Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning

Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning | 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 Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning Khayrul Islam, Ryan F. Forelli, Jianzhong Han, Deven Bhadane, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6247562/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 Precise cell classification is essential in biomedical diagnostics and therapeutic monitoring, particularly for identifying diverse cell types involved in various diseases. Traditional cell classification methods such as flow cytometry depend on molecular labeling which is often costly, time-intensive, and can alter cell integrity. To overcome these limitations, we present a label-free machine learning framework for cell classification, designed for real-time sorting applications using bright-field microscopy images. This approach leverages a teacher-student model architecture enhanced by knowledge distillation, achieving high efficiency and scalability across different cell types. Demonstrated through a use case of classifying lymphocyte subsets, our framework accurately classifies T4, T8, and B cell types with a dataset of 80,000 preprocessed images, accessible via an open-source Python package for easy adaptation. Our teacher model attained 98% accuracy in differentiating T4 cells from B cells and 93% accuracy in zero-shot classification between T8 and B cells. Remarkably, our student model operates with only 0.02% of the teacher model's parameters, enabling field-programmable gate array (FPGA) deployment. Our FPGA-accelerated student model achieves an ultra-low inference latency of just 14.5~$\mu$s and a complete cell detection-to-sorting trigger time of 24.7~$\mu$s, delivering 12x and 40x improvements over the previous state-of-the-art real-time cell analysis algorithm in inference and total latency, respectively, while preserving accuracy comparable to the teacher model. This framework provides a scalable, cost-effective solution for lymphocyte classification, as well as a new SOTA real-time cell sorting implementation for rapid identification of subsets using in situ deep learning on off-the-shelf computing hardware. Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Computational biology and bioinformatics Full Text Additional Declarations There is NO Competing Interest. Supplementary Files supplimentarymat.pdf Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning 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-6247562","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":430378403,"identity":"c7f60544-617c-46bc-b052-7caee9ca1464","order_by":0,"name":"Khayrul Islam","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-2199-553X","institution":"Lehigh University","correspondingAuthor":true,"prefix":"","firstName":"Khayrul","middleName":"","lastName":"Islam","suffix":""},{"id":430378404,"identity":"20da0fcf-efa8-4268-a8b1-97a0d6eac6ca","order_by":1,"name":"Ryan F. Forelli","email":"","orcid":"","institution":"Northwestern University","correspondingAuthor":false,"prefix":"","firstName":"Ryan","middleName":"F.","lastName":"Forelli","suffix":""},{"id":430378405,"identity":"524181ef-f7b8-4fc9-8e1f-7414fdd61fb1","order_by":2,"name":"Jianzhong Han","email":"","orcid":"","institution":"Coriell Institue for Medical Research","correspondingAuthor":false,"prefix":"","firstName":"Jianzhong","middleName":"","lastName":"Han","suffix":""},{"id":430378406,"identity":"dcea2567-c12f-4d00-94a4-dc1e22b39ad3","order_by":3,"name":"Deven Bhadane","email":"","orcid":"","institution":"Lehigh Univeristy","correspondingAuthor":false,"prefix":"","firstName":"Deven","middleName":"","lastName":"Bhadane","suffix":""},{"id":430378407,"identity":"d6add1a1-f691-4b3a-8f33-d4e56c3bb625","order_by":4,"name":"Jian Huang","email":"","orcid":"","institution":"Coriell Institue for Medical Research","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Huang","suffix":""},{"id":430378408,"identity":"5d36fa41-6ec5-4078-b1eb-a8950e4dfbc2","order_by":5,"name":"Josh Agar","email":"","orcid":"https://orcid.org/0000-0001-5411-4693","institution":"Drexel University","correspondingAuthor":false,"prefix":"","firstName":"Josh","middleName":"","lastName":"Agar","suffix":""},{"id":430378409,"identity":"aa72e3ad-5fd0-4b2d-8415-7751971f0832","order_by":6,"name":"Nhan Tran","email":"","orcid":"","institution":"Fermi National Accelerator Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Nhan","middleName":"","lastName":"Tran","suffix":""},{"id":430378410,"identity":"cb54793b-7050-4008-806c-0c349f5c3a08","order_by":7,"name":"Seda Ogrenci","email":"","orcid":"","institution":"Northwestern University","correspondingAuthor":false,"prefix":"","firstName":"Seda","middleName":"","lastName":"Ogrenci","suffix":""},{"id":430378411,"identity":"ccd321a9-655c-4cef-9064-737bb06d17e4","order_by":8,"name":"Yaling Liu","email":"","orcid":"https://orcid.org/0000-0002-4519-3358","institution":"Lehigh University","correspondingAuthor":false,"prefix":"","firstName":"Yaling","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-03-17 20:41:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6247562/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6247562/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":78896934,"identity":"6ad5b3a5-a6a7-4a72-86c9-f12aacafae5b","added_by":"auto","created_at":"2025-03-20 12:39:38","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":13783591,"visible":true,"origin":"","legend":"Article File","description":"","filename":"main.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6247562/v1_covered_35b00af1-bda0-413f-82bf-bd3990ed4cb5.pdf"},{"id":78823849,"identity":"2c13025e-466d-49ff-b83d-d3ff52149bbd","added_by":"auto","created_at":"2025-03-19 12:08:48","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":70085,"visible":true,"origin":"","legend":"Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning","description":"","filename":"supplimentarymat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6247562/v1/bdb83848c154621b1f061548.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning","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":"","lastPublishedDoi":"10.21203/rs.3.rs-6247562/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6247562/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Precise cell classification is essential in biomedical diagnostics and therapeutic monitoring, particularly for identifying diverse cell types involved in various diseases. Traditional cell classification methods such as flow cytometry depend on molecular labeling which is often costly, time-intensive, and can alter cell integrity. To overcome these limitations, we present a label-free machine learning framework for cell classification, designed for real-time sorting applications using bright-field microscopy images. This approach leverages a teacher-student model architecture enhanced by knowledge distillation, achieving high efficiency and scalability across different cell types. Demonstrated through a use case of classifying lymphocyte subsets, our framework accurately classifies T4, T8, and B cell types with a dataset of 80,000 preprocessed images, accessible via an open-source Python package for easy adaptation. Our teacher model attained 98\\% accuracy in differentiating T4 cells from B cells and 93\\% accuracy in zero-shot classification between T8 and B cells. Remarkably, our student model operates with only 0.02\\% of the teacher model's parameters, enabling field-programmable gate array (FPGA) deployment. Our FPGA-accelerated student model achieves an ultra-low inference latency of just 14.5~$\\mu$s and a complete cell detection-to-sorting trigger time of 24.7~$\\mu$s, delivering 12x and 40x improvements over the previous state-of-the-art real-time cell analysis algorithm in inference and total latency, respectively, while preserving accuracy comparable to the teacher model. This framework provides a scalable, cost-effective solution for lymphocyte classification, as well as a new SOTA real-time cell sorting implementation for rapid identification of subsets using in situ deep learning on off-the-shelf computing hardware.","manuscriptTitle":"Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-19 12:08:19","doi":"10.21203/rs.3.rs-6247562/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":"167d9f9f-53f0-4288-b19d-9e39334eee9e","owner":[],"postedDate":"March 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":45836808,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":45836809,"name":"Biological sciences/Computational biology and bioinformatics"}],"tags":[],"updatedAt":"2025-03-20T12:30:59+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-19 12:08:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6247562","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6247562","identity":"rs-6247562","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}