Introduction

High-resolution tomography (HRCT) is the main imaging test in the diagnosis and monitoring of interstitial lung diseases (ILD). Radiomics, which includes methods such as texture-based and histogram analysis along with machine learning, is a promising new approach in the field of ILD diagnosis and prognosis. The aim of the study was to analyze the correlation between the respiratory function tests and the values obtained using lung texture analysis software (SATP) in a group of patients from the ILD unit of Hospital Clinic.

Material and methods

A total of 121 patients from the ILD Unit of the Hospital Clínic were evaluated. Pulmonary function studies (FVC, DLco, and Kco) and high-resolution quantified tomography were performed. The following parameters were obtained using a SATP: Lung Volume (Vol), Hyperlucency, Ground Glass, Reticulations, Honeycombing and pulmonary vascular volume (PVV).

Results

Of the 121 patients, 67 were men (55%) and 54 women (45%) with a mean age of 69±11 years, who were grouped into: 20 with idiopathic pulmonary fibrosis (IPF) (17%), 83 with non-IPF ILD (69%) and 18 with Sarcoidosis (15%). The correlation analysis was good with an inverse relationship between PVV and DLco (r=-0.60, p=<0.001) (see figure 1), FVC (r=-0.46, p=<0.001) and Kco (r=-0.40, p=<0.001). The correlation between FVC and Vol was moderate with a positive linear relationship (r=0.53, p=<0.001).

Conclusions

Our study shows that there is a correlation between FVC and DL_CO and the values obtained by SATP (Vol/PVV) in patients with ILD. The use of quantified HRCT can be an objective assessment tool in the diagnosis and follow-up of patients with ILD.