A new visual State Space Model for low-dose CT denoising.

BACKGROUND: Low-dose computed tomography (LDCT) can mitigate potential health risks to the public. However, the severe noise and artifacts in LDCT images can impede subsequent clinical diagnosis and analysis. Convolutional neural networks (CNNs) and Transformers stand out as the two most popular backbones in LDCT denoising. Nonetheless, CNNs suffer from a lack of long-range modeling capabilities, while Transformers are hindered by high computational complexity. PURPOSE: In this study, our main goal is to develop a simple and efficient model that can both focus on local spatial context and model long-range dependencies with linear computational complexity for LDCT denoising. METHODS: In this study, we make the first attempt to apply the State Space Model to LDCT denoising and propose a novel LDCT denoising model named Visual Mamba Encoder-Decoder Network (ViMEDnet). To efficiently and effectively capture both the local and global features, we propose the Mixed State Space Module (MSSM), where the depth-wise convolution, max-pooling, and 2D Selective Scan Module (2DSSM) are coupled together through a partial channel splitting mechanism. 2DSSM is capable of capturing global information with linear computational complexity, while convolution and max-pooling can effectively learn local signals to facilitate detail restoration. Furthermore, the network uses a weighted gradient-sensitive hybrid loss function to facilitate the preservation of image details, improving the overall denoising performance. RESULTS: The performance of our proposed ViMEDnet is compared to five state-of-the-art LDCT denoising methods, including an iterative algorithm, two CNN-based methods, and two Transformer-based methods. The comparative experimental results demonstrate that the proposed ViMEDnet can achieve better visual quality and quantitative assessment outcomes. In visual evaluation, ViMEDnet effectively removes noise and artifacts, while exhibiting superior performance in restoring fine structures and low-contrast structural edges, resulting in minimal deviation of denoised images from NDCT. In quantitative assessment, ViMEDnet obtains the lowest RMSE and the highest PSNR, SSIM, and FSIM scores, further substantiating the superiority of ViMEDnet. CONCLUSIONS: The proposed ViMEDnet possesses excellent LDCT denoising performance and provides a new alternative to LDCT denoising models beyond the existing CNN and Transformer options.

A feasibility study of tumor motion monitoring for SBRT of lung cancer based on 3D point cloud detection and stacking ensemble learning.

PURPOSE: To construct a tumor motion monitoring model for stereotactic body radiation therapy (SBRT) of lung cancer from a feasibility perspective. METHODS: A total of 32 treatment plans for 22 patients were collected, whose planning CT and the centroid position of the planning target volume (PTV) were used as the reference. Images of different respiratory phases in 4DCT were acquired to redefine the targets and obtain the floating PTV centroid positions. In accordance with the planning CT and CBCT registration parameters, data augmentation was accomplished, yielding 2130 experimental recordings for analysis. We employed a stacking multi-learning ensemble approach to fit the 3D point cloud variations of body surface and the change of target position to construct the tumor motion monitoring model, and the prediction accuracy was assess using root mean squared error (RMSE) and R-Square (R2). RESULTS: The prediction displacement of the stacking ensemble model shows a high degree of agreement with the reference value in each direction. In the first layer of model, the X direction (RMSE =0.019 ∼ 0.145mm, R2 =0.9793∼0.9996) and the Z direction (RMSE = 0.051 ∼ 0.168 mm, R2 = 0.9736∼0.9976) show the best results, while the Y direction ranked behind (RMSE = 0.088 ∼ 0.224 mm, R2 = 0.9553∼ 0.9933). The second layer model summarizes the advantages of unit models of first layer, and RMSE of 0.015 mm, 0.083 mm, 0.041 mm, and R2 of 0.9998, 0.9931, 0.9984 respectively for X, Y, Z were obtained. CONCLUSIONS: The tumor motion monitoring method for SBRT of lung cancer has potential application of non-ionization, non-invasive, markerless, and real-time.

Dosimetric impact of CT metal artifact reduction for spinal implants in stereotactic body radiotherapy planning.

Background: Metal artifacts due to spinal implants may affect the accuracy of dose calculation for radiotherapy. However, the dosimetric impact of metal artifact reduction (MAR) for spinal implants in stereotactic body radiotherapy (SBRT) plans has not been well studied. The objective of this study was to evaluate the dosimetric impact of MAR in spinal SBRT planning with three clinically common dose calculation algorithms. Methods: Gammex phantom and 10 patients' computed tomography (CT) images were studied to investigate the effects of titanium implants. A commercial orthopedic MAR algorithm was employed to reduce artifacts. Dose calculations for SBRT were conducted on both artifact-corrected and uncorrected images using three commercial algorithms [analytical anisotropic algorithm (AAA), Acuros XB (AXB), and Monte Carlo (MC)]. Dose discrepancies between artifact-corrected and uncorrected cases were appraised using a dose-volume histogram (DVH) and 3-dimensional (3D) gamma analysis with different distance to agreement (DTA) and dose difference criteria. The gamma agreement index (GAI) was denoted as G(∆D, DTA). Statistical analysis of t-test was utilized to evaluate the dose differences of different algorithms. Results: The phantom study demonstrated that titanium metal artifacts can be effectively reduced. The patient cases study showed that dose differences between the artifact-corrected and uncorrected datasets were small evaluated by gamma index and DVH. Gamma analysis found that even the strict criterion local G(1,1) had average values ≥93.9% for the three algorithms. For all DVH metrics, average differences did not exceed 0.7% in planning target volume (PTV) and 2.1% in planning risk volume of spinal cord (PRV-SC). Statistical analysis showed that the observed dose differences of MC method were significantly larger than those of AAA (P<0.01 for D98% of PTV and P<0.001 for D0.1cc of spinal cord) and AXB methods (P<0.001 for D98% and P<0.0001 for D0.1cc). Conclusions: Dosimetric impact of artifacts caused by titanium implants is not significant in spinal SBRT planning, which indicates that dose calculation algorithms might not be very sensitive to CT number variation caused by titanium inserts.

Define dose field to assess the modulation complexity of intensity-modulated radiation therapy.

PURPOSE: Introduce a new concept of dose field to assess the modulation complexity (MC) of intensity-modulated radiation therapy (IMRT). METHODS: A total of 91 IMRT plans for different diseases were retrospectively retrieved randomly from treatment database. The dose field of plans were calculated and feature values such as force magnitude and diversity were defined and extracted. Correlation analysis between these feature values and execution cost, delivery accuracy of plans was performed, to verify the validity of dose field in characterizing the MC. RESULTS: The feature values of dose field in different disease own significant differences (p < 0.001). For correlation analysis, number of control point (CP) and cumulative perimeter of CP have the highest correlation with angle entropy (0.815 and 0.848 respectively), while the correlation between number of monitor units(MU), cumulative area of CP and force, force entropy is higher than others (0.797-0.909). However, complexity of CP shape is almost irrelevant to all the dose field features. The gamma passing rate and the dose field features shows a weak negative correlation trend. CONCLUSIONS: Dose field can be used as a tool to assess the MC of IMRT.