Tuberculosis: can AI become a reliable method for diagnosis and screening?

Congress:

ECR 2020

Poster Number:

C-14393

Type:

Educational Exhibit

Keywords:

Artificial Intelligence and Machine Learning, Artificial Intelligence, Computer applications, Thorax, Conventional radiography, Teleradiology, Computer Applications-Detection, diagnosis, Computer Applications-General, Computer Applications-Teleradiology, Cavitation, Image verification, Outcomes, Retrospective, Diagnostic or prognostic study, Performed at one institution

Authors:

A.-E. Moldovan, C. Avramescu, B. Bercean, S. Iarca, A. Tenescu, F. Birsasteanu; Timisoara/RO

DOI:

10.26044/ecr2020/C-14393

DOI-Link:

https://dx.doi.org/10.26044/ecr2020/C-14393

Learning objectives

The purpose of this work is to: emphasize the importance of chest X-ray imaging in establishing the diagnosis of tuberculosis (TB) develop and prove the reliability of an artificial intelligence (AI) software used for disease screening

Background

Tuberculosis is a multisystemic disease, being among the leading causes of infectious disease–related mortality worldwide [1]. During the twentieth century, the overall incidence and mortality rate for tuberculosis declined considerably over the years due to new treatment [2]. Unfortunately, despite advances in diagnosis and treatment, tuberculosis infection began spreading after the 1980s in developed countries, partly because of the emergence of HIV, and still has a high incidence in developing countries, which makes it necessary to improve screening methods for detection, optimal prevention and eradication...

Findings and procedure details

Eastern Europe is one of the worldwide regions most affected by tuberculosis, with Romania being one of the countries where the disease is widely spread [5]. The “Marius Nasta” Pneumology Institute initiated the “Organizing Programmes for Early Screening, Diagnosis and Treatment of Tuberculosis” project. The project: aims to help combat tuberculosis in Romania by early detection and constant treatment; focuses on people with limited access to medical diagnosis or treatment; total target: 75.010 individuals. Four caravans containing X-ray machines and acting as mobile scanner points...

Conclusion

Despite current treatments, tuberculosis remains a disease with high mortality rate, especially if not recognized and left untreated. When compared to radiologists’ performance, the AI algorithm has lower precision but has a high level of sensitivity for the detection of TB lesions. Therefore an AI-based screening of pulmonary tuberculosis is very useful for detection, prevention and choice of adequate therapeutic conduct against this disease.

Personal information and conflict of interest

A.-E. Moldovan; Timisoara/RO - nothing to disclose C. Avramescu; Timisoara/RO - nothing to disclose B. Bercean; Timisoara/RO - nothing to disclose S. Iarca; Timisoara/RO - nothing to disclose A. Tenescu; Timisoara/RO - nothing to disclose F. Birsasteanu; Timisoara/RO - nothing to disclose

References

Jeong YJ, Lee KS. Pulmonary tuberculosis: Up-to-date imaging and management. AJR Am J Roentgenol. 2008;191:834–44. Van Dyck P, Vanhoenacker FM, Van den Brande P, De Schepper AM. Imaging of pulmonary tuberculosis. Eur Radiol. 2003;13:1771–85. Burrill J, Williams CJ, Bain G, Conder G, Hin AL, Misra RR. 2007. Tuberculosis: A radiological review. Radiographics 27: 1255–1273. McAdams HP, Erasmus J, Winter JA. Radiologic manifestations of pulmonary tuberculosis. Radiol Clin North Am. 1995;33:655–78. Golli, A. L., et al. “Tuberculosis Remains a Public Health Problem in Romania.” The International...