An Extreme Learning Machine based Approach to Detect the Alzheimer's Disease

Ensemble Extreme Learning Algorithms for Alzheimer’s Disease Detection

“…The same Preprocessing steps used in our work earlier [1] are used for processing the images. The steps involved are initially Preprocessing, segmenting and feature extracting leading the vectors into our desired ELM classifier.…”

“…This paper is an extension of the originally submitted work in the conference "2021 IEEE 4th International Conference on Computing, Power and Communication Technologies (GUCON). [1] We used Modified Extreme Learning Machine algorithms to study the training computational requirements in comparison to conventional machine learning methods. [2,3] The variations of ELMs used were, Regularized ELM (RELM) and MLPELM [4,5].…”

“…Multiclass Adaboost ELM (MAELM), Bagging Stepwise ELM and Negative Correlation ELM are suitable for imbalanced data. They are simple, cost effective, supporting many kernels or feature mapping functions for multi class data.Knowing ELMs require lesser time than conventional networks for training from our study performed earlier [1], our objective here is to incorporate the above modifications to ELMs and study the classification performance on the same OASIS Dataset.We also have compared the performance of other state of art techniques like CNNs and SVM. We have studied different Ensemble ELMs like ridge bagging, boosting, (Adaboost ELM )AELM and Negative Correlation ELMs(Incremental ELM) IELM.…”

“…It can be implemented and tested on PET (Positron Emission Tomography) Data. Can further work on improving on timing constraints [1].The time required for execution can be optimized.…”