Synthetic data generation in histopathology faces unique challenges: preserving tissue heterogeneity, capturing subtle morphological features, and scaling to unannotated datasets. We present a latent diffusion model that generates realistic heterogeneous histopathology images through a novel dual-conditioning approach combining semantic segmentation maps with tissue-specific visual crops. Unlike existing methods that rely on text prompts or abstract visual embeddings, our approach preserves critical morphological details by directly incorporating raw tissue crops from corresponding semantic regions. For annotated datasets (i.e., Camelyon16, Panda), we extract patches ensuring 20-80% tissue heterogeneity. For unannotated data (i.e., TCGA), we introduce a self-supervised extension that clusters whole-slide images into 100 tissue types using foundation model embeddings, automatically generating pseudo-semantic maps for training. Our method synthesizes high-fidelity images with precise region-wise annotations, achieving superior performance on downstream segmentation tasks. When evaluated on annotated datasets, models trained on our synthetic data show competitive performance to those trained on real data, demonstrating the utility of controlled heterogeneous tissue generation. In quantitative evaluation, prompt‐guided synthesis reduces Fréchet Distance by up to 6× on Camelyon16 (from 430.1 to 72.0) and yields 2–3× lower FD across Panda and TCGA. Downstream DeepLabv3+ models trained solely on synthetic data attain test IoU of 0.71 and 0.95 on Camelyon16 and Panda, within 1–2% of real‐data baselines (0.72 and 0.96). By scaling to 11,765 TCGA whole‐slide images without manual annotations, our framework offers a practical solution for an urgent need for generating diverse, annotated histopathology data, addressing a critical bottleneck in computational pathology.

This paper addresses complex challenges in histopatho-logical image analysis through three key contributions. Firstly, it introduces a fast patch selection method, FPS, for whole-slide image (WSI) analysis, significantly reducing computational cost while maintaining accuracy. Sec-ondly, it presents PathDino, a lightweight histopathol-ogy feature extractor with a minimal configuration of five Transformer blocks and only$\approx 9$million parameters, markedly fewer than alternatives. Thirdly, it introduces a rotation-agnostic representation learning paradigm using self-supervised learning, effectively mitigating overfuting. We also show that our compact model outperforms existing state-of-the-art histopathology-specific vision transformers on 12 diverse datasets, including both internal datasets spanning four sites (breast, liver, skin, and colorectal) and seven public datasets (PANDA, CAMELYON16, BRACS, DigestPath, Kather, PanNuke, and WSSS4LUAD). Notably, even with a training dataset of$\approx 6$million histopathol-ogy patches from The Cancer Genome Atlas (TCGA), our approach demonstrates an average 8.5% improvement in patch-level majority vote performance. These contributions provide a robust framework for enhancing image analysis in digital pathology, rigorously validated through extensive evaluation.11The project page: https://KimiaLabMayo.github.io/PathDino-Page/

One of the main obstacles of adopting digital pathology is the challenge of efficient processing of hyperdimensional digitized biopsy samples, called whole slide images (WSIs). Exploiting deep learning and introducing compact WSI representations are urgently needed to accelerate image analysis and facilitate the visualization and interpretability of pathology results in a postpandemic world. In this article, we introduce a new evolutionary approach for WSI representation based on large-scale multiobjective optimization (LSMOP) of deep embeddings. We start with patch-based sampling to feed KimiaNet, a histopathology-specialized deep network, and to extract a multitude of feature vectors. Coarse multiobjective feature selection uses the reduced search space strategy guided by the classification accuracy and the number of features. In the second stage, the frequent features histogram (FFH), a novel WSI representation, is constructed by multiple runs of coarse LSMOP. Fine evolutionary feature selection is then applied to find a compact (short-length) feature vector based on the FFH and contributes to a more robust deep-learning approach to digital pathology supported by the stochastic power of evolutionary algorithms. We validate the proposed schemes using The Cancer Genome Atlas (TCGA) images in terms of WSI representation, classification accuracy, and feature quality. Furthermore, a novel decision space for multicriteria decision making in the LSMOP field is introduced. Finally, a patch-level visualization approach is proposed to increase the interpretability of deep features. The proposed evolutionary algorithm finds a very compact feature vector to represent a WSI (almost 14000 times smaller than the original feature vectors) with 8% higher accuracy compared to the codes provided by the state-of-the-art methods

Medical centers and healthcare providers have concerns and hence restrictions around sharing data with external collaborators. Federated learning, as a privacy-preserving method, involves learning a site-independent model without having direct access to patient-sensitive data in a distributed collaborative fashion. The federated approach relies on decentralized data distribution from various hospitals and clinics. The collaboratively learned global model is supposed to have acceptable performance for the individual sites. However, existing methods focus on minimizing the average of the aggregated loss functions, leading to a biased model that performs perfectly for some hospitals while exhibiting undesirable performance for other sites. In this paper, we improve model "fairness" among participating hospitals by proposing a novel federated learning scheme called Proportionally Fair Federated Learning, short Prop-FFL. Prop-FFL is based on a novel optimization objective function to decrease the performance variations among participating hospitals. This function encourages a fair model, providing us with more uniform performance across participating hospitals. We validate the proposed Prop-FFL on two histopathology datasets as well as two general datasets to shed light on its inherent capabilities. The experimental results suggest promising performance in terms of learning speed, accuracy, and fairness.

In the era of big data, methods for improving memory and computational efficiency have become crucial for the successful deployment of technology. Hashing is one of the most effective approaches to deal with the computational limitations associated with big data. One natural way to formulate this problem is spectral hashing, which directly incorporates an affinity to learning binary codes. However, owing to the binary constraints, the optimization becomes intractable. To mitigate this challenge, different relaxation approaches have been proposed to reduce the computational load required to obtain binary codes and still attain a good solution. The problem with all existing relaxation methods involves the use of one or more additional auxiliary variables to attain high-quality binary codes while relaxing the problem. The existence of auxiliary variables leads to the coordinate descent approach, which increases the computational complexity. We argue that the introduction of these variables is unnecessary. To this end, we propose a novel relaxed formulation for spectral hashing that adds no additional variables to the problem. Furthermore, instead of solving the problem in the original space where the number of variables is equal to the data points, we solve the problem in a much smaller space and retrieve the binary codes from this solution. This technique reduces both the memory and computational complexity simultaneously. We apply two optimization techniques, namely, the projected gradient and optimization on the manifold, to obtain the solution. Using comprehensive experiments on four public datasets, we show that the proposed efficient spectral hashing (ESH) algorithm achieves a highly competitive retrieval performance compared with the state-of-the-art algorithms at low complexity.

An effective unsupervised hashing algorithm leads to compact binary codes preserving the neighborhood structure of data as much as possible. One of the most established schemes for unsupervised hashing is to reduce the dimensionality of data and then find a rigid (neighborhood-preserving) transformation that reduces the quantization error. Although employing rigid transformations is effective, we may not reduce quantization loss to the ultimate limits. As well, reducing dimensionality and quantization loss in two separate steps seems to be sub-optimal. Motivated by these shortcomings, we propose to employ both rigid and non-rigid transformations to reduce quantization error and dimensionality simultaneously. We relax the orthogonality constraint on the projection in a PCA-formulation and regularize this by a quantization term. We show that both the non-rigid projection matrix and rotation matrix contribute towards minimizing quantization loss but in different ways. A scalable nested coordinate descent approach is proposed to optimize this mixed-integer optimization problem. We evaluate the proposed method on five public benchmark datasets providing almost half a million images. Comparative results indicate that the proposed method mostly outperforms state-of-art linear methods and competes with end-to-end deep solutions.

This paper investigates the effect of magnification on content-based image search in digital pathology archives and proposes to use multi-magnification image representation. Image search in large archives of digital pathology slides provides researchers and medical professionals with an opportunity to match records of current and past patients and learn from evidently diagnosed and treated cases. When working with microscopes, pathologists switch between different magnification levels while examining tissue specimens to find and evaluate various morphological features. Inspired by the conventional pathology workflow, we have investigated several magnification levels in digital pathology and their combinations to minimize the gap between AI-enabled image search methods and clinical settings. The proposed searching framework does not rely on any regional annotation and potentially applies to millions of unlabelled (raw) whole slide images. This paper suggests two approaches for combining magnification levels and compares their performance. The first approach obtains a single-vector deep feature representation for a digital slide, whereas the second approach works with a multi-vector deep feature representation. We report the search results of 20×, 10×, and 5× magnifications and their combinations on a subset of The Cancer Genome Atlas (TCGA) repository. The experiments verify that cell-level information at the highest magnification is essential for searching for diagnostic purposes. In contrast, low-magnification information may improve this assessment depending on the tumor type. Our multi-magnification approach achieved up to 11% F1-score improvement in searching among the urinary tract and brain tumor subtypes compared to the single-magnification image search.

Visual similarity estimation for histopathology images plays a key role in many medical imaging tasks, especially in image search and retrieval. All image similarity evaluation approaches employ distance-based metrics to quantify the degree of (dis) similarity. However, it has always been challenging to numerically estimate the similarity between two images, which is compatible with subjective assessment of the human operator, i.e., physicians such as radiologists and pathologists. Relying only on distance calculations through Euclidean, Manhattan, Hamming, and cosine distances does not provide us with the result that can be translated to human judgment in linguistic terms and/or in a normalized range. There is a need for a reliable image similarity measurement compatible with the human assessment with minimum possible conflict. This work proposes a new scheme that evaluates the similarity between a pair of histopathology images close to human reasoning using a fuzzy-logic approach. To this end, we developed a web application to interface with users and to collect descriptive image similarity data for training and testing purposes. We designed an adaptive neuro-fuzzy inference system (ANFIS) to model the vague and uncertain nature of user image assessment for the histopathology image comparison task. The experimental results show that the trained ANFIS can estimate the image similarity with acceptable accuracy and consistent with the user evaluations.

Learning to hash is generating similarity-preserving binary representations of images, which is, among others, an efficient way for fast image retrieval. Two-step hashing has become a common approach because it simplifies the learning by separating binary code inference from hash function training. However, the binary code inference typically leads to an intractable optimization problem with binary constraints. Different relaxation methods, which are generally based on complicated optimization techniques, have been proposed to address this challenge. In this paper, a simple relaxation scheme based on the projected gradient is proposed. To this end in each iteration, we try to update the optimization variable as if there is no binary constraint and then project the updated solution to the feasible set. We formulate the projection step as fining closet binary matrix to the updated matrix and take advantage of the closed-form solution for the projection step to complete our learning algorithm. Inspired by opposition-based learning, pairwise opposite weights between data points are incorporated to impose a stronger penalty on data instances with higher misclassification probability in the proposed objective function. We show that this simple learning algorithm leads to binary codes that achieve competitive results on both CIFAR-10 and NUS-WIDE datasets compared to state-of-the-art benchmarks.

Feature vectors provided by pre-trained deep artificial neural networks have become a dominant source for image representation in recent literature. Their contribution to the performance of image analysis can be improved through fine-tuning. As an ultimate solution, one might even train a deep network from scratch with the domain-relevant images, a highly desirable option which is generally impeded in pathology by lack of labeled images and the computational expense. In this study, we propose a new network, namely KimiaNet, that employs the topology of the DenseNet with four dense blocks, fine-tuned and trained with histopathology images in different configurations. We used more than 240,000 image patches with 1000×1000 pixels acquired at 20× magnification through our proposed "high-cellularity mosaic" approach to enable the usage of weak labels of 7126 whole slide images of formalin-fixed paraffin-embedded human pathology samples publicly available through The Cancer Genome Atlas (TCGA) repository. We tested KimiaNet using three public datasets, namely TCGA, endometrial cancer images, and colorectal cancer images by evaluating the performance of search and classification when corresponding features of different networks are used for image representation. As well, we designed and trained multiple convolutional batch-normalized ReLU (CBR) networks. The results show that KimiaNet provides superior results compared to the original DenseNet and smaller CBR networks when used as feature extractor to represent histopathology images.

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.

Recent advances in medical imaging have created many possibilities for the exploitation of both microscopic images in digital form and the whole slide images (WSIs) for multiple tasks such as classification, prediction, and retrieval. This is mainly due to annotated datasets available through various research organizations. Magnification level is an important factor as pathologist views the biopsy samples at various magnifications to reach a diagnosis. Whereas WSIs generally do contain the magnification information, microscopic snapshots are often captured without attaching the magnification information. In this paper, we introduce a new dataset, Kimia-5MAG, consisting of 33,345 patches at 5 different magnification classes created from WSIs made publicly available by The Cancer Genome Atlas (TCGA). There exists a large number of microscopic snapshots captured from camera-mounted microscopes but are of little use for automatic processing due to lack of magnification information. One direction to make use of these datasets is learning the magnification level from high resolutions captured WSIs and transferring the knowledge to microscopic snapshots. We investigate combinations of several deep networks and classifiers to predict different magnification levels. The proposed framework achieves 93% classification accuracy. We also analyze the effect of rotation on magnification prediction.

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.

During the last decade, the digitization of pathology has gained considerable momentum. Digital pathology offers many advantages including more efficient workflows, easier collaboration as well as a powerful venue for telepathology. At the same time, applying Computer-Aided Diagnosis (CAD) on Whole Slide Images (WSIs) has received substantial attention as a direct result of the digitization. The first step in any image analysis is to extract the tissue. Hence, background removal is an essential prerequisite for efficient and accurate results for many algorithms. In spite of the obvious discrimination for human operator, the identification of tissue regions in WSIs could be challenging for computers mainly due to the existence of color variations and artifacts. Moreover, some cases such as alveolar tissue types, fatty tissues, and tissues with poor staining are difficult to detect. In this paper, we perform experiments on U-Net architecture with different network backbones (different topologies) to remove the background as well as artifacts from WSIs in order to extract the tissue regions. We compare a wide range of backbone networks including MobileNet, VGG16, EfficientNet-B3, ResNet50, ResNext101 and DenseNet121. We trained and evaluated the network on a manually labeled subset of The Cancer Genome Atlas (TCGA) Dataset. EfficientNet-B3 and MobileNet by almost 99% sensitivity and specificity reached the best results.

With the emergence of digital pathology, searching for similar images in large archives has gained considerable attention. Image retrieval can provide pathologists with unprecedented access to the evidence embodied in already diagnosed and treated cases from the past. This paper proposes a search engine specialized for digital pathology, called Yottixel, a portmanteau for "one yotta pixel," alluding to the big-data nature of histopathology images. The most impressive characteristic of Yottixel is its ability to represent whole slide images (WSIs) in a compact manner. Yottixel can perform millions of searches in real-time with a high search accuracy and low storage profile. Yottixel uses an intelligent indexing algorithm capable of representing WSIs with a mosaic of patches which are then converted into barcodes, called "Bunch of Barcodes" (BoB), the most prominent performance enabler of Yottixel. The performance of the prototype platform is qualitatively tested using 300 WSIs from the University of Pittsburgh Medical Center (UPMC) and 2,020 WSIs from The Cancer Genome Atlas Program (TCGA) provided by the National Cancer Institute. Both datasets amount to more than 4,000,000 patches of 1000 × 1000 pixels. We report three sets of experiments that show that Yottixel can accurately retrieve organs and malignancies, and its semantic ordering shows good agreement with the subjective evaluation of human observers.

The colorless biopsied tissue samples are usually stained in order to visualize different microscopic structures for diagnostic purposes. But color variations associated with the process of sample preparation, usage of raw materials, diverse staining protocols, and using different slide scanners may adversely influence both visual inspection and computer-aided image analysis. As a result, many methods are proposed for histopathology image stain normalization in recent years. In this study, we introduce a novel approach for stain normalization based on learning a mixture of multivariate skew-normal distributions for stain clustering and parameter estimation alongside a stain transformation technique. The proposed method, labeled "Class-Agnostic Weighted Normalization" (short CLAW normalization), has the ability to normalize a source image by learning the color distribution of both source and target images within an expectation-maximization framework. The novelty of this approach is its flexibility to quantify the underlying both symmetric and nonsymmetric distributions of the different stain components while it is considering the spatial information. The performance of this new stain normalization scheme is tested on several publicly available digital pathology datasets to compare it against state-of-the-art normalization algorithms in terms of ability to preserve the image structure and information. All in all, our proposed method performed superior more consistently in comparison with existing methods in terms of information preservation, visual quality enhancement, and boosting computer-aided diagnosis algorithm performance.

This paper introduces the "Projectron" as a new neural network architecture that uses Radon projections to both classify and represent medical images. The motivation is to build shallow networks which are more interpretable in the medical imaging domain. Radon transform is an established technique that can reconstruct images from parallel projections. The Projectron first applies global Radon transform to each image using equidistant angles and then feeds these transformations for encoding to a single layer of neurons followed by a layer of suitable kernels to facilitate a linear separation of projections. Finally, the Projectron provides the output of the encoding as an input to two more layers for final classification. We validate the Projectron on five publicly available datasets, a general dataset (namely MNIST) and four medical datasets (namely Emphysema, IDC, IRMA, and Pneumonia). The results are encouraging as we compared the Projectron's performance against MLPs with raw images and Radon projections as inputs, respectively. Experiments clearly demonstrate the potential of the proposed Projectron for representing/classifying medical images.

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%.

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 bar-codes 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.

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.

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.

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.

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.

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<; 180, in order to increase the expressiveness of Radon barcodes. We show examples for the exhaustive search for the simple case when we attempt to find 4 best projections out of 16 equidistant projections and compare it with the evolutionary approach in order to establish the benefit of the latter when operating on a small population size as in the case of micro-DE. We randomly selected 10 different classes from IRMA dataset (14,400 x-ray images in 58 classes) and further randomly selected 5 images per class for our tests.

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 (≈ 80% accuracy for the first hit) was achieved.

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 type-I and 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 opposition mining on the sample data, and then use the mined data to learn the relationship between input x and its opposite x̌. 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.

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 ≈ 81% retrieval accuracy for the first hit.

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.

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.

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.

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.

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 ∈ [a; b] assigns its opposite as x̆I= a + b - x. This, of course, is a very naive estimate of the actual or true (non-linear) opposite x̆II, which has been called type-II opposite in literature. In absence of any knowledge about a function y = f(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 x̆Iaccording 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 x̆II= f(x; y).

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.

Good results on image classification and retrieval using support vector machines (SVM) with local binary patterns (LBPs) as features have been extensively reported in the literature where an entire image is retrieved or classified. In contrast, in medical imaging, not all parts of the image may be equally significant or relevant to the image retrieval application at hand. For instance, in lung x-ray image, the lung region may contain a tumour, hence being highly significant whereas the surrounding area does not contain significant information from medical diagnosis perspective. In this paper, we propose to detect salient regions of images during training and fold the data to reduce the effect of irrelevant regions. As a result, smaller image areas will be used for LBP features calculation and consequently classification by SVM. We use IRMA 2009 dataset with 14,410 xray images to verify the performance of the proposed approach. The results demonstrate the benefits of saliency-based folding approach that delivers comparable classification accuracies with state-of-the-art but exhibits lower computational cost and storage requirements, factors highly important for big data analytics.

Every segmentation algorithm has parameters that need to be adjusted in order to achieve good results. Evolving fuzzy systems for adjustment of segmentation parameters have been proposed recently (Evolving fuzzy image segmentation -- EFIS [1]). However, similar to any other algorithm, EFIS too suffers from a few limitations when used in practice. As a major drawback, EFIS depends on detection of the object of interest for feature calculation, a task that is highly application-dependent. In this paper, a new version of EFIS is proposed to overcome these limitations. The new EFIS, called self-configuring EFIS (SC-EFIS), uses available training data to auto-configure the parameters that are fixed in EFIS. As well, the proposed SCEFIS relies on a feature selection process that does not require the detection of a region of interest (ROI).

The concept of opposition-based learning (OBL) can be categorized into Type-I and Type-II OBL methodologies. The Type-I OBL is based on the opposite points in the variable space while the Type-II OBL considers the opposite of function value on the landscape. In the past few years, many research works have been conducted on development of Type-I OBL-based approaches with application in science and engineering, such as opposition-based differential evolution (ODE). However, compared to Type-I OBL, which cannot address a real sense of opposition in term of objective value, the Type-II OBL is capable to discover more meaningful knowledge about problem's landscape. Due to natural difficulty of proposing a Type-II-based approach, very limited research has been reported in that direction. In this paper, for the first time, the concept of Type-II OBL has been investigated in detail in optimization; also it is applied on the DE algorithm as a case study. The proposed algorithm is called opposition-based differential evolution Type-II (ODE-II) algorithm; it is validated on the testbed proposed for the IEEE Congress on Evolutionary Computation 2013 (IEEE CEC-2013) contest with 28 benchmark functions. Simulation results on the benchmark functions demonstrate the effectiveness of the proposed method as the first step for further developments in Type-II OBL-based schemes.

One of the main disadvantages of population-based evolutionary algorithms (EAs) is their high computational cost due to the nature of evaluation, specially when the population size is large. The micro-algorithms employ a very small number of individuals, which can accelerate the convergence speed of algorithms dramatically, while it highly increases the stagnation risk. One approach to overcome the stagnation problem can be increasing the diversity of the population. To do so, a micro-differential evolution with vectorized random mutation factor (MDEVM) algorithm is proposed in this paper, which utilizes the small size population benefit while preventing stagnation through diversification of the population. The proposed algorithm is tested on the 28 benchmark functions provided at the IEEE congress on evolutionary computation 2013 (CEC-2013). Simulation results on the benchmark functions demonstrate that the proposed algorithm improves the convergence speed of its parent algorithm.

Despite the large number of techniques proposed in recent years, accurate segmentation of digital images remains a challenging task for automated computer algorithms. Approaches based on machine learning hold particular promise in this regard, because in many applications, e.g., medical image analysis, frequent user intervention can be assumed to correct the results, thereby generating valuable feedback for algorithmic learning. In order to learn segmentation of new (unseen) images, such user feedback (correction of current or past results) seems indispensable. In this paper, we propose the formation and evolution of fuzzy rules for user-oriented environments in which feedback is captured by design. The evolving fuzzy image segmentation (EFIS) can be used to adjust the parameters of existing segmentation methods, switch between their results, or fuse their results. Specifically, we propose a single-parametric EFIS (SEFIS), apply its rule evolution to breast ultrasound images, and evaluate the results using three segmentation methods, namely, global thresholding, region growing, and statistical region merging. The results show increased accuracy across all tests and for all methods. For instance, the accuracy of statistical region merging can be improved from 59% ± 30% to 71% ± 22%. We also propose a multiparametric EFIS (MEFIS) for switching between or fusing the results of multiple segmentation methods. Preliminary results indicate that MEFIS can further increase overall segmentation accuracy. Three thresholding methods with accuracies of 62% ± 11%, 64% ± 16%, and 61% ± 9% were combined to reach an overall accuracy of 66% ± 15%. Finally, we compare our SEFIS scheme with five other thresholding methods to evaluate its overall performance.

Normalized cut (N-cut) is a rather recent approach to image segmentation representing the image as a graph and using eigenvalues to partition it. However, this method has several parameters that affect the segmentation accuracy. Using pre-set values for these parameters may generate good results for some images and bad results for others. Thus, to achieve maximum segmentation accuracy, these parameters may be manually finetuned for every set of images. This process, of course, would be impractical and lack generality. In this paper, a method is proposed to automatically determine N-cut parameters for every single image based on the image features using evolving fuzzy sets. The proposed method is applied to magnetic reasoning images (MRI) of bladder.

Image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label are connected and meaningful, and share certain visual characteristics. Pixels in a region are similar with respect to some features or property, such as color, intensity, or texture. Adjacent regions may be significantly different with respect to the same characteristics. Therefore, it is difficult for a static (non-learning) segmentation technique to accurately segment different images with different characteristics. In this paper, an evolving fuzzy system is used to segment medical images. The system uses some training images to build an initial fuzzy system which then evolves online as new images are encountered. Each new image is segmented using the evolved fuzzy system and may contribute to updating the system. This process provides better segmentation results for new images compared to static paradigms. The average of segmentation accuracy for test images is calculated by comparing every segmented image with its gold standard image prepared manually by an expert.

Level set methods have been used for image segmentation. Because partial deferential equations are solved to propagate a curve, level-set image segmentation has a slow convergence speed. The objective of this paper is to propose a method that increases the convergence speed. The proposed approach exploits the benefit of multi-resolutional analysis. Wavelet transform is used to decompose the image into different resolutions. The obtained results show a great improvement in terms of speed and accuracy.

Image thresholding is a very important phase in the image analysis process. In all traditional segmentation schemes, statically calculated thresholds or initial points are used to binarize images. Because of the differences in images characteristics, these techniques may generate high segmentation accuracy for some images and low accuracy for other images. Intelligent segmentation by “dynamic” determination of thresholds based on image properties may be a more robust solution. In this paper, we use the Gabor filter to generate features from regions of interest (ROIs) detected by the the SIFT technique (Scale-Invariant Feature Transform). These features are used to train a neural network for the task of image thresholding. The average of segmentation accuracies for a set of test images is calculated by comparing every segmented image with its gold standard image marked by human experts.

The oppositional and antipodal forms of forces, entities and quantities have been envisaged in the context of practical and applied field of engineering and management science in order to create a more complete picture of reality. The interplay between entities and opposite entities is apparently fundamental for maintaining universal balance. A large number of problems in engineering and science cannot be approached with conventional schemes and are generally handled with intelligent techniques such as evolutionary, neural, reinforcing and swarm-based methods. Visualization of unforeseen financial events is one of those applications, where failure of particular corporate firm can be forecasted based on the combination of several indicators. The hidden artifacts of corporate financial events could also be evaluated with the help of ant-based behavior of pheromone deposition. The learning in the pheromone deposition is subjected to oppositional forces leading towards the equilibrium of the corporate of interest. The motivation of this paper is to initiate the model to analyze and to represent the financial practices of typical clusters, which may cause failure of that firm in near future. In this paper we propose to use opposition based learning and Soft Bergman Based Clustering to implement the proposed model. Brief comparison of results is presented at the end of the proposal.

In many image processing applications, image thresholding is considered to be an important task. Opposition-Based Learning (OBL) was recently introduced and used to enhance different computation algorithms. In this paper, a new thresholding algorithm is proposed by utilizing the concept of opposite fuzzy sets. The algorithm is applied on general set of images and compared with the previous opposition-based thresholding algorithm [1] and a commonly used thresholding method, namely the Otsu method. The most reliable results on the test data are achieved using the proposed algorithm.

In algorithms design, one of the important aspects is to consider efficiency. Many algorithm design paradigms are existed and used in order to enhance algorithms' efficiency. Opposition-based Learning (OBL) paradigm was recently introduced as a new way of thinking during the design of algorithms. The concepts of opposition have already been used and applied in several applications. These applications are from different fields, such as optimization algorithms, learning algorithms and fuzzy logic. The reported results confirm that OBL paradigm was promising to accelerate or to enhance accuracy of soft computing algorithms. In this paper, a survey of existing applications of opposition-based computing is presented.

Image thresholding is a very important phase in the image analysis process. However, different images have different characteristics making the traditional process of thresholding by one algorithm a very challenging task. That is because any thresholding method may be perform well for some images but for sure it will not be suitable for all images. In this paper, intelligent thresholding by training a neural network is proposed. The neural network is trained using a set of features extracted from medical images randomly selected form a sample set and then tested using the remaining medical images. This process is repeated multiple times to verify the generalization ability of the network. The average of segmentation accuracy is calculated by comparing every segmented image with its gold standard image.

Opposition-based computing is the paradigm for incorporating entities along with their opposites within the search, optimization and learning mechanisms. In this work, we introduce the notion of "opposite fuzzy sets" in order to use the entropy difference between a fuzzy set and its opposite to carry out object discrimination in digital images. A quasi-global scheme is used to execute the calculations, which can be employed by any other existing thresholding technique. Results for prostate ultrasound images have been provided to verify the performance whereas expert's markings have been used as gold standard.

Large scale neural networks have many hundreds or thousands of parameters (weights and biases) to learn, and as a result tend to have very long training times. Small scale networks can be trained quickly by using second-order information, but these fail for large architectures due to high computational cost. Other approaches employ local search strategies, which also add to the computational cost. In this paper we present a simple method, based on opposite transfer functions which greatly improve the convergence rate and accuracy of gradient-based learning algorithms. We use two variants of the backpropagation algorithm and common benchmark data to highlight the improvements. We find statistically significant improvements in both convergence speed and accuracy.

Image thresholding is a challenging task in image processing field. Many efforts have already been made to propose universal, robust methods to handle a wide range of images. Previously by the same authors, an optimization-based thresholding approach was introduced. According to the proposed approach, differential evolution (DE) algorithm, minimizes dissimilarity between the input grey-level image and the bi-level (thresholded) image. In the current paper, micro opposition-based differential evolution (micro-ODE), DE with very small population size and opposition-based population initialization, has been proposed. Then, it is compared with a well-known thresholding method, Kittler algorithm and also with its non-opposition-based version (micro-DE). In overall, the proposed approach outperforms Kittler method over 16 challenging test images. Furthermore, the results confirm that the micro-ODE is faster than micro-DE because of embedding the opposition-based population initialization.

Until recent years, terahertz (THz) waves were an undiscovered, or most importantly, an unexploited area of electromagnetic spectrum. This was due to difficulties in generation and detection of THz waves. Recent advances in hardware technology have started to open up the field to new applications such as THz imaging. THz waves can penetrate through diverse materials such that internal structures, invisible to other imaging modalities, can be visulaized. However, automated processing of THz images can be quite challenging. Low contrast and the presence of a widely unknown type of noise make the analysis of these images difficult. In this paper we attempt to detect defects in composite material using a Terahertz imaging system. According to our knowledge this is the first time that this type of materials are being tested under Terahertz cameras using filtering and information fusion. In this preliminary report, we employ stick filter and a simple fuzzy approach to detect defects by fusing information from both amplitude and phase images. The results show that using fuzzy techniques can easily incorporate domain knowledge and assist the defect detection.

This paper presents some investigations on tradeoff between exploration and exploitation of opposition-based Q(lambda) with non-Markovian update (NOQ(lambda)) in a dynamic environment. In the previous work the authors applied NOQ(lambda) to the deterministic GridWorld problem. In this paper, we have implemented the NOQ(lambda) algorithm for a simple elevator control problem to test the behavior of the algorithm for non-deterministic and dynamic environment. We also extend the NOQ(lambda) algorithm by introducing the opposition weight to find a better tradeoff between exploration and exploitation for the NOQ(lambda) technique. The value of the opposition weight increases as the number of steps increases. Hence, it has more positive effects on the Q-value updates for opposite actions as the learning progresses. The performance of NOQ(lambda) method is compared with Q(lambda) technique. The experiments indicate that NOQ(lambda) performs better than Q(lambda).

Numerical condition affects the learning speed and accuracy of most artificial neural network learning algorithms. In this paper, we examine the influence of opposite transfer functions on the conditioning of feedforward neural network architectures. The goal is not to discuss a new training algorithm nor error surface geometry, but rather to present characteristics of opposite transfer functions which can be useful for improving existing or to develop new algorithms. Our investigation examines two situations: (1) network initialization, and (2) early stages of the learning process. We provide theoretical motivation for the consideration of opposite transfer functions as a means to improve conditioning during these situations. These theoretical results are validated by experiments on a subset of common benchmark problems. Our results also reveal the potential for opposite transfer functions in other areas of, and related to neural networks.

Evolutionary algorithms (EAs) are well-known optimization approaches to deal with nonlinear and complex problems. However, these population-based algorithms are computationally expensive due to the slow nature of the evolutionary process. This paper presents a novel algorithm to accelerate the differential evolution (DE). The proposed opposition-based DE (ODE) employs opposition-based learning (OBL) for population initialization and also for generation jumping. In this work, opposite numbers have been utilized to improve the convergence rate of DE. A comprehensive set of 58 complex benchmark functions including a wide range of dimensions is employed for experimental verification. The influence of dimensionality, population size, jumping rate, and various mutation strategies are also investigated. Additionally, the contribution of opposite numbers is empirically verified. We also provide a comparison of ODE to fuzzy adaptive DE (FADE). Experimental results confirm that the ODE outperforms the original DE and FADE in terms of convergence speed and solution accuracy.

In this paper, an enhanced version of the opposition-based differential evolution (ODE) is proposed. ODE utilizes opposite numbers in the population initialization and generation jumping to accelerate differential evolution (DE). Instead of opposite numbers, in this work, quasi opposite points are used. So, we call the new extension quasi- oppositional DE (QODE). The proposed mathematical proof shows that in a black-box optimization problem quasi- opposite points have a higher chance to be closer to the solution than opposite points. A test suite with 15 benchmark functions has been employed to compare performance of DE, ODE, and QODE experimentally. Results confirm that QODE performs better than ODE and DE in overall. Details for the proposed approach and the conducted experiments are provided.

Although reinforcement learning (RL) emerged more than a decade ago, it is still under extensive investigation in application to large problems, where the states and actions are multi-dimensional and continuous and result in the so- called curse of dimensionality. Conventional RL methods are still not efficient enough in huge state-action spaces, while value-function generalization-based approaches require a very large number of good training examples. This paper presents an active exploratory approach to address the challenge of RL in large problems. The core principle of this approach is that the agent does not rush to the next state. Instead, it attempts a number of actions at the current state first, and then selects the action that returns the greatest immediate reward. The state resulting from performing the action is considered as the next state. Four active exploration algorithms for good actions are proposed: random-based search, opposition-based random search, search by cyclical adjustment, and opposition-based cyclical adjustment of each action dimension. The efficiency of these algorithms is determined by a visual-servoing experiment with a 6-axis robot.

This paper presents several extensions to an algorithm in the family of ant colony optimization, the ant colony system. The proposed extensions are based on the idea of opposition and attempt to increase the exploration efficiency of the solution space. The modifications focus on the solution construction phase of the ant colony system. Three of the proposed methods work by pairing the ants and synchronizing their path selection. The two other approaches modify the decisions of the ants by using an opposite-pheromone content. Results on the application of these algorithms on travelling salesman problem instances demonstrate that the concept of opposition is not easily applied to the ant algorithm. Only one of the pheromone-based methods showed performance improvements that were statistically significant. The quality of the solutions increased and more optimal solutions were found. The other extensions showed no clear improvement. Further work must be conducted to explore the successful pheromone-based approach, as well as to determine if opposition should be applied to a different phase of the algorithm

Rule-based classifiers allow rationalization of classifications made. This in turn improves understanding which is essential for effective decision support. As a rule based classifier, the pattern discovery (PD) algorithm functions well in discrete, nominal and continuous data domains. A drawback when using PD as a classifier for decision support is that it has an unbounded decision space that confounds the understanding of the degree of support for a decision. Incorporating PD into a fuzzy inference system (FIS) allows the the degree of support for a decision to be expressed with intuitively understandable terms. In addition, using discrete algorithms in continuous domains can result in reduced accuracy due to quantization. Fuzzification reduces this ldquocost of quantizationrdquo and improves classification performance. In this work, the PD algorithm was used as a source of rules for a series of FISs implemented using different rule weighting and defuzzification schemes, each providing a linguistic basis for rule description and a bounded space for expression of decision support. The output of each FIS consists of a suggested outcome, a strong confidence metric describing suggestions within this space and a linguistic expression of the rules. This constitutes a stronger basis for decision making than that provided by PD alone. A variety of synthetic, continuous class distributions with varying degrees of separation was used to evaluate the performance of fuzzy, PD, back-propagation and Bayesian classifiers. Overall, the accuracy of the fuzzy system was found to be similar, but slightly below, that of the inherently continuous valued classifiers and was somewhat improved with respect to the PD classifiers. For the difficult spiral class distributions studied, the fuzzy classifiers were able to make more classifications than the PD classifiers. The correct classification rates for the fuzzy classifiers were similar across the various rule weighting and defuzzification schemes, demonstrating the strength of the statistical method for rule generation. Analysis of several real-world data sets shows that a PD-based FIS has comparable performance to a neuro-fuzzy system. The use of a PD based FIS however, provides insight into the structure of the data analyzed not available through the other approaches.

Differential evolution (DE) is a simple, reliable, and efficient optimization algorithm. However, it suffers from a weakness, losing the efficiency over optimization of noisy problems. In many real-world optimization problems we are faced with noisy environments. This paper presents a new algorithm to improve the efficiency of DE to cope with noisy optimization problems. It employs opposition-based learning for population initialization, generation jumping, and also improving population's best member. A set of commonly used benchmark functions is employed for experimental verification. The details of proposed algorithm and also conducted experiments are given. The new algorithm outperforms DE in terms of convergence speed.

Evolutionary Algorithms (EAs) are well-known optimization approaches to cope with non-linear, complex problems. These population-based algorithms, however, suffer from a general weakness; they are computationally expensive due to slow nature of the evolutionary process. This paper presents some novel schemes to accelerate convergence of evolutionary algorithms. The proposed schemes employ opposition-based learning for population initialization and also for generation jumping. In order to investigate the performance of the proposed schemes, Differential Evolution (DE), an efficient and robust optimization method, has been used. The main idea is general and applicable to other population-based algorithms such as Genetic algorithms, Swarm Intelligence, and Ant Colonies. A set of test functions including unimodal and multimodal benchmark functions is employed for experimental verification. The details of proposed schemes and also conducted experiments are given. The results are highly promising.

This paper presents an active contour model to accurately detect pupil boundary in order to improve the performance of iris recognition systems. The contour model takes into consideration that an actual pupil boundary is a near-circular contour rather than a perfect circle. Two types of controlling force models, introduced as internal and external forces, are designed to properly activate the contour and locate it over the pupil boundary. The internal forces are designed to smooth the curve as well as to keep it close to a circular shape by pushing the contour vertices to their local radial mean. The external forces, which are responsible for pulling the contour vertices toward the pupil boundary, are designed based on a circular-curve gradient measurement with a proper angular range with respect to the contour center. In addition, an iterative algorithm has been developed in order to capture limbus and eyelids. The developed algorithm iteratively searches the limbus and eyelids boundaries and excludes the detected eyelids areas that cover the iris. Excluding the eyelids leads to a more precise search for limbus in the next iteration and the search is completed when the circular parameters of the limbus converge to fixed values. The eyelid contours are modeled as elliptic curves considering the spherical shape of an eyeball and the search is based on the expected contour in different degrees of eye openness.

This paper is an investigation of the application of ant colony optimization to image thresholding. It presents an approach where ants are assigned to each pixel of an image and they move around the image seeking low grayscale regions. The proposed ant-based method performs better than three other established thresholding algorithms. Further work must be conducted to optimize parameters, select the best cost function, improve the analysis of the pheromone data and reduce computation time. The study indicates that an ant-based approach has the potential of becoming an established image thresholding technique.

A new robust image thresholding technique is introduced in this paper. Comprehensive experiments show that a single thresholding method can not be successful for all kind of images. The proposed approach uses fusion of some well-known thresholding methods by applying weighted voting at the decision level. The main objective is improving robustness of thresholding approach by participating several methods. Although, the proposed approach can not guaranty the best result for all kind of images but it shows higher performance and consistent/smoother behavior in overall. The performance of the new approach and nine well-established thresholding methods are compared by applying to an image set with high image diversity. The comparison results show that the proposed approach outperforms other nine well-established thresholding approaches. The proposed approach has been explained in details and experimental results are provided.

In this paper a new approach to object extraction and recognition based on reinforcement learning is presented. We use this novel idea as a method to optimally segment the image and increase the recognition rate. The success rate is compared with a classical approach. Preliminary results demonstrate increase in recognition rate.

This paper introduces a new method to medical image segmentation using a reinforcement learning scheme. We use this novel idea as an effective way to optimally find the appropriate local thresholding and structuring element values and segment the prostate in ultrasound images. Reinforcement learning agent uses an ultrasound image and its manually segmented version and takes some actions (i.e., different thresholding and structuring element values) to change the environment (the quality of segmented image). The agent is provided with a scalar reinforcement signal determined objectively. The agent uses these objective reward/punishment to explore/exploit the solution space. The values obtained using this way can be used as valuable knowledge to fill a Q-matrix. The reinforcement learning agent can use this knowledge for similar ultrasound images as well. The results demonstrate high potential for applying reinforcement learning in the field of medical image segmentation.

The problem of delayed reward in reinforcement learning is usually tackled by implementing the mechanism of eligibility traces. In this paper we introduce an extension of eligibility traces to solve one of the challenging problems in reinforcement learning. The concept of opposition traces is proposed in this work to deal with large state space problems in reinforcement learning applications. We combine the idea of opposition and eligibility traces to construct the opposition-based Q(lambda). The results are compared with the conventional Watkins' Q(lambda) and reflect a remarkable performance increase.

The backpropagation algorithm is a very popular approach to learning in feed-forward multi-layer perceptron networks. However, in many scenarios the time required to adequately learn the task is considerable. Many existing approaches have improved the convergence rate by altering the learning algorithm. We present a simple alternative approach inspired by opposition-based learning that simultaneously considers each network transfer function and its opposite. The effect is an improvement in convergence rate and over traditional backpropagation learning with momentum. We use four common benchmark problems to illustrate the improvement in convergence time.

In this paper a novel method for automatic prostate segmentation in transrectal ultrasound images is presented. Morphological grey level transformations are first used to generate an image with enough bright intensity around the prostate. This image is then thresholded to produce a binary image. Then by finding and using a point as the inside point for the prostate, a Kalman estimator is used to isolate the prostate boundary from any irrelevant parts and produce a roughly segmented version (as coarse estimation). Consequently, a fuzzy inference system describing regional and gray level information is employed to enhance the contrast of the prostate with respect to the background. Using strong edges obtained from this enhanced image and information from pixels gradients and also the characteristics in the vicinity of the coarse estimation, the final boundary is extracted. A number of experiments are conducted to validate this method.

Image quality evaluation by human observers is heavily subjective in nature. Individual observers judge the image quality differently. In previous works, an observer-dependent system for subjective image enhancement, which is based on fusion of different algorithms, was introduced. In this paper, the system configuration for sharpness/smoothness is discussed and experimental results are provided.

In many image-processing applications the image quality should be improved to support the human perception. Image quality evaluation by human observers is, however, heavily subjective in nature. Individual observers judge the image quality differently. In many cases, the quality of the relevant part of image information, which is perceived by the observer, should reach a maximum. In previous works, an overall system for subjective image enhancement, which is based on fusion of different algorithms, was introduced. In this paper, more details of the overall-system structure are provided. Furthermore, the test results for contrast and sharpness/smoothness as interesting image qualities are also presented.

The Electronic Portal Imaging Device (EPID) uses a high-energy treatment beam to project the body interior of a patient on to a fluorescent screen that is scanned by a camera. Because of the imaging physics, the unprocessed images are very poor in quality. This paper presents an approach that combines an associative restoration algorithm with a fuzzy image enhancement technique. By fusing the electronic portal image with a pre-treatment captured simulator image, a higher image quality than by conventional techniques is achieved.

Medical images in radiation therapy, especially electronic portal images, are often very poor in quality because of imaging physics. For a reliable patient set-up verification by tracking of relevant features, better in-treatment images are necessary. In this work, we present the prototype of an additive fuzzy system for a locally adaptive image enhancement and a modified associative memory for image restoration. Using a-priori knowledge and the advantages of this hybrid neural-fuzzy system, a much better quality for the in-treatment image can be achieved.