Artificial intelligence assisted Chest Radiograph assessment as screening tool for COVID-19

Subashree Chandrasekaran; Subathra Devi A; Sinddhu M

Authors

Dr. Subhashree Chandrasekaran Professor, Department of Radiodiagnosis, Government Medical College and ESIC Hospital, Coimbatore, India.
Dr. Subathra Devi A Department of Radiology, Government Medical College and ESIC Hospital, Coimbatore, India.
Dr. Sinddhu M Department of Radiology, Government Medical College and ESIC Hospital, Coimbatore, India.

Keywords:

Artificial Intelligence based software, Chest Radiograph, Radiodiagnosis, COVID-19, CHOCO, Chest X-ray, Digital Radiography

Abstract

Aim and Objective: To study whether trained convolutional networks in artificial intelligence can be useful in assessing Chest radiographs of COVID -19 patients, to detect abnormalities and classify them as ‘Normal’, ‘Likely COVID-19’ and ‘Other abnormalities’ and detecting the Region of Interest (ROI) in Chest radiograph. Materials and Methods: The study was done on RTPCR positive Covid-19 patients. The study period being March to May 2020 in ICU and July 2020 to Feb 2021 ,in Department of Radiodiagnosis in Govt Medical College and ESI hospital, Coimbatore, India. 611 Chest radiographs taken during this period were initially reported by 2 senior radiologists. Then the same radiographs were processed into a prior trained and tested Artificial Intelligence software named CHOCO. The software detected and classified the radiographs into 3 sets –Normal, Likely COVID-19 and Other abnormalities respectively. The performance of AI software was compared with diagnostic performance of expert radiologists. Statistical Analysis and Results: Performance of the AI software was found to have 94.27% sensitivity, 93.79% specificity and 93.94% accuracy for detecting and classifying the abnormality. Detecting the Region of interest (ROI) in Chest radiograph was also done by the software with sensitivity of 92%, specificity of 86% and accuracy of 87.1%. Conclusion: The study shows that Artificial Intelligence software performs reasonably well with CXR image and can be used for rapid screening for a huge population.

References

World Health Organization. Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus2019.

Junhui Zhai, Thomas Briese, et al. Real-Time Polymerase Chain Reaction for Detecting SARS Coronavirus, Beijing, 2003, DOI: 10.3201/eid1002.030799 PMCID: PMC3322935 PMID: 15030701.

Shannon L. Emery, Dean D. Erdman, et al. Real-Time Polymerase Chain Reaction Assay for SARS-associated Coronavirus, DOI: 10.3201/eid1002.030759 PCID: PMC3322901 PMID:15030703.

Borkowski AA, Wilson CP, Borkowski SA, Deland LA,Mastorides SM. Using Apple machine learning algorithms to detect and subclassify non-small cell lung cancer. http://arxiv.org/abs/1808.08230. Updated January 15, 2019.

Borkowski AA, Wilson CP, Borkowski SA, et al. Comparing artificial intelligence platforms for histopathologic cancer diagnosis. Fed Pract. 2019;36(10):456-463.

Dr.Vimal Raj, Role of Chest Radiograph (CXR) in COVID-19 Diagnosis and Management, Journal of the Indian Medical Association, Vol 118, No 05, May 2020.

Wong HYF, Lam HYS, Fong AH, et al. Frequency and distribution of chest radiographic findings in patients positive for COVID-19. Radiology. 2020;296(2): E72-E78. doi:10.1148/radiol.2020201160

Jacobi A, Chung M, Bernheim A, Eber C. Portable chest X-ray in coronavirus disease-19 (COVID-19): a pictorial review. Clin Imaging. 2020; 64:35-42. doi: 10.1016/j.clinimag.2020.04.001

Qin ZZ, Sander MS, Rai B — Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems. Sci Rep 2019;9(1):15000. doi: 10.1038/s41598-019-51503-3.

Puxuan Lu, Boping Zhou, et al. Chest X-ray Imaging of Patients with SARS, 2003 July PMID:12890364 Chinese medical journal.

Xuanyang Xie, Xi Li, Shouhong Wan, Yuchang Gong, Mining X-Ray Images of SARS Patients, Department of Computer Science and Technology, University of Science and Technology of China, DOI: 10.1007/11677437_22.

Pranav Rajpurkar, Jeremy Irvin, et al. CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning, 14 Nov 2017, Computer Vision and Pattern Recognition arXiv:1711.05225v3[cs.CV].

Borkowski AA, Viswanadhan NA, Thomas LB, Guzman RD, Deland LA, Mastorides SM. Using Artificial Intelligence for COVID-19 Chest X-ray Diagnosis. Fed Pract. 2020;37(9):398-404. doi: 10.12788/fp.0045. PMID: 33029064; PMCID: PMC7535959.

Liu X, Faes L, Kale AU, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digit Heal. 2019;1(6): E271-E297. doi:10.1016/S2589-7500(19)30123-2

Bai HX, Wang R, Xiong Z, et al. Artificial intelligence augmentation of radiologist performance in distinguishing COVID-19 from pneumonia of other origin at chest CT. Radiology. 2020;296(3): E156-E165. doi: 10.1148/radiol.2020201491

Kermany DS, Goldbaum M, Cai W, et al. Identifying medical diagnoses and treatable diseases by image based deep learning. Cell. 2018;172(5):1122-1131.e9. doi: 10.1016/j.cell.2018.02.010