Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

RED-CNN

A Residual Encoder-Decoder Convolutional Neural Network (RED-CNN) is a deep learning architecture used in medical image processing to reduce noise and enhance image quality, particularly in low-dose computed tomography scans. The architecture pairs an encoder path of convolutional layers that extract high-level feature representations with a decoder path of deconvolutional layers that reconstruct the spatial image. Symmetrical residual shortcut connections link corresponding encoder and decoder stages, which facilitates model training and allows the network to preserve fine anatomical details, subtle lesions, and sharp structural edges while suppressing noise artifacts. By performing restoration directly on reconstructed image data, RED-CNN provides effective image denoising without requiring access to raw scanner projection data.

1 item

Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)

Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)

Hu Chen, Yi Zhang, Mannudeep K. Kalra, Feng Lin, Yang Chen, Peixi Liao, Jiliu Zhou, Ge Wang

OrganizationsMassachusetts General HospitalRensselaer Polytechnic InstituteSichuan UniversitySoutheast UniversityThe Sixth People's Hospital of Chengdu

Why you should read this

Introduces a residual encoder-decoder convolutional neural network (RED-CNN) that effectively suppresses noise, preserves anatomical details, and improves lesion detection directly from reconstructed low-dose CT images without requiring proprietary raw projection data.

Given the potential X-ray radiation risk to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. The current main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction, but they need to access original raw data whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, the deconvolution network, and shortcut connections into the residual encoder-decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation and lesion detection.

Added

2026-09-24