TY - JOUR
T1 - Repeatability of AI-quantified incidental findings on lung cancer screening CT scans in the NELSON trial
AU - Bunk, Stijn
AU - Slot, Thijs Bruins
AU - Bennink, Edwin
AU - Sidorenkov, Grigory
AU - van der Velden, Nils
AU - Schurink, Niels
AU - Lades, Felix
AU - Sebald, Markus
AU - Heuvelmans, Marjolein A.
AU - Gietema, Hester A.
AU - Aerts, Joachim G.
AU - de Bock, Geertruida H.
AU - Schaefer-Prokop, Cornelia
AU - de Jong, Pim A.
AU - Vliegenthart, Rozemarijn
AU - Mohamed Hoesein, Firdaus A. A.
AU - Cornelissen, Robin
AU - Stadhouders, Ralph
AU - van Rooij, Jeroen G. J.
AU - Trap, Lianne
AU - Nackaerts, Kristiaan
AU - de Wever, Walter
AU - Prokop, Mathias
AU - Jacobs, Colin
AU - Antonissen, Noa
AU - Zhong, Danrong
AU - Groen, Harry J. M.
AU - Downward, George S.
AU - Vermeulen, Roel C. H.
PY - 2026
Y1 - 2026
N2 - ObjectivesComputed tomography (CT) scans for lung cancer screening provide the opportunity of quantifying incidental findings. We evaluated the repeatability of AI-based measurements of incidental findings using short-term repeat CT scan pairs. Materials and methodsAI-Rad Companion Chest CT software was applied to low-dose non-contrast CT scans from the NELSON lung cancer screening trial to measure aorta diameters, coronary artery calcium volume (CACV), vertebral height and radiodensity, and emphysema (low attenuation area percentage, LAA%). Categories (absent/present) of aortic dilatation and osteopenia, and severity (none/mild/moderate/severe) of CACV and emphysema were calculated. We included subjects who had a short-term repeat CT scan pair with a maximum interval of 120 days. We analyzed repeatability with absolute and relative differences, and agreement with the intraclass correlation coefficient (ICC) and Cohen's kappa. Results1436 subjects were included, with age (mean +/- SD) 59.7 +/- 5.7 years, 86.3% men, 55.9% currently smoking, and scan interval 85 +/- 20 days. Mean absolute differences were 0.7 to 1.5 mm for aorta diameters, 26 mm3 for CACV, 0.3 mm to 0.4 mm for vertebral height, and 7.6 HU for vertebral radiodensity. Median absolute difference was 0.8% for LAA%. Aorta diameters showed good (0.75 < ICC < 0.9) to excellent (ICC > 0.9) agreement. Agreement for the rest was excellent. Categorization Cohen's kappa between the first and second measurements was 0.68 for aortic dilatation, 0.63 for CACV, 0.84 for osteopenia, and 0.70 for emphysema. ConclusionIn a lung cancer screening cohort with short-term repeat CT, the repeatability and agreement of automated AI measurements of aortic diameters, coronary calcium, vertebral height and radiodensity, and emphysema was good to excellent. Key Points QuestionCan AI measurements of incidental findings beyond lung nodules be repeatably performed on pairs of lung cancer screening chest CT scans of the same subject? FindingsThe repeatability and agreement of AI-based measurements of aorta diameter, coronary calcium, vertebrae, and emphysema were generally excellent. Clinical relevanceMeasurement by radiologists of incidental findings on lung cancer screening chest CT scans imposes a high workload. AI algorithms allow for automatic measurement of incidental findings on lung cancer screening chest CT with high repeatability and agreement.
AB - ObjectivesComputed tomography (CT) scans for lung cancer screening provide the opportunity of quantifying incidental findings. We evaluated the repeatability of AI-based measurements of incidental findings using short-term repeat CT scan pairs. Materials and methodsAI-Rad Companion Chest CT software was applied to low-dose non-contrast CT scans from the NELSON lung cancer screening trial to measure aorta diameters, coronary artery calcium volume (CACV), vertebral height and radiodensity, and emphysema (low attenuation area percentage, LAA%). Categories (absent/present) of aortic dilatation and osteopenia, and severity (none/mild/moderate/severe) of CACV and emphysema were calculated. We included subjects who had a short-term repeat CT scan pair with a maximum interval of 120 days. We analyzed repeatability with absolute and relative differences, and agreement with the intraclass correlation coefficient (ICC) and Cohen's kappa. Results1436 subjects were included, with age (mean +/- SD) 59.7 +/- 5.7 years, 86.3% men, 55.9% currently smoking, and scan interval 85 +/- 20 days. Mean absolute differences were 0.7 to 1.5 mm for aorta diameters, 26 mm3 for CACV, 0.3 mm to 0.4 mm for vertebral height, and 7.6 HU for vertebral radiodensity. Median absolute difference was 0.8% for LAA%. Aorta diameters showed good (0.75 < ICC < 0.9) to excellent (ICC > 0.9) agreement. Agreement for the rest was excellent. Categorization Cohen's kappa between the first and second measurements was 0.68 for aortic dilatation, 0.63 for CACV, 0.84 for osteopenia, and 0.70 for emphysema. ConclusionIn a lung cancer screening cohort with short-term repeat CT, the repeatability and agreement of automated AI measurements of aortic diameters, coronary calcium, vertebral height and radiodensity, and emphysema was good to excellent. Key Points QuestionCan AI measurements of incidental findings beyond lung nodules be repeatably performed on pairs of lung cancer screening chest CT scans of the same subject? FindingsThe repeatability and agreement of AI-based measurements of aorta diameter, coronary calcium, vertebrae, and emphysema were generally excellent. Clinical relevanceMeasurement by radiologists of incidental findings on lung cancer screening chest CT scans imposes a high workload. AI algorithms allow for automatic measurement of incidental findings on lung cancer screening chest CT with high repeatability and agreement.
KW - Artificial intelligence
KW - Coronary artery disease
KW - Emphysema
KW - Thoracic vertebrae
KW - Tomography (x-ray computed)
KW - CHEST COMPUTED-TOMOGRAPHY
KW - OSTEOPOROSIS
KW - RELIABILITY
U2 - 10.1007/s00330-026-12669-3
DO - 10.1007/s00330-026-12669-3
M3 - Article
SN - 0938-7994
JO - European Radiology
JF - European Radiology
ER -