International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 10 | Oct 2024
p-ISSN: 2395-0072
www.irjet.net
ALZHEIMER’S DISEASE DETECTION USING DEEP LEARNING Mrs. K. Soniya Lakshmi1, Varshini. S2, Varshini. V3, Sowmiya. S4, Kanmani. T5, Mrs. B. Ananthi6 1,6Assistant professor, Vivekanandha College of Engineering for Women, Tiruchengode, Tamilnadu(India), 2,3,4,5 UG Student of Vivekanandha College of Engineering for Women, Tiruchengode, TamilNadu(India).
---------------------------------------------------------------------***--------------------------------------------------------------------will analyze a dataset consisting of EEG recordings and brain Abstract— Alzheimer's disease, a gradual and degenerative
MRI images from healthy and Alzheimer's disease affected individuals Using metrics such as sensitivity, specificity, and area under the curve, we will compare the accuracy of various machine learning algorithms for detecting Alzheimer's disease. Our findings will contribute to the development of more objective and accurate methods for early detection and diagnosis of Alzheimer's disease, leading to improved treatment and management of this debilitating disease
brain illness, can cause cognitive decline as well behavioral changes. The condition affects not only the sufferer, but also those who care for them and society as a whole. This study uses convolutional neural networks (CNN) for MRI data and machine learning methods for EEG (electroencephalogram) data to classify new data as healthy or pathological. Our findings imply that integrating EEG and MRI (Magnetic Resonance Imagining ) data with data augmentation techniques can lead to more accurate and reliable strategies for early identification and diagnosis of Alzheimer's. Alzheimer's disease (AD) is a progressive neurological ailment that affects a large number of people worldwide. Early identification of Alzheimer's disease is critical for effective treatment and management. Deep learning approaches have showed potential in predicting Alzheimer's disease development using diverse indicators such as neuroimaging data, genetic markers, and clinical information. This paper provides a deep learning approach for predicting Alzheimer's disease utilizing multimodal data fusion approaches. To improve prediction accuracy, we present a unique neural network architecture that incorporates data from many modalities. Experimental results on a large dataset show that the suggested method is more effective than existing approaches at accurately forecasting AD progression.
1.1 Deep Learning: Deep learning represents a revolutionary approach to artificial intelligence, leveraging complex neural networks with multiple layers to automatically learn intricate patterns from vast amounts of data. At its heart are artificial neural networks, inspired by the structure of the human brain, with each layer processing information in a hierarchical manner to extract increasingly abstract features. Unlike traditional machine learning methods, deep learning eliminates the need for manual feature engineering by autonomously discovering relevant features from raw data. However, this power comes with computational demands, requiring substantial resources for training and handling large datasets. Despite challenges such as overfitting and interpretability, deep learning has fueled breakthroughs across diverse fields, including computer vision, natural language processing, healthcare, and finance. Its impact is felt in applications ranging from image classification and object detection to language translation and medical diagnosis, ushering in a new era of intelligent technology.
Keywords—CNN, MRI, EEG, ImageNet, Finetuning.
I.
INTRODUCTION
Alzheimer disease cause of dementia among the elderly and affects millions around the world. Alzheimer's disease is incurable, advances gradually or fast, and can result in loss of independence and death. Alzheimer's disease is diagnosed using memory loss, cognitive impairment, medical history, as well as physical and neurological testing. These methods are subjective and may fail to detect early diseases. Thus, Alzheimer's disease detection and diagnosis must be objective and precise. Recent advances in medical imaging and machine learning have enabled the diagnosis of Alzheimer's disease using EEG and MRI data. MRI produces high-resolution brain images, whereas EEG measures brain electrical activity non-invasively. Machine learning algorithms can detect indicators for Alzheimer's disease in large EEG and MRI datasets. These algorithms can recognize new EEG and MRI data as healthy. This paper aims to investigate the use of EEG and MRI data in the detection of Alzheimer's disease using machine learning techniques. We
© 2024, IRJET
|
Impact Factor value: 8.315
1.2 Objectives: Detect subtle patterns and biomarkers indicative of Alzheimer's disease in brain imaging scans, including MRI and PET scans Provide objective and consistent diagnostic predictions, reducing variability and subjectivity in diagnosis. Enable early detection of Alzheimer's disease, allowing for timely intervention and treatment to potentially slow disease progression and improve patient outcomes and Results will be rigorously evaluated using test datasets, with comparisons existing methods and discussions in limitations and potential improvements. 3.The core of the study will focus on designing and implementing a CNN architecture optimized for Alzheimer's detection, with considerations for layers, activation functions, and model evaluation metrics such as accuracy, precision, recall, and F1-score. The training process
|
ISO 9001:2008 Certified Journal
|
Page 766