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: Dec 23, 2025

Modeling Breast Cancer via an Intraductal Injection of Cre-expressing Adenovirus into the Mouse Mammary Gland

Modeling Breast Cancer via an Intraductal Injection of Cre-expressing Adenovirus into the Mouse Mammary Gland

Published on: June 7, 2019

Improving breast mass classification by shared data with domain transformation using a generative adversarial

Chisako Muramatsu¹, Mizuho Nishio², Takuma Goto³

¹Faculty of Data Science, Shiga University, 1-1-1 Banba, Hikone, Shiga, 522-8522, Japan.

Computers in Biology and Medicine

|April 28, 2020

Summary

Related Concept Videos

Aggregates Classification

Aggregates Classification

Aggregate classification is generally based on its size, petrographic characteristics, weight, and source. Size classification ranges from coarse to fine aggregates, defined by the size of the particles. Coarse aggregates are particles that do not pass through ASTM sieve No. 4, and aggregates that pass through the sieve are fine aggregates.
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...

You might also read

Related Articles

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

Sort by

Same author

Generating Findings for Jaw Cysts in Dental Panoramic Radiographs Using a GPT-Based VLM: A Preliminary Study on Building a Two-Stage Self-Correction Loop with a Structured Output (SLSO) Framework.

Diagnostics (Basel, Switzerland)·2026

Same author

Establishing a sustainable protocol for massive transfusion using thawed plasma and cryoprecipitate reconstituted from expired thawed plasma.

Transfusion·2026

Same author

[Site-specific Segmentation of Skeletal Muscles in Body CT Images via Comprehensive Muscular Consideration].

Nihon Hoshasen Gijutsu Gakkai zasshi·2026

Same author

[1. Utilization of Data and AI in Society (1): Ongoing Social Changes].

Nihon Hoshasen Gijutsu Gakkai zasshi·2026

Same author

Atypical periprosthetic femoral fractures around cemented stems: Radiographic features and clinical implications from a multicenter study.

Bone·2025

Same author

Retrieving the intravascular ultrasound catheter in the coronary artery by a gooseneck snare: a case report.

European heart journal. Case reports·2025

Synthetic medical images generated via domain transformation can augment limited datasets for training convolutional neural networks (CNNs). This approach shows promise for improving classification performance in medical imaging tasks.

Area of Science:

Medical Imaging and Artificial Intelligence
Machine Learning in Healthcare

Background:

Training deep learning models like convolutional neural networks (CNNs) typically demands extensive datasets.
Acquiring large, diverse medical image datasets for training is a significant challenge.

Purpose of the Study:

To evaluate the effectiveness of synthetic medical images in training CNNs.
To demonstrate the applicability of unrelated medical images through domain transformation for enhancing CNN training.

Main Methods:

A cycle generative adversarial network (GAN) was employed to create synthetic mammogram data from lung nodule CT scans and existing mammogram datasets (DDSM).
A CNN was trained to classify benign and malignant breast masses using various datasets: original, augmented, synthetic, DDSM, and natural images (ImageNet).

Keywords:

Classification Deep learning Generative adversarial network Mammography ROC curve

Related Experiment Videos

Last Updated: Dec 23, 2025

Modeling Breast Cancer via an Intraductal Injection of Cre-expressing Adenovirus into the Mouse Mammary Gland

Modeling Breast Cancer via an Intraductal Injection of Cre-expressing Adenovirus into the Mouse Mammary Gland

Published on: June 7, 2019

Performance was assessed using classification accuracy and area under the receiver operating characteristic curves (AUC).

Main Results:

Data augmentation improved classification accuracy from 65.7% to 67.1%.
Pretraining with an ImageNet model yielded 79.2% accuracy; pretraining with synthetic or DDSM images resulted in 67.6% and 72.5% accuracy, respectively.
Combining an ImageNet pretrained model with synthetic images slightly improved classification performance to 81.4%, comparable to using DDSM images.

Conclusions:

Synthetic data generated from unrelated lesions via domain transformation can effectively supplement training datasets for CNNs.
This method offers a viable strategy to increase training sample size, potentially improving diagnostic performance in medical image analysis.