Identifying dementia from cognitive footprints in hospital records among Chinese older adults: a machine-learning study

Zhou, Jiayi; Liu, Wenlong; Zhou, Huiquan; Lau, Kui Kai; Wong, Gloria H.Y.; Chan, Wai Chi; Zhang, Qingpeng; Knapp, Martin; Wong, Ian C.K.; Luo, Hao

doi:10.1016/j.lanwpc.2024.101060

Cited by 2 publications

(1 citation statement)

References 53 publications

(62 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 17 MLP, a type of neural network, excels at capturing complex nonlinear interactions in the data, making it well‐suited for predicting clinical outcomes where the relationships between variables are not straightforward. 18 XGBoost, an ensemble learning method, is known for its robustness against overfitting, providing high predictive accuracy in various clinical applications. 19 , 20 This combination of models allows us to leverage their individual strengths for a comprehensive evaluation of our predictive tasks.…”

Section: Discussionmentioning

confidence: 99%

Machine learning‐based prediction of clinical outcomes after traumatic brain injury: Hidden information of early physiological time series

Ding,

Deng,

Wei

et al. 2024

CNS Neurosci Ther

View full text Add to dashboard Cite

AimsTo assess the predictive value of early‐stage physiological time‐series (PTS) data and non‐interrogative electronic health record (EHR) signals, collected within 24 h of ICU admission, for traumatic brain injury (TBI) patient outcomes.MethodsUsing data from TBI patients in the multi‐center eICU database, we focused on in‐hospital mortality, neurological status based on the Glasgow Coma Score (mGCS) motor subscore at discharge, and prolonged ICU stay (PLOS). Three machine learning (ML) models were developed, utilizing EHR features, PTS signals collected 24 h after ICU admission, and their combination. External validation was performed using the MIMIC III dataset, and interpretability was enhanced using the Shapley Additive Explanations (SHAP) algorithm.ResultsThe analysis included 1085 TBI patients. Compared to individual models and existing scoring systems, the combination of EHR and PTS features demonstrated comparable or even superior performance in predicting in‐hospital mortality (AUROC = 0.878), neurological outcomes (AUROC = 0.877), and PLOS (AUROC = 0.835). The model's performance was validated in the MIMIC III dataset, and SHAP algorithms identified six key intervention points for EHR features related to prognostic outcomes. Moreover, the EHR results (All AUROC >0.8) were translated into online tools for clinical use.ConclusionOur study highlights the importance of early‐stage PTS signals in predicting TBI patient outcomes. The integration of interpretable algorithms and simplified prediction tools can support treatment decision‐making, contributing to the development of accurate prediction models and timely clinical intervention.

show abstract