arXiv subjects

Hojjat Salehinejad

Publications and source records attributed to Hojjat Salehinejad.

36 records · Page 2Linked to original sources

arXiv2021

Pruning of Convolutional Neural Networks Using Ising Energy Model

Pruning is one of the major methods to compress deep neural networks. In this paper, we propose an Ising energy model within an optimization framework for pruning convolutional kernels and hidden units. This model is designed to reduce redundancy between weight kernels and detect inactive kernels/hidden units. Our experiments using ResNets, AlexNet, and SqueezeNet on CIFAR-10 and CIFAR-100 datasets show that the proposed method on average can achieve a pruning rate of more than $50\%$ of the trainable parameters with approximately $<10\%$ and $<5\%$ drop of Top-1 and Top-5 classification accuracy, respectively.

cs.NE↗

arXiv2021

A Real-World Demonstration of Machine Learning Generalizability: Intracranial Hemorrhage Detection on Head CT

Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5,965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications.

eess.IV↗

arXiv2021

Deep Sequential Learning for Cervical Spine Fracture Detection on Computed Tomography Imaging

Fractures of the cervical spine are a medical emergency and may lead to permanent paralysis and even death. Accurate diagnosis in patients with suspected fractures by computed tomography (CT) is critical to patient management. In this paper, we propose a deep convolutional neural network (DCNN) with a bidirectional long-short term memory (BLSTM) layer for the automated detection of cervical spine fractures in CT axial images. We used an annotated dataset of 3,666 CT scans (729 positive and 2,937 negative cases) to train and validate the model. The validation results show a classification accuracy of 70.92% and 79.18% on the balanced (104 positive and 104 negative cases) and imbalanced (104 positive and 419 negative cases) test datasets, respectively.

eess.IV↗

arXiv2019

Survey of Dropout Methods for Deep Neural Networks

Dropout methods are a family of stochastic techniques used in neural network training or inference that have generated significant research interest and are widely used in practice. They have been successfully applied in neural network regularization, model compression, and in measuring the uncertainty of neural network outputs. While original formulated for dense neural network layers, recent advances have made dropout methods also applicable to convolutional and recurrent neural network layers. This paper summarizes the history of dropout methods, their various applications, and current areas of research interest. Important proposed methods are described in additional detail.

cs.NE↗

arXiv2019

Ising-Dropout: A Regularization Method for Training and Compression of Deep Neural Networks

Overfitting is a major problem in training machine learning models, specifically deep neural networks. This problem may be caused by imbalanced datasets and initialization of the model parameters, which conforms the model too closely to the training data and negatively affects the generalization performance of the model for unseen data. The original dropout is a regularization technique to drop hidden units randomly during training. In this paper, we propose an adaptive technique to wisely drop the visible and hidden units in a deep neural network using Ising energy of the network. The preliminary results show that the proposed approach can keep the classification performance competitive to the original network while eliminating optimization of unnecessary network parameters in each training cycle. The dropout state of units can also be applied to the trained (inference) model. This technique could compress the network in terms of number of parameters up to 41.18% and 55.86% for the classification task on the MNIST and Fashion-MNIST datasets, respectively.

cs.NE↗

arXiv2018

Cylindrical Transform: 3D Semantic Segmentation of Kidneys With Limited Annotated Images

In this paper, we propose a novel technique for sampling sequential images using a cylindrical transform in a cylindrical coordinate system for kidney semantic segmentation in abdominal computed tomography (CT). The images generated from a cylindrical transform augment a limited annotated set of images in three dimensions. This approach enables us to train contemporary classification deep convolutional neural networks (DCNNs) instead of fully convolutional networks (FCNs) for semantic segmentation. Typical semantic segmentation models segment a sequential set of images (e.g. CT or video) by segmenting each image independently. However, the proposed method not only considers the spatial dependency in the x-y plane, but also the spatial sequential dependency along the z-axis. The results show that classification DCNNs, trained on cylindrical transformed images, can achieve a higher segmentation performance value than FCNs using a limited number of annotated images.

cs.CV↗

arXiv2018

Recent Advances in Recurrent Neural Networks

Recurrent neural networks (RNNs) are capable of learning features and long term dependencies from sequential and time-series data. The RNNs have a stack of non-linear units where at least one connection between units forms a directed cycle. A well-trained RNN can model any dynamical system; however, training RNNs is mostly plagued by issues in learning long-term dependencies. In this paper, we present a survey on RNNs and several new advances for newcomers and professionals in the field. The fundamentals and recent advances are explained and the research challenges are introduced.

cs.NE↗

arXiv2018

Image Augmentation using Radial Transform for Training Deep Neural Networks

Deep learning models have a large number of free parameters that must be estimated by efficient training of the models on a large number of training data samples to increase their generalization performance. In real-world applications, the data available to train these networks is often limited or imbalanced. We propose a sampling method based on the radial transform in a polar coordinate system for image augmentation to facilitate the training of deep learning models from limited source data. This pixel-wise transform provides representations of the original image in the polar coordinate system by generating a new image from each pixel. This technique can generate radial transformed images up to the number of pixels in the original image to increase the diversity of poorly represented image classes. Our experiments show improved generalization performance in training deep convolutional neural networks with radial transformed images.

cs.CV↗

arXiv2018

Generalization of Deep Neural Networks for Chest Pathology Classification in X-Rays Using Generative Adversarial Networks

Medical datasets are often highly imbalanced with over-representation of common medical problems and a paucity of data from rare conditions. We propose simulation of pathology in images to overcome the above limitations. Using chest X-rays as a model medical image, we implement a generative adversarial network (GAN) to create artificial images based upon a modest sized labeled dataset. We employ a combination of real and artificial images to train a deep convolutional neural network (DCNN) to detect pathology across five classes of chest X-rays. Furthermore, we demonstrate that augmenting the original imbalanced dataset with GAN generated images improves performance of chest pathology classification using the proposed DCNN in comparison to the same DCNN trained with the original dataset alone. This improved performance is largely attributed to balancing of the dataset using GAN generated images, where image classes that are lacking in example images are preferentially augmented.

cs.CV↗

arXiv2017

A Convolutional Neural Network for Search Term Detection

Pathfinding in hospitals is challenging for patients, visitors, and even employees. Many people have experienced getting lost due to lack of clear guidance, large footprint of hospitals, and confusing array of hospital wings. In this paper, we propose Halo; An indoor navigation application based on voice-user interaction to help provide directions for users without assistance of a localization system. The main challenge is accurate detection of origin and destination search terms. A custom convolutional neural network (CNN) is proposed to detect origin and destination search terms from transcription of a submitted speech query. The CNN is trained based on a set of queries tailored specifically for hospital and clinic environments. Performance of the proposed model is studied and compared with Levenshtein distance-based word matching.

cs.IR↗

arXiv2017

Opposition based Ensemble Micro Differential Evolution

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

cs.NE↗

arXiv2017

Interpretation of Mammogram and Chest X-Ray Reports Using Deep Neural Networks - Preliminary Results

Radiology reports are an important means of communication between radiologists and other physicians. These reports express a radiologist's interpretation of a medical imaging examination and are critical in establishing a diagnosis and formulating a treatment plan. In this paper, we propose a Bi-directional convolutional neural network (Bi-CNN) model for the interpretation and classification of mammograms based on breast density and chest radiographic radiology reports based on the basis of chest pathology. The proposed approach helps to organize databases of radiology reports, retrieve them expeditiously, and evaluate the radiology report that could be used in an auditing system to decrease incorrect diagnoses. Our study revealed that the proposed Bi-CNN outperforms the random forest and the support vector machine methods.

cs.CV↗

arXiv2016

Diversity Enhancement for Micro-Differential Evolution

The differential evolution (DE) algorithm suffers from high computational time due to slow nature of evaluation. In contrast, micro-DE (MDE) algorithms employ a very small population size, which can converge faster to a reasonable solution. However, these algorithms are vulnerable to a premature convergence as well as to high risk of stagnation. In this paper, MDE algorithm with vectorized random mutation factor (MDEVM) is proposed, which utilizes the small size population benefit while empowers the exploration ability of mutation factor through randomizing it in the decision variable level. The idea is supported by analyzing mutation factor using Monte-Carlo based simulations. To facilitate the usage of MDE algorithms with very-small population sizes, new mutation schemes for population sizes less than four are also proposed. Furthermore, comprehensive comparative simulations and analysis on performance of the MDE algorithms over various mutation schemes, population sizes, problem types (i.e. uni-modal, multi-modal, and composite), problem dimensionalities, and mutation factor ranges are conducted by considering population diversity analysis for stagnation and trapping in local optimum situations. The studies are conducted on 28 benchmark functions provided for the IEEE CEC-2013 competition. Experimental results demonstrate high performance and convergence speed of the proposed MDEVM algorithm.

cs.NE↗

arXiv2016

Learning Over Long Time Lags

The advantage of recurrent neural networks (RNNs) in learning dependencies between time-series data has distinguished RNNs from other deep learning models. Recently, many advances are proposed in this emerging field. However, there is a lack of comprehensive review on memory models in RNNs in the literature. This paper provides a fundamental review on RNNs and long short term memory (LSTM) model. Then, provides a surveys of recent advances in different memory enhancements and learning techniques for capturing long term dependencies in RNNs.

cs.NE↗

arXiv2016

Deep Recurrent Neural Networks for Sequential Phenotype Prediction in Genomics

In analyzing of modern biological data, we are often dealing with ill-posed problems and missing data, mostly due to high dimensionality and multicollinearity of the dataset. In this paper, we have proposed a system based on matrix factorization (MF) and deep recurrent neural networks (DRNNs) for genotype imputation and phenotype sequences prediction. In order to model the long-term dependencies of phenotype data, the new Recurrent Linear Units (ReLU) learning strategy is utilized for the first time. The proposed model is implemented for parallel processing on central processing units (CPUs) and graphic processing units (GPUs). Performance of the proposed model is compared with other training algorithms for learning long-term dependencies as well as the sparse partial least square (SPLS) method on a set of genotype and phenotype data with 604 samples, 1980 single-nucleotide polymorphisms (SNPs), and two traits. The results demonstrate performance of the ReLU training algorithm in learning long-term dependencies in RNNs.

cs.NE↗

arXiv2015

Internet of Things for Residential Areas: Toward Personalized Energy Management Using Big Data

Intelligent management of machines, particularly in a residence area, has been of interest for many years. However, such system design has always been limited to simple control of machines from a local area or remotely from the Internet. In this report, for the first time, an intelligent system is proposed, where not only provides intelligent control ability of machines to user, but also utilizes big data and optimization techniques to provide promotional offers to the user to optimize energy consumption of machines. Since a high traffic communication is involved among the machines and the optimization-big data core of system, the communication core of the proposed system is designed based on cloud, where many challenging issues such as spectrum assignment and resource management are involved. To deal with that, the communication network in the home area network (HAN) is designed based on the cognitive radio system, where a new spectrum assignment method based on the ant colony optimization (ACO) algorithm is proposed to perform spectrum assignment to the machines in the HAN. Performance evaluation of the proposed spectrum assignment method shows its performance in fair spectrum assignment among machines.

cs.NI↗

arXiv2015

Toward Smart Power Grids: Communication Network Design for Power Grids Synchronization

In smart power grids, keeping the synchronicity of generators and the corresponding controls is of great importance. To do so, a simple model is employed in terms of swing equation to represent the interactions among dynamics of generators and feedback control. In case of having a communication network available, the control can be done based on the transmitted measurements by the communication network. The stability of system is denoted by the largest eigenvalue of the weighted sum of the Laplacian matrices of the communication infrastructure and power network. In this work, we use graph theory to model the communication network as a graph problem. Then, Ant Colony System (ACS) is employed for optimum design of above graph for synchronization of power grids. Performance evaluation of the proposed method for the 39-bus New England power system versus methods such as exhaustive search and Rayleigh quotient approximation indicates feasibility and effectiveness of our method for even large scale smart power grids.

cs.NE↗

arXiv2015

Combined A*-Ants Algorithm: A New Multi-Parameter Vehicle Navigation Scheme

In this paper a multi-parameter A*(A- star)-ants based algorithm is proposed in order to find the best optimized multi-parameter path between two desired points in regions. This algorithm recognizes paths, according to user desired parameters using electronic maps. The proposed algorithm is a combination of A* and ants algorithm in which the proposed A* algorithm is the prologue to the suggested ant based algorithm .In fact, this A* algorithm invigorates some paths pheromones in ants algorithm. As one of implementations of this method, this algorithm was applied on a part of Kerman city, Iran as a multi-parameter vehicle navigator. It finds the best optimized multi-parameter direction between two desired junctions based on city traveler parameters. Comparison results between the proposed method and ants algorithm demonstrates efficiency and lower cost function results of the proposed method versus ants algorithm.

cs.NE↗