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

Search research articles

Related Experiment Video

Updated: Nov 19, 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

Wavelet decomposition facilitates training on small datasets for medical image classification by deep learning.

Axel H Masquelin¹, Nicholas Cheney², C Matthew Kinsey³

¹University of Vermont, Burlington, VT, USA.

Histochemistry and Cell Biology

|January 27, 2021

Summary

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

Assessment of pulmonary function in a mouse model of bronchopulmonary dysplasia by unrestrained whole body plethysmography and the forced oscillation technique.

American journal of physiology. Lung cellular and molecular physiology·2026

Same author

Mechanical ventilation energy analysis: Recruitment focuses injurious power in the ventilated lung.

Proceedings of the National Academy of Sciences of the United States of America·2025

Same author

Intratumoral Chemotherapy: The Effects of Drug Concentration and Dose Apportioning on Tumor Cell Injury.

Bioengineering (Basel, Switzerland)·2024

Same author

Nature is resource, playground, and gift: What artificial intelligence reveals about human-Nature relationships.

PloS one·2024

Same author

LDCT image biomarkers that matter most for the deep learning classification of indeterminate pulmonary nodules.

Cancer biomarkers : section A of Disease markers·2024

Same author

Initial Safety and Feasibility Results From a Phase 1, Diagnose-and-Treat Trial of Neoadjuvant Intratumoral Cisplatin for Stage IV NSCLC.

JTO clinical and research reports·2024

Discrete wavelet transforms (DWTs) show promise in improving lung nodule classification for cancer screening. This method offers better performance and faster training than convolutional neural networks (CNNs) in limited datasets.

Area of Science:

Medical Imaging Analysis
Artificial Intelligence in Healthcare
Radiology

Background:

Low-dose computed tomography (LDCT) reduces lung cancer mortality but increases false positives.
Convolutional neural networks (CNNs) show potential for improving nodule detection but struggle with limited data, leading to overfitting and poor generalizability.

Purpose of the Study:

To compare the efficacy of discrete wavelet transforms (DWTs) against CNNs for classifying malignant versus benign lung nodules.
To explore DWTs as an alternative to convolutional layers to reduce model parameters and overfitting risk.

Main Methods:

Utilized computed tomography images from the National Lung Screening Trial (NLST).
Compared multi-level DWTs against standard convolutional layers within a CNN architecture.

Keywords:

Area under the AUC curve Convolutional neural network Learning rate Lung cancer detection

Related Experiment Videos

Last Updated: Nov 19, 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

Evaluated classification performance using area under the receiver-operating curve (AUC).

Main Results:

Multi-level DWTs outperformed convolutional layers, achieving AUCs of 94% and 92%, respectively.
DWTs significantly reduced the number of network parameters compared to CNNs.
DWTs demonstrated a substantially faster convergence rate during training.

Conclusions:

Multi-level DWTs offer a viable alternative to early convolutional layers in deep neural networks (DNNs) for lung nodule classification.
DWT-based approaches may enhance image classification performance in data-limited medical domains.