Related Experiment Video
Updated: Sep 12, 2025
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
2.6K
Automatic Multi-Stage Classification Model for Fetal Ultrasound Images Based on EfficientNet.
Chen-Shen Shih1, Hung-Wen Chiu1
1Graduate Institute of Biomedical Informatics, TMU.
Studies in Health Technology and Informatics
|August 8, 2025
Summary
This study improves fetal ultrasound image classification using EfficientNet, a deep learning model. The AI shows high accuracy, especially for newborn stage images, aiding clinical decisions.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- Accurate fetal ultrasound classification is crucial for prenatal care.
- Deep learning models offer potential for improving diagnostic accuracy.
Purpose of the Study:
- To enhance fetal ultrasound image classification accuracy using EfficientNet.
- To evaluate model performance across early, midterm, and newborn pregnancy stages.
Main Methods:
- Data collection and preprocessing of fetal ultrasound images.
- Training and evaluation of the EfficientNet convolutional neural network.
- Comparative analysis of model performance at different gestational ages.
Main Results:
- EfficientNet demonstrated superior performance in fetal ultrasound image classification.
- The model achieved the highest accuracy in classifying images from the newborn stage.
- Deep learning significantly improved classification performance.
Conclusions:
- EfficientNet is a promising deep learning approach for fetal ultrasound analysis.
- Enhanced classification accuracy supports improved clinical workflows and prenatal diagnostics.
- AI-driven image analysis holds potential for advancing obstetric care.
