The Effect of Lossy Compression on 3D Medical Images Segmentation with Deep Learning
This work addresses storage and transmission costs for medical imaging by showing compression is viable without degrading segmentation performance, though it is incremental as it applies known compression techniques to a new domain.
The study investigated the impact of lossy compression on 3D medical image segmentation using deep learning, finding that compression up to 20 times had no negative effect on segmentation quality across multiple CT and MRI datasets, and models trained on compressed data could predict on uncompressed data without quality loss.
Image compression is a critical tool in decreasing the cost of storage and improving the speed of transmission over the internet. While deep learning applications for natural images widely adopts the usage of lossy compression techniques, it is not widespread for 3D medical images. Using three CT datasets (17 tasks) and one MRI dataset (3 tasks) we demonstrate that lossy compression up to 20 times have no negative impact on segmentation quality with deep neural networks (DNN). In addition, we demonstrate the ability of DNN models trained on compressed data to predict on uncompressed data and vice versa with no quality deterioration.