{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T09:29:42Z","timestamp":1766741382546,"version":"3.48.0"},"posted":{"date-parts":[[2025,12,22]]},"group-title":"Bioinformatics","reference-count":56,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T00:00:00Z","timestamp":1766361600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100015326","name":"NIH Common Fund","doi-asserted-by":"crossref","award":["1R16GM145671"],"award-info":[{"award-number":["1R16GM145671"]}],"id":[{"id":"10.13039\/100015326","id-type":"DOI","asserted-by":"crossref"}]},{"name":"NSF","award":["MCB 2027738"],"award-info":[{"award-number":["MCB 2027738"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,12,22]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Background<\/jats:title>\n                  <jats:p>Super Resolution Microscopy (SRM) is a powerful method in quantitative bioscience that allows interrogation of nanoscale details. These methods require extensive imaging times on the microscope resulting in data sets on the scale of gigabytes. In order to reduce imaging times, the concentration of emitters can be increased, however that results in overlapping emitters rendering the isolation of single emitters extremely difficult. Statistical methods have been developed to deconvolute overlapping emitters, however they require parameter optimization and user expertise. Recently, Machine Learning (ML) has been developed to automate this analysis but often require larger Convolutional Neural Networks (CNN). While powerful, such models require compute and storage that would make pushing these models to compute limited devices difficult. To address this, we investigate if the dense multi-emitter localization capacity of a larger model, Deep Residual Stochastic Optical Reconstruction Microscopy (DRL-STORM), can be transferred to a smaller model, Super Resolution Convolutional Neural Network (SRCNN).<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Knowledge transfer from DRL-STORM to SRCNN did not result in an improvement of multi-emitter localization performance of SRCNN. Hint Learning (HL) was performed to facilitate knowledge transfer in a more deliberate manner. SRCNN demonstrated a limited capacity to learn an intermediate representation of the input image in the same manner as DRL-STORM and resultantly did not perform any better in its task.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>Knowledge transfer was not successful between DRL-STORM and SRCNN, but evidence suggests that it is possible and may require another model besides SRCNN. A future work will investigate if hyper-parameter optimization results in greater knowledge distillation between DRL-STORM and SRCNN. SRCNN may not be ideal for multi-emitter localizations, but it can still prove effective for SRM data analysis at emitter concentrations typical of SRM experiments and is uniquely suited as a neural network model that can be deployed in compute limited settings.<\/jats:p>\n                <\/jats:sec>","DOI":"10.64898\/2025.12.19.694650","type":"posted-content","created":{"date-parts":[[2025,12,23]],"date-time":"2025-12-23T01:35:38Z","timestamp":1766453738000},"source":"Crossref","is-referenced-by-count":0,"title":["Assessing Knowledge Distillation of a Multi-Emitter Localizing Neural Network for Applications in Stochastic Optical Reconstruction Microscopy"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0390-5695","authenticated-orcid":false,"given":"Micheal B.","family":"Reed","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9662-7331","authenticated-orcid":false,"given":"Reza","family":"Zadegan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2025122601250670000_2025.12.19.694650v1.1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-05925-9"},{"key":"2025122601250670000_2025.12.19.694650v1.2","doi-asserted-by":"crossref","first-page":"10205","DOI":"10.1039\/C7NR00928C","article-title":"Metrology of DNA arrays by super-resolution microscopy","volume":"9","year":"2017","journal-title":"Nanoscale"},{"key":"2025122601250670000_2025.12.19.694650v1.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.ymeth.2019.05.018"},{"key":"2025122601250670000_2025.12.19.694650v1.4","doi-asserted-by":"crossref","unstructured":"Feld LG , Shynkarenko Y , Krieg F , Rain\u00f2 G , Kovalenko MV . Perovskite Quantum Dots for Super-Resolution Optical Microscopy: Where Strong Photoluminescence Blinking Matters. Adv Opt Mater. 2021;9:2100620.","DOI":"10.1002\/adom.202100620"},{"key":"2025122601250670000_2025.12.19.694650v1.5","doi-asserted-by":"crossref","unstructured":"Choudhury P , Ranjan Boruah B . Localization and Image Reconstruction in a STORM Based Super-resolution Microscope. Image Process Line. 2024;14:64.","DOI":"10.5201\/ipol.2024.496"},{"issue":"81","key":"2025122601250670000_2025.12.19.694650v1.6","first-page":"46","article-title":"Stochastic optical reconstruction microscopy (STORM)","volume":"12","year":"2017","journal-title":"Curr Protoc Cytom"},{"key":"2025122601250670000_2025.12.19.694650v1.7","doi-asserted-by":"crossref","first-page":"3715","DOI":"10.1364\/BOE.517480","article-title":"Triggered cagedSTORM microscopy","volume":"15","year":"2024","journal-title":"Biomed Opt Express"},{"key":"2025122601250670000_2025.12.19.694650v1.8","doi-asserted-by":"crossref","first-page":"2730","DOI":"10.1093\/bioinformatics\/btab136","article-title":"Blinking statistics and molecular counting in direct stochastic reconstruction microscopy (dSTORM)","volume":"37","year":"2021","journal-title":"Bioinformatics"},{"key":"2025122601250670000_2025.12.19.694650v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/nmeth929"},{"key":"2025122601250670000_2025.12.19.694650v1.10","doi-asserted-by":"crossref","first-page":"5847","DOI":"10.1126\/sciadv.adk5847","article-title":"Comparing divisome organization between vegetative and sporulating Bacillus subtilis at the nanoscale using DNA-PAINT","volume":"10","year":"2024","journal-title":"Sci Adv"},{"key":"2025122601250670000_2025.12.19.694650v1.11","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-23276-8"},{"key":"2025122601250670000_2025.12.19.694650v1.12","doi-asserted-by":"crossref","first-page":"1277","DOI":"10.1364\/OPTICA.4.001277","article-title":"Blind sparse inpainting reveals cytoskeletal filaments with sub-Nyquist localization","volume":"4","year":"2017","journal-title":"Optica"},{"key":"2025122601250670000_2025.12.19.694650v1.13","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-35677-5"},{"key":"2025122601250670000_2025.12.19.694650v1.14","first-page":"1","article-title":"Single-molecule localization microscopy","volume":"1","year":"2021","journal-title":"Nat Rev Methods Primers"},{"key":"2025122601250670000_2025.12.19.694650v1.15","doi-asserted-by":"publisher","DOI":"10.1038\/nmeth.1978"},{"key":"2025122601250670000_2025.12.19.694650v1.16","doi-asserted-by":"publisher","DOI":"10.1038\/nmeth0411-279"},{"key":"2025122601250670000_2025.12.19.694650v1.17","doi-asserted-by":"publisher","DOI":"10.1038\/srep04562"},{"key":"2025122601250670000_2025.12.19.694650v1.18","doi-asserted-by":"crossref","unstructured":"Fazel M , Wester MJ . Analysis of super-resolution single molecule localization microscopy data: A tutorial. AIP Adv. 2022;12:10701.","DOI":"10.1063\/5.0069349"},{"key":"2025122601250670000_2025.12.19.694650v1.19","doi-asserted-by":"crossref","unstructured":"Smith CS , Joseph N , Rieger B , Lidke KA . Fast, single-molecule localization that achieves theoretically minimum uncertainty. Nat Methods. 2010;7:373.","DOI":"10.1038\/nmeth.1449"},{"key":"2025122601250670000_2025.12.19.694650v1.20","doi-asserted-by":"publisher","DOI":"10.1364\/optica.5.000458"},{"key":"2025122601250670000_2025.12.19.694650v1.21","doi-asserted-by":"crossref","first-page":"3647","DOI":"10.1515\/nanoph-2023-0936","article-title":"Digital-SMLM for precisely localizing emitters within the diffraction limit","volume":"13","year":"2024","journal-title":"Nanophotonics"},{"key":"2025122601250670000_2025.12.19.694650v1.22","doi-asserted-by":"crossref","first-page":"1833","DOI":"10.1364\/BOE.484540","article-title":"Deep learning using a residual deconvolutional network enables real-time high-density single-molecule localization microscopy","volume":"14","year":"2023","journal-title":"Biomed Opt Express"},{"key":"2025122601250670000_2025.12.19.694650v1.23","doi-asserted-by":"publisher","DOI":"10.1021\/acs.analchem.1c00431"},{"key":"2025122601250670000_2025.12.19.694650v1.24","doi-asserted-by":"crossref","first-page":"2403","DOI":"10.1021\/acssensors.1c00628","article-title":"Dynamic Imaging and Characterization of Volatile Aerosols in E-Cigarette Emissions Using Deep Learning-Based Holographic Microscopy","volume":"6","year":"2021","journal-title":"ACS Sens"},{"key":"2025122601250670000_2025.12.19.694650v1.25","doi-asserted-by":"crossref","unstructured":"Andani S , Chen B , Ficek-Pascual J , Heinke S , Casanova R , Hild BF , et al. Histopathology-based protein multiplex generation using deep learning. Nat Mach Intell. 2025;:1\u201316.","DOI":"10.1101\/2024.01.26.24301803"},{"key":"2025122601250670000_2025.12.19.694650v1.26","doi-asserted-by":"crossref","first-page":"B182","DOI":"10.1364\/PRJ.415960","article-title":"Deep learning in nano-photonics: inverse design and beyond","volume":"9","year":"2021","journal-title":"Photonics Res"},{"key":"2025122601250670000_2025.12.19.694650v1.27","doi-asserted-by":"publisher","DOI":"10.1364\/oe.392358"},{"key":"2025122601250670000_2025.12.19.694650v1.28","doi-asserted-by":"crossref","first-page":"45750","DOI":"10.1364\/OE.578643","article-title":"EdgePhase: a portable edge-computational microscope for point-of-care quantitative phase imaging","volume":"33","year":"2025","journal-title":"Opt Express"},{"key":"2025122601250670000_2025.12.19.694650v1.29","doi-asserted-by":"crossref","first-page":"8288","DOI":"10.1109\/JIOT.2023.3317878","article-title":"MicrosMobiNet: A Deep Lightweight Network with Hierarchical Feature Fusion Scheme for Microscopy Image Analysis in Mobile-Edge Computing","volume":"11","year":"2024","journal-title":"IEEE Internet Things J"},{"key":"2025122601250670000_2025.12.19.694650v1.30","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pntd.0012117"},{"key":"2025122601250670000_2025.12.19.694650v1.31","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0209827"},{"key":"2025122601250670000_2025.12.19.694650v1.32","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3447085"},{"key":"2025122601250670000_2025.12.19.694650v1.33","unstructured":"Liu Z , Sun M , Zhou T , Huang G , Darrell T . Rethinking the Value of Network Pruning. 7th International Conference on Learning Representations, ICLR 2019. 2018."},{"key":"2025122601250670000_2025.12.19.694650v1.34","unstructured":"Frankle J , Carbin M. The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks. 7th International Conference on Learning Representations, ICLR 2019. 2018."},{"key":"2025122601250670000_2025.12.19.694650v1.35","doi-asserted-by":"crossref","unstructured":"Takamoto M , Morishita Y , Imaoka H . An Efficient Method of Training Small Models for Regression Problems with Knowledge Distillation. Proceedings - 3rd International Conference on Multimedia Information Processing and Retrieval, MIPR 2020. 2020;:67\u201372.","DOI":"10.1109\/MIPR49039.2020.00021"},{"key":"2025122601250670000_2025.12.19.694650v1.36","doi-asserted-by":"crossref","unstructured":"Kang M , Kang S . Knowledge distillation with insufficient training data for regression. Eng Appl Artif Intell. 2024;132:108001.","DOI":"10.1016\/j.engappai.2024.108001"},{"key":"2025122601250670000_2025.12.19.694650v1.37","doi-asserted-by":"crossref","unstructured":"Saputra MRU , Gusmao P , Almalioglu Y , Markham A , Trigoni N . Distilling knowledge from a deep pose regressor network. Proc IEEE Int Conf Comput Vis. 2019;2019-October:263\u201372.","DOI":"10.1109\/ICCV.2019.00035"},{"key":"2025122601250670000_2025.12.19.694650v1.38","doi-asserted-by":"crossref","unstructured":"Dong C , Loy CC , Tang X . Accelerating the super-resolution convolutional neural network. Lect Notes Comput Sci. 2016;9906 LNCS:391\u2013407.","DOI":"10.1007\/978-3-319-46475-6_25"},{"key":"2025122601250670000_2025.12.19.694650v1.39","doi-asserted-by":"crossref","unstructured":"Dong C , Loy CC , He K , Tang X . Learning a deep convolutional network for image super-resolution. Lect Notes Comput Sci. 2014;8692 LNCS PART 4:184\u201399.","DOI":"10.1007\/978-3-319-10593-2_13"},{"key":"2025122601250670000_2025.12.19.694650v1.40","unstructured":"Orhan AE , Pitkow X . Skip Connections Eliminate Singularities. 6th International Conference on Learning Representations, ICLR 2018 - Conference Track Proceedings. 2017."},{"key":"2025122601250670000_2025.12.19.694650v1.41","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , Sun J . Deep Residual Learning for Image Recognition. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2015;2016-December:770\u20138.","DOI":"10.1109\/CVPR.2016.90"},{"key":"2025122601250670000_2025.12.19.694650v1.42","doi-asserted-by":"crossref","first-page":"364","DOI":"10.1016\/j.neucom.2018.05.103","article-title":"Pixel-wise regression using U-Net and its application on pansharpening","volume":"312","year":"2018","journal-title":"Neurocomputing"},{"key":"2025122601250670000_2025.12.19.694650v1.43","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2025122601250670000_2025.12.19.694650v1.44","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btu202"},{"key":"2025122601250670000_2025.12.19.694650v1.45","doi-asserted-by":"crossref","first-page":"953","DOI":"10.1038\/s42256-022-00547-8","article-title":"A fast blind zero-shot denoiser","volume":"4","year":"2022","journal-title":"Nat Mach Intell"},{"key":"2025122601250670000_2025.12.19.694650v1.46","doi-asserted-by":"crossref","unstructured":"Krull A , Buchholz TO , Jug F . Noise2Void - Learning Denoising from Single Noisy Images. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2018;2019-June:2124\u201332.","DOI":"10.1109\/CVPR.2019.00223"},{"key":"2025122601250670000_2025.12.19.694650v1.47","doi-asserted-by":"crossref","unstructured":"Mansour Y , Heckel R . Zero-Shot Noise2Noise: Efficient Image Denoising without any Data. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2023;2023-June:14018\u201327.","DOI":"10.1109\/CVPR52729.2023.01347"},{"key":"2025122601250670000_2025.12.19.694650v1.48","unstructured":"G\u00f6bl R , Hennersperger C , Navab N. Speckle2Speckle: Unsupervised Learning of Ultrasound Speckle Filtering Without Clean Data. 2022."},{"key":"2025122601250670000_2025.12.19.694650v1.49","first-page":"2408","article-title":"3D-Printed Portable Robotic Mobile Microscope for Remote Diagnosis of Global Health Diseases","volume":"10","year":"2021","journal-title":"Electronics (Basel)"},{"key":"2025122601250670000_2025.12.19.694650v1.50","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-35677-5"},{"key":"2025122601250670000_2025.12.19.694650v1.51","doi-asserted-by":"crossref","unstructured":"Elinisa CA , Mduma N . Mobile-Based convolutional neural network model for the early identification of banana diseases. Smart Agric Technol. 2024;7:100423.","DOI":"10.1016\/j.atech.2024.100423"},{"key":"2025122601250670000_2025.12.19.694650v1.52","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1021\/ph400108p","article-title":"Counting Molecules with a Mobile Phone Camera Using Plasmonic Enhancement","volume":"1","year":"2014","journal-title":"ACS Photonics"},{"key":"2025122601250670000_2025.12.19.694650v1.53","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.dcan.2021.06.001","article-title":"A survey on deploying mobile deep learning applications: A systemic and technical perspective","volume":"8","year":"2022","journal-title":"Digit Commun Netw"},{"key":"2025122601250670000_2025.12.19.694650v1.54","doi-asserted-by":"crossref","unstructured":"Kornilova A , Salnikov M , Novitskaya O , Begicheva M , Sevriugov E , Shcherbakov K , et al. Deep learning framework for mobile microscopy. Proc IEEE Int Symp Biomed Imaging. 2021;2021-April:324\u20138.","DOI":"10.1109\/ISBI48211.2021.9434133"},{"key":"2025122601250670000_2025.12.19.694650v1.55","doi-asserted-by":"publisher","DOI":"10.1098\/RSTA.2023.0112"},{"key":"2025122601250670000_2025.12.19.694650v1.56","first-page":"1","article-title":"Imagining the future of optical microscopy: everything, everywhere, all at once","volume":"6","year":"2023","journal-title":"Commun Biol"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2025.12.19.694650","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T09:25:17Z","timestamp":1766741117000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2025.12.19.694650"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,22]]},"references-count":56,"URL":"https:\/\/doi.org\/10.64898\/2025.12.19.694650","relation":{},"subject":[],"published":{"date-parts":[[2025,12,22]]},"subtype":"preprint"}}