IKONOS: an intelligent tool to support diagnosis of COVID-19 by texture analysis of X-ray images
- PDF / 1,564,574 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 9 Downloads / 132 Views
ORIGINAL ARTICLE
IKONOS: an intelligent tool to support diagnosis of COVID-19 by texture analysis of X-ray images Juliana C. Gomes 1 & Valter A. de F. Barbosa 2 & Maíra A. Santana 1 & Jonathan Bandeira 1 & Mêuser Jorge Silva Valença 1 & Ricardo Emmanuel de Souza 2 & Aras Masood Ismael 3 & Wellington P. dos Santos 2 Received: 4 May 2020 / Accepted: 27 August 2020 # Sociedade Brasileira de Engenharia Biomedica 2020
Abstract Purpose In late 2019, the SARS-CoV-2 virus spread worldwide. The virus has high rates of proliferation and causes severe respiratory symptoms, such as pneumonia. The standard diagnostic method for pneumonia is chest X-ray image. There are many advantages to using COVID-19 diagnostic X-rays: low cost, fast, and widely available. Methods We propose an intelligent system to support diagnosis by X-ray images. We tested Haralick and Zernike moments for feature extraction. Experiments with classic classifiers were done. Results Support vector machines stood out, reaching an average accuracy of 89.78%, average sensitivity of 0.8979, and average precision and specificity of 0.8985 and 0.9963, respectively. Conclusion Using features based on textures and shapes combined with classical classifiers, the developed system was able to differentiate COVID-19 from viral and bacterial pneumonia with low computational cost. Keywords COVID-19 . SARS-Cov-2 . X-ray . Diagnosis support . Artificial intelligence . Machine learning
Introduction A new kind of coronavirus crossed the species barrier in December 2019 in China. This virus is related to severe acute respiratory syndrome coronavirus; for this reason, it received the name SARS-Cov-2 (Lin et al. 2020; Okba et al. 2020). SARS-Cov-2 causes the coronavirus disease 19 (COVID-19) that affects the respiratory system causing several health complications as fever, cough, sore throat, and in the most severe cases, it can lead to shortness of breath and death.
* Wellington P. dos Santos [email protected] Juliana C. Gomes [email protected] Aras Masood Ismael [email protected] 1
Polytechnique School of the University of Pernambuco, Recife, Brazil
2
Federal University of Pernambuco, Recife, Brazil
3
Information Technology Department, Technical College of Informatics, Sulaimani Polytechnic University, Sulaymaniyah, Iraq
Until the end of April 2020, the SARS-Cov-2 has spread from 213 countries infecting almost 3 million people worldwide and causing more than 200 thousand deaths (WHO 2020). Thus, there is an urgent need of methods to diagnose the disease in a quick and precise way. Due to the high virus spread rate, tests for diagnosis that are quick and precise are necessary (Beeching et al. 2020). The precise diagnosis in patients with COVID-19 permits they receive medical attention; furthermore, placing these patients in isolation will decrease the disease spread (Beeching et al. 2020). The test standard to the COVID-19 diagnosis is the quantitative polymerase chain reaction (qPCR). However, this exam needs several hours to confirm positivity (Döhla et
Data Loading...
