Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

MRI-based assessment of tumor aggressiveness in nasopharyngeal carcinoma: risk stratification and survival prediction.

European radiology·2026

Same author

Differentiation of hepatocellular carcinoma from hepatic hemangioma using diffusion-derived vessel density map-based radiomics features.

Frontiers in oncology·2026

Same author

Synchronous metastatic differentiated high-grade thyroid carcinoma in lymph node from classic papillary thyroid carcinoma: a case report and literature review.

Frontiers in oncology·2026

Same author

Lung motion estimation from 4D CT using structure-tensor-guided finite-element digital volume correlation.

Physics in medicine and biology·2026

Same author

Constructing ultra-lightweight, thermally insulating and intrinsically flame-retardant all-wood aerogels via a top-down approach.

International journal of biological macromolecules·2026

Same author

Diffusion weighted imaging-based tumor growth rate for predicting long-term survival in nasopharyngeal carcinoma.

Cancer imaging : the official publication of the International Cancer Imaging Society·2026

Search research articles

Related Experiment Video

Updated: Jul 5, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

A two-stage deep-learning framework for CT denoising based on a clinically structure-unaligned paired data set.

Ruibao Hu^1,2, Yongsheng Xie³, Lulu Zhang¹

¹Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China.

Quantitative Imaging in Medicine and Surgery

|January 15, 2024

Summary

This summary is machine-generated.

This study introduces a novel deep learning method for low-dose computed tomography (LDCT) denoising, improving lung cancer screening by enhancing soft tissue visualization. The method effectively removes noise while retaining crucial details in structure-unaligned clinical data.

Keywords:

Computed tomography (CT)Wasserstein generative adversarial network (WGAN)attention mechanism structure-unaligned image

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Related Experiment Videos

Last Updated: Jul 5, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Area of Science:

Medical Imaging
Artificial Intelligence
Radiology

Background:

Low-dose computed tomography (LDCT) lung cancer screening suffers from high noise, obscuring soft tissue details.
Existing deep learning denoising methods often require structurally aligned data, which is not clinically feasible.
This study addresses the need for denoising using clinically available, structure-unaligned paired LDCT and normal-dose CT (NDCT) scans.

Purpose of the Study:

To develop and evaluate a novel deep learning-based method for denoising LDCT images.
To improve the detection and characterization of soft tissue lesions in LDCT lung cancer screening.
To utilize clinically realistic, structure-unaligned paired LDCT and NDCT data for training.

Main Methods:

A two-stage training approach using a Wasserstein generative adversarial network (WGAN) with an attention mechanism was employed.
Initially, Gaussian noise was added to NDCT data to simulate LDCT for generator training.
The model was then fine-tuned on a dataset of structure-unaligned, but paired, clinical LDCT and NDCT scans from the same patients.

Main Results:

The proposed method significantly outperformed existing techniques (CycleGAN, Pixel2Pixel, BM3D) in noise removal and detail retention, achieving approximately 7% improvement in PSNR, SSIM, and RMSE.
The denoised CT output closely matched the probability density profile of reference NDCT scans.
A two-stage WGAN approach demonstrated superior performance over a one-stage model in both objective and subjective evaluations.

Conclusions:

The developed method effectively reduces noise in LDCT scans while preserving essential details.
This technique has the potential to enhance lesion detection and characterization in soft tissues within LDCT lung cancer screening.
The use of structure-unaligned paired data represents a significant advancement for clinically applicable LDCT denoising.