Alzheimer's Disease Detection Research Project

Overview

This project focuses on the detection of Alzheimer's disease using a dataset, employing fuzzy rough set theory along with machine learning algorithms such as Support Vector Machine (SVM) and k-Nearest Neighbors (KNN). The primary objective is to develop a reliable predictive model for early detection of Alzheimer's disease.

Objectives

• To utilize fuzzy rough set theory for feature selection and data preprocessing.
• To implement and compare the performance of SVM and KNN algorithms in predicting Alzheimer's disease.
• To achieve high accuracy in disease prediction, with a benchmark set at 90% accuracy.

Dataset

The dataset used in this research is available on Kaggle and includes various features related to Alzheimer's disease indicators. You can access the dataset here.

Methodology

1. Data Preprocessing:

   • Applied fuzzy rough set theory for feature selection and noise reduction.
   • Normalized and standardized the data for optimal algorithm performance.

2. Machine Learning Algorithms:

   • Support Vector Machine (SVM): Implemented with appropriate kernel functions and parameters to maximize accuracy.
   • k-Nearest Neighbors (KNN): Configured with optimal 'k' value to enhance prediction capabilities.

3. Model Training and Evaluation:

   • Split the dataset into training and testing sets.
   • Trained SVM and KNN models on the training set.
   • Evaluated the models on the testing set to measure performance metrics.

Results

• The SVM model achieved an accuracy of 90% in predicting Alzheimer's disease, meeting the project’s benchmark.
• Comparative analysis showed that SVM outperformed KNN in terms of accuracy and robustness.

Conclusion

The research successfully demonstrates the application of fuzzy rough set theory and machine learning algorithms in detecting Alzheimer's disease with high accuracy. The SVM model, in particular, proved to be an effective tool for early diagnosis, achieving a 90% accuracy rate.

Future Work

• Explore other machine learning algorithms and ensemble methods to further improve prediction accuracy.
• Expand the dataset to include more diverse samples for better generalization.
• Investigate the integration of additional features and data types, such as imaging data, for comprehensive analysis.

Acknowledgments

We extend our gratitude to the data providers and all individuals who contributed to this research project.

Contact

For any questions or further information, please contact:

• Researcher Name: Atharva Narkhede
• Email: [email protected]