Affiliation:
1. Vellore Institute of Technology, India
2. University of Jaffna, Sri Lanka
Abstract
Artificial Neural Networks (ANNs) optimized with Particle Swarm Optimization (PSO) for predicting Alzheimer's disease have demonstrated reliability in estimating mild cognitive impairment (LSM). Traditional ANN training faces challenges such as slow learning rates and difficulty overcoming local minima. Integrating PSO, a Resquare Optimization Algorithm (ROA), enhances ANN performance. In our study, using a dataset of 12,130 preparation records and 51,642 test records, we trained ICA-ANN and ICA-PSO-ROA-ANN models. PSO parameters were optimized to maximize accuracy while minimizing computational load. Evaluation using Root-Mean-Squared Error (RMSE) showed that the ROA-PSO-ANN model consistently outperformed traditional ANN and hybrid models, highlighting its effectiveness in complex medical diagnostics for Alzheimer's disease prediction.
Reference21 articles.
1. ] Alickovic, E., Subasi, A., & Alzheimer’s Disease Neuroimaging Initiative. (2020). Automatic detection of alzheimer disease based on histogram and random forest. In CMBEBIH 2019: Proceedings of the International Conference on Medical and Biological Engineering, 16 ̶̶ 18 May 2019, Banja Luka, Bosnia and Herzegovina (pp. 91-96). Springer International Publishing.
2. Awate, G., Bangare, S., Pradeepini, G., & Patil, S. (2018). Detection of alzheimers disease from mri using convolutional neural network with tensorflow. arXiv preprint arXiv:1806.10170.
3. Badnjevic, A., Škrbić, R., & Pokvić, L. G. (Eds.). (2019). CMBEBIH 2019: Proceedings of the International Conference on Medical and Biological Engineering, 16 ̶̶ 18 May 2019, Banja Luka, Bosnia and Herzegovina (Vol. 73). Springer.
4. Ant colony optimization
5. Multiclass Classification of Brain Cancer with Machine Learning Algorithms