Systemic sclerosis & Raynaud's
Prognostic value of automated assessment of interstitial lung disease on CT in systemic sclerosis
Authors' conclusion
The deep-learning-based, computer-aided quantification of ILD extent on HRCT provides an effective tool for risk stratification in SSc.
Study context
- Population
- Patients with SSc, with or without ILD196 with ILD
- Participants
- 318 patients
- Outcome
- Death during follow-up (prognostic value of automated ILD extent)
- Follow-up
- 10 years
Methodological detailsdeep-learning-based automated quantification of ILD on high-resolution CT (HRCT)
- Deep-learning-based automated quantification of ILD on HRCT
Performance
As reported in the abstract
Risk of death per 1% increase in baseline ILD extent (up to 30% of the lung)
HR per 1% increase in baseline ILD extent
1.04, 95% CI 1.01, 1.07
P = 0.004
Further findings
Mortality rate
- Mortality at 2 years
- 1.6%
- Mortality at 10 years
- 26.3%
- Median follow-up 94 months (IQR 73-111)
Risk prediction model discrimination for 10-year mortality
- C index
- 0.789
- Adding automated ILD quantification improved 10-year survival prediction
- P = 0.007
- Discrimination only marginally improved
Prediction of 2-year mortality with automated ILD quantification added
- Difference in time-dependent AUC
- 0.043, 95% CI 0.002, 0.084, P = 0.040
Original publication
Prognostic value of automated assessment of interstitial lung disease on CT in systemic sclerosis. Rheumatology (Oxford). Published 19 April 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →