Related Experiment Video
Updated: Aug 1, 2025
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
A novel discrete learning-based intelligent methodology for breast cancer classification purposes
Mehdi Khashei1, Negar Bakhtiarvand2
1Department of Industrial and Systems Engineering, Isfahan University of Technology (IUT), Isfahan, Iran; Center for Optimization and Intelligent Decision Making in Healthcare Systems (COID-Health), Isfahan University of Technology (IUT), Isfahan 8415683111, Iran.
This study introduces a novel discrete learning method for classification models, improving accuracy in breast cancer detection. The new discrete learning-based multilayer perceptron (DIMLP) model significantly outperforms traditional continuous learning methods.
Area of Science:
- Data Mining
- Machine Learning
- Artificial Intelligence
Background:
- Existing classification models use continuous cost functions, which are ill-suited for discrete objective functions.
- This mismatch leads to inefficiencies and suboptimal performance in classification tasks.
Purpose of the Study:
- To propose a novel classification methodology using a discrete cost function in the learning process.
- To implement and evaluate this methodology using a multilayer perceptron (MLP).
Main Methods:
- Developed a discrete learning-based MLP (DIMLP) model.
- Applied the DIMLP model to breast cancer classification datasets.
- Compared its performance against the conventional continuous learning-based MLP model.
Main Results:
- The DIMLP model achieved an average classification rate of 94.70%.
- This represents a 6.95% improvement over the traditional MLP model's 88.54% rate.
- DIMLP outperformed the MLP model across all tested datasets.
Conclusions:
- The proposed discrete cost function approach offers a more logical and efficient learning process for classification.
- The DIMLP model demonstrates superior performance, particularly for medical decision-making and other data mining applications requiring high accuracy.
More Related Videos
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
09:53Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography
Published on: August 16, 2020