Detection and classification of lung diseases for pneumonia and Covid-19 using machine and deep learning techniques

Abstract

Since the arrival of the novel Covid-19, several types of researches have been initiated for its accurate prediction across the world. The earlier lung disease pneumonia is closely related to Covid-19, as several patients died due to high chest congestion (pneumonic condition). It is challenging to differentiate Covid-19 and pneumonia lung diseases for medical experts. The chest X-ray imaging is the most reliable method for lung disease prediction. In this paper, we propose a novel framework for the lung disease predictions like pneumonia and Covid-19 from the chest X-ray images of patients. The framework consists of dataset acquisition, image quality enhancement, adaptive and accurate region of interest (ROI) estimation, features extraction, and disease anticipation. In dataset acquisition, we have used two publically available chest X-ray image datasets. As the image quality degraded while taking X-ray, we have applied the image quality enhancement using median filtering followed by histogram equalization. For accurate ROI extraction of chest regions, we have designed a modified region growing technique that consists of dynamic region selection based on pixel intensity values and morphological operations. For accurate detection of diseases, robust set of features plays a vital role. We have extracted visual, shape, texture, and intensity features from each ROI image followed by normalization. For normalization, we formulated a robust technique to enhance the detection and classification results. Soft computing methods such as artificial neural network (ANN), support vector machine (SVM), K-nearest neighbour (KNN), ensemble classifier, and deep learning classifier are used for classification. For accurate detection of lung disease, deep learning architecture has been proposed using recurrent neural network (RNN) with long short-term memory (LSTM). Experimental results show the robustness and efficiency of the proposed model in comparison to the existing state-of-the-art methods.

A Deep Learning and Handcrafted Based Computationally Intelligent Technique for Effective COVID-19 Detection from X-ray/CT-scan Imaging

“…The three best models were selected based on their performance for the heterogeneous stacked ensemble. Goyal and Singh [41] used machine learning and deep learning technique for the early detection of COVID-19. They first used a median filter followed by the histogram equalization to enhance the image quality.…”

CovidConvLSTM: A fuzzy ensemble model for COVID-19 detection from chest X-rays

“…These methods were able to perform well with the use of the deep features. For example, Goyal and Singh (2021) designed fusion and normalization features based recurrent neural network (RNN)-long short-term memory (LSTM) (abbreviated as F-RNN-LSTM) that relied on only handcrafted features comprised of Gray-level Co-occurrence Matrix (GLCM), histogram of oriented gradients (HOG), intensity and geometric features. However, this model was empowered with a deep learner known as RNN-LSTM model.…”

“…In this context, it is to be noted that Rahman, et al (2021) studied, in detail, the effect of various image enhancement techniques on COVID-19 detection using some standard CNN models. Also, Goyal and Singh (2021) used histogram equalization to enhance the quality before applying an adaptive image segmentation method to extract the region of interest, which was then used for classification purposes.…”

Non-smooth Bayesian learning for artificial neural networks

“…During the last two decades, Deep Learning (DL) architectures (Devunooru et al. 2021 ; Goyal and Singh 2021 ) have demonstrated their ability to deal with more voluminous and complex data. Moreover, it has gradually become the most widely used computational approach in the field of ML, achieving outstanding results on several cognitive tasks, matching or even beating those reached by human performance.…”

“…ML research achieved outstanding results on several complex cognitive tasks, including Computer Vision Alsarhan et al (2021), Medical diagnoses Chaabene et al (2021); Sree et al (2021), Signal Processing Jaini et al (2021), etc. During the last two decades, Deep Learning (DL) architectures Devunooru et al (2021); Goyal and Singh (2021) have demonstrated their ability to deal with more voluminous and complex data. Moreover, it has gradually become the most widely used computational approach in the field of ML, achieving outstanding results on several cognitive tasks, matching or even beating those reached by human performance.…”