| Autor: |
Khan, Ruksana, Saurabh, Amit, Bisht, Yashpal Singh, Rawat, Anand Singh, Aheer, Megha, Kumar, Deepak, Khan, Nazam, Thakur, Priya, Sharma, Vedika, Saini, Diksha, Kapoor, Himanshi, Pathania, Mamta |
| Předmět: |
|
| Zdroj: |
Library of Progress-Library Science, Information Technology & Computer; Jul-Dec2024, Vol. 44 Issue 3, p17039-17046, 8p |
| Abstrakt: |
Alzheimer's disease (AD) poses a substantial healthcare concern, particularly given its incidence among those aged 60 and up. Early detection of Alzheimer's disease remains difficult, with few good diagnostic techniques available. Furthermore, clinical trials for AD drugs have a high failure rate, and a definite solution remains elusive. AD evolves through various stages, ranging from extremely mild to severe, making it difficult to discern between them and resulting in increasing health issues, memory loss, and increased reliance on others for everyday activities. Early detection, particularly in mild instances, shows promise for guiding interventions to reduce disease progression and limit brain damage. Deep learning (DL) has emerged as a promising approach for early AD detection, although existing algorithms struggle to detect subtle changes in brain networks among individuals with mild dementia. Nonetheless, researchers are actively developing methods for early identification using MRI images. This study utilizes a dataset of 6400 MRI images and employs a DL algorithm with a neural network classifier for early AD diagnosis, achieving promising results in terms of accuracy, precision, recall, AUC, and F1-score. Comparative analysis with previous studies confirms the superior performance of the proposed model. Overall, this research contributes to the exploration of various machine learning (ML) and DL approaches applicable to AD stage identification. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
