mlsanityData to 14 Oct 2026

CTLformer: A Hybrid Denoising Model Combining Convolutional Layers and Self-Attention for Enhanced CT Image Reconstruction

arXiv 2505.12203 ↗First posted 18 May 2025eess.IV cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Low-dose CT (LDCT) images are often accompanied by significant noise, which negatively impacts image quality and subsequent diagnostic accuracy. To address the challenges of multi-scale feature fusion and diverse noise distribution patterns in LDCT denoising, this paper introduces an innovative model, CTLformer, which combines convolutional structures with transformer architecture. Two key innovations are proposed: a multi-scale attention mechanism and a dynamic attention control mechanism. The multi-scale attention mechanism, implemented through the Token2Token mechanism and self-attention interaction modules, effectively captures both fine details and global structures at different scales, enhancing relevant features and suppressing noise. The dynamic attention control mechanism adapts the attention distribution based on the noise characteristics of the input image, focusing on high-noise regions while preserving details in low-noise areas, thereby enhancing robustness and improving denoising performance. Furthermore, CTLformer integrates convolutional layers for efficient feature extraction and uses overlapping inference to mitigate boundary artifacts, further strengthening its denoising capability. Experimental results on the 2016 National Institutes of Health AAPM Mayo Clinic LDCT Challenge dataset demonstrate that CTLformer significantly outperforms existing methods in both denoising performance and model efficiency, greatly improving the quality of LDCT images. The proposed CTLformer not only provides an efficient solution for LDCT denoising but also shows broad potential in medical image analysis, especially for clinical applications dealing with complex noise patterns.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 8 of its 12 references that are in the map. First posted on arXiv 18 May 2025.

Builds on

8 of 12 references.

References · 8

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2017Non-local Neural Networkscited by 2,293
  5. 2017Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)cited by 224
  6. 2022CTformer: Convolution-free Token2Token Dilated Vision Transformer for Low-dose CT Denoisingcited by 46
  7. 2018On a Sparse Shortcut Topology of Artificial Neural Networkscited by 20
  8. 2020CT Reconstruction with PDF: Parameter-Dependent Framework for Multiple Scanning Geometries and Dose Levelscited by 2

Built on by

Cited by year

Citing papers per year: 2025: 7.

7 papers.

  1. 2025LLM-Driven E-Commerce Marketing Content Optimization: Balancing Creativity and Conversioncited by 7
  2. 2025Research on Personalized Financial Product Recommendation by Integrating Large Language Models and Graph Neural Networkscited by 6
  3. 2025Research on Model Parallelism and Data Parallelism Optimization Methods in Large Language Model-Based Recommendation Systemscited by 6
  4. 2025Meta-Learning for Cold-Start Personalization in Prompt-Tuned LLMscited by 5
  5. 2025Deep Learning Model Acceleration and Optimization Strategies for Real-Time Recommendation Systemscited by 3
  6. 2025Credit Risk Analysis for SMEs Using Graph Neural Networks in Supply Chaincited by 2
  7. 2025LLM-Augmented Symptom Analysis for Cardiovascular Disease Risk Prediction: A Clinical NLPcited by 2