The assembly of misfolded proteins into amyloid fibrils is associated with amyloidosis, including neurodegenerative diseases, such as Alzheimer's, Parkinson's, and prion diseases. The self-propagation of amyloid fibrils is widely observed in the aggregation pathways of numerous amyloidogenic proteins. This propensity with plasticity in primary nucleation allows amyloid fibril polymorphism, which is correlated with the pathology/phenotypes of patients. Because the interference with the nucleation and replication processes of amyloid fibrils can alter the amyloid structure and the outcome of the disease, these processes can be a target for developing clinical drugs. Single-molecule observation of amyloid fibril replication can be an experimental system to provide the kinetic parameters for simulation studies and confirm the effect of clinical drugs. Here, we review single-molecule observation of the amyloid fibril replication process using fluorescence microscopy and time-lapse atomic force microscopy, including high-speed atomic force microscopy. We discussed the amyloid fibril replication process and combined single-molecule observation results with molecular dynamics simulations. Mini Abstract Structural dynamics in amyloid aggregation is related with various Alzheimer's and Parkinson's disease symptoms. Single-molecule observation using high-speed atomic force microscopy can directly visualize the structural dynamics of individual amyloid aggregate assemblies. Here, we review historical and recent studies of single-molecule observation of amyloid aggregation with supportive molecular dynamics simulation.
{"title":"Single-molecule Observation of Self-Propagating Amyloid Fibrils.","authors":"T. Watanabe-Nakayama, K. Ono","doi":"10.1093/jmicro/dfac011","DOIUrl":"https://doi.org/10.1093/jmicro/dfac011","url":null,"abstract":"The assembly of misfolded proteins into amyloid fibrils is associated with amyloidosis, including neurodegenerative diseases, such as Alzheimer's, Parkinson's, and prion diseases. The self-propagation of amyloid fibrils is widely observed in the aggregation pathways of numerous amyloidogenic proteins. This propensity with plasticity in primary nucleation allows amyloid fibril polymorphism, which is correlated with the pathology/phenotypes of patients. Because the interference with the nucleation and replication processes of amyloid fibrils can alter the amyloid structure and the outcome of the disease, these processes can be a target for developing clinical drugs. Single-molecule observation of amyloid fibril replication can be an experimental system to provide the kinetic parameters for simulation studies and confirm the effect of clinical drugs. Here, we review single-molecule observation of the amyloid fibril replication process using fluorescence microscopy and time-lapse atomic force microscopy, including high-speed atomic force microscopy. We discussed the amyloid fibril replication process and combined single-molecule observation results with molecular dynamics simulations. Mini Abstract Structural dynamics in amyloid aggregation is related with various Alzheimer's and Parkinson's disease symptoms. Single-molecule observation using high-speed atomic force microscopy can directly visualize the structural dynamics of individual amyloid aggregate assemblies. Here, we review historical and recent studies of single-molecule observation of amyloid aggregation with supportive molecular dynamics simulation.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41614649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Electron microscopy of biological materials such as bacteria allows multifaceted analysis to understand their structure and function with high resolution, which is difficult to achieve with optical microscopy. However, the samples are damaged or broken by electron beam irradiation and by the vacuum environment. Here, we observed bacteria in a suspension encapsulated in a graphene sandwich that prevents electron beam damage without the need for fixation. Specifically, we demonstrated in situ scanning electron microscopy observation of Escherichia coli in a graphene sandwich containing a perforated membrane as a spacer, encapsulating non-immobilized E. coli between the graphene layers. However, E. coli activity, such as division, was not observed, although the irradiated cells grew slightly when resuspended under optimal culture conditions. Our findings suggest that the graphene sandwich methodology enables the observation of wet E. coli cells by electron microscopy but requires refinement to allow the live imaging of biological materials.
{"title":"Scanning electron microscopy of Escherichia coli encapsulated in a spacerized graphene sandwich.","authors":"Yuki Sasaki, S. Hirayama, R. Nakao","doi":"10.1093/jmicro/dfac010","DOIUrl":"https://doi.org/10.1093/jmicro/dfac010","url":null,"abstract":"Electron microscopy of biological materials such as bacteria allows multifaceted analysis to understand their structure and function with high resolution, which is difficult to achieve with optical microscopy. However, the samples are damaged or broken by electron beam irradiation and by the vacuum environment. Here, we observed bacteria in a suspension encapsulated in a graphene sandwich that prevents electron beam damage without the need for fixation. Specifically, we demonstrated in situ scanning electron microscopy observation of Escherichia coli in a graphene sandwich containing a perforated membrane as a spacer, encapsulating non-immobilized E. coli between the graphene layers. However, E. coli activity, such as division, was not observed, although the irradiated cells grew slightly when resuspended under optimal culture conditions. Our findings suggest that the graphene sandwich methodology enables the observation of wet E. coli cells by electron microscopy but requires refinement to allow the live imaging of biological materials.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43821219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nowadays, sub-50 meV atom-wide electron probes are routinely produced for electron energy loss spectroscopy in transmission electron microscopes due to monochromator technology advances. We review how gradual improvements in energy resolution enabled the study of very low-energy excitations such as lattice phonons, molecular vibrations, infrared plasmons and strongly coupled hybrid modes in nanomaterials. Starting with the theoretical framework needed to treat inelastic electron scattering from phonons in solids, we illustrate contributions in detecting optical surface phonons in photonic structures. We discuss phonon mapping capabilities in real and reciprocal space, and the localized phonon response near nano-/atomic-scale structural features. We also survey the progress of aloof spectroscopy in studying vibrations in organic materials and applications in measuring local temperature and photonic density of states in single nanostructures using phonon scattering. We then turn towards studies on infrared plasmons in metals and semiconductors. Spectroscopy analyses now extend towards probing extremely complex broadband platforms, the effects of defects and nanogaps, and some far-reaching investigations towards uncovering plasmon lifetime and 3D photonic density of states. In doped semiconductors, we review research on the use of the electron probe to correlate local doping concentration and atomic-scale defects with the plasmonic response. Finally, we discuss advances in studying strong coupling phenomena in plasmon-exciton and plasmon-phonon systems. Overall, the wealth of information gained extends our knowledge about nanomaterial properties and elementary excitations, illustrating the powerful capabilities of high-energy resolution scanning transmission electron microscopy-electron energy loss spectrometry.
{"title":"Advances in ultrahigh-energy resolution EELS: phonons, infrared plasmons and strongly coupled modes.","authors":"M. Lagos, I. Bicket, S. Mousavi M, G. Botton","doi":"10.1093/jmicro/dfab050","DOIUrl":"https://doi.org/10.1093/jmicro/dfab050","url":null,"abstract":"Nowadays, sub-50 meV atom-wide electron probes are routinely produced for electron energy loss spectroscopy in transmission electron microscopes due to monochromator technology advances. We review how gradual improvements in energy resolution enabled the study of very low-energy excitations such as lattice phonons, molecular vibrations, infrared plasmons and strongly coupled hybrid modes in nanomaterials. Starting with the theoretical framework needed to treat inelastic electron scattering from phonons in solids, we illustrate contributions in detecting optical surface phonons in photonic structures. We discuss phonon mapping capabilities in real and reciprocal space, and the localized phonon response near nano-/atomic-scale structural features. We also survey the progress of aloof spectroscopy in studying vibrations in organic materials and applications in measuring local temperature and photonic density of states in single nanostructures using phonon scattering. We then turn towards studies on infrared plasmons in metals and semiconductors. Spectroscopy analyses now extend towards probing extremely complex broadband platforms, the effects of defects and nanogaps, and some far-reaching investigations towards uncovering plasmon lifetime and 3D photonic density of states. In doped semiconductors, we review research on the use of the electron probe to correlate local doping concentration and atomic-scale defects with the plasmonic response. Finally, we discuss advances in studying strong coupling phenomena in plasmon-exciton and plasmon-phonon systems. Overall, the wealth of information gained extends our knowledge about nanomaterial properties and elementary excitations, illustrating the powerful capabilities of high-energy resolution scanning transmission electron microscopy-electron energy loss spectrometry.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44622865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
H. Terashima, Keiichiro Minatohara, Hisato Maruoka, S. Okabe
Recent advances in human genetics identified genetic variants involved in causing autism spectrum disorders (ASDs). Mouse models that mimic mutations found in patients with ASD exhibit behavioral phenotypes consistent with ASD symptoms. These mouse models suggest critical biological factors of ASD etiology. Another important implication of ASD genetics is the enrichment of ASD risk genes in molecules involved in developing synapses and regulating neural circuit function. Sophisticated in vivo imaging technologies applied to ASD mouse models identify common synaptic impairments in the neocortex, with genetic-mutation-specific defects in local neural circuits. In this article, we review synapse- and circuit-level phenotypes identified by in vivo two-photon imaging in multiple mouse models of ASD and discuss the contributions of altered synapse properties and neural circuit activity to ASD pathogenesis.
{"title":"Imaging neural circuit pathology of autism spectrum disorders: autism-associated genes, animal models and the application of in vivo two-photon imaging.","authors":"H. Terashima, Keiichiro Minatohara, Hisato Maruoka, S. Okabe","doi":"10.1093/jmicro/dfab039","DOIUrl":"https://doi.org/10.1093/jmicro/dfab039","url":null,"abstract":"Recent advances in human genetics identified genetic variants involved in causing autism spectrum disorders (ASDs). Mouse models that mimic mutations found in patients with ASD exhibit behavioral phenotypes consistent with ASD symptoms. These mouse models suggest critical biological factors of ASD etiology. Another important implication of ASD genetics is the enrichment of ASD risk genes in molecules involved in developing synapses and regulating neural circuit function. Sophisticated in vivo imaging technologies applied to ASD mouse models identify common synaptic impairments in the neocortex, with genetic-mutation-specific defects in local neural circuits. In this article, we review synapse- and circuit-level phenotypes identified by in vivo two-photon imaging in multiple mouse models of ASD and discuss the contributions of altered synapse properties and neural circuit activity to ASD pathogenesis.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45970709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Genetically encoded tags have introduced extensive lines of application from purification of tagged proteins to their visualization at the single molecular, cellular, histological and whole-body levels. Combined with other rapidly developing technologies such as clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) system, proteomics, super-resolution microscopy and proximity labeling, a large variety of genetically encoded tags have been developed in the last two decades. In this review, I focus on the current status of tag development for electron microscopic (EM) visualization of proteins with metal particle labeling. Compared with conventional immunoelectron microscopy using gold particles, tag-mediated metal particle labeling has several advantages that could potentially improve the sensitivity, spatial and temporal resolution, and applicability to a wide range of proteins of interest (POIs). It may enable researchers to detect single molecules in situ, allowing the quantitative measurement of absolute numbers and exact localization patterns of POI in the ultrastructural context. Thus, genetically encoded tags for EM could revolutionize the field as green fluorescence protein did for light microscopy, although we still have many challenges to overcome before reaching this goal.
{"title":"Electron microscopic visualization of single molecules by tag-mediated metal particle labeling.","authors":"R. Shigemoto","doi":"10.1093/jmicro/dfab048","DOIUrl":"https://doi.org/10.1093/jmicro/dfab048","url":null,"abstract":"Genetically encoded tags have introduced extensive lines of application from purification of tagged proteins to their visualization at the single molecular, cellular, histological and whole-body levels. Combined with other rapidly developing technologies such as clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) system, proteomics, super-resolution microscopy and proximity labeling, a large variety of genetically encoded tags have been developed in the last two decades. In this review, I focus on the current status of tag development for electron microscopic (EM) visualization of proteins with metal particle labeling. Compared with conventional immunoelectron microscopy using gold particles, tag-mediated metal particle labeling has several advantages that could potentially improve the sensitivity, spatial and temporal resolution, and applicability to a wide range of proteins of interest (POIs). It may enable researchers to detect single molecules in situ, allowing the quantitative measurement of absolute numbers and exact localization patterns of POI in the ultrastructural context. Thus, genetically encoded tags for EM could revolutionize the field as green fluorescence protein did for light microscopy, although we still have many challenges to overcome before reaching this goal.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"60920361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Yokoyama, H. Takahashi, S. Koshiya, T. Murano, M. Terauchi
Abstract The method deriving the L self-absorption spectrum from Lα,β emission spectra obtained at different accelerating voltages has been optimized for analyzing the chemical state of Fe in solid materials. Fe Lα,β emission spectra obtained are fitted using Pseudo-Voigt functions and normalized by the integrated intensity of each Fe Ll line, which is not affected by L2,3 absorption edge. The self-absorption spectrum is calculated by dividing the normalized intensity profile collected at low accelerating voltage by that collected at a higher accelerating voltage. The obtained profile is referred to as soft X-ray self-absorption structure (SX-SAS). This method is applied to six Fe-based materials (Fe metal, FeO, Fe3O4, Fe2O3, FeS and FeS2) to observe different chemical states of Fe in those materials. By comparing the self-absorption spectra of iron oxides, one can observe the L3 absorption peak structure shows a shift to the higher energy side as ferric (3+) Fe increases with respect to ferrous (+2) Fe. The intensity profiles of self-absorption spectra of metallic Fe and FeS2 shows shoulder structures between the L3 and L2 absorption peaks, which were not observed in spectra of Fe oxides. These results indicate that the SX-SAS technique is useful to examine X-ray absorption structure as a means to understand the chemical states of transition metal elements.
{"title":"Analytical technique for self-absorption structure of iron L-emission spectra obtained by soft X-ray emission spectrometer","authors":"T. Yokoyama, H. Takahashi, S. Koshiya, T. Murano, M. Terauchi","doi":"10.1093/jmicro/dfac009","DOIUrl":"https://doi.org/10.1093/jmicro/dfac009","url":null,"abstract":"Abstract The method deriving the L self-absorption spectrum from Lα,β emission spectra obtained at different accelerating voltages has been optimized for analyzing the chemical state of Fe in solid materials. Fe Lα,β emission spectra obtained are fitted using Pseudo-Voigt functions and normalized by the integrated intensity of each Fe Ll line, which is not affected by L2,3 absorption edge. The self-absorption spectrum is calculated by dividing the normalized intensity profile collected at low accelerating voltage by that collected at a higher accelerating voltage. The obtained profile is referred to as soft X-ray self-absorption structure (SX-SAS). This method is applied to six Fe-based materials (Fe metal, FeO, Fe3O4, Fe2O3, FeS and FeS2) to observe different chemical states of Fe in those materials. By comparing the self-absorption spectra of iron oxides, one can observe the L3 absorption peak structure shows a shift to the higher energy side as ferric (3+) Fe increases with respect to ferrous (+2) Fe. The intensity profiles of self-absorption spectra of metallic Fe and FeS2 shows shoulder structures between the L3 and L2 absorption peaks, which were not observed in spectra of Fe oxides. These results indicate that the SX-SAS technique is useful to examine X-ray absorption structure as a means to understand the chemical states of transition metal elements.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47496946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract It is difficult to discriminate the amorphous state using a transmission electron microscope (TEM). We discriminated different amorphous states on TEM images using persistent homology, which is a mathematical analysis technique that employs the homology concept and focuses on ‘holes’. The structural models of the different amorphous states, that is, amorphous and liquid states, were created using classical molecular dynamic simulation. TEM images in several defocus conditions were simulated by the multi-slice method using the created amorphous and liquid states, and their persistent diagrams were calculated. Finally, logistic regression and support vector classification machine learning algorithms were applied for discrimination. Consequently, we found that the amorphous and liquid phases can be discriminated by more than 85%. Because the contrast of TEM images depends on sample thickness, focus, lens aberration, etc., radial distribution function cannot be classified; however, the persistent homology can discriminate different amorphous states in a wide focus range.
{"title":"Classification for transmission electron microscope images from different amorphous states using persistent homology","authors":"Fumihiko Uesugi, M. Ishii","doi":"10.1093/jmicro/dfac008","DOIUrl":"https://doi.org/10.1093/jmicro/dfac008","url":null,"abstract":"Abstract It is difficult to discriminate the amorphous state using a transmission electron microscope (TEM). We discriminated different amorphous states on TEM images using persistent homology, which is a mathematical analysis technique that employs the homology concept and focuses on ‘holes’. The structural models of the different amorphous states, that is, amorphous and liquid states, were created using classical molecular dynamic simulation. TEM images in several defocus conditions were simulated by the multi-slice method using the created amorphous and liquid states, and their persistent diagrams were calculated. Finally, logistic regression and support vector classification machine learning algorithms were applied for discrimination. Consequently, we found that the amorphous and liquid phases can be discriminated by more than 85%. Because the contrast of TEM images depends on sample thickness, focus, lens aberration, etc., radial distribution function cannot be classified; however, the persistent homology can discriminate different amorphous states in a wide focus range.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45762873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shoko Ishikawa, M. Nikaido, Takahito Otani, Kayoko Ogata, H. Iida, Yuko Inai, Sachio Tamaoki, T. Inai
Abstract Retinoic acid (RA) plays an important role in epithelial homeostasis and influences the morphology, proliferation, differentiation and permeability of epithelial cells. Mouse keratinocytes, K38, reconstituted non-keratinized stratified epithelium in three-dimensional (3D) cultures with serum, which contains retinol (a source of RA), but the morphology was different from in vivo epithelium. The formed epithelium was thick, with loosened cell–cell contacts. Here, we investigated whether the inhibition of RA receptor (RAR)/retinoid X receptor (RXR)-mediated signaling by an RXR antagonist, HX 531, improved K38 3D cultures in terms of morphology and intercellular junctions. The epithelium formed by 0.5 μM HX531 was thin, and the intercellular space was narrowed because of the restoration of the layer-specific distribution of desmoglein (DSG)-1, DSG3 and plakoglobin (PG). Moreover, the levels of desmosomal proteins and tight junction proteins, including DSG1, DSG2, DSG3, PG, claudin (CLDN)-1 and CLDN4 increased, but the adherens junction protein, E-cadherin, did not show any change. Furthermore, CLDN1 was recruited to occludin-positive cell–cell contacts in the superficial cells and transepithelial electrical resistance was increased. Therefore, K38 3D cultures treated with 0.5 μM HX531 provides a useful in vitro model to study intercellular junctions in the non-keratinized epithelium.
{"title":"Inhibition of retinoid X receptor improved the morphology, localization of desmosomal proteins and paracellular permeability in three-dimensional cultures of mouse keratinocytes","authors":"Shoko Ishikawa, M. Nikaido, Takahito Otani, Kayoko Ogata, H. Iida, Yuko Inai, Sachio Tamaoki, T. Inai","doi":"10.1093/jmicro/dfac007","DOIUrl":"https://doi.org/10.1093/jmicro/dfac007","url":null,"abstract":"Abstract Retinoic acid (RA) plays an important role in epithelial homeostasis and influences the morphology, proliferation, differentiation and permeability of epithelial cells. Mouse keratinocytes, K38, reconstituted non-keratinized stratified epithelium in three-dimensional (3D) cultures with serum, which contains retinol (a source of RA), but the morphology was different from in vivo epithelium. The formed epithelium was thick, with loosened cell–cell contacts. Here, we investigated whether the inhibition of RA receptor (RAR)/retinoid X receptor (RXR)-mediated signaling by an RXR antagonist, HX 531, improved K38 3D cultures in terms of morphology and intercellular junctions. The epithelium formed by 0.5 μM HX531 was thin, and the intercellular space was narrowed because of the restoration of the layer-specific distribution of desmoglein (DSG)-1, DSG3 and plakoglobin (PG). Moreover, the levels of desmosomal proteins and tight junction proteins, including DSG1, DSG2, DSG3, PG, claudin (CLDN)-1 and CLDN4 increased, but the adherens junction protein, E-cadherin, did not show any change. Furthermore, CLDN1 was recruited to occludin-positive cell–cell contacts in the superficial cells and transepithelial electrical resistance was increased. Therefore, K38 3D cultures treated with 0.5 μM HX531 provides a useful in vitro model to study intercellular junctions in the non-keratinized epithelium.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2022-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42852543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For quality control of special steels, the microstructure of the steel is visually inspected on the basis of microscopic images. In this study, aiming to eliminate the effect of personal differences between inspectors and reduce inspection costs, a system for automatically estimating quality level (hereafter, “automatic-quality-level-estimation system ‘’) based on machine learning is proposed and evaluated. Collecting the images is a manual task performed by the inspector, and it is difficult to prepare multiple training samples in advance. As for the proposed method, overfitting, which is a problem in training with few samples, is suppressed by data expansion based on variation distribution of correct-answer values. The correct-answer rate for judging quality level by an inspector was about 90%, while the proposed method achieved a rate of 90%, which is sufficient to render the method practically applicable.
{"title":"Machine-learning-based quality-level-estimation system for inspecting steel microstructures","authors":"Hiromi Nishiura, A. Miyamoto, Akira Ito, Shogo Suzuki, Kouhei Fujii, Hiroshi Morifuji, Hiroyuki Takatsuka","doi":"10.1093/jmicro/dfac019","DOIUrl":"https://doi.org/10.1093/jmicro/dfac019","url":null,"abstract":"For quality control of special steels, the microstructure of the steel is visually inspected on the basis of microscopic images. In this study, aiming to eliminate the effect of personal differences between inspectors and reduce inspection costs, a system for automatically estimating quality level (hereafter, “automatic-quality-level-estimation system ‘’) based on machine learning is proposed and evaluated. Collecting the images is a manual task performed by the inspector, and it is difficult to prepare multiple training samples in advance. As for the proposed method, overfitting, which is a problem in training with few samples, is suppressed by data expansion based on variation distribution of correct-answer values. The correct-answer rate for judging quality level by an inspector was about 90%, while the proposed method achieved a rate of 90%, which is sufficient to render the method practically applicable.","PeriodicalId":48655,"journal":{"name":"Microscopy","volume":null,"pages":null},"PeriodicalIF":1.8,"publicationDate":"2021-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44301133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：