Machine Learning Models for Early Prediction and Management of Dementia Related Disorders

Citation

Shubar, Abduelhakem and Ramakrishnan, Kannan and Ho, Chin Kuan (2022) Machine Learning Models for Early Prediction and Management of Dementia Related Disorders. In: Postgraduate Colloquium December 2022, 1-15 December 2022, Multimedia University, Malaysia. (Unpublished)

[img] Text
32.SHUBAR, ABDUELHAKEM G ABDUSALAM.pdf - Published Version
Restricted to Registered users only

Download (540kB)

Abstract

More than 55 million people diagnosed with dementia around the world in 2020. Dementia care costs over US$ 1.3 trillion annually world wide. It is anticipated to increase to US$ 2.8 trillion by 2030. 1 • A report reveals that majority of dementia cases between 2010 and 2050 reside in low and middle income countries. 2 • Majority of research work in regards to developing dementia prediction solutions are based on invasive and costly methods. • A model based on structural and non-structural EMR data achieved accuracy of 80%. 3 • Authors trained a minimal recurrent neural network (RNN) model to predict AD up to 6 years with an average 88.7% accuracy using one to 4 timesteps of multimodal AD and clinical diagnosis. 4 • A support vector machine model for dementia prediction was trained over a data set of personal medical history features. The model achieved F-score of 80.9%. 5 • A gradient boosting machine model was trained with administrative and EHR data to predict onset of dementia from 3 to 8 year.6

Item Type: Conference or Workshop Item (Poster)
Uncontrolled Keywords: Machine Learning
Subjects: Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 20 Dec 2022 03:34
Last Modified: 20 Dec 2022 03:34
URII: http://shdl.mmu.edu.my/id/eprint/10949

Downloads

Downloads per month over past year

View ItemEdit (login required)