Quantitative CT Imaging Features Associated with Stable PRISm using Machine Learning

Leila Lukhumaidze; James C Hogg; Jean Bourbeau; Wan C Tan; Miranda Kirby

doi:10.1016/j.acra.2024.08.030

Quantitative CT Imaging Features Associated with Stable PRISm using Machine Learning

Acad Radiol. 2024 Aug 26:S1076-6332(24)00589-0. doi: 10.1016/j.acra.2024.08.030. Online ahead of print.

Authors

Leila Lukhumaidze¹, James C Hogg², Jean Bourbeau³, Wan C Tan², Miranda Kirby⁴

Affiliations

¹ Toronto Metropolitan University, Toronto, ON, Canada (L.L., M.K.).
² Center for Heart, Lung Innovation, University of British Columbia, Vancouver, BC, Canada (J.C.H., W.C.T.).
³ Montreal Chest Institute of the Royal Victoria Hospital, McGill University Health Centre, Montreal, QC, Canada (J.B.); Respiratory Epidemiology and Clinical Research Unit, Research Institute of McGill University Health Centre, Montreal, QC, Canada (J.B.).
⁴ Toronto Metropolitan University, Toronto, ON, Canada (L.L., M.K.). Electronic address: Miranda.Kirby@torontomu.ca.

PMID: 39191563
DOI: 10.1016/j.acra.2024.08.030

Abstract

Rationale and objectives: The structural lung features that characterize individuals with preserved ratio impaired spirometry (PRISm) that remain stable overtime are unknown. The objective of this study was to use machine learning models with computed tomography (CT) imaging to classify stable PRISm from stable controls and stable COPD and identify discriminative features.

Materials and methods: A total of 596 participants that did not transition between control, PRISm and COPD groups at baseline and 3-year follow-up were evaluated: n = 274 with normal lung function (stable control), n = 22 stable PRISm, and n = 300 stable COPD. Investigated features included: quantitative CT (QCT) features (n = 34), such as total lung volume (%TLC_CT) and percentage of ground glass and reticulation (%GG+Reticulation_texture), as well as Radiomic (n = 102) features, including varied intensity zone distribution grainy texture (GLDZM_ZDV). Logistic regression machine learning models were trained using various feature combinations (Base, Base+QCT, Base+Radiomic, Base+QCT+Radiomic). Model performances were evaluated using area under receiver operator curve (AUC) and comparisons between models were made using DeLong test; feature importance was ranked using Shapley Additive Explanations values.

Results: Machine learning models for all feature combinations achieved AUCs between 0.63-0.84 for stable PRISm vs. stable control, and 0.65-0.92 for stable PRISm vs. stable COPD classification. Models incorporating imaging features outperformed those trained solely on base features (p < 0.05). Compared to stable control and COPD, those with stable PRISm exhibited decreased %TLC_CT and increased %GG+Reticulation_texture and GLDZM_ZDV.

Conclusion: These findings suggest that reduced lung volumes, and elevated high-density and ground glass/reticulation patterns on CT imaging are associated with stable PRISm.

Keywords: Fibrosis; ILA; PRISm; Radiomic; Texture Features.