Background
As a means to extract biomarkers from medical imaging, radiomics has attracted increased attention from researchers. However, reproducibility and performance of radiomics in lowâdose CT scans are still poor, mostly due to noise. Deep learning generative models can be used to denoise these images and in turn improve radiomicsâ reproducibility and performance. However, most generative models are trained on paired data, which can be difficult or impossible to collect.
Purpose
In this article, we investigate the possibility of denoising lowâdose CTs using cycle generative adversarial networks (GANs) to improve radiomics reproducibility and performance based on unpaired datasets.
Methods and materials
Two cycle GANs were trained: (1) from paired data, by simulating lowâdose CTs (i.e., introducing noise) from highâdose CTs and (2) from unpaired real low dose CTs. To accelerate convergence, during GAN training, a sliceâpaired training strategy was introduced. The trained GANs were applied to three scenarios: (1) improving radiomics reproducibility in simulated lowâdose CT images and (2) sameâday repeat low dose CTs (RIDER dataset), and (3) improving radiomics performance in survival prediction. Cycle GAN results were compared with a conditional GAN (CGAN) and an encoderâdecoder network (EDN) trained on simulated paired data.
Results
The cycle GAN trained on simulated data improved concordance correlation coefficients (CCC) of radiomic features from 0.87 (95%CI, [0.833,0.901]) to 0.93 (95%CI, [0.916,0.949]) on simulated noise CT and from 0.89 (95%CI, [0.881,0.914]) to 0.92 (95%CI, [0.908,0.937]) on the RIDER dataset, as well improving the area under the receiver operating characteristic curve (AUC) of survival prediction from 0.52 (95%CI, [0.511,0.538]) to 0.59 (95%CI, [0.578,0.602]). The cycle GAN trained on real data increased the CCCs of features in RIDER to 0.95 (95%CI, [0.933,0.961]) and the AUC of survival prediction to 0.58 (95%CI, [0.576,0.596]).
Conclusion
The results show that cycle GANs trained on both simulated and real data can improve radiomicsâ reproducibility and performance in lowâdose CT and achieve similar results compared to CGANs and EDNs.