Bone age assessment(BAA)helps doctors determine how a child’s bones grow and develop in clinical medicine.Traditional BAA methods rely on clinician expertise,leading to time-consuming predictions and inaccurate resul...Bone age assessment(BAA)helps doctors determine how a child’s bones grow and develop in clinical medicine.Traditional BAA methods rely on clinician expertise,leading to time-consuming predictions and inaccurate results.Most deep learning-based BAA methods feed the extracted critical points of images into the network by providing additional annotations.This operation is costly and subjective.To address these problems,we propose a multi-scale attentional densely connected network(MSADCN)in this paper.MSADCN constructs a multi-scale dense connectivity mechanism,which can avoid overfitting,obtain the local features effectively and prevent gradient vanishing even in limited training data.First,MSADCN designs multi-scale structures in the densely connected network to extract fine-grained features at different scales.Then,coordinate attention is embedded to focus on critical features and automatically locate the regions of interest(ROI)without additional annotation.In addition,to improve the model’s generalization,transfer learning is applied to train the proposed MSADCN on the public dataset IMDB-WIKI,and the obtained pre-trained weights are loaded onto the Radiological Society of North America(RSNA)dataset.Finally,label distribution learning(LDL)and expectation regression techniques are introduced into our model to exploit the correlation between hand bone images of different ages,which can obtain stable age estimates.Extensive experiments confirm that our model can converge more efficiently and obtain a mean absolute error(MAE)of 4.64 months,outperforming some state-of-the-art BAA methods.展开更多
Complex brain diseases seriously endanger human health,and early diagnostic biomarkers and effective treatments are currently lacking.Due to ethical constraints on human research,establishing monkey models is crucial ...Complex brain diseases seriously endanger human health,and early diagnostic biomarkers and effective treatments are currently lacking.Due to ethical constraints on human research,establishing monkey models is crucial to address these issues.With the rapid development of technology,transgenic monkey models of a range of brain diseases,especially autism spectrum disorder(ASD),have been successfully established.However,to establish practical and effective brain disease models and subsequently apply them to disease mechanism and treatment studies,there is still a lack of a standard tool,i.e.,a system for collecting and analyzing the daily behaviors of brain disease model monkeys.Therefore,with the goal of undertaking a comprehensive and quantitative study of behavioral phenotypes,we established a standard daily behavior collection and analysis system,including behavioral data collection protocols and a monkey daily behavior ethogram(MDBE)for rhesus and cynomolgus monkeys,which are the most commonly used non-human primates in model construction.Then,we used ASD as an application example after referring to the Diagnostic and Statistical Manual of Mental Disorders,Fifth Edition,Text Revision(DSM-5-TR),which is widely used in clinical disease diagnosis to obtain ASD core clinical symptoms.We then established a sub-ethogram(ASD monkey core behavior ethogram(MCBE-ASD))specifically for quantitative assessment of the core clinical symptoms of an ASD monkey model based on MDBE.Subsequently,we demonstrated the high reproducibility of the system.展开更多
Objective To introduce an end-to-end automatic segmentation method for organs at risk(OARs)in chest computed tomography(CT)images based on dense connection deep learning and to provide an accurate auto-segmentation mo...Objective To introduce an end-to-end automatic segmentation method for organs at risk(OARs)in chest computed tomography(CT)images based on dense connection deep learning and to provide an accurate auto-segmentation model to reduce the workload on radiation oncologists.Methods CT images of 36 lung cancer cases were included in this study.Of these,27 cases were randomly selected as the training set,six cases as the validation set,and nine cases as the testing set.The left and right lungs,cord,and heart were auto-segmented,and the training time was set to approximately 5 h.The testing set was evaluated using geometric metrics including the Dice similarity coefficient(DSC),95%Hausdorff distance(HD95),and average surface distance(ASD).Thereafter,two sets of treatment plans were optimized based on manually contoured OARs and automatically contoured OARs,respectively.Dosimetric parameters including Dmax and Vx of the OARs were obtained and compared.Results The proposed model was superior to U-Net in terms of the DSC,HD95,and ASD,although there was no significant difference in the segmentation results yielded by both networks(P>0.05).Compared to manual segmentation,auto-segmentation significantly reduced the segmentation time by nearly 40.7%(P<0.05).Moreover,the differences in dose-volume parameters between the two sets of plans were not statistically significant(P>0.05).Conclusion The bilateral lung,cord,and heart could be accurately delineated using the DenseNet-based deep learning method.Thus,feature map reuse can be a novel approach to medical image auto-segmentation.展开更多
The fungal cell wall is the front-line in host-pathogen interactions,which is an essential dynamic structure maintaining cellular integrity and protecting the fungal cell from external aggressors,such as environmental...The fungal cell wall is the front-line in host-pathogen interactions,which is an essential dynamic structure maintaining cellular integrity and protecting the fungal cell from external aggressors,such as environmental stress,or during host infection(Geoghegan et al.,2017;Gow et al.,2017).Its synthesizing and remodeling/reinforcement are controlled by the cell wall integrity(CWl)pathway(Riquelme et al.,2018).The CWI pathway is a conserved signalling transduction cascade and is well characterized in the model yeast Saccharomyces cerevisiae.In phytopathogenic fungi,it exhibits more speciesspecific functions to suit their distinct invasion strategies.For many appressorium-forming plant pathogens such as the rice blast fungus Magnaporthe oryzae and Colletotrichum gloeosporioide,CWl pathway regulates the development and functioning of appressorium,which is required for pathogenicity(Jeon et al.,2008;Yin et al.,2016,2020;Fang et al.,2018).展开更多
Numerous fluorescent marker lines are currently available to visualize microtubule(MT)architecture and dynamics in living plant cells, such as markers expressing p35S::GFP-MBD or p35S::GFP-TUB6.However, these MT marke...Numerous fluorescent marker lines are currently available to visualize microtubule(MT)architecture and dynamics in living plant cells, such as markers expressing p35S::GFP-MBD or p35S::GFP-TUB6.However, these MT marker lines display obvious defects that affect plant growth or produce unstable fluorescent signals. Here, a series of new marker lines were developed, including the pTUB6::VisGreen-TUB6-expressing line in which TUB6 is under the control of its endogenous regulatory elements and e GFP is replaced with VisGreen, a brighter fluorescent protein. Moreover, two different markers were combined into one expression vector and developed two dual-marker lines.These marker lines produce bright, stable fluorescent signals in various tissues, and greatly shorten the screening process for generating dual-marker lines.These new marker lines provide a novel resource for MT research.展开更多
Legumes have evolved a symbiotic relationship with rhizobial bacteria and their roots form unique nitrogen-fixing organs called nodules.Studies have shown that abiotic and biotic stresses alter the profile of gene exp...Legumes have evolved a symbiotic relationship with rhizobial bacteria and their roots form unique nitrogen-fixing organs called nodules.Studies have shown that abiotic and biotic stresses alter the profile of gene expression and transcript mobility in plants.However,little is known about the systemic transport of RNA between roots and shoots in response to rhizobial infection on a genome-wide scale during the formation of legume-rhizobia symbiosis.In our study,we found that two soybean(Glycine max)cultivars,Peking and Williams,show a high frequency of single nucleotide polymorphisms;this allowed us to characterize the origin and mobility of transcripts in hetero-grafts of these two cultivars.We identified 4,552 genes that produce mobile RNAs in soybean,and found that rhizobial infection triggers mass transport of m RNAs between shoots and roots at the early stage of nodulation.The majority of these mRNAs are of relatively low abundance and their transport occurs in a selective manner in soybean plants.Notably,the mRNAs that moved from shoots to roots at the early stage of nodulation were enriched in many nodule-related responsive processes.Moreover,the transcripts of many known symbiosis-related genes that are induced by rhizobial infection can move between shoots and roots.Our findings provide a deeper understanding of endogenous RNA transport in legume-rhizobia symbiotic processes.展开更多
基金This research is partially supported by grant from the National Natural Science Foundation of China(No.72071019)grant from the Natural Science Foundation of Chongqing(No.cstc2021jcyj-msxmX0185)grant from the Chongqing Graduate Education and Teaching Reform Research Project(No.yjg193096).
文摘Bone age assessment(BAA)helps doctors determine how a child’s bones grow and develop in clinical medicine.Traditional BAA methods rely on clinician expertise,leading to time-consuming predictions and inaccurate results.Most deep learning-based BAA methods feed the extracted critical points of images into the network by providing additional annotations.This operation is costly and subjective.To address these problems,we propose a multi-scale attentional densely connected network(MSADCN)in this paper.MSADCN constructs a multi-scale dense connectivity mechanism,which can avoid overfitting,obtain the local features effectively and prevent gradient vanishing even in limited training data.First,MSADCN designs multi-scale structures in the densely connected network to extract fine-grained features at different scales.Then,coordinate attention is embedded to focus on critical features and automatically locate the regions of interest(ROI)without additional annotation.In addition,to improve the model’s generalization,transfer learning is applied to train the proposed MSADCN on the public dataset IMDB-WIKI,and the obtained pre-trained weights are loaded onto the Radiological Society of North America(RSNA)dataset.Finally,label distribution learning(LDL)and expectation regression techniques are introduced into our model to exploit the correlation between hand bone images of different ages,which can obtain stable age estimates.Extensive experiments confirm that our model can converge more efficiently and obtain a mean absolute error(MAE)of 4.64 months,outperforming some state-of-the-art BAA methods.
基金supported by the Key-Area Research and Development Program of Guangdong Province(No.2019B030335001)the STI2030-Major Projects(No.2021ZD0200900)+9 种基金the National Key Research and Development Program of China(Nos.2021YFF0702700 and 2018YFA0801403)the National Natural Science Foundation of China(Nos.81960422,32100801,82101241,82360226,82160923,82260929,82374425,and 32460194)the Strategic Priority Research Program of the Chinese Academy of Sciences(CAS)(No.XDB32060200)the STI2030-Major Projects(Nos.2022ZD0205200 and 2022ZD0212700)the Science and Technology Department of Yunnan Province(Nos.202305AH340006 and 202305AH340007)the Yunnan Fundamental Research Projects(Nos.202201AT070153 and 202201AT070139)the Science and Technology Project of Yunnan Province(No.202101AY070001-001)the Yunnan Revitalization Talent Support Program(No.YNWR-QNBJ2019-043)the CAS“Light of West China”Programthe Yunnan Revitalization Talents Support Plan。
文摘Complex brain diseases seriously endanger human health,and early diagnostic biomarkers and effective treatments are currently lacking.Due to ethical constraints on human research,establishing monkey models is crucial to address these issues.With the rapid development of technology,transgenic monkey models of a range of brain diseases,especially autism spectrum disorder(ASD),have been successfully established.However,to establish practical and effective brain disease models and subsequently apply them to disease mechanism and treatment studies,there is still a lack of a standard tool,i.e.,a system for collecting and analyzing the daily behaviors of brain disease model monkeys.Therefore,with the goal of undertaking a comprehensive and quantitative study of behavioral phenotypes,we established a standard daily behavior collection and analysis system,including behavioral data collection protocols and a monkey daily behavior ethogram(MDBE)for rhesus and cynomolgus monkeys,which are the most commonly used non-human primates in model construction.Then,we used ASD as an application example after referring to the Diagnostic and Statistical Manual of Mental Disorders,Fifth Edition,Text Revision(DSM-5-TR),which is widely used in clinical disease diagnosis to obtain ASD core clinical symptoms.We then established a sub-ethogram(ASD monkey core behavior ethogram(MCBE-ASD))specifically for quantitative assessment of the core clinical symptoms of an ASD monkey model based on MDBE.Subsequently,we demonstrated the high reproducibility of the system.
基金Supported by a grant from the Beijing Municipal Science and Technology Commission Foundation Programme(No.Z181100001718011).
文摘Objective To introduce an end-to-end automatic segmentation method for organs at risk(OARs)in chest computed tomography(CT)images based on dense connection deep learning and to provide an accurate auto-segmentation model to reduce the workload on radiation oncologists.Methods CT images of 36 lung cancer cases were included in this study.Of these,27 cases were randomly selected as the training set,six cases as the validation set,and nine cases as the testing set.The left and right lungs,cord,and heart were auto-segmented,and the training time was set to approximately 5 h.The testing set was evaluated using geometric metrics including the Dice similarity coefficient(DSC),95%Hausdorff distance(HD95),and average surface distance(ASD).Thereafter,two sets of treatment plans were optimized based on manually contoured OARs and automatically contoured OARs,respectively.Dosimetric parameters including Dmax and Vx of the OARs were obtained and compared.Results The proposed model was superior to U-Net in terms of the DSC,HD95,and ASD,although there was no significant difference in the segmentation results yielded by both networks(P>0.05).Compared to manual segmentation,auto-segmentation significantly reduced the segmentation time by nearly 40.7%(P<0.05).Moreover,the differences in dose-volume parameters between the two sets of plans were not statistically significant(P>0.05).Conclusion The bilateral lung,cord,and heart could be accurately delineated using the DenseNet-based deep learning method.Thus,feature map reuse can be a novel approach to medical image auto-segmentation.
基金supported by the National Key Research and Development Program of China(No.2022YFD1200300)by grants from the State Key Laboratory of Plant Genomics.
文摘The fungal cell wall is the front-line in host-pathogen interactions,which is an essential dynamic structure maintaining cellular integrity and protecting the fungal cell from external aggressors,such as environmental stress,or during host infection(Geoghegan et al.,2017;Gow et al.,2017).Its synthesizing and remodeling/reinforcement are controlled by the cell wall integrity(CWl)pathway(Riquelme et al.,2018).The CWI pathway is a conserved signalling transduction cascade and is well characterized in the model yeast Saccharomyces cerevisiae.In phytopathogenic fungi,it exhibits more speciesspecific functions to suit their distinct invasion strategies.For many appressorium-forming plant pathogens such as the rice blast fungus Magnaporthe oryzae and Colletotrichum gloeosporioide,CWl pathway regulates the development and functioning of appressorium,which is required for pathogenicity(Jeon et al.,2008;Yin et al.,2016,2020;Fang et al.,2018).
基金supported by the National Natural Science Foundation of China(31571378 and 31501088)by grants from the State Key Laboratory of Plant Genomics
文摘Numerous fluorescent marker lines are currently available to visualize microtubule(MT)architecture and dynamics in living plant cells, such as markers expressing p35S::GFP-MBD or p35S::GFP-TUB6.However, these MT marker lines display obvious defects that affect plant growth or produce unstable fluorescent signals. Here, a series of new marker lines were developed, including the pTUB6::VisGreen-TUB6-expressing line in which TUB6 is under the control of its endogenous regulatory elements and e GFP is replaced with VisGreen, a brighter fluorescent protein. Moreover, two different markers were combined into one expression vector and developed two dual-marker lines.These marker lines produce bright, stable fluorescent signals in various tissues, and greatly shorten the screening process for generating dual-marker lines.These new marker lines provide a novel resource for MT research.
基金supported by the Key Research Program from the Chinese Academy of Sciences(ZDRW-ZS-2019-2)the“Strategic Priority Research Program”of the Chinese Academy of Sciences(XDA08000000)。
文摘Legumes have evolved a symbiotic relationship with rhizobial bacteria and their roots form unique nitrogen-fixing organs called nodules.Studies have shown that abiotic and biotic stresses alter the profile of gene expression and transcript mobility in plants.However,little is known about the systemic transport of RNA between roots and shoots in response to rhizobial infection on a genome-wide scale during the formation of legume-rhizobia symbiosis.In our study,we found that two soybean(Glycine max)cultivars,Peking and Williams,show a high frequency of single nucleotide polymorphisms;this allowed us to characterize the origin and mobility of transcripts in hetero-grafts of these two cultivars.We identified 4,552 genes that produce mobile RNAs in soybean,and found that rhizobial infection triggers mass transport of m RNAs between shoots and roots at the early stage of nodulation.The majority of these mRNAs are of relatively low abundance and their transport occurs in a selective manner in soybean plants.Notably,the mRNAs that moved from shoots to roots at the early stage of nodulation were enriched in many nodule-related responsive processes.Moreover,the transcripts of many known symbiosis-related genes that are induced by rhizobial infection can move between shoots and roots.Our findings provide a deeper understanding of endogenous RNA transport in legume-rhizobia symbiotic processes.
