Related Experiment Video
Updated: Jan 12, 2026
12:06
Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
Published on: March 3, 2023
4.6K
Deep-learning based model for sperm morphology assessment using the SMD/MSS dataset
Olfa Abdelkefi1,2, Rania Maalej3, Tarek Rebai2
1Reproductive Biology Service, Hedi Chaker University Hospital, Sfax, Tunisia.
Future Science OA
|November 7, 2025
Summary
This study developed an artificial intelligence model to automate sperm morphology assessment, improving standardization and efficiency in semen analysis. The deep learning approach enhances diagnostic accuracy in reproductive biology.
Area of Science:
- Reproductive Biology
- Medical Imaging
- Artificial Intelligence
Background:
- Sperm morphology assessment is subjective and difficult to standardize.
- Operator expertise significantly influences manual evaluation.
- Developing objective methods is crucial for accurate semen analysis.
Purpose of the Study:
- To create a predictive model for sperm morphology evaluation.
- To address the standardization challenges in manual assessment.
- To utilize artificial neural networks for objective analysis.
Main Methods:
- A Convolutional Neural Network (CNN) model was developed.
- The model was trained on the Sperm Morphology Dataset/Medical School of Sfax (SMD/MSS) dataset.
- Data augmentation techniques were used to expand the dataset from 1000 to 6035 images.
Main Results:
- The deep learning model achieved satisfactory accuracy, ranging from 55% to 92%.
- Data augmentation significantly increased the dataset size.
- The CNN model demonstrated effectiveness in spermatozoa classification.
Conclusions:
- The AI approach automates and standardizes semen analysis.
- This technology accelerates the process of sperm morphology evaluation.
- Artificial intelligence holds significant potential in reproductive biology applications.
