Dimensionality reduction and ensemble of LSTMs for antimicrobial resistance prediction
Visualitza/Obre
10.1016/j.artmed.2023.102508
Inclou dades d'ús des de 2022
Cita com:
hdl:2117/383488
Tipus de documentArticle
Data publicació2023-04
EditorElsevier
Condicions d'accésAccés obert
Llevat que s'hi indiqui el contrari, els
continguts d'aquesta obra estan subjectes a la llicència de Creative Commons
:
Reconeixement-NoComercial-SenseObraDerivada 4.0 Internacional
Abstract
Bacterial resistance to antibiotics has been rapidly increasing, resulting in a low antibiotic effectiveness even treating common infections. The presence of resistant pathogens in environments such as a hospital Intensive Care Unit (ICU) exacerbates the critical admission-acquired infections. This work focuses on the prediction of antibiotic resistance in Pseudomonas aeruginosa nosocomial infections at the ICU, using Long Short-Term Memory (LSTM) artificial neural networks as predictive method. The analyzed data were extracted from the Electronic Health Records (EHR) of patients admitted in the University Hospital of Fuenlabrada from 2004 to 2019, and were modeled as Multivariate Time Series. A data-driven dimensionality reduction method is built by adapting three feature importance techniques from the literature to the considered data, and proposing an algorithm for selecting the most appropriate number of features. This is done using LSTM sequential capabilities so that the temporal aspect of features is taken into account. Furthermore, an ensemble of LSTMs is used to reduce the performance variance. Our results indicate that the patient's admission information, the antibiotics administered during the ICU stay, and the previous antimicrobial resistance are the most important features. The proposed dimensionality reduction method dramatically reduces the number of features while considerably increasing the prediction performance. The variance in the performance is reduced by considering the ensemble of classifiers. In essence, the proposed framework achieve, in a computationally cost efficient manner, promising results for supporting decisions in this clinical task, characterized by high dimensionality, data scarcity and concept drift.
CitacióHernàndez-Carnerero, À. [et al.]. Dimensionality reduction and ensemble of LSTMs for antimicrobial resistance prediction. "Artificial intelligence in medicine", Abril 2023, vol. 138, article 102508, p. 1-17.
ISSN1873-2860
Versió de l'editorhttps://www.sciencedirect.com/science/article/pii/S0933365723000222
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
1-s2.0-S0933365723000222-main.pdf | 2,377Mb | Visualitza/Obre |