Progression Modeling of Cognitive Disease Using Temporal Data Mining: Research Landscape, Gaps and Solution Design

Abstract

Dementia is a cognitive disorder whose diagnosis and progression monitoring is very difficult due to a very slow onset and progression. It is difficult to detect whether cognitive decline is due to ageing process or due to some form of dementia as MRI scans of the brain cannot reliably differentiate between ageing related volume loss and pathological changes. Laboratory tests on blood or CSF samples have also not proved very useful. Alzheimer�s disease (AD) is recognized as the most common cause of dementia. Development of sensitive and reliable tool for evaluation in terms of early diagnosis and progression monitoring of AD is required. Since there is an absence of specific markers for predicting AD progression, there is a need to learn more about specific attributes and their temporal relationships that lead to this disease and determine progression from mild cognitive impairment to full blown AD. Various stages of disease and transitions from one stage to the have be modelled based on longitudinal patient data. This paper provides a critical review of the methods to understand disease progression modelling and determine factors leading to progression of AD from initial to final stages. Then the design of a machine learning based solution is proposed to handle the gaps in current research.