Related Experiment Video
Updated: Jun 9, 2025
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
1.8K
Anomaly Detection in Embryo Development and Morphology Using Medical Computer Vision-Aided Swin Transformer with
Alanoud Al Mazroa1, Mashael Maashi2, Yahia Said3
1Department of Information Systems, College of Computer and Information Sciences, Princess Nourah Bint Abdulrahman University (PNU), P.O. Box 84428, Riyadh 11671, Saudi Arabia.
Bioengineering (Basel, Switzerland)
|October 25, 2024
Summary
This study introduces a new AI technique, EDMCV-STBDTO, for assessing embryo development in fertility treatments. The method improves accuracy and efficiency over traditional embryo evaluation, aiding in better assisted reproduction outcomes.
Area of Science:
- Reproductive biology
- Artificial intelligence in medicine
- Computer vision for medical imaging
Background:
- Infertility affects many individuals, with in vitro fertilization (IVF) being a key assisted reproduction technology.
- Current embryo assessment relies on manual microscopic evaluation, which is time-consuming, labor-intensive, and prone to subjective bias.
- Advancements in artificial intelligence (AI) and computer vision (CV) offer potential solutions to enhance the accuracy and efficiency of embryo evaluation.
Purpose of the Study:
- To develop and validate an advanced AI-driven technique, EDMCV-STBDTO, for accurate and efficient detection of human embryo development and morphology.
- To improve the quality assessment of embryos for successful IVF outcomes and advance developmental biology research.
- To overcome the limitations of traditional manual embryo assessment methods.
Main Methods:
- The proposed Embryo Development and Morphology Using a Computer Vision-Aided Swin Transformer with a Boosted Dipper-Throated Optimization (EDMCV-STBDTO) technique was employed.
- Image preprocessing utilized a bilateral filter (BF) for noise reduction.
- Feature extraction was performed using the swin transformer method, followed by variational autoencoder (VAE) for embryo development classification.
- Hyperparameter optimization for the VAE was achieved using the boosted dipper-throated optimization (BDTO) technique.
Main Results:
- The EDMCV-STBDTO technique demonstrated superior performance in accurately detecting embryo development compared to existing methods.
- Comprehensive validation using a benchmark dataset confirmed the method's effectiveness and efficiency.
- The AI-driven approach showed significant improvements over traditional manual embryo assessment.
Conclusions:
- The EDMCV-STBDTO technique presents a robust and efficient AI-based solution for embryo development assessment in assisted reproduction.
- This advanced computer vision approach has the potential to significantly improve IVF success rates and contribute to developmental biology research.
- The study highlights the transformative impact of AI and deep learning in modern fertility treatments.
