ABSTRACT
This article aims to improve and diversify signal processing techniques to execute a brain-computer interface (BCI) based on neurological phenomena observed when performing motor tasks using motor imagery (MI). The noise present in the original data, such as intermodulation noise, crosstalk, and other unwanted noise, is removed by Modify Least Mean Square (M-LMS) in the pre-processing stage. Traditional LMSs were unable to extract all the noise from the images. After pre-processing, the required features, such as statistical features, entropy features, etc., were extracted using Common Spatial Pattern (CSP) and Pearson's Correlation Coefficient (PCC) instead of the traditional single feature extraction model. The arithmetic optimization algorithm cannot select the features accurately and fails to reduce the feature dimensionality of the data. Thus, an Extended Arithmetic operation optimization (ExAo) algorithm is used to select the most significant attributes from the extracted features. The proposed model uses Double Attention Convolutional Neural Networks (DAttnConvNet) to classify the types of EEG signals based on optimal feature selection. Here, the attention mechanism is used to select and optimize the features to improve the classification accuracy and efficiency of the model. In EEG motor imagery datasets, the proposed model has been analyzed under class, which obtained an accuracy of 99.98% in class Baseline (B), 99.82% in class Imagined movement of a right fist (R) and 99.61% in class Imagined movement of both fists (RL). In the EEG dataset, the proposed model can obtain a high accuracy of 97.94% compared to EEG datasets of other models.
ABSTRACT
Recently, the Covid-19 pandemic has affected several lives of people globally, and there is a need for a massive number of screening tests to diagnose the existence of coronavirus. For the medical specialist, detecting COVID-19 cases is a difficult task. There is a need for fast, cheap and accurate diagnostic tools. The chest X-ray and the computerized tomography (CT) play a significant role in the COVID-19 diagnosis. The advancement of deep learning (DL) approaches helps to introduce a COVID diagnosis system to achieve maximum detection rate with minimum time complexity. This research proposed a discrete wavelet optimized network model for COVID-19 diagnosis and feature extraction to overcome these problems. It consists of three stages pre-processing, feature extraction and classification. The raw images are filtered in the pre-processing phase to eliminate unnecessary noises and improve the image quality using the MMG hybrid filtering technique. The next phase is feature extraction, in this stage, the features are extracted, and the dimensionality of the features is diminished with the aid of a modified discrete wavelet based Mobile Net model. The third stage is the classification here, the convolutional Aquila COVID detection network model is developed to classify normal and COVID-19 positive cases from the collected images of the COVID-CT and chest X-ray dataset. Finally, the performance of the proposed model is compared with some of the existing models in terms of accuracy, specificity, sensitivity, precision, f-score, negative predictive value (NPV) and positive predictive value (PPV), respectively. The proposed model achieves the performance of 99%, 100%, 98.5%, and 99.5% for the CT dataset, and the accomplished accuracy, specificity, sensitivity, and precision values of the proposed model for the X-ray dataset are 98%, 99%, 98% and 97% respectively. In addition, the statistical and cross validation analysis is conducted to validate the effectiveness of the proposed model.
