Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Media Statistika
Ramifidiosa, Lucius
Unknown Affiliation
Decision Sciences, Operations Research & Management

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
PERFORMANCE OF NEURAL NETWORK IN PREDICTING MENTAL HEALTH STATUS OF PATIENTS WITH PULMONARY TUBERCULOSIS: A LONGITUDINAL STUDY Rahmanda, Lalu Ramzy; Fernandes, Adji Achmad Rinaldo; Solimun, Solimun; Ramifidiosa, Lucius; Zamelina, Armando Jacquis Federal
MEDIA STATISTIKA Vol 16, No 2 (2023): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/medstat.16.2.124-135

Abstract

Comorbidity between pulmonary tuberculosis and mental health status requires effective psychiatric treatment. This study aims to predict anxiety and depression levels in patients with pulmonary tuberculosis and consider future mental health treatment for patients. A sample of 60 pulmonary tuberculosis patients in Malang were involved and evaluated longitudinally every two weeks over 13 periods. In this study, we use the Generalized Neural Network Mixed Model (GNMM) to obtain better results in predicting anxiety and depression levels in patients with pulmonary tuberculosis and compare the results with the Generalized Linear Mixed Model (GLMM). The flexibility of GLMM in modeling longitudinal data, and the power of neural network in performing a prediction makes GNMM a powerful tool for predicting longitudinal data. The result shows that neural network's prediction performance is better than the classical GLMM with a smaller MSPE and fairly accurate prediction. The MSPEs of the three compared models: 1-Layer GNMM, 2-Layer, and GLMM, respectively are 0.0067, 0.0075, 0.0321 for the anxiety levels, and 0.0071, 0.0002, and 0.0775 for the depression levels. Furthermore, future research needs to investigate the data with a larger sample size or high dimensional data with large network architectures to prove the robustness of GNMM.
Co-Authors Fernandes, Adji Achmad Rinaldo Rahmanda, Lalu Ramzy Solimun, Solimun Zamelina, Armando Jacquis Federal