article · Frontiers in Computational Neuroscience
Alzheimer's disease causes progressive deterioration of memory and cognitive capabilities such as logic and recall. Standard diagnostic imaging techniques including CT, MRI, and PET scans can be time-consuming and prone to inaccuracies. To address these limitations, a deep learning approach was developed using a modified Inception architecture combined with transfer learning. The framework incorporates data pre-processing through normalisation and data addition to enhance image analysis. The model was trained and evaluated on an open Kaggle dataset containing 6,200 images categorised into non-demented, very mild demented, mild demented, and moderate demented stages. During testing, the architecture attained an accuracy of 94.92 percent and a sensitivity of 94.94 percent, exceeding the performance of existing state-of-the-art models. The system demonstrates the capability of automated computer vision to assist in identifying indicators of Alzheimer's disease from medical scans.
Conventional clinical diagnostics for Alzheimer's disease can be slow and susceptible to error. Applying deep learning models to neuroimaging allows faster and more reliable evaluation of cognitive decline. Demonstrating high accuracy across diverse disease stages supports earlier detection, which is vital for managing neurodegenerative progression and planning timely patient care.
This research is at an applied and tested stage using an open image dataset. It could enable the development of clinical decision-support software for radiologists and neurologists analysing MRI scans. Commercial translation would require further validation on real-world clinical datasets, but the algorithm provides a functional base for automated diagnostic screening tools.
AI-generated from the published abstract. Always read the original work before citing.
Alzheimer's disease (AD) is a neurodegenerative ailment, which gradually deteriorates memory and weakens the cognitive functions and capacities of the body, such as recall and logic. To diagnose this disease, CT, MRI, PET, etc. are used. However, these methods are time-consuming and sometimes yield inaccurate results. Thus, deep learning models are utilized, which are less time-consuming and yield results with better accuracy, and could be used with ease. This article proposes a transfer learning-based modified inception model with pre-processing methods of normalization and data addition. The proposed model achieved an accuracy of 94.92 and a sensitivity of 94.94. It is concluded from the results that the proposed model performs better than other state-of-the-art models. For training purposes, a Kaggle dataset was used comprising 6,200 images, with 896 mild demented (M.D) images, 64 moderate demented (Mod.D) images, and 3,200 non-demented (N.D) images, and 1,966 veritably mild demented (V.M.D) images. These models could be employed for developing clinically useful results that are suitable to descry announcements in MRI images.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3389/fncom.2022.1000435
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
