Evolving Deep Learning Convolutional Neural Networks for Early COVID-19 Detection in Chest X-ray Images

Abstract

This article proposes a framework that automatically designs classifiers for the early detection of COVID-19 from chest X-ray images. To do this, our approach repeatedly makes use of a heuristic for optimisation to efficiently find the best combination of the hyperparameters of a convolutional deep learning model. The framework starts with optimising a basic convolutional neural network which represents the starting point for the evolution process. Subsequently, at most two additional convolutional layers are added, at a time, to the previous convolutional structure as a result of a further optimisation phase. Each performed phase maximises the the accuracy of the system, thus requiring training and assessment of the new model, which gets gradually deeper, with relevant COVID-19 chest X-ray images. This iterative process ends when no improvement, in terms of accuracy, is recorded. Hence, the proposed method evolves the most performing network with the minimum number of convolutional layers. In this light, we simultaneously achieve high accuracy while minimising the presence of redundant layers to guarantee a fast but reliable model. Our results show that the proposed implementation of such a framework achieves accuracy up to 99.11%, thus being particularly suitable for the early detection of COVID-19.

Artificial Intelligence-Enabled Electrocardiography Predicts Left Ventricular Dysfunction and Future Cardiovascular Outcomes: A Retrospective Analysis

“…Third, novel optimization techniques proposed recently were not applied in this study, such as the Whale Optimizer or chimp optimization algorithm. These optimizers could provide better performance and increase the reliability of the network while maintaining its capability [ 15 , 16 , 60 ]. Finally, the ECG-ECHO pairs were not simultaneously acquired.…”

“…In cardiology, Shah et al used a machine learning model to recognize cardiac arrest risk and survival probability [ 14 ]. During the COVID-19 pandemic, the real-time diagnosis of COVID-19 was important and was assisted with DLM-based chest X-ray images with an accuracy of 99% [ 15 , 16 , 17 , 18 ].…”

Evolving deep convolutional neural networks by IP-based marine predator algorithm for COVID-19 diagnosis using chest CT scans

“…All experiments were carried out in MATLAB R2019b on a computer provided with an Intel Core i7-7700HQ processor running at a maximum frequency of 3.8 GHz, Windows 10, and 16 GB RAM. This paper compares IPMPA’s performance to that of IPPSO (Wang et al, 2018 ), variable-length genetic algorithm (VLGA) (Qiongbing and Lixin, 2016 ), variable-length NSGA-II (VLNSGA-II) (Pal et al, 2021 ), variable-length brain storm optimization algorithm (VLBSO) (Cheng et al, 2021 ), IP-Modified PSO (IPMPSO) (Abbas, 2018 ), variable-length biogeography-based optimizer (VLBBO) (Khishe et al, 2021 ), and variable-length ant colony optimization (VLACO) (Liao et al, 2013 ) on the two datasets used in the study. The MPA and other benchmark models’ parameters are summarized in Table 6 .…”

“…However, when the disease is in its early stages, a visual diagnosis of this condition is accompanied by some doubt. COVID-19 lung damage has recently been identified using X-rays and CT scans (Khishe et al, 2021 ). However, the diagnostic accuracy is dependent on the expert’s assessment (Zhang et al, 2022 ).…”

Remaining Useful Life Estimation of Aircraft Engines Using Differentiable Architecture Search

“…Thus, great achievements have been realized in signal processing [6] and motion captures [7]. Several advanced networks have been proposed based on CNN and RNN (for instance, deep convolution neural networks (DCNN) [8,9], echo state networks [10], and long short-term memory networks (LSTM) [11]).…”