Deep Learning-based Outcome Prediction in Progressive Fibrotic Lung Disease Using High-Resolution Computed Tomography.

Journal: American journal of respiratory and critical care medicine
Published Date:

Abstract

Reliable outcome prediction in patients with fibrotic lung disease using baseline high-resolution computed tomography (HRCT) data remains challenging. To evaluate the prognostic accuracy of a deep learning algorithm (SOFIA [Systematic Objective Fibrotic Imaging Analysis Algorithm]), trained and validated in the identification of usual interstitial pneumonia (UIP)-like features on HRCT (UIP probability), in a large cohort of well-characterized patients with progressive fibrotic lung disease drawn from a national registry. SOFIA and radiologist UIP probabilities were converted to Prospective Investigation of Pulmonary Embolism Diagnosis (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 was assessed using Cox proportional hazards modeling. In multivariable analysis adjusting for age, sex, guideline-based radiologic diagnosis, anddisease severity (using total interstitial lung disease [ILD] extent on HRCT, percent predicted FVC, Dl, or the composite physiologic index), only SOFIA UIP probability PIOPED categories predicted survival. SOFIA-PIOPED UIP probability categories remained prognostically significant in patients considered indeterminate ( = 83) by expert radiologist consensus (hazard ratio, 1.73;  < 0.0001; 95% confidence interval, 1.40-2.14). In patients undergoing surgical lung biopsy ( = 86), after adjusting for guideline-based histologic pattern and total ILD extent on HRCT, only SOFIA-PIOPED probabilities were predictive of mortality (hazard ratio, 1.75;  < 0.0001; 95% confidence interval, 1.37-2.25). Deep learning-based UIP probability on HRCT provides enhanced outcome prediction in patients with progressive fibrotic lung disease when compared with expert radiologist evaluation or guideline-based histologic pattern. In principle, this tool may be useful in multidisciplinary characterization of fibrotic lung disease. The utility of this technology as a decision support system when ILD expertise is unavailable requires further investigation.

Authors

  • Simon L F Walsh
    Department of Radiology, King's College Hospital Foundation Trust, London, UK. Electronic address: slfwalsh@gmail.com.
  • John A Mackintosh
    Queensland Lung Transplant Service, The Prince Charles Hospital, Brisbane, Queensland, Australia.
  • Lucio Calandriello
    Department of Radiology, Fondazione Policlinico Universitario A Gemelli IRCCS, Rome, Italy.
  • Mario Silva
    Department of Medicine and Surgery, University of Parma, Parma, Italy.
  • Nicola Sverzellati
    Department of Medicine and Surgery, University of Parma, Parma, Italy.
  • Anna Rita Larici
    Section of Radiology, Department of Radiological and Hematological Sciences, Università Cattolica del Sacro Cuore, Rome, Italy.
  • Stephen M Humphries
    Quantitative Imaging Laboratory, Department of Radiology, National Jewish Health, Denver, CO, USA.
  • David A Lynch
    National Jewish Health, Denver, CO, USA.
  • Helen E Jo
    Respiratory Medicine, Royal Prince Alfred Hospital, New South Wales, Australia.
  • Ian Glaspole
    Department of Allergy and Respiratory Medicine, Alfred Hospital, Melbourne, Victoria, Australia.
  • Christopher Grainge
    Department of Respiratory Medicine, New Lambton Heights, John Hunter Hospital, New South Wales, Australia.
  • Nicole Goh
    Yale-NUS College, Singapore, Singapore.
  • Peter M A Hopkins
    Queensland Lung Transplant Service, The Prince Charles Hospital, Brisbane, Queensland, Australia.
  • Yuben Moodley
    School of Medicine & Pharmacology, University of Western Australia, Perth, Western Australia, Australia.
  • Paul N Reynolds
    Lung Research, Hanson Institute, Adelaide, South Australia, Australia.
  • Christopher Zappala
    Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia.
  • Gregory Keir
    Department of Respiratory Medicine, Princess Alexandra Hospital, Brisbane, Queensland, Australia.
  • Wendy A Cooper
    Tissue Pathology and Diagnostic Oncology, New South Wales Health Pathology, Royal Prince Alfred Hospital, Sydney, New South Wales, Australia.
  • Annabelle M Mahar
    Tissue Pathology and Diagnostic Oncology, New South Wales Health Pathology, Royal Prince Alfred Hospital, Sydney, New South Wales, Australia.
  • Samantha Ellis
    Department of Radiology, Alfred Health, Melbourne, Victoria, Australia; and.
  • Athol U Wells
    Interstitial Lung Disease Unit, Royal Brompton Hospital, London, UK.
  • Tamera J Corte
    Respiratory Medicine, Royal Prince Alfred Hospital, New South Wales, Australia.