Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 2, 2026

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Differentiation between nodules and end-on vessels using a convolution neural network architecture

J S Lin1, A Hasegawa, M T Freedman

  • 1Radiology Department, Georgetown University Medical Center, Washington, DC 20007, USA.

Journal of Digital Imaging
|August 1, 1995
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Overview of the Vascular System01:20

Overview of the Vascular System

The vascular system comprises an extensive network of arteries, capillaries, and veins. The vascular system can be broadly divided into the blood and lymphatic systems. Typically, blood vessels can be categorized into three histological regions: tunica intima, tunica media, and tunica adventitia. The tunica intima consists of a single layer of endothelial cells attached to the basal lamina. Underlying the basal lamina is a connective tissue layer and an elastic lamina that gives stability and...

You might also read

Related Articles

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

Sort by
Same author

Synthesis of sialyl Lewis X ganglioside analogues containing modified L-fucose residues.

Carbohydrate research·1995
Same author

[Blood levels of cyclosporine, acute rejections and the prognosis of the allografts in pediatric renal allograft recipients].

Nihon Hinyokika Gakkai zasshi. The japanese journal of urology·1995
Same author

Synthetic and structural studies of alpha-sialyl-(2-->6) and alpha-sialyl-(2-->3) 1-deoxynojirimycin derivatives potentially useful for biomedical applications.

Carbohydrate research·1995
Same author

Neuronal expression of a minor monosialosyl ganglioside GM1b in rat brain: immunochemical characterization using a specific monoclonal antibody.

Neuroscience research·1995
Same author

Synthesis of a sialyl Lewis X ganglioside analogue containing N-glycolyl in place of the N-acetyl group in the N-acetylneuraminic acid residue.

Bioscience, biotechnology, and biochemistry·1995
Same author

Reversal of lipopolysaccharide-analog-induced antibody suppression by anti-transforming growth factor beta and indomethacin.

Infection and immunity·1995
Same journal

Bayesian Convolutional Neural Networks in Medical Imaging Classification: A Promising Solution for Deep Learning Limits in Data Scarcity Scenarios.

Journal of digital imaging·2023
Same journal

Detecting and Characterizing Inferior Vena Cava Filters on Abdominal Computed Tomography with Data-Driven Computational Frameworks.

Journal of digital imaging·2023
Same journal

DMCA-GAN: Dual Multilevel Constrained Attention GAN for MRI-Based Hippocampus Segmentation.

Journal of digital imaging·2023
Same journal

Left Ventricular Myocardial Dysfunction Evaluation in Thalassemia Patients Using Echocardiographic Radiomic Features and Machine Learning Algorithms.

Journal of digital imaging·2023
Same journal

Public Imaging Datasets of Gastrointestinal Endoscopy for Artificial Intelligence: a Review.

Journal of digital imaging·2023
Same journal

External Validation of Robust Radiomic Signature to Predict 2-Year Overall Survival in Non-Small-Cell Lung Cancer.

Journal of digital imaging·2023
See all related articles

A novel convolution neural network (CNN) significantly improved lung nodule detection by differentiating nodules from end-on vessels on radiographs. This AI approach outperformed expert radiologists, offering a promising tool for medical image analysis.

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Computer-aided diagnosis (CAD) systems aim to enhance radiologist accuracy in detecting lung nodules on digital chest radiographs.
  • Reducing false-positive detections, particularly differentiating nodules from end-on vessels, remains a significant challenge in CAD development.
  • Traditional two-stage pattern recognition approaches struggle with defining and extracting optimal features for nodule classification.

Purpose of the Study:

  • To propose and evaluate a novel convolution neural network (CNN) architecture for improved lung nodule detection and differentiation from end-on vessels.
  • To address the limitations of conventional feature extraction methods in computer-aided diagnosis of lung nodules.
  • To compare the diagnostic performance of the proposed CNN with that of an expert radiologist.

Related Experiment Videos

Last Updated: Jul 2, 2026

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Main Methods:

  • A convolution neural network (CNN) architecture was developed, directly trained on raw radiographic images using locally responsive activation functions.
  • The CNN's performance was evaluated using the receiver operating characteristics (ROC) method.
  • Area under the curve (Az) was used as the primary performance index for comparison.

Main Results:

  • The CNN achieved a superior performance index (Az = 0.99) compared to an expert radiologist (Az = 0.83).
  • The CNN demonstrated significant effectiveness in differentiating lung nodules from challenging features like end-on vessels.
  • The proposed method overcomes difficulties in defining and extracting image features inherent in traditional CAD systems.

Conclusions:

  • The developed CNN architecture shows superior performance in lung nodule detection and differentiation compared to expert radiologists.
  • CNNs offer a powerful alternative to conventional pattern recognition methods, especially when image features are difficult to define.
  • This approach has potential applications in other medical imaging tasks, such as mammography for detecting microcalcifications or film defects.