Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data

Depression and anxiety are common comorbidities of stroke. Research has shown that about 30% of stroke survivors develop depression and about 20% develop anxiety. Stroke survivors with such adverse mental outcomes are often attributed to poorer health outcomes, such as higher mortality rates. The ob...

Full description

Saved in:
Bibliographic Details
Main Authors: Chien Wei Oei, Eddie Yin Kwee Ng, Matthew Hok Shan Ng, Yam Meng Chan, Vinithasree Subbhuraam, Lai Gwen Chan, U. Rajendra Acharya
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Bioengineering
Subjects:
Online Access:https://www.mdpi.com/2306-5354/12/5/517
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850257354381066240
author Chien Wei Oei
Eddie Yin Kwee Ng
Matthew Hok Shan Ng
Yam Meng Chan
Vinithasree Subbhuraam
Lai Gwen Chan
U. Rajendra Acharya
author_facet Chien Wei Oei
Eddie Yin Kwee Ng
Matthew Hok Shan Ng
Yam Meng Chan
Vinithasree Subbhuraam
Lai Gwen Chan
U. Rajendra Acharya
author_sort Chien Wei Oei
collection DOAJ
description Depression and anxiety are common comorbidities of stroke. Research has shown that about 30% of stroke survivors develop depression and about 20% develop anxiety. Stroke survivors with such adverse mental outcomes are often attributed to poorer health outcomes, such as higher mortality rates. The objective of this study is to use deep learning (DL) methods to predict the risk of a stroke survivor experiencing post-stroke depression and/or post-stroke anxiety, which is collectively known as post-stroke adverse mental outcomes (PSAMO). This study studied 179 patients with stroke, who were further classified into PSAMO versus no PSAMO group based on the results of validated depression and anxiety questionnaires, which are the industry’s gold standard. This study collected demographic and sociological data, quality of life scores, stroke-related information, medical and medication history, and comorbidities. In addition, sequential data such as daily lab results taken seven consecutive days after admission are also collected. The combination of using DL algorithms, such as multi-layer perceptron (MLP) and long short-term memory (LSTM), which can process complex patterns in the data, and the inclusion of new data types, such as sequential data, helped to improve model performance. Accurate prediction of PSAMO helps clinicians make early intervention care plans and potentially reduce the incidence of PSAMO.
format Article
id doaj-art-c40240dad8ce4c0a982af844e331ad3b
institution OA Journals
issn 2306-5354
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Bioengineering
spelling doaj-art-c40240dad8ce4c0a982af844e331ad3b2025-08-20T01:56:25ZengMDPI AGBioengineering2306-53542025-05-0112551710.3390/bioengineering12050517Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential DataChien Wei Oei0Eddie Yin Kwee Ng1Matthew Hok Shan Ng2Yam Meng Chan3Vinithasree Subbhuraam4Lai Gwen Chan5U. Rajendra Acharya6Management Information Department, Office of Clinical Epidemiology, Analytics and kNowledge (OCEAN), Tan Tock Seng Hospital, Singapore 308433, SingaporeSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeRehabilitation Research Institute of Singapore, Nanyang Technological University, Singapore 308232, SingaporeDepartment of General Surgery, Vascular Surgery Service, Tan Tock Seng Hospital, Singapore 308433, SingaporeThe Digital Health Hub, Austin, TX 78944, USADepartment of Psychiatry, Tan Tock Seng Hospital, Singapore 308433, SingaporeSchool of Mathematics, Physics and Computing, University of Southern Queensland, Brisbane, QLD 4305, AustraliaDepression and anxiety are common comorbidities of stroke. Research has shown that about 30% of stroke survivors develop depression and about 20% develop anxiety. Stroke survivors with such adverse mental outcomes are often attributed to poorer health outcomes, such as higher mortality rates. The objective of this study is to use deep learning (DL) methods to predict the risk of a stroke survivor experiencing post-stroke depression and/or post-stroke anxiety, which is collectively known as post-stroke adverse mental outcomes (PSAMO). This study studied 179 patients with stroke, who were further classified into PSAMO versus no PSAMO group based on the results of validated depression and anxiety questionnaires, which are the industry’s gold standard. This study collected demographic and sociological data, quality of life scores, stroke-related information, medical and medication history, and comorbidities. In addition, sequential data such as daily lab results taken seven consecutive days after admission are also collected. The combination of using DL algorithms, such as multi-layer perceptron (MLP) and long short-term memory (LSTM), which can process complex patterns in the data, and the inclusion of new data types, such as sequential data, helped to improve model performance. Accurate prediction of PSAMO helps clinicians make early intervention care plans and potentially reduce the incidence of PSAMO.https://www.mdpi.com/2306-5354/12/5/517artificial intelligencedeep learningmachine learningneural networkpost-stroke anxietypost-stroke depression
spellingShingle Chien Wei Oei
Eddie Yin Kwee Ng
Matthew Hok Shan Ng
Yam Meng Chan
Vinithasree Subbhuraam
Lai Gwen Chan
U. Rajendra Acharya
Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
Bioengineering
artificial intelligence
deep learning
machine learning
neural network
post-stroke anxiety
post-stroke depression
title Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
title_full Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
title_fullStr Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
title_full_unstemmed Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
title_short Automated Risk Prediction of Post-Stroke Adverse Mental Outcomes Using Deep Learning Methods and Sequential Data
title_sort automated risk prediction of post stroke adverse mental outcomes using deep learning methods and sequential data
topic artificial intelligence
deep learning
machine learning
neural network
post-stroke anxiety
post-stroke depression
url https://www.mdpi.com/2306-5354/12/5/517
work_keys_str_mv AT chienweioei automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT eddieyinkweeng automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT matthewhokshanng automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT yammengchan automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT vinithasreesubbhuraam automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT laigwenchan automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata
AT urajendraacharya automatedriskpredictionofpoststrokeadversementaloutcomesusingdeeplearningmethodsandsequentialdata