人工智能
计算机科学
神经影像学
机器学习
人工神经网络
模式识别(心理学)
医学
疾病
鉴定(生物学)
特征(语言学)
医学影像学
作者
Sharmeen Saba M. Ahmed,Saundarya Vinodrao Raut,Hrushikesh Madhukar Panchabudhe,Prithvi Chawla,Ms. Sampa Barman,Praveen Kumar Dhankar
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2025-11-28
卷期号:: 1202-1207
标识
DOI:10.1109/sisimpact67725.2025.11439462
摘要
Alzheimer disease (AD) is a brain disorder that leads to cognitive impairment in the long run. Early diagnosis and provision of the correct treatment to individuals is extremely difficult. It is also an area of study interest to predict how diseases will deteriorate further as individuals shift to Alzheimer disease and early intervention can greatly determine their performance. It is now possible to combine dissimilar forms of neuroimaging data, such as structural MRI data, functional MRI data, and PET scan data, to observe the way diseases move in both clinical and functional ways using recent work in machine learning and deep learning. We develop models based on high-tech designs and fusion of multimodal features to forecast the progression of Alzheimer. One uses public datasets such as ADNI, and the process of editing the data involves eliminating noise, normalizing the data, and aligning the images. The removal and combination of multimodal features is done through fusion techniques, and it is thus capable of presenting anatomical, functional and metabolic signals in a robust manner. Experimentation results indicate that multimodal fusion-based mixed deep learning models fare better than machine learning baselines. They are also more successful in the detection of early changes and in predicting cognitive decline scores. This paper demonstrates how the use of prediction models based on machine learning would assist doctors in providing accurate information concerning the prognosis of patients with Alzheimer, classifying them into groups, and designing their own care.
科研通智能强力驱动
Strongly Powered by AbleSci AI