ResSpikingNet: residual spiking neural network for early detection of Alzheimer's disease.
Authors
Affiliations (3)
Affiliations (3)
- Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Karunya University, Coimbatore, Tamil Nadu, India. Electronic address: [email protected].
- Department of Data Science and Cyber Security, Karunya Institute of Technology and Sciences, Karunya University, Coimbatore, Tamil Nadu, India.
- Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Karunya University, Coimbatore, Tamil Nadu, India.
Abstract
The common form of dementia is known as Alzheimer's disease (AD) that mostly affects the elderly individuals also, it is incurable and degenerative brain disease. According to progression level of AD, it led to memory loss initially and it affects functionality in different degrees. Even before the appearances of symptoms, AD cause severe damage in brain tissue and cell. Identifying AD at its earliest stages offers affected patients chance to get together with their healthcare providers, other support network members and loved ones to create personalized care plans, ensuring a more effective method to handle their condition. Here, ResSpikingNet has been developed to detect AD earlier using MRI image. Initially, the input image is given into the pre-processing process that is performed using median filter. After that, brain area segmentation is done by employing O-Segnet. Furthermore, the feature extraction techniques are employed to extract specific features, including ALDP with info gain, entropy, and statistical features. Finally, early detection of AD is done utilizing ResSpikingNet that is developed by the combination of Spiking ResNet and DRN. In addition, developed ResSpikingNet obtained maximal value of accuracy as 91.648%, sensitivity as 91.780%, and specificity as 91.938%.