation Surpasses Human Performance and Uncovers Labelling Errors in Public FTIR Data | 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 Article Machine Learning Microplastic Characterisation Surpasses Human Performance and Uncovers Labelling Errors in Public FTIR Data Wye-Khay Fong, Frithjof Herb, Mario Boley This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5081019/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 Microplastics are ubiquitous and appear to be harmful, however, the full extent to which these inflict harm has not been fully elucidated. Analysing environmental sample data is challenging, as the complexity in data makes both automated and manual analysis either unreliable or time-consuming. To address challenges in data analysis, we explored the use of a dense feed-forward neural network on a diverse dataset of Fourier transform infrared spectroscopic data. The model makes predictions that are probabilistic multi-category vectors, emulating onehot encodings for each of the classes, thereby indicating the probability of a given sample belonging to a known class. Our results indicate that this model outperforms human classification performance for environmental microplastic FTIR data. This is assumed as the model broadly reproduces human decisions, while also revealing systematic errors in human interpretations. These uncovered errors indicate that for a model making informed and reliable decisions, there exists an artificial limit to the realistically achievable performance in metrics, where metrics measure agreement between human and model predictions. This work indicates an enormous potential for a small and highly efficient dense feed-forward neural networks in making reliable and fast analysis of large volumes of complex FTIR data accessible. Earth and environmental sciences/Environmental sciences/Environmental chemistry Physical sciences/Mathematics and computing/Scientific data Physical sciences/Chemistry/Analytical chemistry Scientific community and society/Water resources Physical sciences/Mathematics and computing/Software Microplastics Neural Networks FTIR Analysis Automated Classification Deep Learning Spectroscopic Data Analysis Full Text Additional Declarations There is NO Competing Interest. Supplementary Files ManuscriptSI.pdf supplementary information 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-5081019","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":354590126,"identity":"f53642e6-78a2-4c63-84ba-6950995d6b49","order_by":0,"name":"Wye-Khay Fong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYBACCQkQaQPE7A1ggQQgNiBCSxoQ8xwgWYtEApFaJGc3P5NgSLDJk498nfi4oKYuj4G9eZsEPi3SMsfMgFrSig1v5242nnHscDEDz7EyvFrkJBLMJBh/HE7cODt3mzQP24HEBokcMwJa0r8BbQFqmXkWqOVfXWKD/Bv8WqTBZgK1zJfg3SbN28YMtIUHvxbJGTnFFgkJaYkbeIB+4e07nNjGk1ZsgU+LxI30jTc+JNgkzm8/u/Exz7e6xH72wxtv4NMCBCzgGDE4AOWyEVAOAswfQKR8AxFKR8EoGAWjYGQCAK7iR7TvqWQHAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-0437-9332","institution":"Monash University","correspondingAuthor":true,"prefix":"","firstName":"Wye-Khay","middleName":"","lastName":"Fong","suffix":""},{"id":354590127,"identity":"ae96fbda-822d-4d65-8f64-480e06a217e0","order_by":1,"name":"Frithjof Herb","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"prefix":"","firstName":"Frithjof","middleName":"","lastName":"Herb","suffix":""},{"id":354590128,"identity":"dd1c781b-5118-4805-8bed-70a5890fed11","order_by":2,"name":"Mario Boley","email":"","orcid":"https://orcid.org/0000-0002-0704-4968","institution":"Monash University","correspondingAuthor":false,"prefix":"","firstName":"Mario","middleName":"","lastName":"Boley","suffix":""}],"badges":[],"createdAt":"2024-09-13 04:55:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5081019/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5081019/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":64941826,"identity":"8fb0a755-fd11-42f8-b681-d8ba8b4d510e","added_by":"auto","created_at":"2024-09-20 15:47:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8157194,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscriptmain.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5081019/v1_covered_4e4a4771-6353-458a-8231-023c3feddc27.pdf"},{"id":64939653,"identity":"78ed50ee-58c3-46bc-8b1d-2acf65df1d8d","added_by":"auto","created_at":"2024-09-20 15:22:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":45743805,"visible":true,"origin":"","legend":"\u003cp\u003esupplementary information\u003c/p\u003e","description":"","filename":"ManuscriptSI.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5081019/v1/20560181cb6ad2ea5022aaba.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Machine Learning Microplastic Characterisation Surpasses Human Performance and Uncovers Labelling Errors in Public FTIR Data","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"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":"Microplastics, Neural Networks, FTIR Analysis, Automated Classification, Deep Learning, Spectroscopic Data Analysis","lastPublishedDoi":"10.21203/rs.3.rs-5081019/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5081019/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Microplastics are ubiquitous and appear to be harmful, however, the full extent\r\nto which these inflict harm has not been fully elucidated. Analysing environmental\r\nsample data is challenging, as the complexity in data makes both automated\r\nand manual analysis either unreliable or time-consuming. To address challenges\r\nin data analysis, we explored the use of a dense feed-forward neural network on\r\na diverse dataset of Fourier transform infrared spectroscopic data. The model\r\nmakes predictions that are probabilistic multi-category vectors, emulating onehot\r\nencodings for each of the classes, thereby indicating the probability of a given\r\nsample belonging to a known class. Our results indicate that this model outperforms\r\nhuman classification performance for environmental microplastic FTIR\r\ndata. This is assumed as the model broadly reproduces human decisions, while\r\nalso revealing systematic errors in human interpretations. These uncovered errors\r\nindicate that for a model making informed and reliable decisions, there exists an\r\nartificial limit to the realistically achievable performance in metrics, where metrics\r\nmeasure agreement between human and model predictions. This work indicates\r\nan enormous potential for a small and highly efficient dense feed-forward neural\r\nnetworks in making reliable and fast analysis of large volumes of complex FTIR\r\ndata accessible.","manuscriptTitle":"Machine Learning Microplastic Characterisation Surpasses Human Performance and Uncovers Labelling Errors in Public FTIR Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-20 15:22:47","doi":"10.21203/rs.3.rs-5081019/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":"b66c0959-b352-4362-a5df-05f8a95aff3d","owner":[],"postedDate":"September 20th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":37682866,"name":"Earth and environmental sciences/Environmental sciences/Environmental chemistry"},{"id":37682867,"name":"Physical sciences/Mathematics and computing/Scientific data"},{"id":37682868,"name":"Physical sciences/Chemistry/Analytical chemistry"},{"id":37682869,"name":"Scientific community and society/Water resources"},{"id":37682870,"name":"Physical sciences/Mathematics and computing/Software"}],"tags":[],"updatedAt":"2024-09-20T15:22:49+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-20 15:22:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5081019","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5081019","identity":"rs-5081019","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}