The role of artificial intelligence in assessing the progression of fibrosing lung diseases

Cover Page

Cite item

Full Text

Abstract

Introduction. The widespread use of artificial intelligence (AI) programs during the COVID-19 pandemic to assess the exact volume of lung tissue damage has allowed them to train a large number of radiologists. The simplicity of the program for determining the volume of the affected lung tissue in acute interstitial pneumonia, which has density indicators in the range from -200 HU to -730 HU, which includes the density indicators of "ground glass" and reticulation (the main radiation patterns in COVID-19) allows you to accurately determine the degree of prevalence process. The characteristics of chronic interstitial pneumonia, which are progressive in nature, fit into the same density framework.

Аim. To аssess AI's ability to assess the progression of fibrosing lung disease using lung volume counting programs used for COVID-19 and chronic obstructive pulmonary disease.

Results. Retrospective analysis of computed tomography data during follow-up of 75 patients with progressive fibrosing lung disease made it possible to assess the prevalence and growth of interstitial lesions.

Conclusion. Using the experience of using AI programs to assess acute interstitial pneumonia in COVID-19 can be applied to chronic interstitial pneumonia.

About the authors

Aleksandra A. Speranskaia

Pavlov First Saint Petersburg State Medical University

Author for correspondence.
Email: a.spera@mail.ru
ORCID iD: 0000-0001-8322-4509

д-р мед. наук, проф., проф. каф. рентгенологии и радиационной медицины с рентгенологическим и радиологическим отделениями

Russian Federation, Saint Petersburg

References

  1. Inui S. Radiology: Cardiothoracic Imaging, 8 April 2020. Available at: https://www.researchgate.net/journal/Radiology-Cardiothoracic-Imaging-2638-6135. Accessed at: 24.02.2022.
  2. Han X, Fan Y, Alwalid O, et al. Six-month Follow-up Chest CT Findings after Severe COVID-19 Pneumonia. Radiology. 2021;299(1):E177-86. doi: 10.1148/radiol.2021203153
  3. Flaherty KR, Wells AU, Cottin V, et al. Nintedanib in Progressive Fibrosing Interstitial Lung Diseases. N Engl J Med. 2019;381(18):1718-27. doi: 10.1056/nejmoa1908681
  4. Cottin V, Wollin L, Fischer A, et al. Fibrosing interstitial lung diseases: knowns and unknowns. Eur Respir Rev. 2019;28(151). doi: 10.1183/16000617.0100-2018
  5. Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmona ry fibrosis; evidence based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183:788-824.
  6. Сперанская А.А., Новикова Л.Н., Двораковская И.В., и др. Лучевая и морфологическая картина фиброзирующих болезней легких: от ранних признаков до исхода. Лучевая диагностика и терапия. 2020;11(2):89-98 [Speranskaya AA, Novikova LN, Dvorakovskaya IV, et al. Radiation and morphological picture of fibrosing lung diseases: from early signs to outcome. Diagnostic Radiology and Radiotherapy. 2020;11(2):89-98 (in Russian)]. doi: 10.22328/2079-5343-2020-11-2-89-98
  7. Maldonado F, Moua T, Rajagopalan S, et al. Automated quantification of radiological patterns predicts survival in idiopathic pulmonary fibrosis. Eur Respir J. 2014;43(1):204-12. doi: 10.1183/09031936.00071812
  8. Jacob J, Bartholmai B, Rajagopalan S, et al. Mortality prediction in idiopathic pulmonary fibrosis: evaluation of automated computer-based CT analysis with conventional severity measures. Eur Respir J. 2017;49:1601011. doi: 10.1183/13993003.01011-2016

Supplementary files

Supplementary Files
Action
1. JATS XML
Download
2. Fig. 1. Patient M., 65 years old. Idiopathic pulmonary fibrosis (IPF). Computer tomography (CT) with an assessment of the extent of the lesion from 03.03.2020 – the volume of the affected lung tissue – 18.07%, which is 0.98 liters out of 5.45 liters (total lung capacity) (a, b, c). Control CT from 13.07.2020 – an increase in clinical symptoms, an increase in the volume of the affected lung tissue – 19.62%, 1.05 liters out of 5.35 liters (d, e, f). Control CT from 22.12.2020 – an increase in the volume of the affected lung tissue – 21.92%, 1.08 liters from 4.92 liters (g, h, i). Control CT from 06.09.2021 – stabilization of the process against the background of therapy – 20.97%, 1.12 liters out of 5.37 liters (j, k, l). In the period from 03.2020 to 12.2020, there is a decrease in DLCO by 16% (from 70% from D to 54% from D).

Download (338KB)
Indexing metadata
3. Fig. 2. Patient L., 63 years old. IPF. CT with an assessment of the extent of the lesion dated 21.01.2017 – visually (a), using the artificial intelligenc (AI) program: the volume of interstitial damage of the lung tissue (в, d) – 28.67%, which is 1.20 liters out of 4.18 liters (total lung capacity), the volume of cystic-bullous transformation of lung tissue (c, d) – 3.98%, which is 0.16 liters of 4.18 liters (total lung capacity), the total volume of the lesion is 32.65%. Control CT with an assessment of the extent of the lesion from 14.09.2021 – visually (e), using the AI program: the volume of interstitial damage to the lung tissue (f, h) – 33.24%, which is 1.46 liters out of 4.40 liters (total lung capacity), the volume of cystic-bullous transformation of lung tissue (g, h) – 6.41%, which is 0.28 liters of 4.40 liters (total lung capacity), the total lesion volume is 39.65%. In the period from 2017 to 2021, there is a decrease in DLCO by 18% (from 41% from D to 23% from D).

Download (347KB)
Indexing metadata
4. Fig. 3. The graphic image of life has been used at all times: Pushkin's graphic self-portraits – aging and the appearance of traits of wisdom.

Download (120KB)
Indexing metadata

Copyright (c) 2022 Consilium Medicum

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
 
 


This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies