Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge

The identification of latent risk factors that can induce to health risks or an abnormal status is an important task in healthcare data analyses. In recent years, health analyses based on neural network models have been applied widely. However, such analysis processes are blackbox and the results la...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruichen Cong, Shoji Nishimura, Atsushi Ogihara, Qun Jin
Format: Article
Language:English
Published: Tsinghua University Press 2025-04-01
Series:Big Data Mining and Analytics
Subjects:
Online Access:https://www.sciopen.com/article/10.26599/BDMA.2024.9020081
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850069245620125696
author Ruichen Cong
Shoji Nishimura
Atsushi Ogihara
Qun Jin
author_facet Ruichen Cong
Shoji Nishimura
Atsushi Ogihara
Qun Jin
author_sort Ruichen Cong
collection DOAJ
description The identification of latent risk factors that can induce to health risks or an abnormal status is an important task in healthcare data analyses. In recent years, health analyses based on neural network models have been applied widely. However, such analysis processes are blackbox and the results lack explainability. Some approaches by constructing a domain model may tackle these issues. However, domain knowledge from an expert is required. In this study, we propose an exploratory and interpretable approach to estimating latent health risk factors without relying on domain knowledge, in which feature selection and causal discovery are used to construct a domain model for uncovering complex relationships in health and medical data. An evaluation experiment conducted on two datasets by comparing the proposed approach with four baselines demonstrated that the proposed approach outperformed the baselines in terms of model fitness. Furthermore, the number of model parameters in our method was smaller than that in the baselines, which reduced model complexity. Moreover, the analysis process of the proposed approach was visible and explainable, which improved the interpretability of the analysis processes.
format Article
id doaj-art-471db1d04ee647dca9a12c2c5a0e8b57
institution DOAJ
issn 2096-0654
2097-406X
language English
publishDate 2025-04-01
publisher Tsinghua University Press
record_format Article
series Big Data Mining and Analytics
spelling doaj-art-471db1d04ee647dca9a12c2c5a0e8b572025-08-20T02:47:49ZengTsinghua University PressBig Data Mining and Analytics2096-06542097-406X2025-04-018244745710.26599/BDMA.2024.9020081Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain KnowledgeRuichen Cong0Shoji Nishimura1Atsushi Ogihara2Qun Jin3Graduate School of Human Sciences, Waseda University, Tokorozawa 359-1192, JapanFaculty of Human Sciences, Waseda University, Tokorozawa 359-1192, JapanFaculty of Human Sciences, Waseda University, Tokorozawa 359-1192, JapanFaculty of Human Sciences, Waseda University, Tokorozawa 359-1192, JapanThe identification of latent risk factors that can induce to health risks or an abnormal status is an important task in healthcare data analyses. In recent years, health analyses based on neural network models have been applied widely. However, such analysis processes are blackbox and the results lack explainability. Some approaches by constructing a domain model may tackle these issues. However, domain knowledge from an expert is required. In this study, we propose an exploratory and interpretable approach to estimating latent health risk factors without relying on domain knowledge, in which feature selection and causal discovery are used to construct a domain model for uncovering complex relationships in health and medical data. An evaluation experiment conducted on two datasets by comparing the proposed approach with four baselines demonstrated that the proposed approach outperformed the baselines in terms of model fitness. Furthermore, the number of model parameters in our method was smaller than that in the baselines, which reduced model complexity. Moreover, the analysis process of the proposed approach was visible and explainable, which improved the interpretability of the analysis processes.https://www.sciopen.com/article/10.26599/BDMA.2024.9020081health data analysislatent factor explorationinterpretable approachhealth risk estimation
spellingShingle Ruichen Cong
Shoji Nishimura
Atsushi Ogihara
Qun Jin
Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
Big Data Mining and Analytics
health data analysis
latent factor exploration
interpretable approach
health risk estimation
title Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
title_full Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
title_fullStr Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
title_full_unstemmed Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
title_short Exploratory and Interpretable Approach to Estimating Latent Health Risk Factors Without Using Domain Knowledge
title_sort exploratory and interpretable approach to estimating latent health risk factors without using domain knowledge
topic health data analysis
latent factor exploration
interpretable approach
health risk estimation
url https://www.sciopen.com/article/10.26599/BDMA.2024.9020081
work_keys_str_mv AT ruichencong exploratoryandinterpretableapproachtoestimatinglatenthealthriskfactorswithoutusingdomainknowledge
AT shojinishimura exploratoryandinterpretableapproachtoestimatinglatenthealthriskfactorswithoutusingdomainknowledge
AT atsushiogihara exploratoryandinterpretableapproachtoestimatinglatenthealthriskfactorswithoutusingdomainknowledge
AT qunjin exploratoryandinterpretableapproachtoestimatinglatenthealthriskfactorswithoutusingdomainknowledge