Late Breaking Abstract - Deep learning-based outcome prediction in pulmonary fibrosos using synthetic HRCT

Generative adversarial networks synthesise realistic synthetic images. We trained 2 deep learning models to identify UIP-like features on HRCT. Each model outputs a 5-point UIP PIOPED score (a likelihood of definite UIP on HRCT). One model was trained on real HRCT data (output: RealCT UIP score) using 500 unique 4-slice montages from 264 HRCTs. The second model (output: SynthCT UIP score) was trained on 500 real and 500 synthetic 4-slice montages from the same 264 HRCTs. We compared model performance to expert radiologist evaluation (RadiologistCT UIP score) using Cox regression to assess model accuracy on a national registry of 504 patients with suspected IPF. Synthetic HRCTs can be used to augment the training of prognostic deep learning models with better-than-human performance