No file available
To read the file of this research,
you can request a copy directly from the authors.
Genetic U-Net: Automatically Designing Lightweight U-shaped CNN Architectures Using the Genetic Algorithm for Retinal Vessel Segmentation
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
Many previous works based on deep learning for retinal vessel segmentation have achieved promising performance by manually designing U-shaped convolutional neural networks (CNNs). However, the manual design of these CNNs is time-consuming and requires extensive empirical knowledge. To address this problem, we propose a novel method using genetic algorithms (GAs) to automatically design a lightweight U-shaped CNN for retinal vessel segmentation, called Genetic U-Net. Here we first design a special search space containing the structure of U-Net and its corresponding operations, and then use genetic algorithm to search for superior architectures in this search space. Experimental results show that the proposed method outperforms the existing methods on three public datasets, DRIVE, CHASE_DB1 and STARE. In addition, the architectures obtained by the proposed method are more lightweight but robust than the state-of-the-art models.
ResearchGate has not been able to resolve any citations for this publication.
Deep learning (DL)-based semantic segmentation methods have been providing state-of-the-art performance in the past few years. More specifically, these techniques have been successfully applied in medical image classification, segmentation, and detection tasks. One DL technique, U-Net, has become one of the most popular for these applications. We propose a recurrent U-Net model and a recurrent residual U-Net model, which are named RU-Net and R2U-Net, respectively. The proposed models utilize the power of U-Net, residual networks, and recurrent convolutional neural networks. There are several advantages to using these proposed architectures for segmentation tasks. First, a residual unit helps when training deep architectures. Second, feature accumulation with recurrent residual convolutional layers ensures better feature representation for segmentation tasks. Third, it allows us to design better U-Net architectures with the same number of network parameters with better performance for medical image segmentation. The proposed models are tested on three benchmark datasets, such as blood vessel segmentation in retinal images, skin cancer segmentation, and lung lesion segmentation. The experimental results show superior performance on segmentation tasks compared to equivalent models, including a variant of a fully connected convolutional neural network called SegNet, U-Net, and residual U-Net. © 2019 Society of Photo-Optical Instrumentation Engineers (SPIE).
Neural architecture search (NAS) has significant progress in improving the accuracy of image classification. Recently, some works attempt to extend NAS to image segmentation which shows preliminary feasibility. However, all of them focus on searching architecture for semantic segmentation in natural scenes. In this paper, we design three types of primitive operation set on search space to automatically find two cell architecture DownSC and UpSC for semantic image segmentation especially medical image segmen- tation. Inspired by the U-net architecture and its variants successfully applied to various medical image segmentation, we propose NAS-Unet which is stacked by the same number of DownSC and UpSC on a U- like backbone network. The architectures of DownSC and UpSC updated simultaneously by a differential architecture strategy during search stage. We demonstrate the well segmentation results of the proposed method on Promise12, Chaos and ultrasound nerve datasets, which collected by Magnetic Resonance Imaging (MRI), Computed Tomography (CT), and ultrasound respectively. Without any pretraining, our architecture searched on PASCAL VOC2012, attains better performances and much less parameters (about 0.8M) than U-net and one of its variants when evaluated on above three types of medical image datasets.
Objective:
Deep learning based methods for retinal vessel segmentation are usually trained based on pixel-wise losses, which treat all vessel pixels with equal importance in pixel-to-pixel matching between a predicted probability map and the corresponding manually annotated segmentation. However, due to the highly imbalanced pixel ratio between thick and thin vessels in fundus images, a pixel-wise loss would limit deep learning models to learn features for accurate segmentation of thin vessels, which is an important task for clinical diagnosis of eye-related diseases.
Methods:
In this paper we propose a new segment-level loss which emphasizes more on the thickness consistency of thin vessels in the training process. By jointly adopting both the segment-level and the pixel-wise losses, the importance between thick and thin vessels in the loss calculation would be more balanced. As a result, more effective features can be learned for vessel segmentation without increasing the overall model complexity.
Results:
Experimental results on public datasets demonstrate that the model trained by the joint losses outperforms the current state-of-the-art methods in both separate-training and cross-training evaluations.
Conclusion:
Compared to the pixel-wise loss, utilizing the proposed joint loss framework is able to learn more distinguishable features for vessel segmentation. In addition, the segment-level loss can bring consistent performance improvement for both deep and shallow network architectures.
Significance:
The findings from this study of using joint losses can be applied to other deep learning models for performance improvement without significantly changing the network architectures.
Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output. In this paper, we embrace this observation and introduce the Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion. Whereas traditional convolutional networks with L layers have L connections - one between each layer and its subsequent layer - our network has L(L+1)/2 direct connections. For each layer, the feature-maps of all preceding layers are used as inputs, and its own feature-maps are used as inputs into all subsequent layers. DenseNets have several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters. We evaluate our proposed architecture on four highly competitive object recognition benchmark tasks (CIFAR-10, CIFAR-100, SVHN, and ImageNet). DenseNets obtain significant improvements over the state-of-the-art on most of them, whilst requiring less memory and computation to achieve high performance. Code and models are available at https://github.com/liuzhuang13/DenseNet.
Goals:
In this work, we present an extensive description and evaluation of our method for blood vessel segmentation in fundus images based on a discriminatively trained, fully connected conditional random field model.
Methods:
Standard segmentation priors such as a Potts model or total variation usually fail when dealing with thin and elongated structures. We overcome this difficulty by using a conditional random field model with more expressive potentials, taking advantage of recent results enabling inference of fully connected models almost in real-time. Parameters of the method are learned automatically using a structured output support vector machine, a supervised technique widely used for structured prediction in a number of machine learning applications.
Results:
Our method, trained with state of the art features, is evaluated both quantitatively and qualitatively on four publicly available data sets: DRIVE, STARE, CHASEDB1 and HRF. Additionally, a quantitative comparison with respect to other strategies is included.
Conclusion:
The experimental results show that this approach outperforms other techniques when evaluated in terms of sensitivity, F1-score, G-mean and Matthews correlation coefficient. Additionally, it was observed that the fully connected model is able to better distinguish the desired structures than the local neighborhood based approach.
Significance:
Results suggest that this method is suitable for the task of segmenting elongated structures, a feature that can be exploited to contribute with other medical and biological applications.
Retinal images can be used to detect and follow up several important chronic diseases. The classification of retinal images requires an experienced ophthalmologist. This has been a bottleneck to implement routine screenings performed by general physicians. It has been proposed to create automated systems that can perform such task with little intervention from humans, with partial success. In this work, we report advances in such endeavor, by using a Lattice Neural Network with Dendritic Processing (LNNDP). We report results using several metrics, and comparing against well known methods such as Support Vector Machines (SVM) and Multilayer Perceptrons (MLP). Our proposal shows better performance than other approaches reported in literature. An additional advantage is that unlike those other tools, LNNDP requires no parameters, and it automatically constructs its structure to solve a particular problem. The proposed methodology requires four steps: (1) Pre-processing, (2) Feature computation, (3) Classification and (4) Post-processing. The Hotelling T2 control chart was used to reduce the dimensionality of the feature vector, from 7 that were used before to 5 in this work. The experiments were run on images of DRIVE and STARE databases.The results show that on average, F1-Score is better in LNNDP, compared with SVM and MLP implementations. Same improvement is observed for MCC and the accuracy.
The ImageNet Large Scale Visual Recognition Challenge is a benchmark in
object category classification and detection on hundreds of object categories
and millions of images. The challenge has been run annually from 2010 to
present, attracting participation from more than fifty institutions.
This paper describes the creation of this benchmark dataset and the advances
in object recognition that have been possible as a result. We discuss the
challenges of collecting large-scale ground truth annotation, highlight key
breakthroughs in categorical object recognition, provide detailed a analysis of
the current state of the field of large-scale image classification and object
detection, and compare the state-of-the-art computer vision accuracy with human
accuracy. We conclude with lessons learned in the five years of the challenge,
and propose future directions and improvements.
Many of the common systemic diseases present characteristic changes in the fundus of the eye, but fundoscopy is often performed by an ophthalmologist. Our purpose was to assess the value of fundoscopy for the general practitioners (GPs) regarding the diagnosis and management of the cases which they face in daily practice.
689 patients were referred by GPs to the outpatient ophthalmology department for fundoscopy during the year 2010. The causes of this referral, the results of ophthalmoscopy and its significance in the final diagnosis were recorded and analyzed.
In 22 patients (3.1%), fundoscopy revealed optic disc edema. In 7 patients with head trauma (9.7%), fundoscopy revealed intravitreous haemorrhage and Berlin edema. From the patients with photopsias or floaters, 5 (10.2%) had retinal detachment. Finally, in cases with diabetes mellitus or hypertension, ophthalmoscopy was very important to detect the existence and grade the degree of diabetic or hypertensive retinopathy, if they appeared, and as a result to evaluate the prognosis of the disease.
Fundoscopy is fundamental for the GP, as it may help to confirm or exclude the diagnosis of many common diseases. Nevertheless, there are clinical entities where ophthalmoscopy should be performed by an ophthalmologist, in order to be more specific and accurate, and GP should be able to recognise these cases.
We describe an automated method to locate and outline blood vessels in images of the ocular fundus. Such a tool should prove useful to eye care specialists for purposes of patient screening, treatment evaluation, and clinical study. Our method differs from previously known methods in that it uses local and global vessel features cooperatively to segment the vessel network. We evaluate our method using hand-labeled ground truth segmentations of 20 images. A plot of the operating characteristic shows that our method reduces false positives by as much as 15 times over basic thresholding of a matched filter response (MFR), at up to a 75% true positive rate. For a baseline, we also compared the ground truth against a second hand-labeling, yielding a 90% true positive and a 4% false positive detection rate, on average. These numbers suggest there is still room for a 15% true positive rate improvement, with the same false positive rate, over our method. We are making all our images and hand labelings publicly available for interested researchers to use in evaluating related methods.
A method is presented for automated segmentation of vessels in two-dimensional color images of the retina. This method can be used in computer analyses of retinal images, e.g., in automated screening for diabetic retinopathy. The system is based on extraction of image ridges, which coincide approximately with vessel centerlines. The ridges are used to compose primitives in the form of line elements. With the line elements an image is partitioned into patches by assigning each image pixel to the closest line element. Every line element constitutes a local coordinate frame for its corresponding patch. For every pixel, feature vectors are computed that make use of properties of the patches and the line elements. The feature vectors are classified using a kappaNN-classifier and sequential forward feature selection. The algorithm was tested on a database consisting of 40 manually labeled images. The method achieves an area under the receiver operating characteristic curve of 0.952. The method is compared with two recently published rule-based methods of Hoover et al. and Jiang et al. The results show that our method is significantly better than the two rule-based methods (p < 0.01). The accuracy of our method is 0.944 versus 0.947 for a second observer.
Recurrent neural networks can be used to map input sequences to output sequences, such as for recognition, production or prediction problems. However, practical difficulties have been reported in training recurrent neural networks to perform tasks in which the temporal contingencies present in the input/output sequences span long intervals. We show why gradient based learning algorithms face an increasingly difficult problem as the duration of the dependencies to be captured increases. These results expose a trade-off between efficient learning by gradient descent and latching on information for long periods. Based on an understanding of this problem, alternatives to standard gradient descent are considered.
In this paper, a concept of degree of population diversity is introduced to quantitatively characterize and theoretically analyze the problem of premature convergence in genetic algorithms (GAs) within the framework of Markov chain. Under the assumption that the mutation probability is zero, the search ability of the GAs is discussed. It is proved that the degree of population diversity converges to zero with probability 1 so that the search ability of a genetic algorithm (GA) decreases and premature convergence occurs. Moreover, an explicit formula for the conditional probability of allele loss at a certain bit position is established to show the relationships between premature convergence and the GA parameters, such as population size, mutation probability, and some population statistics. The formula also partly answers the questions of to where a GA most likely converges. The theoretical results are all supported by the simulation experiments.
The detection of retinal vessel is of great importance in the diagnosis and treatment of many ocular diseases. Many methods have been proposed for vessel detection. However, most of the algorithms neglect the connectivity of the vessels, which plays an important role in the diagnosis. In this paper, we propose a novel method for retinal vessel detection. The proposed method includes a dense dilated network to get an initial detection of the vessels and a probability regularized walk algorithm to address the fracture issue in the initial detection. The dense dilated network integrates newly proposed dense dilated feature extraction blocks into an encoder-decoder structure to extract and accumulate features at different scales. A multi-scale Dice loss function is adopted to train the network. To improve the connectivity of the segmented vessels, we also introduce a probability regularized walk algorithm to connect the broken vessels. The proposed method has been applied on three public data sets: DRIVE, STARE and CHASE_DB1. The results show that the proposed method outperforms the state-of-the-art methods in accuracy, sensitivity, specificity and also area under receiver operating characteristic curve.
Due to the complex morphology of fine vessels, it remains challenging for most of existing models to accurately segment them, particularly the capillaries in color fundus retinal images. In this paper, we propose a novel and lightweight deep learning model called Vessel-Net for retinal vessel segmentation. First, we design an efficient inception-residual convolutional block to combine the advantages of the Inception model and residual module for improved feature representation. Next, we embed the inception-residual blocks inside a U-like encoder-decoder architecture for vessel segmentation. Then, we introduce four supervision paths, including the traditional supervision path, a richer feature supervision path, and two multi-scale supervision paths to preserve the rich and multi-scale deep features during model optimization. We evaluated our Vessel-Net against several recent methods on two benchmark retinal databases and achieved the new state-of-the-art performance (i.e. AUC of 98.21%/98.60% on the DRIVE and CHASE databases, respectively). Our ablation studies also demonstrate that the proposed inception-residual block and the multi-path supervision both can produce impressive performance gains for retinal vessel segmentation.
This paper introduces NSGA-Net --- an evolutionary approach for neural architecture search (NAS). NSGA-Net is designed with three goals in mind: (1) a procedure considering multiple and conflicting objectives, (2) an efficient procedure balancing exploration and exploitation of the space of potential neural network architectures, and (3) a procedure finding a diverse set of trade-off network architectures achieved in a single run. NSGA-Net is a population-based search algorithm that explores a space of potential neural network architectures in three steps, namely, a population initialization step that is based on prior-knowledge from hand-crafted architectures, an exploration step comprising crossover and mutation of architectures, and finally an exploitation step that utilizes the hidden useful knowledge stored in the entire history of evaluated neural architectures in the form of a Bayesian Network. Experimental results suggest that combining the dual objectives of minimizing an error metric and computational complexity, as measured by FLOPs, allows NSGA-Net to find competitive neural architectures. Moreover, NSGA-Net achieves error rate on the CIFAR-10 dataset on par with other state-of-the-art NAS methods while using orders of magnitude less computational resources. These results are encouraging and shows the promise to further use of EC methods in various deep-learning paradigms.
Automatic segmentation of retinal vessels in fundus images plays an important role in the diagnosis of some diseases such as diabetes and hypertension. In this paper, we propose Deformable U-Net (DUNet), which exploits the retinal vessels’ local features with a U-shape architecture, in an end to end manner for retinal vessel segmentation. Inspired by the recently introduced deformable convolutional networks, we integrate the deformable convolution into the proposed network. The DUNet, with upsampling operators to increase the output resolution, is designed to extract context information and enable precise localization by combining low-level features with high-level ones. Furthermore, DUNet captures the retinal vessels at various shapes and scales by adaptively adjusting the receptive fields according to vessels’ scales and shapes. Public datasets: DRIVE, STARE, CHASE_DB1 and HRF are used to test our models. Detailed comparisons between the proposed network and the deformable neural network, U-Net are provided in our study. Results show that more detailed vessels can be extracted by DUNet and it exhibits state-of-the-art performance for retinal vessel segmentation with a global accuracy of 0.9566/0.9641/0.9610/0.9651 and AUC of 0.9802/0.9832/0.9804/0.9831 on DRIVE, STARE, CHASE_DB1 and HRF respectively. Moreover, to show the generalization ability of the DUNet, we use another two retinal vessel data sets, i.e., WIDE and SYNTHE, to qualitatively and quantitatively analyze and compare with other methods. Extensive cross-training evaluations are used to further assess the extendibility of DUNet. The proposed method has the potential to be applied to the early diagnosis of diseases.
Photoacoustic imaging is an emerging imaging modality that is based upon the photoacoustic effect. In photoacoustic tomography (PAT), the induced acoustic pressure waves are measured by an array of detectors and used to reconstruct an image of the initial pressure distribution. A common challenge faced in PAT is that the measured acoustic waves can only be sparsely sampled. Reconstructing sparsely sampled data using standard methods results in severe artifacts that obscure information within the image. We propose a modified convolutional neural network (CNN) architecture termed Fully Dense UNet (FD-UNet) for removing artifacts from 2D PAT images reconstructed from sparse data and compare the proposed CNN with the standard UNet in terms of reconstructed image quality.
Rectified activation units (rectifiers) are essential for state-of-the-art neural networks. In this work, we study rectifier neural networks for image classification from two aspects. First, we propose a Parametric Rectified Linear Unit (PReLU) that generalizes the traditional rectified unit. PReLU improves model fitting with nearly zero extra computational cost and little overfitting risk. Second, we derive a robust initialization method that particularly considers the rectifier nonlinearities. This method enables us to train extremely deep rectified models directly from scratch and to investigate deeper or wider network architectures. Based on our PReLU networks (PReLU-nets), we achieve 4.94% top-5 test error on the ImageNet 2012 classification dataset. This is a 26% relative improvement over the ILSVRC 2014 winner (GoogLeNet, 6.66%). To our knowledge, our result is the first to surpass human-level performance (5.1%, Russakovsky et al.) on this visual recognition challenge.
Purpose:
Changes in the appearance of retinal blood vessels are an important indicator for various ophthalmologic and cardiovascular diseases, including diabetes, hypertension, arteriosclerosis, and choroidal neovascularization. Vessel segmentation from retinal images is very challenging because of low blood vessel contrast, intricate vessel topology, and the presence of pathologies such as microaneurysms and hemorrhages. To overcome these challenges, we propose a neural network-based method for vessel segmentation.
Methods:
A deep supervised fully convolutional network is developed by leveraging multi-level hierarchical features of the deep networks. To improve the discriminative capability of features in lower layers of the deep network and guide the gradient back propagation to overcome gradient vanishing, deep supervision with auxiliary classifiers is incorporated in some intermediate layers of the network. Moreover, the transferred knowledge learned from other domains is used to alleviate the issue of insufficient medical training data. The proposed approach does not rely on hand-crafted features and needs no problem-specific preprocessing or postprocessing, which reduces the impact of subjective factors.
Results:
We evaluate the proposed method on three publicly available databases, the DRIVE, STARE, and CHASE_DB1 databases. Extensive experiments demonstrate that our approach achieves better or comparable performance to state-of-the-art methods with a much faster processing speed, making it suitable for real-world clinical applications. The results of cross-training experiments demonstrate its robustness with respect to the training set.
Conclusions:
The proposed approach segments retinal vessels accurately with a much faster processing speed and can be easily applied to other biomedical segmentation tasks.
Automated detection of blood vessel structures is becoming a crucial interest for better management of vascular disease. In this paper, we propose an algorithm for vessel segmentation in digital retinal images based on integral channel features and random forests. In the first stage, preprocessing is performed to obtain the candidate pixels of vessels, then a host of simple features are extracted for each candidate pixels based on integral channels. Furthermore random forests is used to classify the candidate pixels as vessels or not. Finally, postprocessing is applied to fill pixel gaps in classified blood vessels. The proposed algorithm achieves an average accuracy of 0.9614, 0.9588, sensitivity of 0.7191, 0.6996 and specificity of 0.9849, 0.9787 on two public databases DRIVE and STARE respectively.
Retinal blood vessel structure is an important indicator of many retinal and systemic diseases, which has motivated the development of various image segmentation methods for the blood vessels. In this study, two supervised and three unsupervised segmentation methods with a publicly available implementation are reviewed and quantitatively compared with each other on five public databases with ground truth segmentation of the vessels.Each method is tested under consistent conditions with two types of preprocessing, and the parameters of the methods are optimized for each database. Additionally, possibility to predict the parameters of the methods by the linear regression model is tested for each database. Resolution of the input images and amount of the vessel pixels in the ground truth are used as predictors.The results show the positive influence of preprocessing on the performance of the unsupervised methods. The methods show similar performance for segmentation accuracy, with the best performance achieved by the method by Azzopardi et al. (Acc 94.0) on ARIADB, the method by Soares et al. (Acc 94.6, 94.7) on CHASEDB1 and DRIVE, and the method by Nguyen et al. (Acc 95.8, 95.5) on HRF and STARE. The method by Soares et al. performed better with regard to the area under the ROC curve. Qualitative differences between the methods are discussed. Finally, it was possible to predict the parameter settings that give performance close to the optimized performance of each method.
There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at
http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net
.
The condition of the vascular network of human eye is an important diagnostic factor in ophthalmology. Its segmentation in fundus imaging is a nontrivial task due to variable size of vessels, relatively low contrast, and potential presence of pathologies like microaneurysms and hemorrhages. Many algorithms, both unsupervised and supervised, have been proposed for this purpose in the past. We propose a supervised segmentation technique that uses a deep neural network trained on a large (up to 400; 000) sample of examples preprocessed with global contrast normalization, zero-phase whitening, and augmented using geometric transformations and gamma corrections. Several variants of the method are considered, including structured prediction, where a network classifies multiple pixels simultaneously. When applied to standard benchmarks of fundus imaging, the DRIVE, STARE, and CHASE databases, the networks significantly outperform the previous algorithms on the area under ROC curve measure (up to > 0:99) and accuracy of classification (up to > 0:97). The method is also resistant to the phenomenon of central vessel reflex, sensitive in detection of fine vessels (sensitivity > 0:87), and fares well on pathological cases.
This paper presents a new supervised method for vessel segmentation in retinal images. This method remolds the task of segmentation as a problem of cross-modality data transformation from retinal image to vessel map. A wide and deep neural network with strong induction ability is proposed to model the transformation, and an efficient training strategy is presented. Instead of a single label of the center pixel, the network can output the label map of all pixels for a given image patch. Our approach outperforms reported state-of-the-art methods in terms of sensitivity, specificity and accuracy. The result of cross-training evaluation indicates its robustness to the training set. The approach needs no artificially designed feature and no preprocessing step, reducing the impact of subjective factors. The proposed method has the potential for application in image diagnosis of ophthalmologic diseases, and it may provide a new, general, high-performance computing framework for image segmentation.
Retinal vessel segmentation algorithms are a fundamental component of automatic retinal disease screening systems. This work examines the blood vessel segmentation methodologies in two dimensional retinal images acquired from a fundus camera and a survey of techniques is presented. The aim of this paper is to review, analyze and categorize the retinal vessel extraction algorithms, techniques and methodologies, giving a brief description, highlighting the key points and the performance measures. We intend to give the reader a framework for the existing research; to introduce the range of retinal vessel segmentation algorithms; to discuss the current trends and future directions and summarize the open problems. The performance of algorithms is compared and analyzed on two publicly available databases (DRIVE and STARE) of retinal images using a number of measures which include accuracy, true positive rate, false positive rate, sensitivity, specificity and area under receiver operating characteristic (ROC) curve.
Many important eye diseases as well as systemic diseases manifest themselves in the retina. While a number of other anatomical structures contribute to the process of vision, this review focuses on retinal imaging and image analysis. Following a brief overview of the most prevalent causes of blindness in the industrialized world that includes age-related macular degeneration, diabetic retinopathy, and glaucoma, the review is devoted to retinal imaging and image analysis methods and their clinical implications. Methods for 2-D fundus imaging and techniques for 3-D optical coherence tomography (OCT) imaging are reviewed. Special attention is given to quantitative techniques for analysis of fundus photographs with a focus on clinically relevant assessment of retinal vasculature, identification of retinal lesions, assessment of optic nerve head (ONH) shape, building retinal atlases, and to automated methods for population screening for retinal diseases. A separate section is devoted to 3-D analysis of OCT images, describing methods for segmentation and analysis of retinal layers, retinal vasculature, and 2-D/3-D detection of symptomatic exudate-associated derangements, as well as to OCT-based analysis of ONH morphology and shape. Throughout the paper, aspects of image acquisition, image analysis, and clinical relevance are treated together considering their mutually interlinked relationships.
Restricted Boltzmann machines were developed using binary stochastic hidden units. These can be generalized by replacing each binary unit by an infinite number of copies that all have the same weights but have progressively more negative biases. The learning and inference rules for these “Stepped Sigmoid Units ” are unchanged. They can be approximated efficiently by noisy, rectified linear units. Compared with binary units, these units learn features that are better for object recognition on the NORB dataset and face verification on the Labeled Faces in the Wild dataset. Unlike binary units, rectified linear units preserve information about relative intensities as information travels through multiple layers of feature detectors. 1.
This paper describes a methodology of verification of individuals based on a retinal biometric pattern. The pattern consists in feature points of the retinal vessel tree, namely bifurcations and crossovers. These landmarks are detected and characterised adding semantic information to the biometric pattern. The typical authentication process of a person once extracted the biometric pattern includes matching it with the stored pattern for the authorised user obtaining a similarity value between them. A matching algorithm and a deep analysis of similarity metrics performance is presented. The semantic information added for the feature points allows to reduce the computation load in the matching process as only points classified equally can be matched. The system is capable of establishing a safe confidence band in the similarity measure space between scores for patterns of the same individual and between different individuals.
Genetic algorithms are playing an increasingly important role in studies of complex adaptive systems, ranging from adaptive agents in economic theory to the use of machine learning techniques in the design of complex devices such as aircraft turbines and integrated circuits. Adaptation in Natural and Artificial Systems is the book that initiated this field of study, presenting the theoretical foundations and exploring applications.
In its most familiar form, adaptation is a biological process, whereby organisms evolve by rearranging genetic material to survive in environments confronting them. In this now classic work, Holland presents a mathematical model that allows for the nonlinearity of such complex interactions. He demonstrates the model's universality by applying it to economics, physiological psychology, game theory, and artificial intelligence and then outlines the way in which this approach modifies the traditional views of mathematical genetics.
Initially applying his concepts to simply defined artificial systems with limited numbers of parameters, Holland goes on to explore their use in the study of a wide range of complex, naturally occuring processes, concentrating on systems having multiple factors that interact in nonlinear ways. Along the way he accounts for major effects of coadaptation and coevolution: the emergence of building blocks, or schemata, that are recombined and passed on to succeeding generations to provide, innovations and improvements.
Bradford Books imprint
Peer Reviewed http://deepblue.lib.umich.edu/bitstream/2027.42/46947/1/10994_2005_Article_422926.pdf
To examine the agreement of a novel computer program measuring retinal vessel tortuosity with subjective assessment of tortuosity in school-aged children.
Cross-sectional study of 387 retinal vessels (193 arterioles, 194 veins) from 28 eyes of 14 children (aged 10 years). Retinal digital images were analyzed using the Computer Assisted Image Analysis of the Retina (CAIAR) program, including 14 measures of tortuosity. Vessels were graded (from 0 = none; to 5 = tortuous) independently by two observers. Interobserver agreement was assessed by using kappa statistics. Agreement with all 14 objective measures was assessed with correlation/regression analyses. Intersession repeatability (comparing morning and afternoon sessions) of tortuosity indices was calculated.
Interobserver agreement of vessel tortuosity within one grade was high (kappa = 0.97), with total agreement in 56% of grades and 42% differing by +/-1 grade. Tortuosity indices based on subdivided chord length methods showed strong log-linear associations with agreed subjective grades (typically r > 0.6; P < 0.001). An approach that averages the distance from the vessel to chord length along the length of the vessel showed best agreement (r = 0.8; P < 0.0001). Tortuosity measures based on curvature performed less well. Intersession repeatability of the vessel to chord technique was good, differing by values equivalent to <1 in subjective grade.
Tortuosity indices based on changes in subdivided chord lengths showed optimal agreement with subjective assessment. The relation of these indices to ethnicity and cardiovascular risk factors in childhood should be examined further, as these indices may be a useful indicator of early vascular function.
To develop protocols to photograph and evaluate retinal vascular abnormalities in the Atherosclerosis Risk in Communities (ARIC) Study; to test reproducibility of the grading system; and to explore the relationship of these microvascular changes with blood pressure.
Population-based, cross-sectional study.
Among 4 examination centers, 11,114 participants (48-73 years of age) at their third triennial examination, after excluding persons with diabetes from this analysis.
One eye of each participant was photographed by technicians with nonmydriatic fundus cameras. Reading center graders evaluated focal arteriolar narrowing, arteriovenous (AV) nicking, and retinopathy by examining slides on a light box and measured diameters of all vessels in a zone surrounding the optic disc on enhanced digitized images. To gauge generalized narrowing, vessel diameters were combined into central arteriolar and venular equivalents with formulas adjusting for branching, and the ratio of equivalents (A/V ratio) was calculated.
Retinal vascular abnormalities, mean arteriolar blood pressure (MABP).
Among 11,114 participants, photographs were obtained of 99%, with quality sufficient to perform retinal evaluations in 81%. In the 9040 subjects with usable photographs, A/V ratio (lower values indicate generalized arteriolar narrowing) ranged from 0.57 to 1.22 (median = 0.84, interquartile range = 0.10), focal arteriolar narrowing was found in 7%, AV nicking in 6%, and retinopathy in 4%. Because of attrition of subjects and limitation of methods, prevalence of abnormality was likely underestimated. Controlling for gender, race, age, and smoking status, these retinal changes were associated with higher blood pressure. For every 10-mmHg increase in MABP, A/V ratio decreased by 0.02 unit (P < 0.0001), focal arteriolar narrowing had an odds ratio (OR) of 2.00 (95% confidence interval [CI] = 1.87-2.14), AV nicking had an OR of 1.25 (95% CI = 1.16-1.34), and retinopathy had an OR of 1.25 (95% CI = 1.15-1.37). For any degree of generalized narrowing, individuals with focal narrowing had MABP approximately 8 mmHg higher than those without (P < 0.0001). Masked replicate assessment of a sample found the following reproducibility: for A/V ratio, correlation coefficient = 0.79 and median absolute difference = 0.03; for focal arteriolar narrowing, kappa = 0.45; for AV nicking, kappa = 0.61; and for retinopathy, kappa = 0.89.
Protocols have been developed for nonmydriatic fundus photography and for evaluation of retinal vascular abnormalities. Several microvascular changes were significantly associated with higher blood pressure; follow-up will show whether these are predictive of later cerebrovascular or cardiovascular disease independently of other known risk factors.
Crossover plays an important role in GA based search. There have
been many empirical comparisons of different crossover operators in the
literature. However, analytical results are limited. No theory has
explained the behaviours of different crossover operators
satisfactorily. The paper analyses crossover from quite a different
point of view from the classical schema theorem. It explains the
behaviours of different crossover operators through the investigation of
crossover's search neighbourhood and search step size. It is shown that
given the binary chromosome encoding scheme, GAs with a large search
step size are better than GAs with a small step size for most problems.
Since uniform crossover's search step size is larger than that of either
one point or two point crossover, uniform crossover is expected to
perform better than the other two. Similarly, two point crossover is
expected to perform better than one point crossover due to its larger
search step size. It is also shown that increasing the number of
crossover points will increase crossover's search step size. The
analytical results are supported by the experimental studies on 12
benchmark function optimisation problems
Tournament selection is a useful and robust selection mechanism commonly used by genetic algorithms. The selection pressure of tournament selection directly varies with the tournament size --- the more competitors, the higher the resulting selection pressure. This article develops a model, based on order statistics, that can be used to quantitatively predict the resulting selection pressure of a tournament of a given size. This model is used to predict the convergence rates of genetic algorithms utilizing tournament selection. While tournament selection is often used in conjunction with noisy (imperfect) fitness functions, little is understood about how the noise affects the resulting selection pressure. The model is extended to quantitatively predict the selection pressure for tournament selection utilizing noisy fitness functions. Given the tournament size and noise level of a noisy fitness function, the extended model is used to predict the resulting selection pressure of t...