Related Experiment Video
Updated: Oct 7, 2025
08:58
Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning
Published on: November 19, 2018
12.7K
Classifying microscopic images as acute lymphoblastic leukemia by Resnet ensemble model and Taguchi method
Yao-Mei Chen1,2, Fu-I Chou3, Wen-Hsien Ho4,5
1School of Nursing, Kaohsiung Medical University, Kaohsiung, 807, Taiwan.
BMC Bioinformatics
|January 12, 2022
Summary
Researchers developed a Resnet101-9 ensemble model for detecting acute lymphoblastic leukemia (ALL) in microscopic images. This AI model demonstrated superior accuracy compared to individual Resnet-101 models, improving ALL classification.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Deep learning and artificial intelligence are being explored for rapid and accurate detection of diseases.
- Acute lymphoblastic leukemia (ALL) detection from microscopic images presents a significant challenge in hematopathology.
Purpose of the Study:
- To develop and evaluate a deep learning ensemble model for classifying acute lymphoblastic leukemia (ALL) in microscopic images.
- To improve the accuracy of ALL detection compared to individual deep learning models.
Main Methods:
- A Resnet101-9 ensemble model was created by combining nine trained Resnet-101 models using a majority voting strategy.
- Transfer learning and the Taguchi experimental method were employed to optimize hyperparameters for the Resnet-101 models.
- The C-NMC dataset was utilized for training and performance evaluation of the models.
Main Results:
- The Resnet101-9 ensemble model achieved an accuracy of 85.11% and an F1-score of 88.94 in classifying ALL.
- The ensemble model demonstrated superior accuracy and performance metrics (precision, recall, specificity) compared to individual Resnet-101 models.
- The proposed ensemble approach effectively enhanced the classification of ALL in microscopic images.
Conclusions:
- The Resnet101-9 ensemble model significantly outperforms individual Resnet-101 models in classifying acute lymphoblastic leukemia (ALL).
- This study highlights the potential of ensemble deep learning methods for accurate and efficient hematological disease detection from microscopic images.
