EfficientNet-B0 outperforms other CNNs in imagebased five-class embryo grading: a comparative analysis

• Published in: Journal of animal reproduction & biotechnology (Online) Vol. 39; no. 4; pp. 267 - 277
• Main Authors: Vincent Jaehyun Shim, Hosup Shim, Sangho Roh
• Format: Journal Article
• Language: English
• Published: 한국동물생명공학회(구 한국수정란이식학회) 31.12.2024
• Subjects: in vitro fertilization; deep learning; blastocyst; embryo; convolutional neural networks
• ISSN: 2671-4639; 2671-4663

Background: Evaluating embryo quality is crucial for the success of in vitro fertilization procedures. Traditional methods, such as the Gardner grading system, rely on subjective human assessment of morphological features, leading to potential inconsistencies and errors. Artificial intelligence-powe...

Table of Contents:

• ABSTRACT INTRODUCTION MATERIALS AND METHODS Dataset preparation Model selection Data preprocessing and augmentation Training procedure Evaluation metrics Grad-CAM visualization RESULTS Model comparison and performance overview Model training and performance evaluation ROC curve-based evaluation of classification models Error analysis using confusion matrices Interpretation of grad-CAM heatmaps and modelperformance DISCUSSION CONCLUSION REFERENCES