Effect of emphysema on AI software and human reader performance in lung nodule detection from low-dose chest CT.
Nikos SourlosGertJan PelgrimHendrik Joost WisselinkXiaofei YangGonda de JongeMieneke RookMathias ProkopGrigory SidorenkovMarcel van TuinenRozemarijn VliegenthartPeter M A van OoijenPublished in: European radiology experimental (2024)
• The sensitivity of nodule detection by AI was similar in emphysema and non-emphysema. • AI had more FPs/scan in emphysema compared to non-emphysema. • Sensitivity and FPs/scan by the human reader were comparable for emphysema and non-emphysema. • Emphysema and non-emphysema representation in benchmark dataset is important for validating AI.
Keyphrases
- chronic obstructive pulmonary disease
- lung function
- pulmonary fibrosis
- idiopathic pulmonary fibrosis
- low dose
- artificial intelligence
- computed tomography
- endothelial cells
- cystic fibrosis
- air pollution
- magnetic resonance imaging
- high dose
- induced pluripotent stem cells
- machine learning
- magnetic resonance
- loop mediated isothermal amplification
- deep learning
- sensitive detection
- label free
- neural network