EARLY DETECTION OF ALZHEIMERS DISEASE USING MACHINE LEARNING TECHNIQUES

Alzheimers Disease (AD) is believed to be the most widely recognized reason for dementia and it is assessed that lone 1-in-4 individuals with Alzheimers are accurately diagnosed in an opportune manner. While no authoritative fix is accessible, when the weakness is still mellow the side effects can be overseen and treatment is best when it is begun before critical downstream harm happens, i.e., at the phase of mild cognitive impairment (MCI) or considerably prior. AD is clinically analyzed by physical and neurological assessment, and through neuropsychological and intellectual tests. There is a need to grow better diagnostic tools, which is the thing that this postulation addresses. OASIS is an open access dataset accessible online for improving the determination strategy of Alzheimers malady. Information gathered at meeting is recorded and one point of the work in this theory is to investigate the utilization of machine learning strategies to produce a classifier that can help with screening new people for various phases of AD. Contrasted with the past work processes, our technique is fit for breaking down different classes in a single setting and requires less marked training samples and insignificant domain earlier information. A notable performance gain on classification of all diagnosis groups was accomplished in our examinations. The model is prepared at first on 416 analyzed cases got from OASIS database. We at that point test our prepared model on the whole arrangement of entries provided by OASIS dataset to affirm the accuracy of discovery by our framework. Our outcomes produce 94.37% exactness in AD detection and classificatio