Biomedical Image Databases Research Articles

Local Bit-Plane Domain 3D Oriented Arbitrary and Circular Shaped Scanning Patterns for Bio-Medical Image Retrieval

A new feature descriptor called local bit-plane domain 3D oriented arbitrary and circular shaped scanning pattern (LB-3D-OACSP) is proposed for biomedical image retrieval in this study. Unlike the circular, zigzag and other scanning structures, the LB-3D-OACSP descriptor calculates the association between reference-pixel and its surrounding pixels in bit-plane domain in a 3D plane using multi-directional 3D arbitrary and 3D circular shaped scanning patterns. In contrast to other scanning structures, the multi-directional 3-D arbitrary shaped patterns provide more continual angular dissimilarity among the sampling positions with the aim to capture more frequent changes in the local textures. The total of sixteen number of discriminative 3D arbitrary and 3D circular shaped patterns oriented in various directions are applied on a 3D plane constructed using respective bit-planes of three multi-scale images which ensures the maximum extraction of inter-scale geometrical information across the scales which very effectively captures not only the uniform but non-uniform textures too. The multi-scale images are generated by processing the input image with Gaussian filter banks generating three multi-scale images. The LB-3D-OACSP descriptor is able to capture most of the very fine to coarse image textures through encoding of bit-planes. The performance of LB-3D-OACSP is tested on three popular biomedical image databases both in terms of % average retrieval precision (ARP) and % average retrieval recall (ARR). The experiments demonstrate an encouraging enhancement in terms of %ARP and %ARR as compared to many existing state of the art descriptors.

Guest Editorial Large-Scale Medical Image and Video Analytics for Clinical Decision Support

The papers in this special section focus on large-scale medical imaging and video analytics for clinical decision support systems. Biomedical images and videos are ubiquitous and overwhelming in volume, amounting to a database that can be measured in zettabytes.With increased access to open image and video datasets and the recent development of effective image and video analysis systems, there is a unique opportunity for the development of artificial intelligence (AI) systems that can be trained and tested on large-scale biomedical image and video databases. The special issue summarizes emerging methods associated with the development of computer-aided diagnostic systems. More specifically, the special issue discusses the development of methods for dealing with small or large or creating new datasets, biomedical image segmentation, and image classification. The development of new dataset methods allows us to develop methods for specific diseases, employ meta-learning for training on small datasets, or develop methods for reducing larger video datasets. Biomedical image segmentation is a primary focus of the special issue.

Pattern of Local Gravitational Force (PLGF): A Novel Local Image Descriptor.

This paper presents a novel local image descriptor called Pattern of Local Gravitational Force (PLGF). It is inspired by Law of Universal Gravitation. PLGF is a hybrid descriptor, which is a combination of two feature components: one is the Pattern of Local Gravitational Force Magnitude (PLGFM), and another is Pattern of Local Gravitational Force Angle (PLGFA). PLGFM encodes the local gravitational force magnitude, and PLGFA encodes the local gravitational force angle that the center pixel exerts on all other pixels within a local neighborhood. We propose a novel noise resistance and the edge-preserving binary pattern called neighbors to center difference binary pattern (NCDBP) for gravitational force magnitude encoding. Finally, the histograms of the two components are concatenated to construct the PLGF descriptor. Experimental results on the existing face recognition databases, texture database, and biomedical image database show that PLGF is an effective image descriptor, and it outperforms other widely used existing descriptors. Even if in complicated variations like noise, and illumination with smaller databases, a combination of PLGF and convolutional neural network (CNN) performs consistently better than other state-of-the-art techniques.

A Robust Feature Descriptor for Biomedical Image Retrieval

Biomedical image retrieval is a crucial side of computer-aided diagnosis. It helps the radiologist and medical specialist to spot and perceive the specific disease. This paper proposed an efficient approach for retrieving similar biomedical images based on the Zernike moment features, curvelet features and histogram of oriented gradients (HOG) feature. The Zernike polynomials based moments set defines the Zernike moment which is a global descriptor and it is capable of extracting both texture and shape information with minimum redundant data. The curvelet transformation is used to compute the edge-based shape information in form of curvelet histogram for the curves with discontinuity and the HOG features calculate the happenings of gradient orientation in the local areas of an image. The experiments were conducted on four benchmark biomedical image databases: HRCT dataset, Emphysema CT database, OASIS MRI database and NEMA MRI database respectively. The performance of the proposed approach was compared with many existing methods and achieved a better retrieval rate on all the four databases.

An extension of local mesh peak valley edge based feature descriptor for image retrieval in bio-medical images

Various texture based approaches have been proposed for image indexing in bio-medical image processing and a precise description of image for indexing in bio-medical image database has always been a challenging task. In this paper, an extension of local mesh peak valley edge pattern (LMePVEP) has been proposed and its effectiveness is experimentally justified. The proposed algorithm explores the relationship of center pixel with the surrounding ones along with the relationship of pixels amongst each other in five different directions. It is then compared with the original LMePVEP as well as a directional local ternary quantized extrema pattern (DLTerQEP) based approach using two bench mark databases viz. ELCAP database for lungs and Wiki cancer data set for thyroid cancer. Further a live dataset for brain tumor is also used for experimental evaluation. The experimental results show that an average improvement of 11.16% in terms of average retrieval rate (ARR) and 5.37% in terms of average retrieval precision (ARP) is observed for proposed enhanced LMePVEP over conventional LMePVEP.

A novel biomedical image indexing and retrieval system via deep preference learning

Background and ObjectivesThe traditional biomedical image retrieval methods as well as content-based image retrieval (CBIR) methods originally designed for non-biomedical images either only consider using pixel and low-level features to describe an image or use deep features to describe images but still leave a lot of room for improving both accuracy and efficiency. In this work, we propose a new approach, which exploits deep learning technology to extract the high-level and compact features from biomedical images. The deep feature extraction process leverages multiple hidden layers to capture substantial feature structures of high-resolution images and represent them at different levels of abstraction, leading to an improved performance for indexing and retrieval of biomedical images. MethodsWe exploit the current popular and multi-layered deep neural networks, namely, stacked denoising autoencoders (SDAE) and convolutional neural networks (CNN) to represent the discriminative features of biomedical images by transferring the feature representations and parameters of pre-trained deep neural networks from another domain. Moreover, in order to index all the images for finding the similarly referenced images, we also introduce preference learning technology to train and learn a kind of a preference model for the query image, which can output the similarity ranking list of images from a biomedical image database. To the best of our knowledge, this paper introduces preference learning technology for the first time into biomedical image retrieval. ResultsWe evaluate the performance of two powerful algorithms based on our proposed system and compare them with those of popular biomedical image indexing approaches and existing regular image retrieval methods with detailed experiments over several well-known public biomedical image databases. Based on different criteria for the evaluation of retrieval performance, experimental results demonstrate that our proposed algorithms outperform the state-of-the-art techniques in indexing biomedical images. ConclusionsWe propose a novel and automated indexing system based on deep preference learning to characterize biomedical images for developing computer aided diagnosis (CAD) systems in healthcare. Our proposed system shows an outstanding indexing ability and high efficiency for biomedical image retrieval applications and it can be used to collect and annotate the high-resolution images in a biomedical database for further biomedical image research and applications.

Spatial Co-Location Pattern Discovery from Fuzzy Objects

A spatial co-location pattern is a group of spatial objects whose instances are frequently located in the same region. The spatial co-location pattern mining problem has been investigated extensively in the past due to its broad range of applications. In this paper we study this problem for fuzzy objects. Fuzzy objects play an important role in many areas, such as the geographical information system and the biomedical image database. In this paper, we propose two new kinds of co-location pattern mining for fuzzy objects, single co-location pattern mining (SCP) and range co-location pattern mining (RCP), to mining co-location patterns at a membership threshold or within a membership range. For efficient SCP mining, we optimize the basic mining algorithm to accelerate the co-location pattern generation. To improve the performance of RCP mining, effective pruning strategies are developed to significantly reduce the search space. The efficiency of our proposed algorithms as well as the optimization techniques are verified with an extensive set of experiments.

AggNet: Deep Learning From Crowds for Mitosis Detection in Breast Cancer Histology Images.

The lack of publicly available ground-truth data has been identified as the major challenge for transferring recent developments in deep learning to the biomedical imaging domain. Though crowdsourcing has enabled annotation of large scale databases for real world images, its application for biomedical purposes requires a deeper understanding and hence, more precise definition of the actual annotation task. The fact that expert tasks are being outsourced to non-expert users may lead to noisy annotations introducing disagreement between users. Despite being a valuable resource for learning annotation models from crowdsourcing, conventional machine-learning methods may have difficulties dealing with noisy annotations during training. In this manuscript, we present a new concept for learning from crowds that handle data aggregation directly as part of the learning process of the convolutional neural network (CNN) via additional crowdsourcing layer (AggNet). Besides, we present an experimental study on learning from crowds designed to answer the following questions. 1) Can deep CNN be trained with data collected from crowdsourcing? 2) How to adapt the CNN to train on multiple types of annotation datasets (ground truth and crowd-based)? 3) How does the choice of annotation and aggregation affect the accuracy? Our experimental setup involved Annot8, a self-implemented web-platform based on Crowdflower API realizing image annotation tasks for a publicly available biomedical image database. Our results give valuable insights into the functionality of deep CNN learning from crowd annotations and prove the necessity of data aggregation integration.

Local Mesh Patterns Versus Local Binary Patterns: Biomedical Image Indexing and Retrieval

In this paper, a new image indexing and retrieval algorithm using local mesh patterns are proposed for biomedical image retrieval application. The standard local binary pattern encodes the relationship between the referenced pixel and its surrounding neighbors, whereas the proposed method encodes the relationship among the surrounding neighbors for a given referenced pixel in an image. The possible relationships among the surrounding neighbors are depending on the number of neighbors, P. In addition, the effectiveness of our algorithm is confirmed by combining it with the Gabor transform. To prove the effectiveness of our algorithm, three experiments have been carried out on three different biomedical image databases. Out of which two are meant for computer tomography (CT) and one for magnetic resonance (MR) image retrieval. It is further mentioned that the database considered for three experiments are OASIS-MRI database, NEMA-CT database, and VIA/I-ELCAP database which includes region of interest CT images. The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP, LBP with Gabor transform, and other spatial and transform domain methods.

Batch Mode Active Sampling Based on Marginal Probability Distribution Matching

Active Learning is a machine learning and data mining technique that selects the most informative samples for labeling and uses them as training data; it is especially useful when there are large amount of unlabeled data and labeling them is expensive. Recently, batch-mode active learning, where a set of samples are selected concurrently for labeling, based on their collective merit, has attracted a lot of attention. The objective of batch-mode active learning is to select a set of informative samples so that a classifier learned on these samples has good generalization performance on the unlabeled data. Most of the existing batch-mode active learning methodologies try to achieve this by selecting samples based on certain criteria. In this article we propose a novel criterion which achieves good generalization performance of a classifier by specifically selecting a set of query samples that minimize the difference in distribution between the labeled and the unlabeled data, after annotation. We explicitly measure this difference based on all candidate subsets of the unlabeled data and select the best subset. The proposed objective is an NP-hard integer programming optimization problem. We provide two optimization techniques to solve this problem. In the first one, the problem is transformed into a convex quadratic programming problem and in the second method the problem is transformed into a linear programming problem. Our empirical studies using publicly available UCI datasets and two biomedical image databases demonstrate the effectiveness of the proposed approach in comparison with the state-of-the-art batch-mode active learning methods. We also present two extensions of the proposed approach, which incorporate uncertainty of the predicted labels of the unlabeled data and transfer learning in the proposed formulation. In addition, we present a joint optimization framework for performing both transfer and active learning simultaneously unlike the existing approaches of learning in two separate stages, that is, typically, transfer learning followed by active learning. We specifically minimize a common objective of reducing distribution difference between the domain adapted source, the queried and labeled samples and the rest of the unlabeled target domain data. Our empirical studies on two biomedical image databases and on a publicly available 20 Newsgroups dataset show that incorporation of uncertainty information and transfer learning further improves the performance of the proposed active learning based classifier. Our empirical studies also show that the proposed transfer-active method based on the joint optimization framework performs significantly better than a framework which implements transfer and active learning in two separate stages.

IMole, a web based image retrieval system from biomedical literature

iMole is a platform that automatically extracts images and captions from biomedical literature. Images are tagged with terms contained in figure captions by means of a sophisticate text-mining tool. Moreover, iMole allows the user to upload directly their own images within the database and manually tag images by curated dictionary. Using iMole the researchers can develop a proper biomedical image database, storing the images extracted from paper of interest, image found on the web repositories, and their own experimental images. In order to show the functioning of the platform, we used iMole to build a 2DE database. Briefly, tagged 2DE gel images were collected and stored in a searchable 2DE gel database, available to users through an interactive web interface. Images were obtained by automatically parsing 16,608 proteomic publications, which yielded more than 16,500 images. The database can be further expanded by users with images of interest trough a manual uploading process. iMole is available with a preloaded set of 2DE gel data at http://imole.biodigitalvalley.com.

Local ternary co-occurrence patterns: A new feature descriptor for MRI and CT image retrieval

This paper presents a novel feature extraction algorithm called local ternary co-occurrence patterns (LTCoP) for biomedical image retrieval. The LTCoP encodes the co-occurrence of similar ternary edges which are calculated based on the gray values of center pixel and its surrounding neighbors. Whereas the standard local derivative pattern (LDP) encodes the co-occurrence between the first-order derivatives in a specific direction. The existing LDP is a specific direction rotational variant feature where as our method is rotational invariant. In addition, the effectiveness of our algorithm is confirmed by combining it with the Gabor transform. To prove the effectiveness of our algorithm, three experiments have been carried out on three different biomedical image databases. Out of which two are meant for computer tomography (CT) and one for magnetic resonance (MR) image retrieval. It is further mentioned that the database considered for three experiments are OASIS−MRI database, NEMA−CT database and VIA/I–ELCAP database which includes region of interest CT images. The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP, LTP, local tetra patterns (LTrP) and LDP with and without Gabor transform.

Spatial query processing for fuzzy objects

Range and nearest neighbor queries are the most common types of spatial queries, which have been investigated extensively in the last decades due to its broad range of applications. In this paper, we study this problem in the context of fuzzy objects that have indeterministic boundaries. Fuzzy objects play an important role in many areas, such as biomedical image databases and GIS communities. Existing research on fuzzy objects mainly focuses on modeling basic fuzzy object types and operations, leaving the processing of more advanced queries largely untouched. In this paper, we propose two new kinds of spatial queries for fuzzy objects, namely single threshold query and continuous threshold query, to determine the query results which qualify at a certain probability threshold and within a probability interval, respectively. For efficient single threshold query processing, we optimize the classical R-tree-based search algorithm by deriving more accurate approximations for the distance function between fuzzy objects and the query object. To enhance the performance of continuous threshold queries, effective pruning rules are developed to reduce the search space and speed up the candidate refinement process. The efficiency of our proposed algorithms as well as the optimization techniques is verified with an extensive set of experiments using both synthetic and real datasets.

141時空間多次元生体情報の解析と可視化

In order to efficiently develop image analysis and visualize software of multi-dimensional biomedical information such as x-ray CT and MRI, we construct integrated development system under Windows 98.For the purpose, imaging software library and multi-dimensional biomedical image database are arranged as the integrated development system by using Ms-Access. All software are developed by object oriented language, Delphi.Object oriented language is effective for modification and improvement of program. We develop dynamic studying algorithms, such as animation display, organ functional imaging for nuclear image and volume rendering software for x-ray CT images.In order to evaluate the effectiveness of the various clinical imaging and the integrated development system, we open the various clinical software and algorithm on our homepage.

Design and implementation of a biomedical image database (BDIM).

We developed a biomedical image database (BDIM) which proposes a standardized representation of value arrays such as images and curves, and of their associated parameters, independently of their acquisition mode to make their transmission and processing easier. It includes three kinds of interactions, oriented to the users. The network concept was kept as a constraint to incorporate the BDIM in a distributed structure and we maintained compatibility with the ACR/NEMA communication protocol. The management of arrays and their associated parameters includes two distinct bases of objects, linked together via a gateway. The first one manages arrays according to their storage mode: long term storage on optionally on-line mass storage devices, and, for consultations, partial copies of long term stored arrays on hard disk. The second one manages the associated parameters and the gateway by means of the relational DBMS ORACLE. Parameters are grouped into relations. Some of them are in agreement with groups defined by the ACR/NEMA. The other relations describe objects resulting from processed initial objects. These new objects are not described by the ACR/NEMA but they can be inserted as shadow groups of ACR/NEMA description. The relations describing the storage and their pathname constitute the gateway. ORACLE distributed tools and the two-level storage technique will allow the integration of the BDIM into a distributed structure, Queries and array (alone or in sequences) retrieval module has access to the relations via a level in which a dictionary managed by ORACLE is included. This dictionary translates ACR/NEMA objects into objects that can be handled by the DBMS.(ABSTRACT TRUNCATED AT 250 WORDS)