Ronaldo Fumio F Hashimoto | Universidade de São Paulo (original) (raw)
Papers by Ronaldo Fumio F Hashimoto
XII Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00481)
Proceedings SIBGRAPI'98. International Symposium on Computer Graphics, Image Processing, and Vision (Cat. No.98EX237)
An important problem in the field of designing automatically morphological operators is to study ... more An important problem in the field of designing automatically morphological operators is to study the transformation of standard morphological representations of the operators learned into other equivalent representations in order to make efficient implementations. In this paper, we study this problem restricted to decomposition of set translation invariant dilations (respectively, erosions) by arbitrarily shaped structuring elements into a minimum number of compositions of dilations (respectively, erosions) by smaller structuring elements that are subsets of the 3×3 square centered at the origin. In this work, we approach this problem by applying combinatorial optimization techniques and using algebraic and geometrical properties of Minkowski additions in order to find pruning strategies
Motivation: A number of models have been proposed for genetic regulatory networks. In principle, ... more Motivation: A number of models have been proposed for genetic regulatory networks. In principle, a network may contain any number of genes, so long as data are available to make inferences about their relationships. Nevertheless, there are two important reasons why the size of a constructed network should be limited. Computationally and mathematically, it is more feasible to model and simulate a network with a small number of genes. In addition, it is more likely that a small set of genes maintains a specific core regulatory mechanism.
Computational Imaging and Vision
... ronaldo@ime.usp.br ... Starting from measurements of morphological data (covariance, speed of... more ... ronaldo@ime.usp.br ... Starting from measurements of morphological data (covariance, speed of erosion and dilation, size distribution, watershed sizing) on grey level images, correspondence analysis is used in a learning procedure to visualize the surfaces in a reduced space ...
2014 IEEE International Conference on Bioinformatics and Bioengineering, 2014
ABSTRACT Many methods have been published to prioritize genes using network theory. By using prot... more ABSTRACT Many methods have been published to prioritize genes using network theory. By using protein-protein interaction (PPI) data, it is possible to use mathematical features to rank and prioritize genes products in the network. Taking into account that genes related to the same diseases tend to connect, in the network structure, the prioritization methods search for candidate genes in the neighborhood of other genes already known to be related to a specific phenotype. Unfortunately, some existing algorithms can not deal well with highly connected genes, and some of them end up being related by chance to the disease being studied. We propose a pure method, with no need of adjustments, based on the hitting time of a random walk in PPI networks. This method captures information of the whole network and can equally prioritize genes regardless of its degree. We tested the efficiency of our method prioritizing candidate genes for Attention-Deficit/Hyperactivity Disorder (ADHD). The proposed method was able to give a good rank to genes that have genetic association with ADHD and was able to prioritize a large proportion of genes prioritized by other random-walk-based methods.
2013 XXVI Conference on Graphics, Patterns and Images, 2013
This work introduces a residual operator called ultimate attribute leveling. We also present an e... more This work introduces a residual operator called ultimate attribute leveling. We also present an efficient algorithm for ultimate attribute leveling computation by using a structure called tree of shapes. Our algorithm for computating ultimate attribute leveling is based on the fact (proved in this work) that levelings can be obtained by pruning nodes from the tree of shapes. This is a novel result, since so far it is known that levelings can be obtained from component trees. Finally, we propose the use of ultimate attribute leveling with shape information to extract contrast using a priori knowledge of an application. Experimental results applied to text location show the potentiality of using ultimate attribute leveling with shape information for solving problems in image processing area. Keywords-residual morphological operator; ultimate attribute opening; ultimate attribute closing; ultimate attribute leveling.
In this work we propose a method for localizing text regions within scene images consisting of tw... more In this work we propose a method for localizing text regions within scene images consisting of two major stages. In the first stage, a set of potential text regions is extracted from the input image using residual operators (such as ultimate attribute opening and closing). In the second stage a set of features is obtained from each potential text region
Gene, 2014
Inference of gene regulatory networks (GRNs) is one of the most challenging research problems of ... more Inference of gene regulatory networks (GRNs) is one of the most challenging research problems of Systems Biology. In this investigation, a new GRNs inference methodology, called Entropic Biological Score (EBS), which linearly combines the mean conditional entropy (MCE) from expression levels and a Biological Score (BS), obtained by integrating different biological data sources, is proposed. The EBS is validated with the Cell Cycle related functional annotation information, available from Munich Information Center for Protein Sequences (MIPS), and compared with some existing methods like MRNET, ARACNE, CLR and MCE for GRNs inference. For real networks, the performance of EBS, which uses the concept of integrating different data sources, is found to be superior to the aforementioned inference methods. The best results for EBS are obtained by considering the weights w1=0.2 and w2=0.8 for MCE and BS values, respectively, where approximately 40% of the inferred connections are found to be correct and significantly better than related methods. The results also indicate that expression profile is able to recover some true connections, that are not present in biological annotations, thus leading to the possibility of discovering new relations between its genes.
ABSTRACT Abstract—This work,presents,a,new,method,based,on morphological,filters for locating tex... more ABSTRACT Abstract—This work,presents,a,new,method,based,on morphological,filters for locating text regions in scene images. Preliminary results of the application of the method,to images from the ICDAR public dataset show,that the method,is useful for locating text regions. Keywords-scene-text localization, connected component ap-
A discrete dynamical system (DDS) (1) is a finite set of equations that describes the sequential ... more A discrete dynamical system (DDS) (1) is a finite set of equations that describes the sequential evolution of a vector of discrete variables, called state. The next state in time t+1 is computed by a function, called transition function, which depends on the states in the previous instants of time t, t-1, t-2, ... The representation of a DDS by
IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2013
Models of gene regulatory networks (GRN) have been proposed along with algorithms for inferring t... more Models of gene regulatory networks (GRN) have been proposed along with algorithms for inferring their structure. By structure, we mean the relationships among the genes of the biological system under study. Despite the large number of genes found in the genome of an organism, it is believed that a small set of genes is responsible for maintaining a specific core regulatory mechanism (small subnetworks). We propose an algorithm for inference of subnetworks of genes from a small initial set of genes called seed and time series gene expression data. The algorithm has two main steps: First, it grows the seed of genes by adding genes to it, and second, it searches for subnetworks that can be biologically meaningful. The seed growing step is treated as a feature selection problem and we used a thresholded Boolean network with a perturbation model to design the criterion function that is used to select the features (genes). Given that the reverse engineering of GRN is a problem that does not necessarily have one unique solution, the proposed algorithm has as output a set of networks instead of one single network. The algorithm also analyzes the dynamics of the networks which can be time-consuming. Nevertheless, the algorithm is suitable when the number of genes is small. The results showed that the algorithm is capable of recovering an acceptable rate of gene interactions and to generate regulatory hypotheses that can be explored in the wet lab.
Nonlinear Image Processing IX, 1998
It is well known that the family of hit-miss operators constitutes a sup-generating family for W-... more It is well known that the family of hit-miss operators constitutes a sup-generating family for W-operators, that is, any W-operator can be represented as the supremum of hit-miss operators. We present here a new sup-generating family for W- operators: compositions of hit-miss operators with dilations. The representation based on this sup-generating family is called compact, since it may use less
F1000Research, 2012
In this work, we used an algorithm based on a constraint solving problem approach to infer robust... more In this work, we used an algorithm based on a constraint solving problem approach to infer robust Boolean networks of the early embryonic phase in kidney organogenesis from gene expression data. A small subset of ten genes, cited in the literature as possible regulators of this biological process, was used to infer the networks. The data consist of a time-series gene expression of the rat (Rattus norvegicus) kidney development. The results show that several gene interactions could be recovered when compared to known interactions in the literature.
Proceedings XIV Brazilian Symposium on Computer Graphics and Image Processing
DNA chips (i.e., microarrays) biotechnology is a hybridization (i.e., DNA matching) based process... more DNA chips (i.e., microarrays) biotechnology is a hybridization (i.e., DNA matching) based process that makes it possible to quantify the relative abundance of mRNA from two distinct samples by analysing their fluorescence signals. This technique requires robotic placement (i.e., spotting) of thousands of cDNAs (i.e., complementary DNA) in an array format on glass microscope slides which provide gene-specific hybridization targets.
2009 XXII Brazilian Symposium on Computer Graphics and Image Processing, 2009
This work presents a new and fast algorithm for binary morphological erosions with arbitrary shap... more This work presents a new and fast algorithm for binary morphological erosions with arbitrary shaped structuring elements inspired by preprocessing techniques that are quite similar to those presented in many fast string matching algorithms (jumps and miss-matchings). The result of these preprocessing techniques is a speed up for computing binary erosions. A time complexity analysis shows that this algorithm has clear advantages over some known implementations. Experimental results confirm this analysis and shows that this algorithm has a good performance and can be a better option for erosions computation.
2008 XXI Brazilian Symposium on Computer Graphics and Image Processing, 2008
Recently, a new Pattern Recognition technique based on straight line segments (SLSs) was presente... more Recently, a new Pattern Recognition technique based on straight line segments (SLSs) was presented. The key issue in this new technique is to find a function based on distances between points and two sets of SLSs that minimizes a certain error or risk criterion. An algorithm for solving this optimization problem is called training algorithm. Although this technique seems to be very promising, the first presented training algorithm is based on a heuristic. In fact, the search for this best function is a hard nonlinear optimization problem. In this paper, we present a new and improved training algorithm for the SLS technique based on gradient descent optimization method. We have applied this new training algorithm to artificial and public data sets and their results confirm the improvement of this methodology.
2006 5th International Conference on Machine Learning and Applications (ICMLA'06), 2006
Page 1. A New Machine Learning Technique Based on Straight Line Segments Joao Henrique Burckas R... more Page 1. A New Machine Learning Technique Based on Straight Line Segments Joao Henrique Burckas Ribeiro Departamento de Ciência da Computaçao Instituto de Matemática e Estatıstica Universidade de Sao Paulo Brazil burckas@ime.usp.br ...
2014 IEEE International Conference on Bioinformatics and Bioengineering, 2014
ABSTRACT Various statistical and machine learning based algorithms have been proposed in literatu... more ABSTRACT Various statistical and machine learning based algorithms have been proposed in literature for selecting an informative subset of genes from microarray data sets. The recent trend is to use functional knowledge to aid the gene selection process. In this paper we propose a clustering algorithm which generates multiple views (clusters) from the microarray expression profiles, each representing a particular facet of the data. Such multiple clusters are found to represent strongly connected regions of the known protein–protein interaction (PPI) networks, perhaps corresponding to those responsible for certain biological processes. Thus we integrate microarray data clustering with PPI knowledge to obtain enriched gene sets. Results on benchmark microarray data sets demonstrate the competitiveness of our method compared to gene selection techniques.
Image Processing: Algorithms and Systems, 2002
The liver and small intestine restrict oral bioavailability of drugs and constitute the main site... more The liver and small intestine restrict oral bioavailability of drugs and constitute the main sites of pharmacokinetic drug-drug interactions. Hence, detailed data on hepatic and intestinal activities of drug metabolizing enzymes is important for modelling drug disposition and optimizing pharmacotherapy in different patient populations. The aim of this study was to determine the activities of seven cytochrome P450 (CYP) enzymes in paired liver and small intestinal samples from patients with obesity. Biopsies were obtained from twenty patients who underwent Rouxen-Y gastric bypass surgery following a three-week low energy diet. Individual hepatic and intestinal microsomes were prepared and specific probe substrates in combined incubations were used for determination of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6 and CYP3A activities. Activities of CYP2C8, CYP2C9, CYP2D6 and CYP3A were quantified in both human liver microsomes (HLM) and human intestinal microsomes (HIM), while activities of CYP1A2, CYP2B6 and CYP2C19 were only quantifiable in HLM. Considerable interindividual variability was present in both HLM (9 to 23-fold) and HIM (5 to 55-fold). The median metabolic HLM/HIM ratios varied from 1.5 for CYP3A to 252 for CYP2C8. Activities of CYP2C9 in paired HLM and HIM were positively correlated (r=0.74, p<0.001), while no interorgan correlations were found for activities of CYP2C8, CYP2D6 and CYP3A (p>0.05). Small intestinal CYP3A activities were higher in females compared with males (p<0.05). Hepatic CYP2B6 activity correlated negatively with body mass index (r=-0.72, p<0.001). These data may be useful for further in vitro-in vivo predictions of drug disposition in patients with obesity.
XII Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00481)
Proceedings SIBGRAPI'98. International Symposium on Computer Graphics, Image Processing, and Vision (Cat. No.98EX237)
An important problem in the field of designing automatically morphological operators is to study ... more An important problem in the field of designing automatically morphological operators is to study the transformation of standard morphological representations of the operators learned into other equivalent representations in order to make efficient implementations. In this paper, we study this problem restricted to decomposition of set translation invariant dilations (respectively, erosions) by arbitrarily shaped structuring elements into a minimum number of compositions of dilations (respectively, erosions) by smaller structuring elements that are subsets of the 3×3 square centered at the origin. In this work, we approach this problem by applying combinatorial optimization techniques and using algebraic and geometrical properties of Minkowski additions in order to find pruning strategies
Motivation: A number of models have been proposed for genetic regulatory networks. In principle, ... more Motivation: A number of models have been proposed for genetic regulatory networks. In principle, a network may contain any number of genes, so long as data are available to make inferences about their relationships. Nevertheless, there are two important reasons why the size of a constructed network should be limited. Computationally and mathematically, it is more feasible to model and simulate a network with a small number of genes. In addition, it is more likely that a small set of genes maintains a specific core regulatory mechanism.
Computational Imaging and Vision
... ronaldo@ime.usp.br ... Starting from measurements of morphological data (covariance, speed of... more ... ronaldo@ime.usp.br ... Starting from measurements of morphological data (covariance, speed of erosion and dilation, size distribution, watershed sizing) on grey level images, correspondence analysis is used in a learning procedure to visualize the surfaces in a reduced space ...
2014 IEEE International Conference on Bioinformatics and Bioengineering, 2014
ABSTRACT Many methods have been published to prioritize genes using network theory. By using prot... more ABSTRACT Many methods have been published to prioritize genes using network theory. By using protein-protein interaction (PPI) data, it is possible to use mathematical features to rank and prioritize genes products in the network. Taking into account that genes related to the same diseases tend to connect, in the network structure, the prioritization methods search for candidate genes in the neighborhood of other genes already known to be related to a specific phenotype. Unfortunately, some existing algorithms can not deal well with highly connected genes, and some of them end up being related by chance to the disease being studied. We propose a pure method, with no need of adjustments, based on the hitting time of a random walk in PPI networks. This method captures information of the whole network and can equally prioritize genes regardless of its degree. We tested the efficiency of our method prioritizing candidate genes for Attention-Deficit/Hyperactivity Disorder (ADHD). The proposed method was able to give a good rank to genes that have genetic association with ADHD and was able to prioritize a large proportion of genes prioritized by other random-walk-based methods.
2013 XXVI Conference on Graphics, Patterns and Images, 2013
This work introduces a residual operator called ultimate attribute leveling. We also present an e... more This work introduces a residual operator called ultimate attribute leveling. We also present an efficient algorithm for ultimate attribute leveling computation by using a structure called tree of shapes. Our algorithm for computating ultimate attribute leveling is based on the fact (proved in this work) that levelings can be obtained by pruning nodes from the tree of shapes. This is a novel result, since so far it is known that levelings can be obtained from component trees. Finally, we propose the use of ultimate attribute leveling with shape information to extract contrast using a priori knowledge of an application. Experimental results applied to text location show the potentiality of using ultimate attribute leveling with shape information for solving problems in image processing area. Keywords-residual morphological operator; ultimate attribute opening; ultimate attribute closing; ultimate attribute leveling.
In this work we propose a method for localizing text regions within scene images consisting of tw... more In this work we propose a method for localizing text regions within scene images consisting of two major stages. In the first stage, a set of potential text regions is extracted from the input image using residual operators (such as ultimate attribute opening and closing). In the second stage a set of features is obtained from each potential text region
Gene, 2014
Inference of gene regulatory networks (GRNs) is one of the most challenging research problems of ... more Inference of gene regulatory networks (GRNs) is one of the most challenging research problems of Systems Biology. In this investigation, a new GRNs inference methodology, called Entropic Biological Score (EBS), which linearly combines the mean conditional entropy (MCE) from expression levels and a Biological Score (BS), obtained by integrating different biological data sources, is proposed. The EBS is validated with the Cell Cycle related functional annotation information, available from Munich Information Center for Protein Sequences (MIPS), and compared with some existing methods like MRNET, ARACNE, CLR and MCE for GRNs inference. For real networks, the performance of EBS, which uses the concept of integrating different data sources, is found to be superior to the aforementioned inference methods. The best results for EBS are obtained by considering the weights w1=0.2 and w2=0.8 for MCE and BS values, respectively, where approximately 40% of the inferred connections are found to be correct and significantly better than related methods. The results also indicate that expression profile is able to recover some true connections, that are not present in biological annotations, thus leading to the possibility of discovering new relations between its genes.
ABSTRACT Abstract—This work,presents,a,new,method,based,on morphological,filters for locating tex... more ABSTRACT Abstract—This work,presents,a,new,method,based,on morphological,filters for locating text regions in scene images. Preliminary results of the application of the method,to images from the ICDAR public dataset show,that the method,is useful for locating text regions. Keywords-scene-text localization, connected component ap-
A discrete dynamical system (DDS) (1) is a finite set of equations that describes the sequential ... more A discrete dynamical system (DDS) (1) is a finite set of equations that describes the sequential evolution of a vector of discrete variables, called state. The next state in time t+1 is computed by a function, called transition function, which depends on the states in the previous instants of time t, t-1, t-2, ... The representation of a DDS by
IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2013
Models of gene regulatory networks (GRN) have been proposed along with algorithms for inferring t... more Models of gene regulatory networks (GRN) have been proposed along with algorithms for inferring their structure. By structure, we mean the relationships among the genes of the biological system under study. Despite the large number of genes found in the genome of an organism, it is believed that a small set of genes is responsible for maintaining a specific core regulatory mechanism (small subnetworks). We propose an algorithm for inference of subnetworks of genes from a small initial set of genes called seed and time series gene expression data. The algorithm has two main steps: First, it grows the seed of genes by adding genes to it, and second, it searches for subnetworks that can be biologically meaningful. The seed growing step is treated as a feature selection problem and we used a thresholded Boolean network with a perturbation model to design the criterion function that is used to select the features (genes). Given that the reverse engineering of GRN is a problem that does not necessarily have one unique solution, the proposed algorithm has as output a set of networks instead of one single network. The algorithm also analyzes the dynamics of the networks which can be time-consuming. Nevertheless, the algorithm is suitable when the number of genes is small. The results showed that the algorithm is capable of recovering an acceptable rate of gene interactions and to generate regulatory hypotheses that can be explored in the wet lab.
Nonlinear Image Processing IX, 1998
It is well known that the family of hit-miss operators constitutes a sup-generating family for W-... more It is well known that the family of hit-miss operators constitutes a sup-generating family for W-operators, that is, any W-operator can be represented as the supremum of hit-miss operators. We present here a new sup-generating family for W- operators: compositions of hit-miss operators with dilations. The representation based on this sup-generating family is called compact, since it may use less
F1000Research, 2012
In this work, we used an algorithm based on a constraint solving problem approach to infer robust... more In this work, we used an algorithm based on a constraint solving problem approach to infer robust Boolean networks of the early embryonic phase in kidney organogenesis from gene expression data. A small subset of ten genes, cited in the literature as possible regulators of this biological process, was used to infer the networks. The data consist of a time-series gene expression of the rat (Rattus norvegicus) kidney development. The results show that several gene interactions could be recovered when compared to known interactions in the literature.
Proceedings XIV Brazilian Symposium on Computer Graphics and Image Processing
DNA chips (i.e., microarrays) biotechnology is a hybridization (i.e., DNA matching) based process... more DNA chips (i.e., microarrays) biotechnology is a hybridization (i.e., DNA matching) based process that makes it possible to quantify the relative abundance of mRNA from two distinct samples by analysing their fluorescence signals. This technique requires robotic placement (i.e., spotting) of thousands of cDNAs (i.e., complementary DNA) in an array format on glass microscope slides which provide gene-specific hybridization targets.
2009 XXII Brazilian Symposium on Computer Graphics and Image Processing, 2009
This work presents a new and fast algorithm for binary morphological erosions with arbitrary shap... more This work presents a new and fast algorithm for binary morphological erosions with arbitrary shaped structuring elements inspired by preprocessing techniques that are quite similar to those presented in many fast string matching algorithms (jumps and miss-matchings). The result of these preprocessing techniques is a speed up for computing binary erosions. A time complexity analysis shows that this algorithm has clear advantages over some known implementations. Experimental results confirm this analysis and shows that this algorithm has a good performance and can be a better option for erosions computation.
2008 XXI Brazilian Symposium on Computer Graphics and Image Processing, 2008
Recently, a new Pattern Recognition technique based on straight line segments (SLSs) was presente... more Recently, a new Pattern Recognition technique based on straight line segments (SLSs) was presented. The key issue in this new technique is to find a function based on distances between points and two sets of SLSs that minimizes a certain error or risk criterion. An algorithm for solving this optimization problem is called training algorithm. Although this technique seems to be very promising, the first presented training algorithm is based on a heuristic. In fact, the search for this best function is a hard nonlinear optimization problem. In this paper, we present a new and improved training algorithm for the SLS technique based on gradient descent optimization method. We have applied this new training algorithm to artificial and public data sets and their results confirm the improvement of this methodology.
2006 5th International Conference on Machine Learning and Applications (ICMLA'06), 2006
Page 1. A New Machine Learning Technique Based on Straight Line Segments Joao Henrique Burckas R... more Page 1. A New Machine Learning Technique Based on Straight Line Segments Joao Henrique Burckas Ribeiro Departamento de Ciência da Computaçao Instituto de Matemática e Estatıstica Universidade de Sao Paulo Brazil burckas@ime.usp.br ...
2014 IEEE International Conference on Bioinformatics and Bioengineering, 2014
ABSTRACT Various statistical and machine learning based algorithms have been proposed in literatu... more ABSTRACT Various statistical and machine learning based algorithms have been proposed in literature for selecting an informative subset of genes from microarray data sets. The recent trend is to use functional knowledge to aid the gene selection process. In this paper we propose a clustering algorithm which generates multiple views (clusters) from the microarray expression profiles, each representing a particular facet of the data. Such multiple clusters are found to represent strongly connected regions of the known protein–protein interaction (PPI) networks, perhaps corresponding to those responsible for certain biological processes. Thus we integrate microarray data clustering with PPI knowledge to obtain enriched gene sets. Results on benchmark microarray data sets demonstrate the competitiveness of our method compared to gene selection techniques.
Image Processing: Algorithms and Systems, 2002
The liver and small intestine restrict oral bioavailability of drugs and constitute the main site... more The liver and small intestine restrict oral bioavailability of drugs and constitute the main sites of pharmacokinetic drug-drug interactions. Hence, detailed data on hepatic and intestinal activities of drug metabolizing enzymes is important for modelling drug disposition and optimizing pharmacotherapy in different patient populations. The aim of this study was to determine the activities of seven cytochrome P450 (CYP) enzymes in paired liver and small intestinal samples from patients with obesity. Biopsies were obtained from twenty patients who underwent Rouxen-Y gastric bypass surgery following a three-week low energy diet. Individual hepatic and intestinal microsomes were prepared and specific probe substrates in combined incubations were used for determination of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6 and CYP3A activities. Activities of CYP2C8, CYP2C9, CYP2D6 and CYP3A were quantified in both human liver microsomes (HLM) and human intestinal microsomes (HIM), while activities of CYP1A2, CYP2B6 and CYP2C19 were only quantifiable in HLM. Considerable interindividual variability was present in both HLM (9 to 23-fold) and HIM (5 to 55-fold). The median metabolic HLM/HIM ratios varied from 1.5 for CYP3A to 252 for CYP2C8. Activities of CYP2C9 in paired HLM and HIM were positively correlated (r=0.74, p<0.001), while no interorgan correlations were found for activities of CYP2C8, CYP2D6 and CYP3A (p>0.05). Small intestinal CYP3A activities were higher in females compared with males (p<0.05). Hepatic CYP2B6 activity correlated negatively with body mass index (r=-0.72, p<0.001). These data may be useful for further in vitro-in vivo predictions of drug disposition in patients with obesity.