FQGA-single: Towards Fewer Training Epochs and Fewer Model Parameters for Image-to-Image Translation Tasks
This work addresses efficiency in medical imaging synthesis, offering potential time and resource savings for healthcare applications, though it appears incremental as it builds on CycleGAN.
The paper tackles the problem of inefficient training for image-to-image translation in medical CT synthesis by proposing FQGA-single, which achieves higher quality synthetic CT images with fewer training epochs and parameters, outperforming CycleGAN in both single-epoch and multi-epoch settings.
This paper proposes a novel model inspired by CycleGAN: FQGA-single to produce high quality medical synthetic CT (sCT) generated images more efficiently. Evaluations were done on the SynthRAD Grand Challenge dataset with the CycleGAN model used for benchmarking and for comparing the quality of CBCT-to-sCT generated images from both a quantitative and qualitative perspective. Finally, this paper also explores ideas from the paper "One Epoch Is All You Need" to compare models trained on a single epoch versus multiple epochs. Astonishing results from FQGA-single were obtained during this exploratory experiment, which show that the performance of FQGA-single when trained on a single epoch surpasses itself when trained on multiple epochs. More surprising is that its performance also surpasses CycleGAN's multiple-epoch and single-epoch models, and even a modified version of CycleGAN.