Diagnosis of COVID-19 by Wavelet Renyi Entropy and Three-Segment Biogeography-Based Optimization

Abstract

Corona virus disease 2019 (COVID-19) is an acute infectious pneumonia and its pathogen is novel and was not previously found in humans. As a diagnostic method for COVID-19, chest computed tomography (CT) is more sensitive than reverse transcription polymerase chain reaction. However, the interpretation of COVID-19 based on chest CT is mainly done manually by radiologists and takes about 5 to 15 minutes for one patient. To shorten the time of interpreting the CT image and improve the reliability of identification of COVID-19. In this paper, a novel chest CT-based method for the automatic detection of COVID-19 was proposed. Our algorithm is a hybrid method composed of (i) wavelet Renyi entropy, (ii) feedforward neural network, and (iii) a proposed three-segment biogeography-based optimization (3SBBO) algorithm. The wavelet Renyi entropy is used to extract the image features. The novel optimization method of 3SBBO can optimize weights, biases of the network, and Renyi entropy order. Finally, we used 296 chest CT images to evaluate the detection performance of our proposed method. In order to reduce randomness and get unbiased result, the 10 runs of 10-fold cross validation are introduced. Experimental outcomes show that our proposed method is superior to state-of-the-art approaches in terms of sensitivity, specificity, precision, accuracy, and F1.

Keywords

• Wavelet Renyi entropy
• three-segment biogeography-based optimization
• feedforward neural network
• COVID-19
• diagnosis