Early Diagnosis of Alzheimer’s Disease Using ACO Optimized Deep CNN Classifier

Abstract

Alzheimer's disease (AD) is a human brain disease that remains as a common cause of dementia, which occurs mainly in middle-aged or grown-up individuals. AD results in cognitive decline and memory loss. AD is caused by the decomposition of plaques around the nerves of brains or around the brain cells, where the brain cells get neurofibrillary tangled and result in various instability and mental illness. AD is a chronic and irreversible disease; the reasons and disease identification are still not known, but research says it can be identified during the early stages. In spite of that fact, this research work has proposed a computer-aided Alzheimer's classification method that will classify the class of an image either in normal class or demented class. The method uses the hybrid strategy of ant colony optimization (ACO) and feed forward convolutional neural network (CNN or ConvNet); however, identifying the architecture of CNN requires lots of expertise and is time-consuming. Henceforth, this research work has used the bio-inspired optimization strategy, which will identify the optimal combination of hyper-parameters, i.e. it recommends the configuration for the CNN model, and with that, configuration of hyper-parameters with the CNN model is trained with the training dataset, and CNN performs feature extraction alongside classification for arranging the gatherings possibly, when the model undergoes validation, where the performance metric of the model is evaluated and to identify whether the validating images are falling in the category of normal class (i.e. non-demented) or Alzheimer's class (i.e. demented class) with good results or not, the classification error is measured during this phase and is backpropagated to ACO optimizer, iteratively ACO is used to minimize the classification error by tuning the hyper-parameters, and after few iterations, a CNN architecture with optimal hyper-parameters combinations is obtained to result in least classification errors. The method was applied to the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, which constitutes the fMRI images of Alzheimer's affected patients and resulted in developing efficient and state-of-the-art method for the classification of Alzheimer's disease. The proposed method performance metrics were recorded

[Retracted] Ant Colony Optimization‐Enabled CNN Deep Learning Technique for Accurate Detection of Cervical Cancer

“…According to ACO, which is one of the most recent approaches to approximation optimization, actual ants, which use the chemical signaling agent known as pheromone to communicate with one another, may use to find the shortest pathways between their nest and food sources. Pheromones are used by ants to communicate with one another [ 23 ].…”

Image classification and identification for rice leaf diseases based on improved WOACW_SimpleNet

“…ACO was used to optimize the hyperparameters in CNN, and the specific way was to back propagate the classification error in the iterative training of CNN model to the ACO, and finally obtained the CNN structure with the optimal combination of hyperparameters. This method was applied to Alzheimer’s disease neuroimaging initiative dataset by researchers, and the classification accuracy can reach 98.67% ( Singh and Janghel, 2022 ). Compared with other swarm intelligence optimization algorithms, the whale optimization algorithm (WOA) used in this paper is easy to implement, has fast convergence speed, high convergence accuracy, and is not easy to fall into the local optimal solution, so it is widely used in various fields.…”