Publications by authors named "Seiichi Uchida"

27 Publications

  • Page 1 of 1

Editorial for special issue on "Advanced Topics in Document Analysis and Recognition".

Int J Doc Anal Recognit 2021 Aug 10:1-3. Epub 2021 Aug 10.

Department of Advanced Information Technology, Faculty of Information Science and Electrical Engineering, Kyushu University, Kyushu University, 744 Motooka Nishi-ku, Fukuoka, 819-0395 Japan.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s10032-021-00385-1DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8353931PMC
August 2021

An empirical survey of data augmentation for time series classification with neural networks.

PLoS One 2021 15;16(7):e0254841. Epub 2021 Jul 15.

Department of Advanced Information Technology, Kyushu University, Fukuoka, Japan.

In recent times, deep artificial neural networks have achieved many successes in pattern recognition. Part of this success can be attributed to the reliance on big data to increase generalization. However, in the field of time series recognition, many datasets are often very small. One method of addressing this problem is through the use of data augmentation. In this paper, we survey data augmentation techniques for time series and their application to time series classification with neural networks. We propose a taxonomy and outline the four families in time series data augmentation, including transformation-based methods, pattern mixing, generative models, and decomposition methods. Furthermore, we empirically evaluate 12 time series data augmentation methods on 128 time series classification datasets with six different types of neural networks. Through the results, we are able to analyze the characteristics, advantages and disadvantages, and recommendations of each data augmentation method. This survey aims to help in the selection of time series data augmentation for neural network applications.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0254841PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8282049PMC
November 2021

Soft and self constrained clustering for group-based labeling.

Med Image Anal 2021 08 12;72:102097. Epub 2021 May 12.

Department of Advanced Information Technology, Kyushu University, Fukuoka, Japan; Research Center for Medical Bigdata, National Institute of Informatics, Tokyo, Japan.

When using deep neural networks in medical image classification tasks, it is mandatory to prepare a large-scale labeled image set, and this often requires significant effort by medical experts. One strategy to reduce the labeling cost is group-based labeling, where image samples are clustered and then a label is attached to each cluster. The efficiency of this strategy depends on the purity of the clusters. Constrained clustering is an effective way to improve the purity of the clusters if we can give appropriate must-links and cannot-links as constraints. However, for medical image clustering, the conventional constrained clustering methods encounter two issues. The first issue is that constraints are not always appropriate due to the gap between semantic and visual similarities. The second issue is that attaching constraints requires extra effort from medical experts. To deal with the first issue, we propose a novel soft-constrained clustering method, which has the ability to ignore inappropriate constraints. To deal with the second issue, we propose a self-constrained clustering method that utilizes prior knowledge about the target images to set the constraints automatically. Experiments with the endoscopic image datasets demonstrated that the proposed methods give clustering results with higher purity.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.media.2021.102097DOI Listing
August 2021

STIM-Orai1 signaling regulates fluidity of cytoplasm during membrane blebbing.

Nat Commun 2021 01 20;12(1):480. Epub 2021 Jan 20.

Department of Biology, Faculty of Sciences, Kyushu University, Fukuoka, 819-0395, Japan.

The cytoplasm in mammalian cells is considered homogeneous. In this study, we report that the cytoplasmic fluidity is regulated in the blebbing cells; the cytoplasm of rapidly expanding membrane blebs is more disordered than the cytoplasm of retracting blebs. The increase of cytoplasmic fluidity in the expanding bleb is caused by a sharp rise in the calcium concentration. The STIM-Orai1 pathway regulates this rapid and restricted increase of calcium in the expanding blebs. Conversely, activated ERM protein binds to Orai1 to inhibit the store-operated calcium entry in retracting blebs, which results in decreased in cytoplasmic calcium, rapid reassembly of the actin cortex.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41467-020-20826-5DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7817837PMC
January 2021

Explainable Deep Learning Reproduces a 'Professional Eye' on the Diagnosis of Internal Disorders in Persimmon Fruit.

Plant Cell Physiol 2020 Dec;61(11):1967-1973

Field Science Education and Research Center, Kyoto University, Kyoto, 606-8502 Japan.

Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid phenotyping. In this study, beyond simple optimizations of phenotyping, we propose an application of deep neural networks to make an image-based internal disorder diagnosis that is hard even for experts, and to visualize the reasons behind each diagnosis to provide biological interpretations. Here, we exemplified classification of calyx-end cracking in persimmon fruit by using five convolutional neural network models with various layer structures and examined potential analytical options involved in the diagnostic qualities. With 3,173 visible RGB images from the fruit apex side, the neural networks successfully made the binary classification of each degree of disorder, with up to 90% accuracy. Furthermore, feature visualizations, such as Grad-CAM and LRP, visualize the regions of the image that contribute to the diagnosis. They suggest that specific patterns of color unevenness, such as in the fruit peripheral area, can be indexes of calyx-end cracking. These results not only provided novel insights into indexes of fruit internal disorders but also proposed the potential applicability of deep neural networks in plant biology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/pcp/pcaa111DOI Listing
December 2020

Guided neural style transfer for shape stylization.

PLoS One 2020 4;15(6):e0233489. Epub 2020 Jun 4.

Department of Advanced Information Technology, Kyushu University, Fukuoka, Japan.

Designing logos, typefaces, and other decorated shapes can require professional skills. In this paper, we aim to produce new and unique decorated shapes by stylizing ordinary shapes with machine learning. Specifically, we combined parametric and non-parametric neural style transfer algorithms to transfer both local and global features. Furthermore, we introduced a distance-based guiding to the neural style transfer process, so that only the foreground shape will be decorated. Lastly, qualitative evaluation and ablation studies are provided to demonstrate the usefulness of the proposed method.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0233489PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7272093PMC
August 2020

Coordinated changes in cell membrane and cytoplasm during maturation of apoptotic bleb.

Mol Biol Cell 2020 04 12;31(8):833-844. Epub 2020 Feb 12.

Department of Biology, Faculty of Sciences, Kyushu University, Fukuoka 819-0395, Japan.

Apoptotic cells form membrane blebs, but little is known about how the formation and dynamics of membrane blebs are regulated. The size of blebs gradually increases during the progression of apoptosis, eventually forming large extracellular vesicles called apoptotic bodies that have immune-modulating activities. In this study, we investigated the molecular mechanism involved in the differentiation of blebs into apoptotic blebs by comparing the dynamics of the bleb formed during cell migration and the bleb formed during apoptosis. We revealed that the enhanced activity of ROCK1 is required for the formation of small blebs in the early phase of apoptosis, which leads to the physical disruption of nuclear membrane and the degradation of Lamin A. In the late phase of apoptosis, the loss of asymmetry in phospholipids distribution caused the enlargement of blebs, which enabled translocation of damage-associated molecular patterns to the bleb cytoplasm and maturation of functional apoptotic blebs. Thus, changes in cell membrane dynamics are closely linked to cytoplasmic changes during apoptotic bleb formation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1091/mbc.E19-12-0691DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7185959PMC
April 2020

Endoscopic Image Clustering with Temporal Ordering Information Based on Dynamic Programming.

Annu Int Conf IEEE Eng Med Biol Soc 2019 Jul;2019:3681-3684

In this paper, we propose a clustering method with temporal ordering information for endoscopic image sequences. It is difficult to collect a sufficient amount of endoscopic image datasets to train machine learning techniques by manual labeling. The clustering of endoscopic images leads to group-based labeling, which is useful for reducing the cost of dataset construction. Therefore, in this paper, we propose a clustering method where the property of endoscopic image sequences is fully utilized. For the proposed method, a deep neural network was used to extract features from endoscopic images, and clustering with temporal ordering information was solved by dynamic programming. In the experiments, we clustered the esophagogastroduodenoscopy images. From the results, we confirmed that the performance was improved by using the sequential property.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2019.8857011DOI Listing
July 2019

Scribbles for Metric Learning in Histological Image Segmentation.

Annu Int Conf IEEE Eng Med Biol Soc 2019 Jul;2019:1026-1030

Segmentation is a fundamental process in biomedical image analysis that enables various types of analysis. Segmenting organs in histological microscopy images is problematic because the boundaries between regions are ambiguous, the images have various appearances, and the amount of training data is limited. To address these difficulties, supervised learning methods (e.g., convolutional neural networking (CNN)) are insufficient to predict regions accurately because they usually require a large amount of training data to learn the various appearances. In this paper, we propose a semi-automatic segmentation method that effectively uses scribble annotations for metric learning. Deep discriminative metric learning re-trains the representation of the feature space so that the distances between the samples with the same class labels are reduced, while those between ones with different class labels are enlarged. It makes pixel classification easy. Evaluation of the proposed method in a heart region segmentation task demonstrated that it performed better than three other methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2019.8856465DOI Listing
July 2019

Automatic Generation of Typographic Font From Small Font Subset.

IEEE Comput Graph Appl 2020 Jan-Feb;40(1):99-111. Epub 2019 Jul 31.

The automated generation of fonts containing a large number of characters is in high demand. For example, a typical Japanese font requires over 1000 characters. Unfortunately, professional typographers create the majority of fonts, resulting in significant financial and time investments for font generation. The main contribution of this article is the development of a method that automatically generates a target typographic font containing thousands of characters, from a small subset of character images in the target font. We generate characters other than the subset so that a complete font is obtained. We propose a novel font generation method with the capability to deal with various fonts, including a font composed of distinctive strokes, which are difficult for existing methods to handle. We demonstrated the proposed method by generating 2965 characters in 47 fonts. Moreover, objective and subjective evaluations verified that the generated characters are similar to the original characters.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/MCG.2019.2931431DOI Listing
March 2020

Biosignal Data Augmentation Based on Generative Adversarial Networks.

Annu Int Conf IEEE Eng Med Biol Soc 2018 Jul;2018:368-371

In this paper, we propose a synthetic generationmethod for time-series data based on generative adversarial networks (GANs) and apply it to data augmentation for biosinal classification. GANs are a recently proposed framework for learning a generative model, where two neural networks, one generating synthetic data and the other discriminating synthetic and real data, are trained while competing with each other. In the proposed method, each neural network in GANs is developed based on a recurrent neural network using long short-term memories, thereby allowing the adaptation of the GANs framework to time-series data generation. In the experiments, we confirmed the capability of the proposed method for generating synthetic biosignals using the electrocardiogram and electroencephalogram datasets. We also showed the effectiveness of the proposed method for data augmentation in the biosignal classification problem.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2018.8512396DOI Listing
July 2018

The cytoplasmic region of the amyloid β-protein precursor (APP) is necessary and sufficient for the enhanced fast velocity of APP transport by kinesin-1.

FEBS Lett 2018 08 12;592(16):2716-2724. Epub 2018 Aug 12.

Laboratory of Neuroscience, Graduate School of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan.

Amyloid β-protein precursor (APP) is transported mainly by kinesin-1 and at a higher velocity than other kinesin-1 cargos, such as Alcadein α (Alcα); this is denoted by the enhanced fast velocity (EFV). Interaction of the APP cytoplasmic region with kinesin-1, which is essential for EFV transport, is mediated by JNK-interacting protein 1 (JIP1). To determine the roles of interactions between the APP luminal region and cargo components, we monitored transport of chimeric cargo receptors, Alcα (luminal)-APP (cytoplasmic) and APP (luminal)-Alcα (cytoplasmic). Alcα-APP is transported at the EFV, whereas APP-Alcα is transported at the same velocity as wild-type Alcα. Thus, the cytoplasmic region of APP is necessary and sufficient for the EFV of APP transport by kinesin-1.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/1873-3468.13204DOI Listing
August 2018

Three-dimensional computer graphic animations for studying social approach behaviour in medaka fish: Effects of systematic manipulation of morphological and motion cues.

PLoS One 2017 11;12(4):e0175059. Epub 2017 Apr 11.

Laboratory of Neurophysiology, National Institute for Basic Biology, 5-1 Higashiyama, Myodaiji-cho, Okazaki, Aichi, Japan.

We studied social approach behaviour in medaka fish using three-dimensional computer graphic (3DCG) animations based on the morphological features and motion characteristics obtained from real fish. This is the first study which used 3DCG animations and examined the relative effects of morphological and motion cues on social approach behaviour in medaka. Various visual stimuli, e.g., lack of motion, lack of colour, alternation in shape, lack of locomotion, lack of body motion, and normal virtual fish in which all four features (colour, shape, locomotion, and body motion) were reconstructed, were created and presented to fish using a computer display. Medaka fish presented with normal virtual fish spent a long time in proximity to the display, whereas time spent near the display was decreased in other groups when compared with normal virtual medaka group. The results suggested that the naturalness of visual cues contributes to the induction of social approach behaviour. Differential effects between body motion and locomotion were also detected. 3DCG animations can be a useful tool to study the mechanisms of visual processing and social behaviour in medaka.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0175059PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5388324PMC
September 2017

Endoplasmic-reticulum-mediated microtubule alignment governs cytoplasmic streaming.

Nat Cell Biol 2017 04 13;19(4):399-406. Epub 2017 Mar 13.

Cell Architecture Laboratory, Structural Biology Center, National Institute of Genetics, Mishima 411-8540, Japan.

Cytoplasmic streaming refers to a collective movement of cytoplasm observed in many cell types. The mechanism of meiotic cytoplasmic streaming (MeiCS) in Caenorhabditis elegans zygotes is puzzling as the direction of the flow is not predefined by cell polarity and occasionally reverses. Here, we demonstrate that the endoplasmic reticulum (ER) network structure is required for the collective flow. Using a combination of RNAi, microscopy and image processing of C. elegans zygotes, we devise a theoretical model, which reproduces and predicts the emergence and reversal of the flow. We propose a positive-feedback mechanism, where a local flow generated along a microtubule is transmitted to neighbouring regions through the ER. This, in turn, aligns microtubules over a broader area to self-organize the collective flow. The proposed model could be applicable to various cytoplasmic streaming phenomena in the absence of predefined polarity. The increased mobility of cortical granules by MeiCS correlates with the efficient exocytosis of the granules to protect the zygotes from osmotic and mechanical stresses.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/ncb3490DOI Listing
April 2017

Basal filopodia and vascular mechanical stress organize fibronectin into pillars bridging the mesoderm-endoderm gap.

Development 2017 01;144(2):281-291

Department of Radiology and Developmental Neuroscience Program, Saban Research Institute, Children's Hospital Los Angeles, Los Angeles, CA 90027, USA.

Cells may exchange information with other cells and tissues by exerting forces on the extracellular matrix (ECM). Fibronectin (FN) is an important ECM component that forms fibrils through cell contacts and creates directionally biased geometry. Here, we demonstrate that FN is deposited as pillars between widely separated germ layers, namely the somitic mesoderm and the endoderm, in quail embryos. Alongside the FN pillars, long filopodia protrude from the basal surfaces of somite epithelial cells. Loss-of-function of Ena/VASP, α5β1-integrins or talin in the somitic cells abolished the FN pillars, indicating that FN pillar formation is dependent on the basal filopodia through these molecules. The basal filopodia and FN pillars are also necessary for proper somite morphogenesis. We identified a new mechanism contributing to FN pillar formation by focusing on cyclic expansion of adjacent dorsal aorta. Maintenance of the directional alignment of the FN pillars depends on pulsatile blood flow through the dorsal aortae. These results suggest that the FN pillars are specifically established through filopodia-mediated and pulsating force-related mechanisms.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1242/dev.141259DOI Listing
January 2017

Interphase adhesion geometry is transmitted to an internal regulator for spindle orientation via caveolin-1.

Nat Commun 2016 06 13;7:ncomms11858. Epub 2016 Jun 13.

Department of Cell Biology, Institute for Virus Research, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan.

Despite theoretical and physical studies implying that cell-extracellular matrix adhesion geometry governs the orientation of the cell division axis, the molecular mechanisms that translate interphase adhesion geometry to the mitotic spindle orientation remain elusive. Here, we show that the cellular edge retraction during mitotic cell rounding correlates with the spindle axis. At the onset of mitotic cell rounding, caveolin-1 is targeted to the retracting cortical region at the proximal end of retraction fibres, where ganglioside GM1-enriched membrane domains with clusters of caveola-like structures are formed in an integrin and RhoA-dependent manner. Furthermore, Gαi1-LGN-NuMA, a well-known regulatory complex of spindle orientation, is targeted to the caveolin-1-enriched cortical region to guide the spindle axis towards the cellular edge retraction. We propose that retraction-induced cortical heterogeneity of caveolin-1 during mitotic cell rounding sets the spindle orientation in the context of adhesion geometry.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/ncomms11858DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4910015PMC
June 2016

Protective effects of NSP-116, a novel imidazolyl aniline derivative, against light-induced retinal damage in vitro and in vivo.

Free Radic Biol Med 2016 07 2;96:304-12. Epub 2016 May 2.

Molecular Pharmacology, Department of Biofunctional Evaluation, Gifu Pharmaceutical University, Japan. Electronic address:

In this study, we investigated the protective effects of NSP-116 [4-(4-acetylpiperazin-1-yl)-2-(1H-imidazol-1-yl) aniline], a novel imidazolyl aniline derivative, against light-induced photoreceptor cell damage. In an in vitro experiment, murine photoreceptor (661W) cells were damaged by exposure to light for 24h. Viability of 661W cells after light exposure was assessed by Hoechst 33342/Propidium iodide nuclear staining and a tetrazolium salt (WST-8) assay. Intracellular radical production in 661W cells was evaluated using the reactive oxygen species (ROS) sensitive probe 5-(and 6)-chloromethyl-2, 7-dichlorodihydrofluorescein diacetate acetyl ester (CM-H2DCFDA). NSP-116 significantly suppressed light-induced cell death and ROS production in 661W cells. In an in vivo mouse experiment, retinal damage was induced by exposure to white light at 8000lx for 3h after dark adaptation. Retinal damage was evaluated by recording the electroretinogram and measuring the outer nuclear layer (ONL) thickness at 5 days after light exposure. Single oral administration of NSP-116 before light exposure protected retinal function and ONL thinning after light exposure. Furthermore, the effect of NSP-116 on lipid peroxidation was evaluated using thiobarbituric acid reactive substance (TBARS) assay in porcine retina, and was found to decrease the production of TBARS. Electron spin resonance (ESR) measurements showed that NSP-116 exhibited radical scavenging activities against 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical, superoxide anion radical (∙O2(-)), and hydroxyl radical (∙OH). These findings suggest that NSP-116 has protective effects against light-induced photoreceptor degeneration in vitro and in vivo as a free radical scavenger, and it may be a novel therapeutic agent for retinal degenerative disorders, such as dry age-related macular degeneration (AMD).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.freeradbiomed.2016.03.036DOI Listing
July 2016

A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data.

PLoS One 2016 19;11(4):e0152173. Epub 2016 Apr 19.

Department of Advanced Information Technology, Kyushu University, Fukuoka, Japan.

Anomaly detection is the process of identifying unexpected items or events in datasets, which differ from the norm. In contrast to standard classification tasks, anomaly detection is often applied on unlabeled data, taking only the internal structure of the dataset into account. This challenge is known as unsupervised anomaly detection and is addressed in many practical applications, for example in network intrusion detection, fraud detection as well as in the life science and medical domain. Dozens of algorithms have been proposed in this area, but unfortunately the research community still lacks a comparative universal evaluation as well as common publicly available datasets. These shortcomings are addressed in this study, where 19 different unsupervised anomaly detection algorithms are evaluated on 10 different datasets from multiple application domains. By publishing the source code and the datasets, this paper aims to be a new well-funded basis for unsupervised anomaly detection research. Additionally, this evaluation reveals the strengths and weaknesses of the different approaches for the first time. Besides the anomaly detection performance, computational effort, the impact of parameter settings as well as the global/local anomaly detection behavior is outlined. As a conclusion, we give an advise on algorithm selection for typical real-world tasks.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0152173PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4836738PMC
August 2016

A RhoA and Rnd3 cycle regulates actin reassembly during membrane blebbing.

Proc Natl Acad Sci U S A 2016 Mar 14;113(13):E1863-71. Epub 2016 Mar 14.

Department of Biology, Faculty of Sciences, Kyushu University, Fukuoka 819-0395, Japan; Precursory Research for Embryonic Science and Technology, Japan Science and Technology Agency, Saitama 332-0012, Japan; AMED-PRIME, Japan Agency for Medical Research and Development, Tokyo 100-0004, Japan

The actin cytoskeleton usually lies beneath the plasma membrane. When the membrane-associated actin cytoskeleton is transiently disrupted or the intracellular pressure is increased, the plasma membrane detaches from the cortex and protrudes. Such protruded membrane regions are called blebs. However, the molecular mechanisms underlying membrane blebbing are poorly understood. This study revealed that epidermal growth factor receptor kinase substrate 8 (Eps8) and ezrin are important regulators of rapid actin reassembly for the initiation and retraction of protruded blebs. Live-cell imaging of membrane blebbing revealed that local reassembly of actin filaments occurred at Eps8- and activated ezrin-positive foci of membrane blebs. Furthermore, we found that a RhoA-ROCK-Rnd3 feedback loop determined the local reassembly sites of the actin cortex during membrane blebbing.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1073/pnas.1600968113DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4822640PMC
March 2016

Visual Saliency Models for Text Detection in Real World.

PLoS One 2014 10;9(12):e114539. Epub 2014 Dec 10.

Department of Advanced Information technology, Kyushu University, Fukuoka, Fukuoka, Japan.

This paper evaluates the degree of saliency of texts in natural scenes using visual saliency models. A large scale scene image database with pixel level ground truth is created for this purpose. Using this scene image database and five state-of-the-art models, visual saliency maps that represent the degree of saliency of the objects are calculated. The receiver operating characteristic curve is employed in order to evaluate the saliency of scene texts, which is calculated by visual saliency models. A visualization of the distribution of scene texts and non-texts in the space constructed by three kinds of saliency maps, which are calculated using Itti's visual saliency model with intensity, color and orientation features, is given. This visualization of distribution indicates that text characters are more salient than their non-text neighbors, and can be captured from the background. Therefore, scene texts can be extracted from the scene images. With this in mind, a new visual saliency architecture, named hierarchical visual saliency model, is proposed. Hierarchical visual saliency model is based on Itti's model and consists of two stages. In the first stage, Itti's model is used to calculate the saliency map, and Otsu's global thresholding algorithm is applied to extract the salient region that we are interested in. In the second stage, Itti's model is applied to the salient region to calculate the final saliency map. An experimental evaluation demonstrates that the proposed model outperforms Itti's model in terms of captured scene texts.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0114539PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4262416PMC
October 2017

Quantitative analysis of APP axonal transport in neurons: role of JIP1 in enhanced APP anterograde transport.

Mol Biol Cell 2014 Nov 27;25(22):3569-80. Epub 2014 Aug 27.

Laboratory of Neuroscience, Graduate School of Pharmaceutical Sciences, Hokkaido University, Sapporo 060-0812, Japan

Alzheimer's β-amyloid precursor protein (APP) associates with kinesin-1 via JNK-interacting protein 1 (JIP1); however, the role of JIP1 in APP transport by kinesin-1 in neurons remains unclear. We performed a quantitative analysis to understand the role of JIP1 in APP axonal transport. In JIP1-deficient neurons, we find that both the fast velocity (∼2.7 μm/s) and high frequency (66%) of anterograde transport of APP cargo are impaired to a reduced velocity (∼1.83 μm/s) and a lower frequency (45%). We identified two novel elements linked to JIP1 function, located in the central region of JIP1b, that interact with the coiled-coil domain of kinesin light chain 1 (KLC1), in addition to the conventional interaction of the JIP1b 11-amino acid C-terminal (C11) region with the tetratricopeptide repeat of KLC1. High frequency of APP anterograde transport is dependent on one of the novel elements in JIP1b. Fast velocity of APP cargo transport requires the C11 domain, which is regulated by the second novel region of JIP1b. Furthermore, efficient APP axonal transport is not influenced by phosphorylation of APP at Thr-668, a site known to be phosphorylated by JNK. Our quantitative analysis indicates that enhanced fast-velocity and efficient high-frequency APP anterograde transport observed in neurons are mediated by novel roles of JIP1b.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1091/mbc.E14-06-1111DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4230617PMC
November 2014

A voting-based sequential pattern recognition method.

PLoS One 2013 14;8(10):e76980. Epub 2013 Oct 14.

Faculty of Systems Engineering, Wakayama University, Wakayama-shi, Wakayama, Japan.

We propose a novel method for recognizing sequential patterns such as motion trajectory of biological objects (i.e., cells, organelle, protein molecules, etc.), human behavior motion, and meteorological data. In the proposed method, a local classifier is prepared for every point (or timing or frame) and then the whole pattern is recognized by majority voting of the recognition results of the local classifiers. The voting strategy has a strong benefit that even if an input pattern has a very large deviation from a prototype locally at several points, they do not severely influence the recognition result; they are treated just as several incorrect votes and thus will be neglected successfully through the majority voting. For regularizing the recognition result, we introduce partial-dependency to local classifiers. An important point is that this dependency is introduced to not only local classifiers at neighboring point pairs but also to those at distant point pairs. Although, the dependency makes the problem non-Markovian (i.e., higher-order Markovian), it can still be solved efficiently by using a graph cut algorithm with polynomial-order computations. The experimental results revealed that the proposed method can achieve better recognition accuracy while utilizing the above characteristics of the proposed method.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0076980PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3796540PMC
June 2014

Image processing and recognition for biological images.

Authors:
Seiichi Uchida

Dev Growth Differ 2013 May 7;55(4):523-49. Epub 2013 Apr 7.

Department of Advanced Information Technology, Kyushu University, 744 Motooka, Nishi, Fukuoka, 819-0395, Japan.

This paper reviews image processing and pattern recognition techniques, which will be useful to analyze bioimages. Although this paper does not provide their technical details, it will be possible to grasp their main tasks and typical tools to handle the tasks. Image processing is a large research area to improve the visibility of an input image and acquire some valuable information from it. As the main tasks of image processing, this paper introduces gray-level transformation, binarization, image filtering, image segmentation, visual object tracking, optical flow and image registration. Image pattern recognition is the technique to classify an input image into one of the predefined classes and also has a large research area. This paper overviews its two main modules, that is, feature extraction module and classification module. Throughout the paper, it will be emphasized that bioimage is a very difficult target for even state-of-the-art image processing and pattern recognition techniques due to noises, deformations, etc. This paper is expected to be one tutorial guide to bridge biology and image processing researchers for their further collaboration to tackle such a difficult target.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/dgd.12054DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746120PMC
May 2013

Real-time nonlinear FEM with neural network for simulating soft organ model deformation.

Med Image Comput Comput Assist Interv 2008 ;11(Pt 2):742-9

Digital Medicine Initiative, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan.

This paper presents a new method for simulating the deformation of organ models by using a neural network. The proposed method is based on the idea proposed by Chen et al. that a deformed model can be estimated from the superposition of basic deformation modes. The neural network finds a relationship between external forces and the models deformed by the forces. The experimental results show that the trained network can achieve a real-time simulation while keeping the acceptable accuracy compared with the nonlinear FEM computation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/978-3-540-85990-1_89DOI Listing
December 2008

Attenuation of cataract progression by A-3922, a dihydrobenzofuran derivative, in streptozotocin-induced diabetic rats.

Biol Pharm Bull 2008 Oct;31(10):1959-63

Department of Molecular Pharmacology, School of Pharmaceutical Sciences, Kitasato University, 5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan.

The present study was undertaken to assess whether A-3922, a dihydrobenzofuran derivative that possesses antioxidative effects, had any preventive effect on the onset and/or progression of diabetic cataract. Male Wistar rats were received a bolus intravenous injection of streptozotocin (65 mg/kg) and were given 5% glucose in drinking water for 10 weeks. The diabetic rats were divided into two groups and treated with 30 mg/kg/d A-3922 or vehicle during the experimental period. The opacities of eye lenses were observed by using both our original device and a slit lamp microscope. The lens opacities were initially detected as early as the 2nd week and the cataracts were developed in similar fashion in both A-3922-treated and untreated diabetic rats until 7th week, suggesting that A-3922 did not show any appreciable effect on the onset of diabetic cataract. In the later period (8th week or later), however, progression of cataract was retarded and significant reductions in both the total cataract score and the degree of opacity were apparently observed on 10th week of A-3922-treated diabetic rats. These results suggest that A-3922 can delay the progression but not the onset of diabetic cataract, and it has a possibility to be a candidate for drugs of cataract associated with diabetes.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1248/bpb.31.1959DOI Listing
October 2008

Effect of a dihydrobenzofuran derivative on lipid hydroperoxide-induced rabbit corneal neovascularization.

J Pharmacol Sci 2007 Feb 8;103(2):234-40. Epub 2007 Feb 8.

Department of Ophthalmology, School of Medicine, Showa University, Tokyo 142-8555, Japan.

The aim of this study was to investigate the effect of A-3922, a dihydrobenzofuran derivative, on linoleic acid hydroperoxide (LHP)-induced corneal neovascularization (NV) in a rabbit model. Male New Zealand rabbits received intraperitoneal (i.p.) injections of 10 or 30 mg/kg per day A-3922 or its vehicle as control for 3 days. One day after i.p. injections, LHP was injected with a 30-gauge needle into the corneal stroma of the superior quadrant 4.5-mm below the limbus. Photographs of the vessels were taken for digital analysis with a surgical microscope. Vascular endothelial growth factor (VEGF) was measured using an immunoassay kit, and matrix metalloproteinase (MMP)-9 was measured by gelatin zymography in corneal samples. At 7 days post-LHP injection, the total vessel length was 26.7 +/- 3.8 mm in the control animals (n = 8), 16.1 +/- 0.8 mm in the A-3922 (10 mg/kg)-treated group (n = 5), and 11.4 +/- 2.1 mm in the 30 mg/kg group (n = 8, P<0.01 vs control), respectively. After LHP injection, the content of VEGF and MMP-9 activity were increased in the superior cornea, but these were not influenced by A-3922 treatments. These results indicate that LHP-induced corneal NV is inhibited by treatment with A-3922 and therefore may represent a potential pharmacological intervention for ocular neovascularization disorders.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1254/jphs.fp0061301DOI Listing
February 2007
-->