{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T01:38:35Z","timestamp":1777513115577,"version":"3.51.4"},"reference-count":54,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2024,4,8]],"date-time":"2024-04-08T00:00:00Z","timestamp":1712534400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,4,8]],"date-time":"2024-04-08T00:00:00Z","timestamp":1712534400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["SCHM 997\/42-1"],"award-info":[{"award-number":["SCHM 997\/42-1"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2024,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>The 3D nano\/microstructure of materials can significantly influence their macroscopic properties. In order to enable a better understanding of such structure-property relationships, 3D microscopy techniques can be deployed, which are however often expensive in both time and costs. Often 2D imaging techniques are more accessible, yet they have the disadvantage that the 3D nano\/microstructure of materials cannot be directly retrieved from such measurements. The motivation of this work is to overcome the issues of characterizing 3D structures from 2D measurements for hetero-aggregate materials. For this purpose, a method is presented that relies on machine learning combined with methods of spatial stochastic modeling for characterizing the 3D nano\/microstructure of materials from 2D data. More precisely, a stochastic model is utilized for the generation of synthetic training data. This kind of training data has the advantage that time-consuming experiments for the synthesis of differently structured materials followed by their 3D imaging can be avoided. More precisely, a parametric stochastic 3D model is presented, from which a wide spectrum of virtual hetero-aggregates can be generated. Additionally, the virtual structures are passed to a physics-based simulation tool in order to generate virtual scanning transmission electron microscopy (STEM) images. The preset parameters of the 3D model together with the simulated STEM images serve as a database for the training of convolutional neural networks, which can be used to determine the parameters of the underlying 3D model and, consequently, to predict 3D structures of hetero-aggregates from 2D STEM images. Furthermore, an error analysis is performed with respect to structural descriptors, e.g. the hetero-coordination number. The proposed method is applied to image data of TiO<jats:sub>2<\/jats:sub>-WO<jats:sub>3<\/jats:sub> hetero-aggregates, which are highly relevant in photocatalysis processes. However, the proposed method can be transferred to other types of aggregates and to different 2D microscopy techniques. Consequently, the method is relevant for industrial or laboratory setups in which product quality is to be quantified by means of inexpensive 2D image acquisition.<\/jats:p>","DOI":"10.1088\/2632-2153\/ad38fd","type":"journal-article","created":{"date-parts":[[2024,3,28]],"date-time":"2024-03-28T22:28:37Z","timestamp":1711664917000},"page":"025007","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Using convolutional neural networks for stereological characterization of 3D hetero-aggregates based on synthetic STEM data"],"prefix":"10.1088","volume":"5","author":[{"given":"Lukas","family":"Fuchs","sequence":"first","affiliation":[]},{"given":"Tom","family":"Kirstein","sequence":"additional","affiliation":[]},{"given":"Christoph","family":"Mahr","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1624-6542","authenticated-orcid":true,"given":"Orkun","family":"Furat","sequence":"additional","affiliation":[]},{"given":"Valentin","family":"Baric","sequence":"additional","affiliation":[]},{"given":"Andreas","family":"Rosenauer","sequence":"additional","affiliation":[]},{"given":"Lutz","family":"M\u00e4dler","sequence":"additional","affiliation":[]},{"given":"Volker","family":"Schmidt","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2024,4,8]]},"reference":[{"key":"mlstad38fdbib1","doi-asserted-by":"crossref","first-page":"11840","DOI":"10.1021\/acsomega.0c06054","article-title":"Nanoparticulate double-heterojunction photocatalysts comprising TiO2(Anatase)\/WO3\/TiO2(Rutile) with enhanced photocatalytic activity toward the degradation of methyl orange under near-ultraviolet and visible light","volume":"6","author":"Pinedo-Escobar","year":"2021","journal-title":"ACS Omega"},{"key":"mlstad38fdbib2","doi-asserted-by":"publisher","first-page":"192","DOI":"10.1006\/jcat.1999.2776","article-title":"Photocatalytic behavior of WO3-loaded TiO2 in an oxidation reaction","volume":"191","author":"Tae Kwon","year":"2000","journal-title":"J. Catal."},{"key":"mlstad38fdbib3","doi-asserted-by":"publisher","first-page":"654","DOI":"10.1016\/j.pnsc.2012.11.016","article-title":"Ordered mesoporous tungsten oxide and titanium oxide composites and their photocatalytic degradation behavior","volume":"22","author":"Yan","year":"2012","journal-title":"Prog. Nat. Sci.: Mater. Int."},{"key":"mlstad38fdbib4","doi-asserted-by":"publisher","DOI":"10.1002\/adma.201601694","article-title":"Heterojunction photocatalysts","volume":"29","author":"Low","year":"2017","journal-title":"Adv. Mater."},{"key":"mlstad38fdbib5","doi-asserted-by":"publisher","first-page":"413","DOI":"10.1016\/S0304-3991(03)00105-0","article-title":"3D electron microscopy in the physical sciences: the development of Z-contrast and EFTEM tomography","volume":"96","author":"Midgley","year":"2003","journal-title":"Ultramicroscopy"},{"key":"mlstad38fdbib6","doi-asserted-by":"publisher","first-page":"105","DOI":"10.1016\/0022-5193(72)90180-4","article-title":"Iterative methods for the three-dimensional reconstruction of an object from projections","volume":"36","author":"Gilbert","year":"1972","journal-title":"J. Theor. Biol."},{"key":"mlstad38fdbib7","doi-asserted-by":"publisher","first-page":"1893","DOI":"10.3390\/nano13121893","article-title":"Spray flame synthesis and multiscale characterization of carbon black\u2013silica hetero-aggregates","volume":"13","author":"Buchheiser","year":"2023","journal-title":"Nanomaterials"},{"key":"mlstad38fdbib8","doi-asserted-by":"publisher","DOI":"10.1016\/j.chaos.2022.112246","article-title":"Fractal analysis of aggregates: Correlation between the 2D and 3D box-counting fractal dimension and power law fractal dimension","volume":"160","author":"Wang","year":"2022","journal-title":"Chaos Solitons Fractals"},{"key":"mlstad38fdbib9","doi-asserted-by":"publisher","first-page":"324","DOI":"10.1016\/j.powtec.2019.10.020","article-title":"Image-based size analysis of agglomerated and partially sintered particles via convolutional neural networks","volume":"360","author":"Frei","year":"2020","journal-title":"Powder Technol."},{"key":"mlstad38fdbib10","doi-asserted-by":"publisher","DOI":"10.1002\/nano.202300128","article-title":"Characterization of mixing in nanoparticle hetero-aggregates by convolutional neural networks","volume":"2024","author":"Mahr","year":"2024","journal-title":"Nano Select"},{"key":"mlstad38fdbib11","author":"Chiu","year":"2013","edition":"3rd edn"},{"key":"mlstad38fdbib12","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1038\/s42256-021-00322-1","article-title":"Generating three-dimensional structures from a two-dimensional slice with generative adversarial network-based dimensionality expansion","volume":"3","author":"Kench","year":"2021","journal-title":"Nat. Mach. Intell."},{"key":"mlstad38fdbib13","author":"Goodfellow","year":"2016"},{"key":"mlstad38fdbib14","author":"James","year":"2021","edition":"2nd edn"},{"key":"mlstad38fdbib15","doi-asserted-by":"publisher","first-page":"3397","DOI":"10.3390\/ma16093397","article-title":"CNN-based laue spot morphology predictor for reliable crystallographic descriptor estimation","volume":"16","author":"Kirstein","year":"2023","journal-title":"Materials"},{"key":"mlstad38fdbib16","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014"},{"key":"mlstad38fdbib17","first-page":"pp 96","article-title":"A comparison between VGG16, VGG19 and ResNet50 architecture frameworks for image classification","volume":"vol 1","author":"Mascarenhas","year":"2021"},{"key":"mlstad38fdbib18","first-page":"pp 2818","article-title":"Rethinking the inception architecture for computer vision","author":"Szegedy","year":"2016"},{"key":"mlstad38fdbib19","first-page":"pp 770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"mlstad38fdbib20","doi-asserted-by":"publisher","first-page":"5455","DOI":"10.1007\/s10462-020-09825-6","article-title":"A survey of the recent architectures of deep convolutional neural networks","volume":"53","author":"Khan","year":"2020","journal-title":"Artif. Intell. Rev."},{"key":"mlstad38fdbib21","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijsolstr.2023.112394","article-title":"A data-driven modeling approach to quantify morphology effects on transport properties in nanostructured NMC particles","volume":"280","author":"Neumann","year":"2023","journal-title":"Int. J. Solids Struct."},{"key":"mlstad38fdbib22","doi-asserted-by":"publisher","first-page":"155","DOI":"10.1093\/micmic\/ozac030","article-title":"Modeling curved fibers by fitting R-vine copulas to their frenet representations","volume":"29","author":"Weber","year":"2023","journal-title":"Microscopy Microanal."},{"key":"mlstad38fdbib23","doi-asserted-by":"publisher","first-page":"353","DOI":"10.1016\/j.commatsci.2016.03.013","article-title":"Stochastic 3D modeling of complex three-phase microstructures in SOFC-electrodes with completely connected phases","volume":"118","author":"Neumann","year":"2016","journal-title":"Comput. Mater. Sci."},{"key":"mlstad38fdbib24","doi-asserted-by":"publisher","DOI":"10.1016\/j.commatsci.2020.109934","article-title":"Generating digital twins of mesoporous silica by graph-based stochastic microstructure modeling","volume":"187","author":"Prifling","year":"2021","journal-title":"Comput. Mater. Sci."},{"key":"mlstad38fdbib25","doi-asserted-by":"publisher","DOI":"10.1016\/j.powtec.2023.119122","article-title":"Predicting 3D particles shapes based on 2D images by using convolutional neural network","volume":"432","author":"Giannis","year":"2024","journal-title":"Powder Technol."},{"key":"mlstad38fdbib26","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1007\/s11042-020-09722-8","article-title":"Single image 3D object reconstruction based on deep learning: a review","volume":"80","author":"Fu","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"mlstad38fdbib27","doi-asserted-by":"publisher","first-page":"L109","DOI":"10.1088\/0305-4470\/12\/5\/008","article-title":"Long-range correlations in smoke-particle aggregates","volume":"12","author":"Forrest","year":"1979","journal-title":"J. Phys. A: Math. Gen."},{"key":"mlstad38fdbib28","doi-asserted-by":"publisher","first-page":"1119","DOI":"10.1103\/PhysRevLett.51.1119","article-title":"Formation of fractal clusters and networks by irreversible diffusion-limited aggregation","volume":"51","author":"Meakin","year":"1983","journal-title":"Phys. Rev. Lett."},{"key":"mlstad38fdbib29","doi-asserted-by":"publisher","first-page":"470","DOI":"10.1006\/jcis.1995.1204","article-title":"Analysis of fractal cluster morphology parameters: Structural coefficient and density autocorrelation function cutoff","volume":"171","author":"Cai","year":"1995","journal-title":"J. Colloid Interface Sci."},{"key":"mlstad38fdbib30","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1080\/02786826.2011.631956","article-title":"The structure of agglomerates consisting of polydisperse particles","volume":"46","author":"Eggersdorfer","year":"2012","journal-title":"Aerosol Sci. Technol."},{"key":"mlstad38fdbib31","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1016\/j.apt.2013.10.010","article-title":"Agglomerates and aggregates of nanoparticles made in the gas phase","volume":"25","author":"Eggersdorfer","year":"2014","journal-title":"Adv. Powder Technol."},{"key":"mlstad38fdbib32","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1006\/jcis.2000.7027","article-title":"Fractal-like aggregates: Relation between morphology and physical properties","volume":"229","author":"Filippov","year":"2000","journal-title":"J. Colloid Interface Sci."},{"key":"mlstad38fdbib33","doi-asserted-by":"publisher","DOI":"10.1002\/ppsc.201800177","article-title":"Quantitative characterization of mixing in multicomponent nanoparticle aggregates","volume":"35","author":"Baric","year":"2018","journal-title":"Part. Part. Syst. Charact."},{"key":"mlstad38fdbib34","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1051\/jphyslet:01984004505021100","article-title":"Aggregation by kinetic clustering of clusters in dimensions d > 2","volume":"45","author":"Jullien","year":"1984","journal-title":"J. Phys. Lett."},{"key":"mlstad38fdbib35","doi-asserted-by":"publisher","first-page":"677","DOI":"10.1016\/j.ultramic.2009.01.001","article-title":"Is the frozen phonon model adequate to describe inelastic phonon scattering?","volume":"109","author":"Van Dyck","year":"2009","journal-title":"Ultramicroscopy"},{"key":"mlstad38fdbib36","first-page":"pp 170","article-title":"STEMSIM\u2014a new software tool for simulation of STEM HAADF Z-contrast imaging","author":"Rosenauer","year":"2008"},{"key":"mlstad38fdbib37","doi-asserted-by":"publisher","first-page":"1105","DOI":"10.1107\/S0567740878005014","article-title":"The crystal structure of triclinic WO3","volume":"34","author":"Diehl","year":"1978","journal-title":"Acta Crystallogr. B"},{"key":"mlstad38fdbib38","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1524\/zkri.1972.136.3-4.273","article-title":"Refinement of the structure of anatase at several temperatures","volume":"136","author":"Horn","year":"1972","journal-title":"Z. Kristallogr."},{"key":"mlstad38fdbib39","doi-asserted-by":"publisher","first-page":"1420","DOI":"10.1107\/S0567740869004146","article-title":"Further refinement of the structure of WO3","volume":"25","author":"Loopstra","year":"1969","journal-title":"Acta Crystallogr. B"},{"key":"mlstad38fdbib40","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1524\/zkri.1991.194.14.305","article-title":"The crystal structure of rutile as a function of temperature up to 1600 \u2218C","volume":"194","author":"Sugiyama","year":"1991","journal-title":"Z. Kristallogr."},{"key":"mlstad38fdbib41","author":"Kirkland","year":"1998"},{"key":"mlstad38fdbib42","doi-asserted-by":"publisher","first-page":"146","DOI":"10.1016\/j.ultramic.2015.10.026","article-title":"Effects of instrument imperfections on quantitative scanning transmission electron microscopy","volume":"161","author":"Krause","year":"2016","journal-title":"Ultramicroscopy"},{"key":"mlstad38fdbib43","doi-asserted-by":"publisher","first-page":"1050","DOI":"10.1017\/S1431927613001402","article-title":"Identifying and correcting scan noise and drift in the scanning transmission electron microscope","volume":"19","author":"Jones","year":"2013","journal-title":"Microscopy Microanal."},{"key":"mlstad38fdbib44","first-page":"pp 1778","article-title":"Preprocessing for image classification by convolutional neural networks","author":"Pal","year":"2016"},{"key":"mlstad38fdbib45","first-page":"pp 958","article-title":"Best practices for convolutional neural networks applied to visual document analysis","author":"Simard","year":"2003"},{"key":"mlstad38fdbib46","doi-asserted-by":"publisher","first-page":"213","DOI":"10.1145\/3510413","article-title":"Avoiding overfitting: a survey on regularization methods for convolutional neural networks","volume":"54","author":"Santos","year":"2022","journal-title":"ACM Comput. Surv."},{"key":"mlstad38fdbib47","first-page":"pp 169","article-title":"Compressed residual-vgg16 cnn model for big data places image recognition","author":"Qassim","year":"2018"},{"key":"mlstad38fdbib48","first-page":"pp 448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","author":"Ioffe","year":"2015"},{"key":"mlstad38fdbib49","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2015"},{"key":"mlstad38fdbib50","article-title":"Review and comparison of commonly used activation functions for deep neural networks","author":"Szanda\u0142a","year":"2020"},{"key":"mlstad38fdbib51","first-page":"pp 243","article-title":"Machine learning with oversampling and undersampling techniques: overview study and experimental results","author":"Mohammed","year":"2020"},{"key":"mlstad38fdbib52","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/S0001-8686(00)00078-6","article-title":"On techniques for the measurement of the mass fractal dimension of aggregates","volume":"95","author":"Bushell","year":"2002","journal-title":"Adv. Colloid Interface Sci."},{"key":"mlstad38fdbib53","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.69.011405","article-title":"Method for computing the three-dimensional capacity dimension from two-dimensional projections of fractal aggregates","volume":"69","author":"Maggi","year":"2004","journal-title":"Phys. Rev. E"},{"key":"mlstad38fdbib54","doi-asserted-by":"publisher","DOI":"10.1002\/ppsc.202300048","article-title":"Material discrimination in nanoparticle hetero-aggregates by analysis of scanning transmission electron microscopy images","volume":"40","author":"Gerken","year":"2023","journal-title":"Part. Part. Syst. Charact."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,4,8]],"date-time":"2024-04-08T07:50:05Z","timestamp":1712562605000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad38fd"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,8]]},"references-count":54,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,4,8]]},"published-print":{"date-parts":[[2024,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ad38fd","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,4,8]]},"assertion":[{"value":"Using convolutional neural networks for stereological characterization of 3D hetero-aggregates based on synthetic STEM data","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2024 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2023-10-27","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-03-28","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-04-08","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}