Unsupervised Denoising of Real Clinical Low Dose Liver CT with Perceptual Attention Networks
Jingxi Pu, Tonghua Liu, Zhilin Guan, Siqiao Li, Yang Ming + 3 more
TLDR
An unsupervised deep learning framework with perceptual attention networks effectively denoises real clinical low-dose liver CT images.
Key contributions
- Proposes an end-to-end unsupervised denoising framework for low-dose CT.
- Combines U-Net, attention mechanism, residual network, and perceptual loss.
- Constructs a real low-dose CT dataset and performs extensive comparative experiments.
- Achieves excellent performance, validated by physicians, addressing real clinical data limitations.
Why it matters
Low-dose CT reduces radiation but introduces noise, hindering diagnosis. This unsupervised method allows using real clinical data, previously difficult for supervised learning, to improve image quality. It enhances diagnostic accuracy and patient safety by providing clearer images for physicians.
Original Abstract
With the development of deep learning, medical image processing has been widely used to assist clinical research. This paper focuses on the denoising problem of low-dose computed tomography using deep learning. Although low-dose computed tomography reduces radiation exposure to patients, it also introduces more noise, which may interfere with visual interpretation by physicians and affect diagnostic results. To address this problem, inspired by Cycle-GAN for unsupervised learning, this paper proposes an end-to-end unsupervised low-dose computed tomography denoising framework. The proposed framework combines a U-Net structure for multi-scale feature extraction, an attention mechanism for feature fusion, and a residual network for feature transformation. It also introduces perceptual loss to improve the network for the characteristics of medical images. In addition, we construct a real low-dose computed tomography dataset and design a large number of comparative experiments to validate the proposed method, using both image-based evaluation metrics and medical evaluation criteria. Compared with classical methods, the main advantage of this paper is that it addresses the limitation that real clinical data cannot be directly used for supervised learning, while still achieving excellent performance. The experimental results are also professionally evaluated by imaging physicians and meet clinical needs.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.