Search

In recent times, the performance of computer-aided diagnosis systems in classification of malignancies has significantly improved. Search and retrieval methods are specifically important as they assist physicians in making the right diagnosis in medical imaging owing to their ability of obtaining similar cases for a query image. Supervised classification algorithms are generally more accurate than unsupervised search-based classifications; however, the latter may more easily provide insights into the decision-making process by providing a group of similar cases and their corresponding metadata (i.e., diagnostic reports) and not simply a class probability. In this study, we propose a class-aware search operating on deep image embeddings to increase the accuracy of content-based search. We validate our methodology using two different publicly available datasets, one containing endometrial cancer images and the other containing colorectal cancer images. The proposed class-aware scenarios can enhance the accuracy of the search-based classifier, thereby making them more feasible in practice. With search results providing access to the metadata of retrieved cases (i.e., pathology reports of evidently diagnosed cases), such a combination has clear benefits for assisting experts with explainable results.

Histopathology images exhibit considerable variability, which can make diagnosis prone to uncertainty and errors. Using retrieval systems to locate similar images when a query image is given can assist pathologists in making more reliable decisions when diagnosing, based on accurately diagnosed past cases. Local binary patterns (LBP) have been successfully used as image descriptors for different applications. However, using LBP on histopathology images is still under investigation from different perspectives. The immense texture variability of these images and the lack of labeled training data are among the challenges that must be addressed. In this paper, we propose a new extension of LBP that is explicitly aware of the heterogeneity of local texture patterns through heterogeneity-based weighting. We use both homogeneity and the second moment (variance) of local neighborhoods to extend LBP histograms with heterogeneity information so that they better capture the polymorphism in histopathology images. We applied all concepts at multiple scales and investigated the extensions through both separated and concatenated extended histograms. We applied the proposed method on three publicly available datasets, namely, KimiaPath24, invasive ductal carcinoma (IDC), and BreakHis. The experimental results revealed that our features could retrieve and classify images with the highest accuracy. For KimiaPath24, we achieved 96.4% surpassing both LBP (91.0%) and deep features from VGG network (79.5%). For IDC, we reached the highest $F1$ measure (0.7665). Only for BreakHis, the proposed method was slightly less accurate than a customized convolutional neural network with rule fusion.

Fast diagnosis and treatment of pneumothorax, a collapsed or dropped lung, is crucial to avoid fatalities. Pneumothorax is typically detected on a chest X-ray image through visual inspection by experienced radiologists. However, the detection rate is quite low. Therefore, there is a strong need for automated detection systems to assist radiologists. Despite the high accuracy levels generally reported for deep learning classifiers in many applications, they may not be useful in clinical practice due to the lack of large number of high-quality labelled images as well as a lack of interpretation possibility. Alternatively, searching in the archive of past cases to find matching images may serve as a 'virtual second opinion' through accessing the metadata of matched evidently diagnosed cases. To use image search as a triaging/diagnosis tool, all chest X-ray images must first be tagged with identifiers, i.e., deep features. Then, given a query chest X-ray image, the majority vote among the top k retrieved images can provide a more explainable output. While image search can be clinically more viable, its detection performance needs to be investigated at a scale closer to real-world practice. We combined 3 public datasets to assemble a repository with more than 550,000 chest X-ray images. We developed the Autoencoding Thorax Net (short AutoThorax-Net) for image search in chest radiographs compressing three inputs: the left chest side, the flipped right side, and the entire chest image. Experimental results show that image search based on AutoThorax-Net features can achieve high identification rates providing a path towards real-world deployment. We achieved 92% AUC accuracy for a semi-automated search in 194,608 images (pneumothorax and normal) and 82% AUC accuracy for fully automated search in 551,383 images (normal, pneumothorax and many other chest diseases)., Under review for publication since May 2020

Variants of Triplet networks are robust entities for learning a discriminative embedding subspace. There exist different triplet mining approaches for selecting the most suitable training triplets. Some of these mining methods rely on the extreme distances between instances, and some others make use of sampling. However, sampling from stochastic distributions of data rather than sampling merely from the existing embedding instances can provide more discriminative information. In this work, we sample triplets from distributions of data rather than from existing instances. We consider a multivariate normal distribution for the embedding of each class. Using Bayesian updating and conjugate priors, we update the distributions of classes dynamically by receiving the new mini-batches of training data. The proposed triplet mining with Bayesian updating can be used with any triplet-based loss function, e.g., triplet-loss or Neighborhood Component Analysis (NCA) loss. Accordingly, Our triplet mining approaches are called Bayesian Updating Triplet (BUT) and Bayesian Updating NCA (BUNCA), depending on which loss function is being used. Experimental results on two public datasets, namely MNIST and histopathology colorectal cancer (CRC), substantiate the effectiveness of the proposed triplet mining method., Accepted for presentation at the 25th International Conference on Pattern Recognition (ICPR), IEEE, 2020. The first two authors contributed equally to this work

Fisher Discriminant Analysis (FDA) is a subspace learning method which minimizes and maximizes the intra- and inter-class scatters of data, respectively. Although, in FDA, all the pairs of classes are treated the same way, some classes are closer than the others. Weighted FDA assigns weights to the pairs of classes to address this shortcoming of FDA. In this paper, we propose a cosine-weighted FDA as well as an automatically weighted FDA in which weights are found automatically. We also propose a weighted FDA in the feature space to establish a weighted kernel FDA for both existing and newly proposed weights. Our experiments on the ORL face recognition dataset show the effectiveness of the proposed weighting schemes.

Due to recent advances in technology, digitized histopathology images are now widely available for both clinical and research purposes. Accordingly, research into computerized image analysis algorithms for digital histopathology images has been progressing rapidly. In this work, we focus on image retrieval for digital histopathology images. Image retrieval algorithms can be used to find similar images and can assist pathologists in making quick and accurate diagnoses. Histopathology images are typically stained with dyes to highlight features of the tissue, and as such, an image analysis algorithm for histopathology should be able to process colour images and determine relevant information from the stain colours present. In this study, we are interested in the effect that stain separation into their individual stain components has on image search performance. To this end, we implement a basic k-nearest neighbours (kNN) search algorithm on histopathology images from two publicly available data sets (IDC and BreakHis) which are: a) converted to greyscale, b) digitally stain-separated and c) the original RGB colour images. The results of this study show that using H&E separated images yields search accuracies within one or two percent of those obtained with original RGB images, and that superior performance is observed using the H&E images in most scenarios we tested.

Pneumothorax, a collapsed or dropped lung, is a fatal condition typically detected on a chest X-ray by an experienced radiologist. Due to shortage of such experts, automated detection systems based on deep neural networks have been developed. Nevertheless, applying such systems in practice remains a challenge. These systems, mostly compute a single probability as output, may not be enough for diagnosis. On the contrary, content-based medical image retrieval (CBIR) systems, such as image search, can assist clinicians for diagnostic purposes by enabling them to compare the case they are examining with previous (already diagnosed) cases. However, there is a lack of study on such attempt. In this study, we explored the use of image search to classify pneumothorax among chest X-ray images. All chest X-ray images were first tagged with deep pretrained features, which were obtained from existing deep learning models. Given a query chest X-ray image, the majority voting of the top K retrieved images was then used as a classifier, in which similar cases in the archive of past cases are provided besides the probability output. In our experiments, 551,383 chest X-ray images were obtained from three large recently released public datasets. Using 10-fold cross-validation, it is shown that image search on deep pretrained features achieved promising results compared to those obtained by traditional classifiers trained on the same features. To the best of knowledge, it is the first study to demonstrate that deep pretrained features can be used for CBIR of pneumothorax in half a million chest X-ray images.

Siamese neural network is a very powerful architecture for both feature extraction and metric learning. It usually consists of several networks that share weights. The Siamese concept is topology-agnostic and can use any neural network as its backbone. The two most popular loss functions for training these networks are the triplet and contrastive loss functions. In this paper, we propose two novel loss functions, named Fisher Discriminant Triplet (FDT) and Fisher Discriminant Contrastive (FDC). The former uses anchor-neighbor-distant triplets while the latter utilizes pairs of anchor-neighbor and anchor-distant samples. The FDT and FDC loss functions are designed based on the statistical formulation of the Fisher Discriminant Analysis (FDA), which is a linear subspace learning method. Our experiments on the MNIST and two challenging and publicly available histopathology datasets show the effectiveness of the proposed loss functions., Comment: Accepted (to appear) in International Joint Conference on Neural Networks (IJCNN) 2020, IEEE, in IEEE World Congress on Computational Intelligence (WCCI) 2020

With the availability of whole-slide imaging in pathology, high-resolution images offer a more convenient disease observation but also require content-based retrieval of large scans. The bag-of-visual-words methodology has shown a high ability to describe the image content for recognition and retrieval purposes. In this work, a variant of the bag-of-visual-words with multiple dictionaries for histopathology image classification is proposed and tested on the image dataset Kimia Path24 with more than 27,000 patches of size 1000 × 1000 belonging to 24 different classes. Features are extracted from patches and clustered to form multiple codebooks. The histogram intersection approach and support vector machines are exploited to build multiple classifiers. At last, the majority voting determines the final classification for each patch. The experiments demonstrate the superiority of the proposed method for histopathology images that surpasses deep networks, LBP and other BoW results.

Closing the semantic gap in medical image analysis is critical. Access to large-scale datasets might help to narrow the gap. However, large and balanced datasets may not always be available. On the other side, retrieving similar images from an archive is a valuable task to facilitate better diagnosis. In this work, we concentrate on forming a search space, consisting of the most similar images for a given query, to be used for a similarity-based search technique in a retrieval system. We propose a two-step hierarchical shrinking search space when local binary patterns are used. Transfer learning via convolutional neural networks is utilized for the first stage of search space shrinking, followed by creating a selection pool using Radon transform for further reduction. The difference between two orthogonal Radon projections is considered in the selection pool to extract more information. The IRMA dataset, from ImageCLEF initiative, containing 14,400 X-ray images, is used to validate the proposed scheme. We report a total IRMA error of 168.05 (or 90.30% accuracy) which is the best result compared with existing methods in the literature for this dataset when real-time processing is considered.

Learning and extracting representative features along with similarity measurements in high dimensional feature spaces is a critical task. Moreover, the problem of how to bridge the semantic gap, between the low-level information captured by a machine learning model and the high-level one interpreted by a human operator, is still a practical challenge, especially in medicine. In medical applications, retrieving similar images from archives of past cases can be immensely beneficial in diagnostic imaging. However, large and balanced datasets may not be available for many reasons. Exploring the ways of using deep networks, for classification to retrieval, to fill this semantic gap was a key question for this research. In this work, we propose a parallel deep solution approach based on convolutional neural networks followed by a local search using LBP, HOG and Radon features. The IRMA dataset, from ImageCLEF initiative, containing 14,400 X-ray images, is employed to validate the proposed scheme. With a total IRMA error of 165.55, the performance of our scheme surpasses the dictionary approach and many other learning methods applied on the same dataset.

HighlightsA comprehensive survey on micro-evolutionary algorithms.Study of MDEVM algorithm with population distributed implementation.Mutation factor diversity analysis by Monte-Carlo simulations.Conducted comparative study and analysis of MDE algorithm. Differential evolution (DE) algorithm suffers from high computational time due to slow nature of evaluation. Micro-DE (MDE) algorithms utilize a very small population size, which can converge faster to a reasonable solution. Such algorithms are vulnerable to premature convergence and high risk of stagnation. This paper proposes a MDE algorithm with vectorized random mutation factor (MDEVM), which utilizes the small size population benefit while empowers the exploration ability of mutation factor through randomizing it in the decision variable level. The idea is supported by analyzing mutation factor using Monte-Carlo based simulations. To facilitate the usage of MDE algorithms with very-small population sizes, a new mutation scheme for population sizes less than four is also proposed. Furthermore, comprehensive comparative simulations and analysis on performance of the MDE algorithms over various mutation schemes, population sizes, problem types (i.e. uni-modal, multi-modal, and composite), problem dimensionalities, and mutation factor ranges are conducted by considering population diversity analysis for stagnation and pre-mature convergence. The MDEVM is implemented using a population-based parallel model and studies are conducted on 28 benchmark functions provided for the IEEE CEC-2013 competition. Experimental results demonstrate high performance in convergence speed of the proposed MDEVM algorithm.

In recent years, histopathology images have been increasingly used as a diagnostic tool in the medical field. The process of accurately diagnosing a biopsy sample requires significant expertise in the field, and as such can be time-consuming and is prone to uncertainty and error. With the advent of digital pathology, using image recognition systems to highlight problem areas or locate similar images can aid pathologists in making quick and accurate diagnoses. In this paper, we specifically consider the encoded local projections (ELP) algorithm, which has previously shown some success as a tool for classification and recognition of histopathology images. We build on the success of the ELP algorithm as a means for image classification and recognition by proposing a modified algorithm which captures the local frequency information of the image. The proposed algorithm estimates local frequencies by quantifying the changes in multiple projections in local windows of greyscale images. By doing so we remove the need to store the full projections, thus significantly reducing the histogram size, and decreasing computation time for image retrieval and classification tasks. Furthermore, we investigate the effectiveness of applying our method to histopathology images which have been digitally separated into their hematoxylin and eosin stain components. The proposed algorithm is tested on the publicly available invasive ductal carcinoma (IDC) data set. The histograms are used to train an SVM to classify the data. The experiments showed that the proposed method outperforms the original ELP algorithm in image retrieval tasks. On classification tasks, the results are found to be comparable to state-of-the-art deep learning methods and better than many handcrafted features from the literature.

Encoded Local Projections (ELP) is a recently introduced dense sampling image descriptor which uses projections in small neighbourhoods to construct a histogram/descriptor for the entire image. ELP has shown to be as accurate as other state-of-the-art features in searching medical images while being time and resource efficient. This paper attempts for the first time to utilize ELP descriptor as primary features for facial recognition and compare the results with LBP histogram on the Labeled Faces in the Wild dataset. We have evaluated descriptors by comparing the chi-squared distance of each image descriptor versus all others as well as training Support Vector Machines (SVM) with each feature vector. In both cases, the results of ELP were better than LBP in the same sub-image configuration., Comment: To be published at the 2018 IEEE Symp. Series on Comp. Intelligence (IEEE SSCI 2018), 18-21 NOV, 2018, BENGALURU, INDIA

In recent years, there has been increased interest in video summarization and automatic sports highlights generation. In this work, we introduce a new dataset, called SNOW, for umpire pose detection in the game of cricket. The proposed dataset is evaluated as a preliminary aid for developing systems to automatically generate cricket highlights. In cricket, the umpire has the authority to make important decisions about events on the field. The umpire signals important events using unique hand signals and gestures. We identify four such events for classification namely SIX, NO BALL, OUT and WIDE based on detecting the pose of the umpire from the frames of a cricket video. Pre-trained convolutional neural networks such as Inception V3 and VGG19 net-works are selected as primary candidates for feature extraction. The results are obtained using a linear SVM classifier. The highest classification performance was achieved for the SVM trained on features extracted from the VGG19 network. The preliminary results suggest that the proposed system is an effective solution for the application of cricket highlights generation., Comment: To be published at the 2018 IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2018), 18-21 NOV, 2018, BENGALURU, INDIA

This paper introduces the “encoded local projections” (ELP) as a new dense-sampling image descriptor for search and classification problems. The gradient changes of multiple projections in local windows of gray-level images are encoded to build a histogram that captures spatial projection patterns. Using projections is a conventional technique in both medical imaging and computer vision. Furthermore, powerful dense-sampling methods, such as local binary patterns and the histogram of oriented gradients, are widely used for image classification and recognition. Inspired by many achievements of such existing descriptors, we explore the design of a new class of histogram-based descriptors with particular applications in medical imaging. We experiment with three public datasets (IRMA, Kimia Path24, and CT Emphysema) to comparatively evaluate the performance of ELP histograms. In light of the tremendous success of deep architectures, we also compare the results with deep features generated by pretrained networks. The results are quite encouraging as the ELP descriptor can surpass both conventional and deep descriptors in performance in several experimental settings.

Skin cancer is a widespread, global, and potentially deadly disease, which over the last three decades has afflicted more lives in the USA than all other forms of cancer combined. There have been a lot of promising recent works utilizing deep network architectures, such as FCNs, U-Nets, and ResNets, for developing automated skin lesion segmentation. This paper investigates various pre- and post-processing techniques for improving the performance of U-Nets as measured by the Jaccard Index. The dataset provided as part of the “2017 ISBI Challenges on Skin Lesion Analysis Towards Melanoma Detection” was used for this evaluation and the performance of the finalist competitors was the standard for comparison. The pre-processing techniques employed in the proposed system included contrast enhancement, artifact removal, and vignette correction. More advanced image transformations, such as local binary patterns and wavelet decomposition, were also employed to augment the raw grayscale images used as network input features. While the performance of the proposed system fell short of the winners of the challenge, it was determined that using wavelet decomposition as an early transformation step improved the overall performance of the system over pre- and post-processing steps alone.

Image classification and retrieval systems have gained more attention because of easier access to high-tech medical imaging. However, the lack of availability of large-scaled balanced labelled data in medicine is still a challenge. Simplicity, practicality, efficiency, and effectiveness are the main targets in medical domain. To achieve these goals, Radon transformation, which is a well-known technology in medical field, is utilized along with a deep network to propose a retrieval system for a highly imbalanced medical benchmark. The main contribution of this study is to propose a deep model which is trained on the Radon-based transformed input data. The experimental results show that applying this transformation as input to feed into a convolutional neural network, significantly increases the performance, compared with other retrieval systems. The proposed scheme clearly increases the retrieval performance, compared with almost all models which use Radon transformation to retrieve medical images.

Differential Evolution (DE) is a popular global optimization algorithm, mostly due to its high performance, easy implementation, and utilization of a few control parameters. The mutation scheme is one of the important steps of DE, which selects a number of individuals from the population as parents to generate the next population during its evolutionary process. The parents are traditionally selected randomly and in some mutation schemes the best member of population is selected as one of the parents. In this paper, we propose the centroid-based differential evolution (CenDE) algorithm, which uses the centroid of top three individuals in the population in terms of objective function value performance as the base parent. The experiments are conducted for high and low dimensional problems with small and standard population sizes on CEC Black-Box Optimization Benchmark problems 2015 (CEC-BBOB 2015). Our experiments show that the center of best three individuals plays an important role in generating candidate individuals with better objective values for the next generation, resulting in a faster convergence of the DE algorithm.

We investigate the concept of deep barcodes and propose two methods to generate them in order to expedite the process of classification and retrieval of histopathology images. Since binary search is computationally less expensive, in terms of both speed and storage, deep barcodes could be useful when dealing with big data retrieval. Our experiments use the dataset Kimia Path24 to test three pre-trained networks for image retrieval. The dataset consists of 27,055 training images in 24 different classes with large variability, and 1,325 test images for testing. Apart from the high-speed and efficiency, results show a surprising retrieval accuracy of 71.62% for deep barcodes, as compared to 68.91% for deep features and 68.53% for compressed deep features., Accepted for publication in proceedings of the IEEE World Congress on Computational Intelligence (IEEE WCCI), Rio de Janeiro, Brazil, 8-3 July, 2018

Every year, thousands of people receive consumer product related injuries. Research indicates that online customer reviews can be processed to autonomously identify product safety issues. Early identification of safety issues can lead to earlier recalls, and thus fewer injuries and deaths. A dataset of product reviews from Amazon.com was compiled, along with SaferProducts.gov complaints and recall descriptions from the Consumer Product Safety Commission (CPSC) and European Commission Rapid Alert system. A system was built to clean the collected text and to extract relevant features. Dimensionality reduction was performed by computing feature relevance through a Random Forest and discarding features with low information gain. Various classifiers were analyzed, including Logistic Regression, SVMs, Naive-Bayes, Random Forests, and an Ensemble classifier. Experimentation with various features and classifier combinations resulted in a logistic regression model with 66% precision in the top 50 reviews surfaced. This classifier outperforms all benchmarks set by related literature and consumer product safety professionals.

Medical image analysis has become a topic under the spotlight in recent years. There is a significant progress in medical image research concerning the usage of machine learning. However, there are still numerous questions and problems awaiting answers and solutions, respectively. In the present study, comparison of three classification models is conducted using features extracted using local binary patterns, the histogram of gradients, and a pre-trained deep network. Three common image classification methods, including support vector machines, decision trees, and artificial neural networks are used to classify feature vectors obtained by different feature extractors. We use KIMIA Path960, a publicly available dataset of $960$ histopathology images extracted from $20$ different tissue scans to test the accuracy of classification and feature extractions models used in the study, specifically for the histopathology images. SVM achieves the highest accuracy of $90.52\%$ using local binary patterns as features which surpasses the accuracy obtained by deep features, namely $81.14\%$., Comment: Accepted for publication in proceedings of the IEEE World Congress on Computational Intelligence (IEEE WCCI), Rio de Janeiro, Brazil, 8-3 July, 2018

Differential evolution (DE) algorithm with a small population size is called Micro-DE (MDE). A small population size decreases the computational complexity but also reduces the exploration ability of DE by limiting the population diversity. In this paper, we propose the idea of combining ensemble mutation scheme selection and opposition-based learning concepts to enhance the diversity of population in MDE at mutation and selection stages. The proposed algorithm enhances the diversity of population by generating a random mutation scale factor per individual and per dimension, randomly assigning a mutation scheme to each individual in each generation, and diversifying individuals selection using opposition-based learning. This approach is easy to implement and does not require the setting of mutation scheme selection and mutation scale factor. Experimental results are conducted for a variety of objective functions with low and high dimensionality on the CEC BlackBox Optimization Benchmarking 2015 (CEC-BBOB 2015). The results show superior performance of the proposed algorithm compared to the other micro-DE algorithms.

We explore the problem of classification within a medical image data-set based on a feature vector extracted from the deepest layer of pre-trained Convolution Neural Networks. We have used feature vectors from several pre-trained structures, including networks with/without transfer learning to evaluate the performance of pre-trained deep features versus CNNs which have been trained by that specific dataset as well as the impact of transfer learning with a small number of samples. All experiments are done on Kimia Path24 dataset which consists of 27,055 histopathology training patches in 24 tissue texture classes along with 1,325 test patches for evaluation. The result shows that pre-trained networks are quite competitive against training from scratch. As well, fine-tuning does not seem to add any tangible improvement for VGG16 to justify additional training while we observed considerable improvement in retrieval and classification accuracy when we fine-tuned the Inception structure., To appear in proceedings of the 7th International Conference on Image Processing Theory, Tools and Applications (IPTA 2017), Nov 28-Dec 1, Montreal, Canada

Content-based image retrieval (CBIR) of medical images in large datasets to identify similar images when a query image is given can be very useful in improving the diagnostic decision of the clinical experts and as well in educational scenarios. In this paper, we used two stage classification and retrieval approach to retrieve similar images. First, the Gabor filters are applied to Radon-transformed images to extract features and to train a multi-class SVM. Then based on the classification results and using an extracted Gabor barcode, similar images are retrieved. The proposed method was tested on IRMA dataset which contains more than 14,000 images. Experimental results show the efficiency of our approach in retrieving similar images compared to other Gabor-Radon-oriented methods., Comment: To appear in proceedings of The IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2017), Honolulu, Hawaii, USA, Nov. 27 -- Dec 1, 2017

Radon transform and its inverse operation are important techniques in medical imaging tasks. Recently, there has been renewed interest in Radon transform for applications such as content-based medical image retrieval. However, all studies so far have used Radon transform as a global or quasi-global image descriptor by extracting projections of the whole image or large sub-images. This paper attempts to show that the dense sampling to generate the histogram of local Radon projections has a much higher discrimination capability than the global one. In this paper, we introduce Local Radon Descriptor (LRD) and apply it to the IRMA dataset, which contains 14,410 x-ray images as well as to the INRIA Holidays dataset with 1,990 images. Our results show significant improvement in retrieval performance by using LRD versus its global version. We also demonstrate that LRD can deliver results comparable to well-established descriptors like LBP and HOG.

Despite the progress made in the field of medical imaging, it remains a large area of open research, especially due to the variety of imaging modalities and disease-specific characteristics. This paper is a comparative study describing the potential of using local binary patterns (LBP), deep features and the bag-of-visual words (BoVW) scheme for the classification of histopathological images. We introduce a new dataset, \emph{KIMIA Path960}, that contains 960 histopathology images belonging to 20 different classes (different tissue types). We make this dataset publicly available. The small size of the dataset and its inter- and intra-class variability makes it ideal for initial investigations when comparing image descriptors for search and classification in complex medical imaging cases like histopathology. We investigate deep features, LBP histograms and BoVW to classify the images via leave-one-out validation. The accuracy of image classification obtained using LBP was 90.62\% while the highest accuracy using deep features reached 94.72\%. The dictionary approach (BoVW) achieved 96.50\%. Deep solutions may be able to deliver higher accuracies but they need extensive training with a large number of (balanced) image datasets., To appear in proceedings of The IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2017), Honolulu, Hawaii, USA, Nov. 27 -- Dec 1, 2017

Content-based image retrieval is an effective and efficient technique to retrieve images from a big dataset with similar images. To have a robust retrieval system, a proper and accurate classification scheme is required to categorise the information of shape, texture, and colours. In this paper, a deep convolutional neural network is proposed to classify the information of radiology images. Deep networks need millions of data, but the lack of availability of balanced large datasets in medical domain motivates us to trust on even the second prediction category rather than just the best one. Hence the best predicted categories are considered for a query test, followed by a similarity-based search technique. This results in a proper classification performance. Moreover, as Radon transformation is famous in medical domain, this conversion technique is utilized for a similarity-based search scheme, after measuring by a k-nearest neighbours algorithm. The experimental results and comparison show that this strategy not only improve the performance, but also save the computational costs.

Many automatic skin lesion diagnosis systems use segmentation as a preprocessing step to diagnose skin conditions because skin lesion shape, border irregularity, and size can influence the likelihood of malignancy. This paper presents, examines and compares two different approaches to skin lesion segmentation. The first approach uses U-Nets and introduces a histogram equalization based preprocessing step. The second approach is a C-Means clustering based approach that is much simpler to implement and faster to execute. The Jaccard Index between the algorithm output and hand segmented images by dermatologists is used to evaluate the proposed algorithms. While many recently proposed deep neural networks to segment skin lesions require a significant amount of computational power for training (i.e., computer with GPUs), the main objective of this paper is to present methods that can be used with only a CPU. This severely limits, for example, the number of training instances that can be presented to the U-Net. Comparing the two proposed algorithms, U-Nets achieved a significantly higher Jaccard Index compared to the clustering approach. Moreover, using the histogram equalization for preprocessing step significantly improved the U-Net segmentation results., Comment: To appear in proceedings of The IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2017), Honolulu, Hawaii, USA, Nov. 27 -- Dec 1, 2017

The idea of Radon barcodes (RBC) has been introduced recently. In this paper, we propose a content-based image retrieval approach for big datasets based on Radon barcodes. Our method (Single Projection Radon Barcode, or SP-RBC) uses only a few Radon single projections for each image as global features that can serve as a basis for weak learners. This is our most important contribution in this work, which improves the results of the RBC considerably. As a matter of fact, only one projection of an image, as short as a single SURF feature vector, can already achieve acceptable results. Nevertheless, using multiple projections in a long vector will not deliver anticipated improvements. To exploit the information inherent in each projection, our method uses the outcome of each projection separately and then applies more precise local search on the small subset of retrieved images. We have tested our method using IRMA 2009 dataset a with 14,400 x-ray images as part of imageCLEF initiative. Our approach leads to a substantial decrease in the error rate in comparison with other non-learning methods.

Using content-based binary codes to tag digital images has emerged as a promising retrieval technology. Recently, Radon barcodes (RBCs) have been introduced as a new binary descriptor for image search. RBCs are generated by binarization of Radon projections and by assembling them into a vector, namely the barcode. A simple local thresholding has been suggested for binarization. In this paper, we put forward the idea of "autoencoded Radon barcodes". Using images in a training dataset, we autoencode Radon projections to perform binarization on outputs of hidden layers. We employed the mini-batch stochastic gradient descent approach for the training. Each hidden layer of the autoencoder can produce a barcode using a threshold determined based on the range of the logistic function used. The compressing capability of autoencoders apparently reduces the redundancies inherent in Radon projections leading to more accurate retrieval results. The IRMA dataset with 14,410 x-ray images is used to validate the performance of the proposed method. The experimental results, containing comparison with RBCs, SURF and BRISK, show that autoencoded Radon barcode (ARBC) has the capacity to capture important information and to learn richer representations resulting in lower retrieval errors for image retrieval measured with the accuracy of the first hit only., o appear in proceedings of the 23rd International Conference on Pattern Recognition (ICPR 2016), Cancun, Mexico, December 2016

Many research works have successfully extended algorithms such as evolutionary algorithms, reinforcement agents and neural networks using "opposition-based learning" (OBL). Two types of the "opposites" have been defined in the literature, namely \textit{type-I} and \textit{type-II}. The former are linear in nature and applicable to the variable space, hence easy to calculate. On the other hand, type-II opposites capture the "oppositeness" in the output space. In fact, type-I opposites are considered a special case of type-II opposites where inputs and outputs have a linear relationship. However, in many real-world problems, inputs and outputs do in fact exhibit a nonlinear relationship. Therefore, type-II opposites are expected to be better in capturing the sense of "opposition" in terms of the input-output relation. In the absence of any knowledge about the problem at hand, there seems to be no intuitive way to calculate the type-II opposites. In this paper, we introduce an approach to learn type-II opposites from the given inputs and their outputs using the artificial neural networks (ANNs). We first perform \emph{opposition mining} on the sample data, and then use the mined data to learn the relationship between input $x$ and its opposite $\breve{x}$. We have validated our algorithm using various benchmark functions to compare it against an evolving fuzzy inference approach that has been recently introduced. The results show the better performance of a neural approach to learn the opposites. This will create new possibilities for integrating oppositional schemes within existing algorithms promising a potential increase in convergence speed and/or accuracy., To appear in proceedings of the 23rd International Conference on Pattern Recognition (ICPR 2016), Cancun, Mexico, December 2016

Differential evolution (DE) is a high performance and easy to implement evolutionary algorithm. The DE algorithm with small population size (i.e., micro-DE) can further increase the efficiency of the algorithm. However, it also decreases its exploration capability, causing stagnation and pre-mature convergence. In this paper, the idea of exploration enhancement at the mutation level is proposed. The proposed algorithm randomly generates the mutation scale factor for each individual and each dimension of the problem using a uniform distribution. Each individual can select a mutation scheme uniformly and randomly from a pool of mutation schemes in each generation, instead of using a fixed mutation scheme for all individuals during generations. The proposed idea is simple and easy to implement, without changing the algorithm complexity or adding overhead running time. This approach relaxes setting of mutation scheme control parameter. In this paper, we provide a detail analysis about the exploration capability of four variants of micro-DE versions, namely classical micro-DE, micro-DE with vectorized random mutation factor, micro-DE with ensemble mutation scheme, and micro-DE with vectorized random mutation factor and ensemble mutation scheme. Experimental results for various dimensions between 30 to 1000 on the CEC BlackBox Optimization Benchmarking 2015 (CEC-BBOB 2015) show superior performance of the proposed approach compared to the micro-DE and micro-DE with randomized mutation factor algorithms.

For more than two decades, research has been performed on content-based image retrieval (CBIR). By combining Radon projections and the support vector machines (SVM), a content-based medical image retrieval method is presented in this work. The proposed approach employs the normalized Radon projections with corresponding image category labels to build an SVM classifier, and the Radon barcode database which encodes every image in a binary format is also generated simultaneously to tag all images. To retrieve similar images when a query image is given, Radon projections and the barcode of the query image are generated. Subsequently, the k-nearest neighbor search method is applied to find the images with minimum Hamming distance of the Radon barcode within the same class predicted by the trained SVM classifier that uses Radon features. The performance of the proposed method is validated by using the IRMA 2009 dataset with 14,410 x-ray images in 57 categories. The results demonstrate that our method has the capacity to retrieve similar responses for the correctly identified query image and even for those mistakenly classified by SVM. The approach further is very fast and has low memory requirement.

Content-based image retrieval (CBIR) in large medical image archives is a challenging and necessary task. Generally, different feature extraction methods are used to assign expressive and invariant features to each image such that the search for similar images comes down to feature classification and/or matching. The present work introduces a new image retrieval method for medical applications that employs a convolutional neural network (CNN) with recently introduced Radon barcodes. We combine neural codes for global classification with Radon barcodes for the final retrieval. We also examine image search based on regions of interest (ROI) matching after image retrieval. The IRMA dataset with more than 14,000 x-rays images is used to evaluate the performance of our method. Experimental results show that our approach is superior to many published works., Comment: To appear in proceedings of The 2016 IEEE International Joint Conference on Neural Networks (IJCNN 2016), July 24-29, 2016, Vancouver, Canada

A Content-Based Image Retrieval (CBIR) system which identifies similar medical images based on a query image can assist clinicians for more accurate diagnosis. The recent CBIR research trend favors the construction and use of binary codes to represent images. Deep architectures could learn the non-linear relationship among image pixels adaptively, allowing the automatic learning of high-level features from raw pixels. However, most of them require class labels, which are expensive to obtain, particularly for medical images. The methods which do not need class labels utilize a deep autoencoder for binary hashing, but the code construction involves a specific training algorithm and an ad-hoc regularization technique. In this study, we explored using a deep de-noising autoencoder (DDA), with a new unsupervised training scheme using only backpropagation and dropout, to hash images into binary codes. We conducted experiments on more than 14,000 x-ray images. By using class labels only for evaluating the retrieval results, we constructed a 16-bit DDA and a 512-bit DDA independently. Comparing to other unsupervised methods, we succeeded to obtain the lowest total error by using the 512-bit codes for retrieval via exhaustive search, and speed up 9.27 times with the use of the 16-bit codes while keeping a comparable total error. We found that our new training scheme could reduce the total retrieval error significantly by 21.9%. To further boost the image retrieval performance, we developed Radon Autoencoder Barcode (RABC) which are learned from the Radon projections of images using a de-noising autoencoder. Experimental results demonstrated its superior performance in retrieval when it was combined with DDA binary codes., To appear in proceedings of The 2016 International Joint Conference on Neural Networks (IJCNN 2016), July 24-29, 2016, Vancouver, Canada

Recently, Radon transformation has been used to generate barcodes for tagging medical images. The under-sampled image is projected in certain directions, and each projection is binarized using a local threshold. The concatenation of the thresholded projections creates a barcode that can be used for tagging or annotating medical images. A small number of equidistant projections, e.g., 4 or 8, is generally used to generate short barcodes. However, due to the diverse nature of digital images, and since we are only working with a small number of projections (to keep the barcode short), taking equidistant projections may not be the best course of action. In this paper, we proposed to find $n$ optimal projections, whereas $n\!, Comment: To appear in proceedings of The 2016 IEEE Congress on Evolutionary Computation (IEEE CEC 2016), July 24-29, 2016, Vancouver, Canada

Quantitative ultrasound (QUS) methods provide a promising framework that can non-invasively and inexpensively be used to predict or assess the tumour response to cancer treatment. The first step in using the QUS methods is to select a region of interest (ROI) inside the tumour in ultrasound images. Manual segmentation, however, is very time consuming and tedious. In this paper, a semi-automated approach will be proposed to roughly localize an ROI for a tumour in ultrasound images of patients with locally advanced breast cancer (LABC). Content-based barcodes, a recently introduced binary descriptor based on Radon transform, were used in order to find similar cases and estimate a bounding box surrounding the tumour. Experiments with 33 B-scan images resulted in promising results with an accuracy of $81\%$., To appear in proceedings of The International Symposium on Biomedical Imaging (ISBI), April 13-16, 2016, Prague, Czech Republic

Quantifying the accuracy of segmentation and manual delineation of organs, tissue types and tumors in medical images is a necessary measurement that suffers from multiple problems. One major shortcoming of all accuracy measures is that they neglect the anatomical significance or relevance of different zones within a given segment. Hence, existing accuracy metrics measure the overlap of a given segment with a ground-truth without any anatomical discrimination inside the segment. For instance, if we understand the rectal wall or urethral sphincter as anatomical zones, then current accuracy measures ignore their significance when they are applied to assess the quality of the prostate gland segments. In this paper, we propose an anatomy-aware measurement scheme for segmentation accuracy of medical images. The idea is to create a ``master gold'' based on a consensus shape containing not just the outline of the segment but also the outlines of the internal zones if existent or relevant. To apply this new approach to accuracy measurement, we introduce the anatomy-aware extensions of both Dice coefficient and Jaccard index and investigate their effect using 500 synthetic prostate ultrasound images with 20 different segments for each image. We show that through anatomy-sensitive calculation of segmentation accuracy, namely by considering relevant anatomical zones, not only the measurement of individual users can change but also the ranking of users' segmentation skills may require reordering., To appear in SPIE Medical Imaging 2016

In recent years, with the explosion of digital images on the Web, content-based retrieval has emerged as a significant research area. Shapes, textures, edges and segments may play a key role in describing the content of an image. Radon and Gabor transforms are both powerful techniques that have been widely studied to extract shape-texture-based information. The combined Radon-Gabor features may be more robust against scale/rotation variations, presence of noise, and illumination changes. The objective of this paper is to harness the potentials of both Gabor and Radon transforms in order to introduce expressive binary features, called barcodes, for image annotation/tagging tasks. We propose two different techniques: Gabor-of-Radon-Image Barcodes (GRIBCs), and Guided-Radon-of-Gabor Barcodes (GRGBCs). For validation, we employ the IRMA x-ray dataset with 193 classes, containing 12,677 training images and 1,733 test images. A total error score as low as 322 and 330 were achieved for GRGBCs and GRIBCs, respectively. This corresponds to $\approx 81\%$ retrieval accuracy for the first hit., Comment: To appear in proceedings of the 23rd International Conference on Pattern Recognition (ICPR 2016), Cancun, Mexico, December 2016

Medical images can be a valuable resource for reliable information to support medical diagnosis. However, the large volume of medical images makes it challenging to retrieve relevant information given a particular scenario. To solve this challenge, content-based image retrieval (CBIR) attempts to characterize images (or image regions) with invariant content information in order to facilitate image search. This work presents a feature extraction technique for medical images using stacked autoencoders, which encode images to binary vectors. The technique is applied to the IRMA dataset, a collection of 14,410 x-ray images in order to demonstrate the ability of autoencoders to retrieve similar x-rays given test queries. Using IRMA dataset as a benchmark, it was found that stacked autoencoders gave excellent results with a retrieval error of 376 for 1,733 test images with a compression of 74.61%.

In recent years, advances in medical imaging have led to the emergence of massive databases, containing images from a diverse range of modalities. This has significantly heightened the need for automated annotation of the images on one side, and fast and memory-efficient content-based image retrieval systems on the other side. Binary descriptors have recently gained more attention as a potential vehicle to achieve these goals. One of the recently introduced binary descriptors for tagging of medical images are Radon barcodes (RBCs) that are driven from Radon transform via local thresholding. Gabor transform is also a powerful transform to extract texture-based information. Gabor features have exhibited robustness against rotation, scale, and also photometric disturbances, such as illumination changes and image noise in many applications. This paper introduces Gabor Barcodes (GBCs), as a novel framework for the image annotation. To find the most discriminative GBC for a given query image, the effects of employing Gabor filters with different parameters, i.e., different sets of scales and orientations, are investigated, resulting in different barcode lengths and retrieval performances. The proposed method has been evaluated on the IRMA dataset with 193 classes comprising of 12,677 x-ray images for indexing, and 1,733 x-rays images for testing. A total error score as low as $351$ ($\approx 80\%$ accuracy for the first hit) was achieved., Comment: To appear in proceedings of The 2016 IEEE International Conference on Image Processing (ICIP 2016), Sep 25-28, 2016, Phoenix, Arizona, USA

Content-based medical image retrieval can support diagnostic decisions by clinical experts. Examining similar images may provide clues to the expert to remove uncertainties in his/her final diagnosis. Beyond conventional feature descriptors, binary features in different ways have been recently proposed to encode the image content. A recent proposal is “Radon barcodes” that employ binarized Radon projections to tag/annotate medical images with content-based binary vectors, called barcodes. In this paper, MinMax Radon barcodes are introduced which are superior to “local thresholding” scheme suggested in the literature. Using IRMA dataset with 14,410 x-ray images from 193 different classes, the advantage of using MinMax Radon barcodes over thresholded Radon barcodes are demonstrated. The retrieval error for direct search drops by more than 15%. As well, SURF, as a well-established non-binary approach, and BRISK, as a recent binary method are examined to compare their results with MinMax Radon barcodes when retrieving images from IRMA dataset. The results demonstrate that MinMax Radon barcodes are faster and more accurate when applied on IRMA images.

This paper proposes to generate and to use barcodes to annotate medical images and/or their regions of interest such as organs, tumors and tissue types. A multitude of efficient feature-based image retrieval methods already exist that can assign a query image to a certain image class. Visual annotations may help to increase the retrieval accuracy if combined with existing feature-based classification paradigms. Whereas with annotations we usually mean textual descriptions, in this paper barcode annotations are proposed. In particular, Radon barcodes (RBC) are introduced. As well, local binary patterns (LBP) and local Radon binary patterns (LRBP) are implemented as barcodes. The IRMA x-ray dataset with 12,677 training images and 1,733 test images is used to verify how barcodes could facilitate image retrieval., To be published in proceedings of The IEEE International Conference on Image Processing (ICIP 2015), September 27-30, 2015, Quebec City, Canada

The idea of opposition-based learning was introduced 10 years ago. Since then a noteworthy group of researchers has used some notions of oppositeness to improve existing optimization and learning algorithms. Among others, evolutionary algorithms, reinforcement agents, and neural networks have been reportedly extended into their opposition-based version to become faster and/or more accurate. However, most works still use a simple notion of opposites, namely linear (or type- I) opposition, that for each $x\in[a,b]$ assigns its opposite as $\breve{x}_I=a+b-x$. This, of course, is a very naive estimate of the actual or true (non-linear) opposite $\breve{x}_{II}$, which has been called type-II opposite in literature. In absence of any knowledge about a function $y=f(\mathbf{x})$ that we need to approximate, there seems to be no alternative to the naivety of type-I opposition if one intents to utilize oppositional concepts. But the question is if we can receive some level of accuracy increase and time savings by using the naive opposite estimate $\breve{x}_I$ according to all reports in literature, what would we be able to gain, in terms of even higher accuracies and more reduction in computational complexity, if we would generate and employ true opposites? This work introduces an approach to approximate type-II opposites using evolving fuzzy rules when we first perform opposition mining. We show with multiple examples that learning true opposites is possible when we mine the opposites from the training data to subsequently approximate $\breve{x}_{II}=f(\mathbf{x},y)$., Accepted for publication in The 2015 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2015), August 2-5, 2015, Istanbul, Turkey