Coh-Metrix-Dementia: Automated Analysis of Language Impairment in Dementia using Natural Language Processing

• Grant number: 13/16182-0
• Support type: Scholarships in Brazil - Master
• Effective date (Start): February 01, 2014
• Effective date (End): March 29, 2015
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal researcher: Sandra Maria Aluísio
• Grantee: Andre Luiz Verucci da Cunha
• Home Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

According to the World Health Association, dementia is a costly social problem whose management will be a challenge on the coming decades. One of the possible forms of dementia is Primary Progressive Aphasia (PPA), a neurodegenerative disorder that predominantly affects language, and which comes in three variants: Progressive Non-Fluent Aphasia, Semantic Dementia and Logopenic Aphasia. Its diagnosis is based on patient speech analysis. Classical tests include fluency, naming and repetition; however, recent research increasingly recognizes discourse production analysis as a more suitable alternative. While qualitative discourse analysis can reveal the type of the disease presented by the patient, quantitative analysis can reveal the intensity of the existing brain damage. The greatest difficulty of such analysis is its effort requirement: the analysis of oral production is very laborious, which hinders its widespread adoption. In this scenario, computerized analysis emerge as a potential solution. Automated discourse analysis tools for language-affecting dementias already exist for the English language; however, to date no work in this direction has been undertaken in Portuguese. This project aims at creating an environment, called Coh-Metrix-Dementia, which will make use of resources and tools for Natural Language Processing (NLP) and Machine Learning to enable the automated analysis and recognition of PPA and its variants. This tool will supply a significant gap in tools for NLP in Brazilian Portuguese and provide physicians and researchers rapid and systematic discourse analysis; thus, fostering research in dementia detection and management.

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

CUNHA, Andre Luiz Verucci da. Coh-Metrix-Dementia: automatic analysis of language impairment in dementia using Natural Language Processing. 2015. Master's Dissertation - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação São Carlos.