LAUSR

RATIONALE Reliable outcome prediction in patients with fibrotic lung disease using baseline high-resolution computed tomography (HRCT) data remains challenging. OBJECTIVES To evaluate the prognostic accuracy of a deep learning algorithm (SOFIA), trained and validated in the identification of UIP-like features on HRCT (UIP probability), in a large cohort of well characterised patients with progressive fibrotic lung disease, drawn from a national registry. METHODS SOFIA and radiologist-UIP probabilities were converted to PIOPED-based UIP probability categories (UIP not included in the differential: 0-4%, low probability of UIP: 5-29%, intermediate probability of UIP: 30-69%, high probability of UIP: 70-94%, and pathognomonic for UIP:95-100%) and their prognostic utility assessed using Cox proportional hazards modelling. MEASUREMENTS AND MAIN RESULTS On multivariable analysis adjusting for age, gender, guideline based radiologic diagnosis and disease severity (using total ILD extent on HRCT, %predicted FVC, DLco or the CPI), only SOFIA-UIP probability PIOPED categories predicted survival. SOFIA-PIOPED UIP probability categories remained prognostically significant in patients considered indeterminate (n=83) by expert radiologist consensus (HR1.73, p<0.0001, 95%CI 1.40-2.14). In patients undergoing surgical lung biopsy (SLB) (n=86), after adjusting for guideline-based histologic pattern and total ILD extent on HRCT, only SOFIA-PIOPED probabilities were predictive of mortality (HR1.75, p<0.0001, 95%CI 1.37-2.25). CONCLUSIONS Deep learning-based UIP probability on HRCT provides enhanced outcome prediction in patients with progressive fibrotic lung disease when compared to expert radiologist evaluation or guideline-based histologic pattern. In principle this tool may be useful in multidisciplinary characterisation of fibrotic lung disease. The utility of this technology as a decision support system when ILD expertise is unavailable requires further investigation.