Predicting new-onset post-stroke depression from real-world data using machine learning algorithm

doi:10.3389/fpsyt.2023.1195586

CNU IR > College of Smart Living > Dept. of Hospital and Health (including master's program) > Periodical Articles > Item 310902800/34874

Please use this identifier to cite or link to this item: https://ir.cnu.edu.tw/handle/310902800/34874

Title:	Predicting new-onset post-stroke depression from real-world data using machine learning algorithm
Authors:	Chen, Yu-Ming Chen, Po-Cheng Lin, Wei-Che Hung, Kuo-Chuan Chen, Yang-Chieh Brian Hung, Chi-Fa Wang, Liang-Jen Wu, Ching-Nung Hsu, Chih-Wei Kao, Hung-Yu
Contributors:	Chang Gung Univ, Kaohsiung Chang Gung Mem Hosp, Dept Psychiat, Coll Med Chang Gung Univ, Kaohsiung Chang Gung Mem Hosp, Dept Phys Med & Rehabil, Coll Med Chang Gung Univ, Kaohsiung Chang Gung Mem Hosp, Dept Diagnost Radiol, Coll Med Chi Mei Med Ctr, Dept Anesthesiol Chia Nan Univ Pharm & Sci, Dept Hosp & Hlth Care Adm, Coll Recreat & Hlth Management Natl Sun Yat Sen Univ, Coll Med, Sch Med, Kaohsiung Natl Pingtung Univ Sci & Technol, Coll Humanities & Social Sci Chang Gung Univ, Kaohsiung Chang Gung Mem Hosp, Dept Child & Adolescent Psychiat, Coll Med Chang Gung Univ, Kaohsiung Chang Gung Mem Hosp, Dept Otolaryngol, Coll Med Natl Cheng Kung Univ, Coll Med, Dept Publ Hlth Natl Cheng Kung Univ, Dept Comp Sci & Informat Engn
Keywords:	artificial intelligence depressive disorder electronic medical record feature importance prediction
Date:	2023
Issue Date:	2024-12-25 11:04:55 (UTC+8)
Publisher:	FRONTIERS MEDIA SA
Abstract:	IntroductionPost-stroke depression (PSD) is a serious mental disorder after ischemic stroke. Early detection is important for clinical practice. This research aims to develop machine learning models to predict new-onset PSD using real-world data. MethodsWe collected data for ischemic stroke patients from multiple medical institutions in Taiwan between 2001 and 2019. We developed models from 61,460 patients and used 15,366 independent patients to test the models' performance by evaluating their specificities and sensitivities. The predicted targets were whether PSD occurred at 30, 90, 180, and 365 days post-stroke. We ranked the important clinical features in these models. ResultsIn the study's database sample, 1.3% of patients were diagnosed with PSD. The average specificity and sensitivity of these four models were 0.83-0.91 and 0.30-0.48, respectively. Ten features were listed as important features related to PSD at different time points, namely old age, high height, low weight post-stroke, higher diastolic blood pressure after stroke, no pre-stroke hypertension but post-stroke hypertension (new-onset hypertension), post-stroke sleep-wake disorders, post-stroke anxiety disorders, post-stroke hemiplegia, and lower blood urea nitrogen during stroke. DiscussionMachine learning models can provide as potential predictive tools for PSD and important factors are identified to alert clinicians for early detection of depression in high-risk stroke patients.
Relation:	Frontiers in Psychiatry, v.14, Article 1195586
Appears in Collections:	[Dept. of Hospital and Health (including master's program)] Periodical Articles

Files in This Item:

File	Description	Size	Format
fpsyt.2023.1195586.pdf		645Kb	Adobe PDF	45	View/Open
index.html		0Kb	HTML	155	View/Open

Loading...