HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Design of data driven decision support systems for the early detection of subjects at risk to develop Alzheimer’s disease

Manon Ansart 1, 2
1 ARAMIS - Algorithms, models and methods for images and signals of the human brain
SU - Sorbonne Université, Inria de Paris, ICM - Institut du Cerveau et de la Moëlle Epinière = Brain and Spine Institute
Abstract : The goal of this thesis is to design data-driven methods to identify subjects at risk to develop Alzheimer's disease. As it is a progressive disease, subtle signs can appear several years before the first clinical symptoms. Identifying subjects who show these signs, and who are likely to develop the disease in the coming years, is a crucial point that could allow researchers to better study the disease mechanism, select patients for clinical trials and tailor patient care. In the first chapter, we conduct a review of methods predicting the future diagnosis of subjects suffering from mild cognitive impairment. We quantitatively and qualitatively study these methods, and take a critical view point by identifying several methodological issues. In the second chapter, we propose our own method to predict the future diagnosis by using a two-step approach: we first predict the future subject characteristics, and then use this result to predict the corresponding diagnosis. In the third chapter, we propose an automatic method to select subjects with a positive biomarker for clinical trials, so as to minimize the recruitment cost. In the last chapter, we analyze prescription patterns before and after diagnosis using a medical record database. We use them to predict if a patient will develop Alzheimer's disease in the next five or ten years. Across these works, we show the importance to take into account the adoption of these methods and the settings in which they can be used, especially regarding the test cohort, the data types and the interpretability of the method.
Complete list of metadata

Cited literature [319 references]  Display  Hide  Download

Contributor : Manon Ansart Connect in order to contact the contributor
Submitted on : Tuesday, December 31, 2019 - 10:15:53 AM
Last modification on : Friday, May 20, 2022 - 11:06:50 AM
Long-term archiving on: : Wednesday, April 1, 2020 - 12:46:16 PM


Files produced by the author(s)


  • HAL Id : tel-02425735, version 1


Manon Ansart. Design of data driven decision support systems for the early detection of subjects at risk to develop Alzheimer’s disease. Machine Learning [stat.ML]. Sorbonne Université, 2019. English. ⟨tel-02425735⟩



Record views


Files downloads