Deep Learning Architectures for Medical Image Denoising: A Comparative Study of CNN-DAE, CADTra, and DCMIEDNet
This study addresses noise reduction in medical imaging to improve diagnostic accuracy, but it is incremental as it compares existing architectures without introducing new methods.
This paper tackled the problem of noise contamination in medical MRI brain images by comparing three deep learning architectures for denoising, finding that DCMIEDNet performed best at lower noise levels with PSNR values up to 32.921 dB, while CADTra was more robust under severe noise with a PSNR of 27.671 dB, and all models outperformed traditional methods by 5-8 dB.
Medical imaging modalities are inherently susceptible to noise contamination that degrades diagnostic utility and clinical assessment accuracy. This paper presents a comprehensive comparative evaluation of three state-of-the-art deep learning architectures for MRI brain image denoising: CNN-DAE, CADTra, and DCMIEDNet. We systematically evaluate these models across multiple Gaussian noise intensities ($σ= 10, 15, 25$) using the Figshare MRI Brain Dataset. Our experimental results demonstrate that DCMIEDNet achieves superior performance at lower noise levels, with PSNR values of $32.921 \pm 2.350$ dB and $30.943 \pm 2.339$ dB for $σ= 10$ and $15$ respectively. However, CADTra exhibits greater robustness under severe noise conditions ($σ= 25$), achieving the highest PSNR of $27.671 \pm 2.091$ dB. All deep learning approaches significantly outperform traditional wavelet-based methods, with improvements ranging from 5-8 dB across tested conditions. This study establishes quantitative benchmarks for medical image denoising and provides insights into architecture-specific strengths for varying noise intensities.