A computed tomography-based radiomics nomogram for predicting overall survival in patients with connective tissue disease-associated interstitial lung disease.
Eur J Radiol
; 165: 110963, 2023 Aug.
Article
in En
| MEDLINE
| ID: mdl-37437436
ABSTRACT
OBJECTIVES:
Accurate prognostic prediction is beneficial for the management of patients with connective tissue disease-associated interstitial lung disease (CTD-ILD). The purpose of the present study was to develop and validate a nomogram using clinical features and computed tomography (CT) based radiomics features to predict overall survival (OS) in patients with CTD-ILD, and to assess the incremental prognostic value the radiomics might add to clinical risk factors. MATERIALS &METHODS:
Patients from two clinical centers with CTD-ILD were enrolled in the present retrospective study. A radiomics signature, a clinical model and a combined nomogram were developed and assessed in the cohorts. The incremental value of radiomics signature to the clinical independent risk factors in survival prediction was evaluated. The models were externally validated to evaluate the model generalization ability.RESULTS:
A total of 215 patients (mean age, 53 years ± 14 [standard deviation], 45 men) were evaluated. Patients with higher radiomics scores had higher mortality risk than those with lower radiomics scores (Hazard ratio, 12.396; 95% CI, 3.364-45.680; P < 0.001). The combined nomogram showed better predictive capability than the clinical model did with higher C-indices (0.800, 0.738, 0.742 vs. 0.747, 0.631, 0.587 in the training, internal- and external-validation cohort, respectively), time-AUCs and overall net-benefit.CONCLUSION:
The radiomics signature is a potential prognostic biomarker of CTD-ILD and add incremental value to the clinical independent risk factors. The combined nomogram can provide a more accurate estimation of OS than the clinical model for CTD-ILD patients. CLINICAL RELEVANCE STATEMENT The developed combined nomogram showed accurate prognostic prediction performance, which is beneficial for the management of CTD-ILD patients. It also proved radiomics could extract prognostic information from CT images.Key words
Full text:
1
Collection:
01-internacional
Database:
MEDLINE
Main subject:
Lung Diseases, Interstitial
/
Connective Tissue Diseases
Type of study:
Observational_studies
/
Prognostic_studies
/
Risk_factors_studies
Limits:
Humans
/
Male
/
Middle aged
Language:
En
Journal:
Eur J Radiol
Year:
2023
Document type:
Article
Affiliation country:
China