Related Experiment Video
Updated: Aug 13, 2025
09:11
Automation of the Micronucleus Assay Using Imaging Flow Cytometry and Artificial Intelligence
Published on: January 27, 2023
2.2K
Artificial Hummingbird Algorithm with Transfer-Learning-Based Mitotic Nuclei Classification on Histopathologic Breast
Areej A Malibari1, Marwa Obayya2, Abdulbaset Gaddah3
1Department of Industrial and Systems Engineering, College of Engineering, Princess Nourah bint Abdulrahman University, Riyadh 11671, Saudi Arabia.
Bioengineering (Basel, Switzerland)
|January 21, 2023
Summary
This study introduces an AI method for classifying mitotic nuclei in breast cancer images, improving cancer grading and diagnosis accuracy. The artificial hummingbird algorithm enhances classification performance for better medical applications.
Area of Science:
- Medical Image Processing
- Artificial Intelligence in Oncology
Background:
- Accurate mitotic nuclei estimation is crucial for breast cancer prognosis and grading.
- Automated analysis is challenging due to nuclei similarity and varied forms.
Purpose of the Study:
- To develop an artificial hummingbird algorithm with transfer-learning-based mitotic nuclei classification (AHBATL-MNC) for histopathologic breast cancer images.
- To accurately identify mitotic and nonmitotic nuclei in histopathology images (HIs).
Main Methods:
- Utilized PSPNet for HI segmentation to identify candidate mitotic patches.
- Employed Residual Network (ResNet) as a feature extractor and Extreme Gradient Boosting (XGBoost) as a classifier.
- Optimized XGBoost classifier parameters using the Artificial Hummingbird Algorithm (AHBA) for enhanced performance.
Main Results:
- The AHBATL-MNC system demonstrated enhanced outcomes compared to existing methods on medical imaging datasets.
- Achieved improved accuracy in classifying mitotic and nonmitotic nuclei.
Conclusions:
- The AHBATL-MNC technique offers a promising approach for automated mitotic nuclei identification in breast cancer histopathology.
- This AI-driven method can significantly aid in computer-aided diagnosis and improve cancer grading accuracy.
