98%
921
2 minutes
20
Low-dose CT (LDCT) significantly reduces the radiation dose received by patients, however, dose reduction introduces additional noise and artifacts. Currently, denoising methods based on convolutional neural networks (CNNs) face limitations in long-range modeling capabilities, while Transformer-based denoising methods, although capable of powerful long-range modeling, suffer from high computational complexity. Furthermore, the denoised images predicted by deep learning-based techniques inevitably exhibit differences in noise distribution compared to normal-dose CT (NDCT) images, which can also impact the final image quality and diagnostic outcomes. This paper proposes CT-Mamba, a hybrid convolutional State Space Model for LDCT image denoising. The model combines the local feature extraction advantages of CNNs with Mamba's strength in capturing long-range dependencies, enabling it to capture both local details and global context. Additionally, we introduce an innovative spatially coherent Z-shaped scanning scheme to ensure spatial continuity between adjacent pixels in the image. We design a Mamba-driven deep noise power spectrum (NPS) loss function to guide model training, ensuring that the noise texture of the denoised LDCT images closely resembles that of NDCT images, thereby enhancing overall image quality and diagnostic value. Experimental results have demonstrated that CT-Mamba performs excellently in reducing noise in LDCT images, enhancing detail preservation, and optimizing noise texture distribution, and exhibits higher statistical similarity with the radiomics features of NDCT images. The proposed CT-Mamba demonstrates outstanding performance in LDCT denoising and holds promise as a representative approach for applying the Mamba framework to LDCT denoising tasks.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1016/j.compmedimag.2025.102595 | DOI Listing |
Sci Rep
July 2025
Department of Computer Science, Norwegian University of Science and Technology, Trondheim, Trøndelag, Norway.
Low-dose computed tomography (LDCT) has gained considerable attention for its ability to minimize patients' exposure to radiation thereby reducing the associated cancer risks. However, this reduction in radiation dose often results in degraded image quality due to the presence of noise and artifacts. To address this challenge, the present study proposes an LDCT image denoising method that leverages a pixel-level nonlocal self-similarity (NSS) prior in combination with a nonlocal means algorithm.
View Article and Find Full Text PDFComput Med Imaging Graph
September 2025
Tianmushan Laboratory, Hangzhou, China; School of Physics, Beihang University, Beijing, China; State Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou, China. Electronic address:
Low-dose CT (LDCT) significantly reduces the radiation dose received by patients, however, dose reduction introduces additional noise and artifacts. Currently, denoising methods based on convolutional neural networks (CNNs) face limitations in long-range modeling capabilities, while Transformer-based denoising methods, although capable of powerful long-range modeling, suffer from high computational complexity. Furthermore, the denoised images predicted by deep learning-based techniques inevitably exhibit differences in noise distribution compared to normal-dose CT (NDCT) images, which can also impact the final image quality and diagnostic outcomes.
View Article and Find Full Text PDFIEEE Trans Image Process
December 2024
Reducing the radiation dose in CT scanning is important to alleviate the damage to the human health in clinical scenes. A promising way is to replace the normal-dose CT (NDCT) imaging by low-dose CT (LDCT) imaging with lower tube voltage and tube current. This often brings severe noise to the LDCT images, which adversely affects the diagnosis accuracy.
View Article and Find Full Text PDFJ Appl Clin Med Phys
December 2024
Department of Radiation Oncology, The First Affiliated Hospital of Soochow University, Suzhou, China.
Objective: We investigated the feasibility of deep learning-based ultra-low dose kV-fan-beam computed tomography (kV-FBCT) image enhancement algorithm for clinical application in abdominal and pelvic tumor radiotherapy.
Methods: A total of 76 patients of abdominal and pelvic tumors were prospectively selected. The Catphan504 was acquired with the same conditions as the standard phantom test set.
Med Phys
December 2024
Department of Obstetrics and Gynecology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.