-
Automatic Diagnosis of COVID-19 from CT Images using CycleGAN and Transfer Learning
Authors:
Navid Ghassemi,
Afshin Shoeibi,
Marjane Khodatars,
Jonathan Heras,
Alireza Rahimi,
Assef Zare,
Ram Bilas Pachori,
J. Manuel Gorriz
Abstract:
The outbreak of the corona virus disease (COVID-19) has changed the lives of most people on Earth. Given the high prevalence of this disease, its correct diagnosis in order to quarantine patients is of the utmost importance in steps of fighting this pandemic. Among the various modalities used for diagnosis, medical imaging, especially computed tomography (CT) imaging, has been the focus of many pr…
▽ More
The outbreak of the corona virus disease (COVID-19) has changed the lives of most people on Earth. Given the high prevalence of this disease, its correct diagnosis in order to quarantine patients is of the utmost importance in steps of fighting this pandemic. Among the various modalities used for diagnosis, medical imaging, especially computed tomography (CT) imaging, has been the focus of many previous studies due to its accuracy and availability. In addition, automation of diagnostic methods can be of great help to physicians. In this paper, a method based on pre-trained deep neural networks is presented, which, by taking advantage of a cyclic generative adversarial net (CycleGAN) model for data augmentation, has reached state-of-the-art performance for the task at hand, i.e., 99.60% accuracy. Also, in order to evaluate the method, a dataset containing 3163 images from 189 patients has been collected and labeled by physicians. Unlike prior datasets, normal data have been collected from people suspected of having COVID-19 disease and not from data from other diseases, and this database is made available publicly.
△ Less
Submitted 24 April, 2021;
originally announced April 2021.
-
EEG Signal Classification using Variational Mode Decomposition
Authors:
Akshith Ullal,
Ram Bilas Pachori
Abstract:
Epilepsy affects about 1% of the population every year, and is characterized by abnormal and sudden hyper-synchronous excitation of the neurons in the brain. The electroencephalogram(EEG) is the most widely used method to record brain signals and diagnose epilepsy and seizure cases. In this paper we use the method of Variational Mode Decomposition (VMD) in our analysis to classify seizure/seizure…
▽ More
Epilepsy affects about 1% of the population every year, and is characterized by abnormal and sudden hyper-synchronous excitation of the neurons in the brain. The electroencephalogram(EEG) is the most widely used method to record brain signals and diagnose epilepsy and seizure cases. In this paper we use the method of Variational Mode Decomposition (VMD) in our analysis to classify seizure/seizure free signals. This technique uses variational non recursive mode decomposition, in comparison to other methods like Empirical Mode (EMD) and Hilbert-Huang transform which recursively decompose the signals, making them more susceptible to noise and sampling rate. VMD decomposes a signal into its components which are called principal modes. In our analysis, 4 features of the decomposed signals namely Renyi Entropy, second order difference plot (SODP), fourth order difference plot(FODP) and average amplitude are investigated, both individually and using a ranking methodology considering all 4 features at the same time. The SODP of decomposed signal modes is an elliptical structure. The 95% confidence ellipse area measured from the SODP of the decomposed signal modes has been used as a feature in order to discriminate seizure-free EEG signals from the epileptic seizure EEG signal. For the classification, a Multilayer Perceptron(MLP) with back propagation algorithm as the training method was used. A high percentage of accuracy was obtained when the features were used individually for classification and an even higher degree of accuracy was obtained when the ranking methodology was used.
△ Less
Submitted 27 March, 2020;
originally announced March 2020.
-
CNN-based approach for glaucoma diagnosis using transfer learning and LBP-based data augmentation
Authors:
Shishir Maheshwari,
Vivek Kanhangad,
Ram Bilas Pachori
Abstract:
Glaucoma causes an irreversible damage to retinal nerve fibers which results in vision loss, if undetected in early stage. Therefore, diagnosis of glaucoma in its early stage may prevent further vision loss. In this paper, we propose a convolutional neural network (CNN) based approach for automated glaucoma diagnosis by employing retinal fundus images. This approach employs transfer learning techn…
▽ More
Glaucoma causes an irreversible damage to retinal nerve fibers which results in vision loss, if undetected in early stage. Therefore, diagnosis of glaucoma in its early stage may prevent further vision loss. In this paper, we propose a convolutional neural network (CNN) based approach for automated glaucoma diagnosis by employing retinal fundus images. This approach employs transfer learning technique and local binary pattern (LBP) based data augmentation. In the proposed approach, we employ Alexnet as a pre-trained CNN model which is used for transfer learning. Initially, the proposed approach divides the fundus image dataset into training and testing data. Further, the color fundus images in training and testing data are separated into red (R), green (G), and blue (B) channels. Additionally, the LBP-based data augmentation is performed on training data. Specifically, we compute LPBs for each of the channel. Finally, the augmented training data is used to train the CNN model via transfer learning. In testing stage, the R, G, and B channels of test image are fed to the trained CNN model which generates 3 decisions. We employ a decision level fusion technique to combine the decisions obtained from the trained CNN model. The experimental evaluation of the proposed approach on the public RIM-ONE fundus image database, achieves state-of-the-art performance for glaucoma diagnosis.
△ Less
Submitted 19 February, 2020;
originally announced February 2020.
