Predicting the conversion of mild cognitive impairment to Alzheimer's disease using feature engineering and deep learning on structural magnetic resonance images

• Grant number: 23/08307-0
• Support Opportunities: Regular Research Grants
• Start date: October 01, 2023
• End date: September 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Ricardo José Ferrari
• Grantee: Ricardo José Ferrari
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Associated researchers: Paulo Henrique Ferreira Bertolucci ; Roger Tam
• Associated scholarship(s): 25/02685-8 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT; 25/01795-4 - Web platform for simplified access to processing and analyzing neurological magnetic resonance imaging (MRI) images, designed to aid in the diagnosis of Alzheimer's disease., BP.TT; 25/05342-4 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT; + associated scholarships 24/04584-1 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT; 24/04437-9 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT; 24/04549-1 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT; 24/04595-3 - Web platform for simplified access to processing and analysis of neurological magnetic resonance images aimed at assisting in the diagnosis of Alzheimer's disease, BP.TT - associated scholarships

Abstract

With the increase in people's life expectancy, dementia has become a pressing global public health issue. Among the various types of dementia, Alzheimer's disease (AD) is the most common, accounting for nearly 70% of cases. The World Health Organization estimates that 35.6 million people had dementia in 2010, and this number is projected to double by 2030 (to 65.7 million) and reach 131.5 million by 2050. Currently, in Brazil, the number of people with dementia is estimated to be 1.2 million. Early detection and diagnosis of AD is crucial to implement appropriate treatments and improve the quality of life for patients. Mild cognitive impairment (MCI) is a condition characterized by subtle cognitive decline and is considered a prodromal stage of AD. Individuals with MCI have an increased risk of developing AD. Therefore, accurately recognizing patients with MCI who will develop AD in subsequent years is very important, as early identification of these patients will allow for early interventions and better management of the disease. Magnetic resonance imaging (MRI), a non-invasive imaging technique that can provide structural and functional information about the human brain, has been extensively used as a supportive diagnostic tool for AD, with findings suggesting that brain structural changes occur before the onset of cognitive symptoms. The proposed research will use a longitudinal approach to follow a cohort of individuals with MCI over a designated period of time, with the goal of assessing the conversion rate to AD. Structural MRI images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, at baseline and at regular intervals, will be used in this study. The images will undergo preprocessing, including noise reduction, bias field correction, and contrast and spatial standardization using a template image as a reference. Subsequently, feature engineering and deep learning techniques will be employed to extract relevant information from the images, ultimately leading to the development of prediction models. The proposed research will contribute to the understanding of the early brain structural changes that occur in AD and the potential of MR imaging as a diagnostic tool for early detection and prediction of conversion from MCI to AD. The findings of this study have the potential to inform early intervention strategies and improve the quality of life for individuals with MCI and AD. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

CARVALHO, JOANA; ABDOLLAHZADEH, ALI; FERRARI, RICARDO JOSE. Editorial: Deep learning and neuroimage processing in understanding neurological diseases. FRONTIERS IN COMPUTATIONAL NEUROSCIENCE, v. 18, p. 3-pg., 2024-12-19. (23/08307-0)