Fusing Feature Distribution Entropy with R-MAC Features in Image Retrieval
Ontology highlight
ABSTRACT:
SUBMITTER: Liu P
PROVIDER: S-EPMC7514342 | biostudies-literature | 2019 Oct
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Analysis of high-dimensional data is a challenge in machine learning and data mining. Feature selection plays an important role in dealing with high-dimensional data for improvement of predictive accuracy, as well as better interpretation of the data. Frequently used evaluation functions for feature selection include resampling methods such as cross-validation, which show an advantage in predictive accuracy. However, these conventional methods are not only computationally expensive, but also tend to be over-optimistic. We propose a novel cross-entropy which is based on beta distribution for feature selection. In beta distribution-based cross-entropy (BetaDCE) for feature selection, the probability density is estimated by beta distribution and the cross-entropy is computed by the expected value of beta distribution, so that the generalization ability can be estimated more precisely than conventional methods where the probability density is learnt from data. Analysis of the generalization ability of BetaDCE revealed that it was a trade-off between bias and variance. The robustness of BetaDCE was demonstrated by experiments on three types of data. In the exclusive or-like (XOR-like) dataset, the false discovery rate of BetaDCE was significantly smaller than that of other methods. For the leukemia dataset, the area under the curve (AUC) of BetaDCE on the test set was 0.93 with only four selected features, which indicated that BetaDCE not only detected the irrelevant and redundant features precisely, but also more accurately predicted the class labels with a smaller number of features than the original method, whose AUC was 0.83 with 50 features. In the metabonomic dataset, the overall AUC of prediction with features selected by BetaDCE was significantly larger than that by the original reported method. Therefore, BetaDCE can be used as a general and efficient framework for feature selection.
| S-EPMC7515297 | biostudies-literature
Project description:Lung cancer manifests itself in the form of lung nodules, the diagnosis of which is essential to plan the treatment. Automated retrieval of nodule cases will assist the budding radiologists in self-learning and differential diagnosis. This paper presents a content-based image retrieval (CBIR) system for lung nodules using optimal feature sets and learning to enhance the performance of retrieval. The classifiers with more features suffer from the curse of dimensionality. Like classification schemes, we found that the optimal feature set selected using the minimal-redundancy-maximal-relevance (mRMR) feature selection technique improves the precision performance of simple distance-based retrieval (SDR). The performance of the classifier is always superior to SDR, which leans researchers towards conventional classifier-based retrieval (CCBR). While CCBR improves the average precision and provides 100% precision for correct classification, it fails for misclassification leading to zero retrieval precision. The class membership-based retrieval (CMR) is found to bridge this gap for texture-based retrieval. Here, CMR is proposed for nodule retrieval using shape-, margin-, and texture-based features. It is found again that optimal feature set is important for the classifier used in CMR as well as for the feature set used for retrieval, which may lead to different feature sets. The proposed system is evaluated using two independent databases from two continents: a public database LIDC/IDRI and a private database PGIMER-IITKGP, using three distance metrics, i.e., Canberra, City block, and Euclidean. The proposed CMR-based retrieval system with optimal feature sets performs better than CCBR and SDR with optimal features in terms of average precision. Apart from average precision and standard deviation of precision, the fraction of queries with zero precision retrieval is also measured.
| S-EPMC6499853 | biostudies-literature
Project description:An invariant feature matching method is proposed as a spatially invariant feature matching approach. Deformation effects, such as affine and homography, change the local information within the image and can result in ambiguous local information pertaining to image points. New method based on dissimilarity values, which measures the dissimilarity of the features through the path based on Eigenvector properties, is proposed. Evidence shows that existing matching techniques using similarity metrics--such as normalized cross-correlation, squared sum of intensity differences and correlation coefficient--are insufficient for achieving adequate results under different image deformations. Thus, new descriptor's similarity metrics based on normalized Eigenvector correlation and signal directional differences, which are robust under local variation of the image information, are proposed to establish an efficient feature matching technique. The method proposed in this study measures the dissimilarity in the signal frequency along the path between two features. Moreover, these dissimilarity values are accumulated in a 2D dissimilarity space, allowing accurate corresponding features to be extracted based on the cumulative space using a voting strategy. This method can be used in image registration applications, as it overcomes the limitations of the existing approaches. The output results demonstrate that the proposed technique outperforms the other methods when evaluated using a standard dataset, in terms of precision-recall and corner correspondence.
| S-EPMC4795769 | biostudies-literature
Project description:Machine learning-based cell classifiers use cell images to automate cell-type discrimination, which is increasingly becoming beneficial in biological studies and biomedical applications. Brightfield or fluorescence images are generally employed as the classifier input variables. We propose to use Raman spectral images and a method to extract features from these spatial patterns and explore the value of this information for cell discrimination. Raman images provide information regarding distribution of chemical compounds of the considered biological entity. Since each spectral wavelength can be used to reconstruct the distribution of a given compound, spectral images provide multiple channels of information, each representing a different pattern, in contrast to brightfield and fluorescence images. Using a dataset of single living cells, we demonstrate that the spatial information can be ranked by a Fisher discriminant score, and that the top-ranked features can accurately classify cell types. This method is compared with the conventional Raman spectral analysis. We also propose to combine the information from whole spectral analyses and selected spatial features and show that this yields higher classification accuracy. This method provides the basis for a novel and systematic analysis of cell-type investigation using Raman spectral imaging, which may benefit several studies and biomedical applications.
| S-EPMC6079059 | biostudies-literature
Project description:Since the assessment of wheat diseases (e.g., leaf rust and tan spot) via visual observation is subjective and inefficient, this study focused on developing an automatic, objective, and efficient diagnosis approach. For each plant, color, and color-infrared (CIR) images were collected in a paired mode. An automatic approach based on the image processing technique was developed to crop the paired images to have the same region, after which a developed semiautomatic webtool was used to expedite the dataset creation. The webtool generated the dataset from either image and automatically built the corresponding dataset from the other image. Each image was manually categorized into one of the three groups: control (disease-free), disease light, and disease severity. After the image segmentation, handcrafted features (HFs) were extracted from each format of images, and disease diagnosis results demonstrated that the parallel feature fusion had higher accuracy over features from either type of image. Performance of deep features (DFs) extracted through different deep learning (DL) models (e.g., AlexNet, VGG16, ResNet101, GoogLeNet, and Xception) on wheat disease detection was compared, and those extracted by ResNet101 resulted in the highest accuracy, perhaps because deep layers extracted finer features. In addition, parallel deep feature fusion generated a higher accuracy over DFs from a single-source image. DFs outperformed HFs in wheat disease detection, and the DFs coupled with parallel feature fusion resulted in diagnosis accuracies of 75, 84, and 71% for leaf rust, tan spot, and leaf rust + tan spot, respectively. The methodology developed directly for greenhouse applications, to be used by plant pathologists, breeders, and other users, can be extended to field applications with future tests on field data and model fine-tuning.
| S-EPMC8965652 | biostudies-literature
Project description:Retinal image registration is important to assist diagnosis and monitor retinal diseases, such as diabetic retinopathy and glaucoma. However, registering retinal images for various registration applications requires the detection and distribution of feature points on the low-quality region that consists of vessels of varying contrast and sizes. A recent feature detector known as Saddle detects feature points on vessels that are poorly distributed and densely positioned on strong contrast vessels. Therefore, we propose a multiresolution difference of Gaussian pyramid with Saddle detector (D-Saddle) to detect feature points on the low-quality region that consists of vessels with varying contrast and sizes. D-Saddle is tested on Fundus Image Registration (FIRE) Dataset that consists of 134 retinal image pairs. Experimental results show that D-Saddle successfully registered 43% of retinal image pairs with average registration accuracy of 2.329 pixels while a lower success rate is observed in other four state-of-the-art retinal image registration methods GDB-ICP (28%), Harris-PIIFD (4%), H-M (16%), and Saddle (16%). Furthermore, the registration accuracy of D-Saddle has the weakest correlation (Spearman) with the intensity uniformity metric among all methods. Finally, the paired t-test shows that D-Saddle significantly improved the overall registration accuracy of the original Saddle.
| S-EPMC5674727 | biostudies-other
Project description:Serial endoscopic examinations of a patient are important for early diagnosis of malignancies in the gastrointestinal tract. However, retargeting for optical biopsy is challenging due to extensive tissue variations between examinations, requiring the method to be tolerant to these changes whilst enabling real-time retargeting.This work presents an image retrieval framework for inter-examination retargeting. We propose both a novel image descriptor tolerant of long-term tissue changes and a novel descriptor matching method in real time. The descriptor is based on histograms generated from regional intensity comparisons over multiple scales, offering stability over long-term appearance changes at the higher levels, whilst remaining discriminative at the lower levels. The matching method then learns a hashing function using random forests, to compress the string and allow for fast image comparison by a simple Hamming distance metric.A dataset that contains 13 in vivo gastrointestinal videos was collected from six patients, representing serial examinations of each patient, which includes videos captured with significant time intervals. Precision-recall for retargeting shows that our new descriptor outperforms a number of alternative descriptors, whilst our hashing method outperforms a number of alternative hashing approaches.We have proposed a novel framework for optical biopsy in serial endoscopic examinations. A new descriptor, combined with a novel hashing method, achieves state-of-the-art retargeting, with validation on in vivo videos from six patients. Real-time performance also allows for practical integration without disturbing the existing clinical workflow.
| S-EPMC5541128 | biostudies-other
Project description:Content-based medical image retrieval continues to gain attention for its potential to assist radiological image interpretation and decision making. Many approaches have been proposed to improve the performance of medical image retrieval system, among which visual features such as SIFT, LBP, and intensity histogram play a critical role. Typically, these features are concatenated into a long vector to represent medical images, and thus traditional dimension reduction techniques such as locally linear embedding (LLE), principal component analysis (PCA), or laplacian eigenmaps (LE) can be employed to reduce the "curse of dimensionality". Though these approaches show promising performance for medical image retrieval, the feature-concatenating method ignores the fact that different features have distinct physical meanings. In this paper, we propose a new method called multiview locally linear embedding (MLLE) for medical image retrieval. Following the patch alignment framework, MLLE preserves the geometric structure of the local patch in each feature space according to the LLE criterion. To explore complementary properties among a range of features, MLLE assigns different weights to local patches from different feature spaces. Finally, MLLE employs global coordinate alignment and alternating optimization techniques to learn a smooth low-dimensional embedding from different features. To justify the effectiveness of MLLE for medical image retrieval, we compare it with conventional spectral embedding methods. We conduct experiments on a subset of the IRMA medical image data set. Evaluation results show that MLLE outperforms state-of-the-art dimension reduction methods.
| S-EPMC3862625 | biostudies-literature
Project description:Acute infections of the middle ear are the most commonly treated childhood diseases. Because complications affect children's language learning and cognitive processes, it is essential to diagnose these diseases in a timely and accurate manner. The prevailing literature suggests that it is difficult to accurately diagnose these infections, even for experienced ear, nose, and throat (ENT) physicians. Advanced care practitioners (e.g., nurse practitioners, physician assistants) serve as first-line providers in many primary care settings and may benefit from additional guidance to appropriately determine the diagnosis and treatment of ear diseases. For this purpose, we designed a content-based image retrieval (CBIR) system (called OtoMatch) for normal, middle ear effusion, and tympanostomy tube conditions, operating on eardrum images captured with a digital otoscope. We present a method that enables the conversion of any convolutional neural network (trained for classification) into an image retrieval model. As a proof of concept, we converted a pre-trained deep learning model into an image retrieval system. We accomplished this by changing the fully connected layers into lookup tables. A database of 454 labeled eardrum images (179 normal, 179 effusion, and 96 tube cases) was used to train and test the system. On a 10-fold cross validation, the proposed method resulted in an average accuracy of 80.58% (SD 5.37%), and maximum F1 score of 0.90 while retrieving the most similar image from the database. These are promising results for the first study to demonstrate the feasibility of developing a CBIR system for eardrum images using the newly proposed methodology.
| S-EPMC7228122 | biostudies-literature
Project description:Recent advances of single-cell RNA sequencing (scRNA-seq) technologies have led to extensive study of cellular heterogeneity and cell-to-cell variation. However, the high frequency of dropout events and noise in scRNA-seq data confounds the accuracy of the downstream analysis, i.e. clustering analysis, whose accuracy depends heavily on the selected feature genes. Here, by deriving an entropy decomposition formula, we propose a feature selection method, i.e. an intrinsic entropy (IE) model, to identify the informative genes for accurately clustering analysis. Specifically, by eliminating the 'noisy' fluctuation or extrinsic entropy (EE), we extract the IE of each gene from the total entropy (TE), i.e. TE = IE + EE. We show that the IE of each gene actually reflects the regulatory fluctuation of this gene in a cellular process, and thus high-IE genes provide rich information on cell type or state analysis. To validate the performance of the high-IE genes, we conduct computational analysis on both simulated datasets and real single-cell datasets by comparing with other representative methods. The results show that our IE model is not only broadly applicable and robust for different clustering and classification methods, but also sensitive for novel cell types. Our results also demonstrate that the intrinsic entropy/fluctuation of a gene serves as information rather than noise in contrast to its total entropy/fluctuation.
| S-EPMC9175189 | biostudies-literature