scholarly journals Metadata-Guided Visual Representation Learning for Biomedical Images

2019 ◽  
Author(s):  
Stephan Spiegel ◽  
Imtiaz Hossain ◽  
Christopher Ball ◽  
Xian Zhang
Keyword(s):  
Learning Strategies ◽  
Visual Representation ◽  
Morphological Characteristics ◽  
Representation Learning ◽  
Biological Data ◽  
Biomedical Images ◽  
Visual Appearance ◽  
Extra Effort ◽  
Novel Approach ◽  
Implicit And Explicit

AbstractMotivationThe clustering of biomedical images according to their phenotype is an important step in early drug discovery. Modern high-content-screening devices easily produce thousands of cell images, but the resulting data is usually unlabelled and it requires extra effort to construct a visual representation that supports the grouping according to the presented morphological characteristics.ResultsWe introduce a novel approach to visual representation learning that is guided by metadata. In high-context-screening, meta-data can typically be derived from the experimental layout, which links each cell image of a particular assay to the tested chemical compound and corresponding compound concentration. In general, there exists a one-to-many relationship between phenotype and compound, since various molecules and different dosage can lead to one and the same alterations in biological cells.Our empirical results show that metadata-guided visual representation learning is an effective approach for clustering biomedical images. We have evaluated our proposed approach on both benchmark and real-world biological data. Furthermore, we have juxtaposed implicit and explicit learning techniques, where both loss function and batch construction differ. Our experiments demonstrate that metadata-guided visual representation learning is able to identify commonalities and distinguish differences in visual appearance that lead to meaningful clusters, even without image-level annotations.NotePlease refer to the supplementary material for implementation details on metadata-guided visual representation learning strategies.

W-MMP2Vec: Topic-driven network embedding model for link prediction in content-based heterogeneous information network

2021 ◽  
Vol 25 (3) ◽  
pp. 711-738
Author(s):  
Phu Pham ◽  
Phuc Do
Keyword(s):  
Link Prediction ◽  
Representation Learning ◽  
Information Network ◽  
Network Embedding ◽  
Heterogeneous Information Network ◽  
Heterogeneous Information ◽  
Learning Framework ◽  
Novel Approach ◽  
Proposed Model ◽  
Meta Path

Link prediction on heterogeneous information network (HIN) is considered as a challenge problem due to the complexity and diversity in types of nodes and links. Currently, there are remained challenges of meta-path-based link prediction in HIN. Previous works of link prediction in HIN via network embedding approach are mainly focused on exploiting features of node rather than existing relations in forms of meta-paths between nodes. In fact, predicting the existence of new links between non-linked nodes is absolutely inconvincible. Moreover, recent HIN-based embedding models also lack of thorough evaluations on the topic similarity between text-based nodes along given meta-paths. To tackle these challenges, in this paper, we proposed a novel approach of topic-driven multiple meta-path-based HIN representation learning framework, namely W-MMP2Vec. Our model leverages the quality of node representations by combining multiple meta-paths as well as calculating the topic similarity weight for each meta-path during the processes of network embedding learning in content-based HINs. To validate our approach, we apply W-TMP2Vec model in solving several link prediction tasks in both content-based and non-content-based HINs (DBLP, IMDB and BlogCatalog). The experimental outputs demonstrate the effectiveness of proposed model which outperforms recent state-of-the-art HIN representation learning models.

scholarly journals Efficient Rank-Based Diffusion Process with Assured Convergence

2021 ◽  
Vol 7 (3) ◽  
pp. 49
Author(s):  
Daniel Carlos Guimarães Pedronette ◽  
Lucas Pascotti Valem ◽  
Longin Jan Latecki
Keyword(s):  
Diffusion Process ◽  
Learning Strategies ◽  
State Of The Art ◽  
Representation Learning ◽  
Theoretical Background ◽  
High Dimensional ◽  
Visual Features ◽  
Learning Approaches ◽  
Previous Decade ◽  
Asymptotic Complexity

Visual features and representation learning strategies experienced huge advances in the previous decade, mainly supported by deep learning approaches. However, retrieval tasks are still performed mainly based on traditional pairwise dissimilarity measures, while the learned representations lie on high dimensional manifolds. With the aim of going beyond pairwise analysis, post-processing methods have been proposed to replace pairwise measures by globally defined measures, capable of analyzing collections in terms of the underlying data manifold. The most representative approaches are diffusion and ranked-based methods. While the diffusion approaches can be computationally expensive, the rank-based methods lack theoretical background. In this paper, we propose an efficient Rank-based Diffusion Process which combines both approaches and avoids the drawbacks of each one. The obtained method is capable of efficiently approximating a diffusion process by exploiting rank-based information, while assuring its convergence. The algorithm exhibits very low asymptotic complexity and can be computed regionally, being suitable to outside of dataset queries. An experimental evaluation conducted for image retrieval and person re-ID tasks on diverse datasets demonstrates the effectiveness of the proposed approach with results comparable to the state-of-the-art.

A novel approach to sagittal balance restoration following iatrogenic sacral fracture and resulting sacral kyphotic deformity

2007 ◽  
Vol 6 (4) ◽  
pp. 368-372 ◽  
Author(s):  
Patrick C. Hsieh ◽  
Stephen L. Ondra ◽  
Robert J. Wienecke ◽  
Brian A. O'Shaughnessy ◽  
Tyler R. Koski
Keyword(s):  
Sagittal Balance ◽  
Pelvic Incidence ◽  
Morphological Characteristics ◽  
Pedicle Subtraction Osteotomy ◽  
Sacral Fracture ◽  
Novel Approach ◽  
Sagittal Spinal Alignment ◽  
Fixed Parameter ◽  
Lumbar Hyperlordosis ◽  
Sagittal Deformity

✓The authors describe the use of sacral pedicle subtraction osteotomy (PSO) with multiple sacral alar osteotomies for the correction of sacral kyphosis and pelvic incidence and for achieving sagittal balance correction in cases of fixed sagittal deformity after a sacral fracture. In this paper, the authors report on a novel technique using a series of sacral osteotomies and a sacral PSO to correct a fixed sagittal deformity in a patient with a sacral fracture that had healed in a kyphotic position. The patient sustained this fracture after a previous surgery for multilevel instrumented fusion. Preoperative and postoperative radiographic studies are reviewed and the clinical course and outcome are presented. Experts agree that the pelvic incidence is a fixed parameter that dictates the morphological characteristics of the pelvis and affects spinopelvic orientation and sagittal spinal alignment. An increased pelvic incidence is associated with a higher degree of spondylolisthesis in the lumbosacral junction, and increased shear forces across this junction. The authors demonstrate that the pelvic incidence can be altered and corrected with a series of sacral osteotomies to improve sacral kyphosis, compensatory lumbar hyperlordosis, and sagittal balance.

Patterns of polyembryony and frequency of surviving multiple embryos of the Brazilian pine Araucaria angustifolia

2011 ◽  
Vol 59 (8) ◽  
pp. 749 ◽  
Author(s):  
Sarah Zanon Agapito-Tenfen ◽  
Neusa Steiner ◽  
Miguel Pedro Guerra ◽  
Rubens Onofre Nodari
Keyword(s):  
Microsatellite Markers ◽  
Developmental Stages ◽  
Rare Event ◽  
Morphological Characteristics ◽  
Immature Embryos ◽  
Araucaria Angustifolia ◽  
Embryo Survival ◽  
Major Mechanism ◽  
Novel Approach ◽  
Nuclear Microsatellite

The development of polyembryony is a common reproductive strategy in conifers. Multiple embryos are observed during early seed developmental stages. However, upon seed maturation, only the dominant embryo survives, with few exceptions. Although programmed cell death has been reported as the major mechanism responsible for elimination of subordinate embryos, the genetics of surviving embryos and the probabilities of survival remain unclear. The aim of this study is to determine patterns of polyembryony and survival frequency in Araucaria angustifolia (Bert) O. Ktze. Thus, we investigate the morphogenetic parameters that might be related to embryo survival using nuclear microsatellite markers and morphological characteristics of immature embryos and seedlings. Our novel approach couples genotype frequency analysis with the number of surviving embryos, presence of embryo dominance and number of cotyledons present within a single seed. Polyembryonic seedling frequency was low (0.022%) and 91% of surviving embryos were monozygotic. From all monozygotic embryos, 98% showed differences in growth rate (height) in relation to each other. Concrescent tissues were common in the monozygotic polyembryony patterns observed (80%) but not for those with polyzygotic polyembryony. We demonstrate that the survival of multiple embryos is a rare event in A. angustifolia seeds. To the best of our knowledge this study represents the first evidence of cleavage polyembryony in immature embryos and seedlings from A. angustifolia. Our novel approach using a combined set of morphological parameters and microsatellite markers was successful in investigating polyembryony patterns and survival.

scholarly journals Towards CNN Representations for Small Mass Spectrometry Data Classification: From Transfer Learning to Cumulative Learning

2020 ◽  
Author(s):  
Khawla Seddiki ◽  
Philippe Saudemont ◽  
Frédéric Precioso ◽  
Nina Ogrinc ◽  
Maxence Wisztorski ◽  
...  
Keyword(s):  
Mass Spectrometry ◽  
Learning Strategies ◽  
Transfer Learning ◽  
Classification Accuracy ◽  
Data Classification ◽  
Representation Learning ◽  
Mass Spectrometry Data ◽  
Training Dataset ◽  
Require Time ◽  
Training Samples

AbstractRapid and accurate clinical diagnosis of pathological conditions remains highly challenging. A very important component of diagnosis tool development is the design of effective classification models with Mass spectrometry (MS) data. Some popular Machine Learning (ML) approaches have been investigated for this purpose but these ML models require time-consuming preprocessing steps such as baseline correction, denoising, and spectrum alignment to remove non-sample-related data artifacts. They also depend on the tedious extraction of handcrafted features, making them unsuitable for rapid analysis. Convolutional Neural Networks (CNNs) have been found to perform well under such circumstances since they can learn efficient representations from raw data without the need for costly preprocessing. However, their effectiveness drastically decreases when the number of available training samples is small, which is a common situation in medical applications. Transfer learning strategies extend an accurate representation model learnt usually on a large dataset containing many categories, to a smaller dataset with far fewer categories. In this study, we first investigate transfer learning on a 1D-CNN we have designed to classify MS data, then we develop a new representation learning method when transfer learning is not powerful enough, as in cases of low-resolution or data heterogeneity. What we propose is to train the same model through several classification tasks over various small datasets in order to accumulate generic knowledge of what MS data are, in the resulting representation. By using rat brain data as the initial training dataset, a representation learning approach can have a classification accuracy exceeding 98% for canine sarcoma cancer cells, human ovarian cancer serums, and pathogenic microorganism biotypes in 1D clinical datasets. We show for the first time the use of cumulative representation learning using datasets generated in different biological contexts, on different organisms, in different mass ranges, with different MS ionization sources, and acquired by different instruments at different resolutions. Our approach thus proposes a promising strategy for improving MS data classification accuracy when only small numbers of samples are available as a prospective cohort. The principles demonstrated in this work could even be beneficial to other domains (astronomy, archaeology…) where training samples are scarce.

scholarly journals Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity

2019 ◽  
Author(s):  
Yunsheng Bai ◽  
Hao Ding ◽  
Yang Qiao ◽  
Agustin Marinovic ◽  
Ken Gu ◽  
...  
Keyword(s):  
Neural Network ◽  
Vector Space ◽  
General Framework ◽  
Representation Learning ◽  
Generation Mechanism ◽  
Graph Visualization ◽  
Graph Classification ◽  
Training Set ◽  
Multi Scale ◽  
Novel Approach

We introduce a novel approach to graph-level representation learning, which is to embed an entire graph into a vector space where the embeddings of two graphs preserve their graph-graph proximity. Our approach, UGraphEmb, is a general framework that provides a novel means to performing graph-level embedding in a completely unsupervised and inductive manner. The learned neural network can be considered as a function that receives any graph as input, either seen or unseen in the training set, and transforms it into an embedding. A novel graph-level embedding generation mechanism called Multi-Scale Node Attention (MSNA), is proposed. Experiments on five real graph datasets show that UGraphEmb achieves competitive accuracy in the tasks of graph classification, similarity ranking, and graph visualization.

scholarly journals A novel approach to study the morphology and chemistry of pollen in a phylogenetic context, applied to the halophytic taxon Nitraria L.(Nitrariaceae)

PeerJ ◽  
2018 ◽  
Vol 6 ◽  
pp. e5055 ◽  
Author(s):  
Amber Woutersen ◽  
Phillip E. Jardine ◽  
Raul Giovanni Bogotá-Angel ◽  
Hong-Xiang Zhang ◽  
Daniele Silvestro ◽  
...  
Keyword(s):  
Chemical Composition ◽  
Evolutionary History ◽  
Morphological Characteristics ◽  
Pollen Wall ◽  
Secondary Process ◽  
The Tibetan Plateau ◽  
Plant Family ◽  
Novel Approach ◽  
History Of ◽  
Components Analysis

Nitraria is a halophytic taxon (i.e., adapted to saline environments) that belongs to the plant family Nitrariaceae and is distributed from the Mediterranean, across Asia into the south-eastern tip of Australia. This taxon is thought to have originated in Asia during the Paleogene (66–23 Ma), alongside the proto-Paratethys epicontinental sea. The evolutionary history of Nitraria might hold important clues on the links between climatic and biotic evolution but limited taxonomic documentation of this taxon has thus far hindered this line of research. Here we investigate if the pollen morphology and the chemical composition of the pollen wall are informative of the evolutionary history of Nitraria and could explain if origination along the proto-Paratethys and dispersal to the Tibetan Plateau was simultaneous or a secondary process. To answer these questions, we applied a novel approach consisting of a combination of Fourier Transform Infrared spectroscopy (FTIR), to determine the chemical composition of the pollen wall, and pollen morphological analyses using Light Microscopy (LM) and Scanning Electron Microscopy (SEM). We analysed our data using ordinations (principal components analysis and non-metric multidimensional scaling), and directly mapped it on the Nitrariaceae phylogeny to produce a phylomorphospace and a phylochemospace. Our LM, SEM and FTIR analyses show clear morphological and chemical differences between the sister groups Peganum and Nitraria. Differences in the morphological and chemical characteristics of highland species (Nitraria schoberi, N. sphaerocarpa, N. sibirica and N. tangutorum) and lowland species (Nitraria billardierei and N. retusa) are very subtle, with phylogenetic history appearing to be a more important control on Nitraria pollen than local environmental conditions. Our approach shows a compelling consistency between the chemical and morphological characteristics of the eight studied Nitrariaceae species, and these traits are in agreement with the phylogenetic tree. Taken together, this demonstrates how novel methods for studying fossil pollen can facilitate the evolutionary investigation of living and extinct taxa, and the environments they represent.

scholarly journals THE ROLE OF LEARNING STRATEGIES IN THE PROCESS OF ESP ACQUISITION OF INTENDING EDUCATORS

2021 ◽  
Vol 5 ◽  
pp. 535-546
Author(s):  
Natalia Dmitrenko ◽  
Anastasiia Petrova ◽  
Olena Podzygun
Keyword(s):  
Learning Strategies ◽  
Research Topic ◽  
Explicit Model ◽  
Mental Processes ◽  
Experimental Teaching ◽  
Implicit And Explicit ◽  
Experimental Training

The article is dedicated to the problem of using learning strategies in the process of ESP acquisition of intending educators. The authors analyzed the methodological literature on the research topic, on the basis of which the definitions and main characteristics of learning strategies, as well as the implicit and explicit ways of using the strategies in the process of ESP acquisition of intending educators, are presented. R. Oxford’s explicit model of using the instruction-based learning strategies is considered in detail: an algorithm for its use is presented, as well as the basic rules that should be followed in the process of using the model. During the experimental teaching of ESP of intending teachers, R. Oxford’s explicit model of using the instruction-based learning strategies was tested. At the beginning and the end of the experimental training, the questionnaires were conducted to establish the level of proficiency in the learning strategies, as well as the control testing was performed to define the level of ESP acquisition of intending educators. The results of the research confirmed the effectiveness of the used model, which showed an increase in the level of proficiency in the strategies: effective memorization, mental processes, compensatory ability, organization of education and its assessment, management of emotions and collaboration with other students. In turn, the increase in the level of proficiency in the learning strategies was positively reflected in the level of ESP acquisition of intending educators. 

Sign in / Sign up

Export Citation Format

Share Document