GAN-based survival prediction model from CT images of patients with idiopathic pulmonary fibrosis

Background.Recent reports indicate that matrix metalloproteinase-7 (MMP-7) and CC-chemokine ligand 18 (CCL18) are potential disease markers of idiopathic pulmonary fibrosis (IPF). The objective of this study was to perform direct comparisons of these two biomarkers with three well-investigated serum markers of IPF, Krebs von den Lungen-6 (KL-6), surfactant protein-A (SP-A), and SP-D.Methods.The serum levels of MMP-7, CCL18, KL-6, SP-A, and SP-D were evaluated in 65 patients with IPF, 31 patients with bacterial pneumonia, and 101 healthy controls. The prognostic performance of these five biomarkers was evaluated in patients with IPF.Results.The serum levels of MMP-7, KL-6, and SP-D in patients with IPF were significantly elevated compared to those in patients with bacterial pneumonia and in the healthy controls. Multivariate survival analysis showed that serum MMP-7 and KL-6 levels were independent predictors in IPF patients. Moreover, elevated levels of both KL-6 and MMP-7 were associated with poorer survival rates in IPF patients, and the combination of both markers provided the best risk discrimination using the C statistic.Conclusions.The present results indicated that MMP-7 and KL-6 were promising prognostic markers of IPF, and the combination of the two markers might improve survival prediction in patients with IPF.

Download Full-text

Standardised 3D-CT Lung Volumes for Patients with Idiopathic Pulmonary Fibrosis

10.21203/rs.3.rs-1129054/v1 ◽

2021 ◽

Author(s):

Yuko Tanaka ◽

Yuzo Suzuki ◽

Hirotsugu Hasegawa ◽

Koshi Yokomura ◽

Atsuki Fukada ◽

...

Keyword(s):

Computed Tomography ◽

Idiopathic Pulmonary Fibrosis ◽

Pulmonary Fibrosis ◽

Pulmonary Function Tests ◽

Three Dimensional ◽

Ct Images ◽

3D Ct ◽

Pulmonary Physiology ◽

Lung Physiology ◽

Lower Lung

Abstract Background: The assessment of lung physiology via pulmonary function tests (PFTs) is essential for patients with idiopathic pulmonary fibrosis (IPF). However, PFTs require active participation, which can be challenging for patients with severe respiratory failure, such as during acute exacerbations (AE) of IPF. Recently advances enabled to re-construct of 3-dimensional computed-tomography (3D-CT) images. Methods: This is a retrospective multi-center cohort study. This study established a standardisation method and quantitative analysis of lung volume (LV) based on anthropometry using three-dimensional computed tomography (3D-CT) images. The standardised 3D-CT LV in patients with IPF at diagnosis (n=140) and during AE (cohort1; n=61 and cohort2; n=50) and those of controls (n=53) were measured. Results: The standardised 3D-CT LVs at IPF diagnosis were less than those of control patients, especially in the lower lung lobes. The standardised 3D-CT LVs were correlated with forced vital capacity (FVC) and validated using the modified Gender-Age-Physiology (GAP) index. The standardised 3D-CT LVs at IPF diagnosis were independently associated with prognosis. During AE, PFTs were difficult to perform, 3D-CT analyses revealed reduced lung capacity in both the upper and lower lobes compared to those obtained at diagnosis. Lower standardised 3D-CT LVs during AE were independently associated with worse outcomes in independent two cohorts. Particularly, volume loss in the upper lobe at AE had prognostic values.Conclusion: A novel image quantification method for assessing pulmonary physiology using standardised 3D-CT-derived LVs was developed. This method successfully predicts mortality in patients with IPF and AE of IPF, and may be a useful alternative to PFTs when PFTs cannot be performed.

Download Full-text

A study design for statistical learning technique to predict radiological progression with an application of idiopathic pulmonary fibrosis using chest CT images

Contemporary Clinical Trials ◽

10.1016/j.cct.2021.106333 ◽

2021 ◽

Vol 104 ◽

pp. 106333

Author(s):

Grace Hyun J. Kim ◽

Yu Shi ◽

Wenxi Yu ◽

Weng Kee Wong

Keyword(s):

Idiopathic Pulmonary Fibrosis ◽

Pulmonary Fibrosis ◽

Statistical Learning ◽

Study Design ◽

Ct Images ◽

Chest Ct ◽

Radiological Progression ◽

Learning Technique

Download Full-text

Unsupervised survival prediction model from CT images of patients with COVID-19

Medical Imaging 2021: Computer-Aided Diagnosis ◽

10.1117/12.2581929 ◽

2021 ◽

Author(s):

Tomoki Uemura ◽

Janne Näppi ◽

Chinatsu Watari ◽

Tohru Kamiya ◽

Hiroyuki Yoshida

Keyword(s):

Prediction Model ◽

Ct Images ◽

Survival Prediction

Download Full-text

Assessment of Prognosis of Patients With Idiopathic Pulmonary Fibrosis by Computer-aided Analysis of CT Images

Journal of Thoracic Imaging ◽

10.1097/rti.0b013e3181a6527d ◽

2009 ◽

Vol 24 (3) ◽

pp. 216-222 ◽

Cited By ~ 41

Author(s):

Tae Iwasawa ◽

Akira Asakura ◽

Fumikazu Sakai ◽

Tetu Kanauchi ◽

Toshiyuki Gotoh ◽

...

Keyword(s):

Idiopathic Pulmonary Fibrosis ◽

Pulmonary Fibrosis ◽

Ct Images ◽

Computer Aided Analysis ◽

Computer Aided

Download Full-text

Investigation of a Hypoxia-Immune-Related Microenvironment Gene Signature and Prediction Model for Idiopathic Pulmonary Fibrosis

Frontiers in Immunology ◽

10.3389/fimmu.2021.629854 ◽

2021 ◽

Vol 12 ◽

Author(s):

Xinyu Li ◽

Haozheng Cai ◽

Yufeng Cai ◽

Quyan Zhang ◽

Yinghe Ding ◽

...

Keyword(s):

Idiopathic Pulmonary Fibrosis ◽

Pulmonary Fibrosis ◽

Peripheral Blood Mononuclear Cell ◽

Prediction Model ◽

Peripheral Blood ◽

Immune Status ◽

Peripheral Blood Mononuclear ◽

Independent Validation ◽

Geo Database

BackgroundThere is growing evidence found that the role of hypoxia and immune status in idiopathic pulmonary fibrosis (IPF). However, there are few studies about the role of hypoxia and immune status in the lung milieu in the prognosis of IPF. This study aimed to develop a hypoxia-immune-related prediction model for the prognosis of IPF.MethodsHypoxia and immune status were estimated with microarray data of a discovery cohort from the GEO database using UMAP and ESTIMATE algorithms respectively. The Cox regression model with the LASSO method was used for identifying prognostic genes and developing hypoxia-immune-related genes. Cibersort was used to evaluate the difference of 22 kinds of immune cell infiltration. Three independent validation cohorts from GEO database were used for external validation. Peripheral blood mononuclear cell (PBMC) and bronchoalveolar lavage fluid (BALF) were collected to be tested by Quantitative reverse transcriptase-PCR (qRT-PCR) and flow cytometry from 22 clinical samples, including 13 healthy controls, six patients with non-fibrotic pneumonia and three patients with pulmonary fibrosis.ResultsHypoxia and immune status were significantly associated with the prognosis of IPF patients. High hypoxia and high immune status were identified as risk factors for overall survival. CD8+ T cell, activated CD4+ memory T cell, NK cell, activated mast cell, M1 and M0 macrophages were identified as key immune cells in hypoxia-immune-related microenvironment. A prediction model for IPF prognosis was established based on the hypoxia-immune-related one protective and nine risk DEGs. In the independent validation cohorts, the prognostic prediction model performed the significant applicability in peripheral whole blood, peripheral blood mononuclear cell, and lung tissue of IPF patients. The preliminary clinical specimen validation suggested the reliability of most conclusions.ConclusionsThe hypoxia-immune-based prediction model for the prognosis of IPF provides a new idea for prognosis and treatment.

Download Full-text