Intelligence-Based Recommendation System for Critical Stroke Management in Intensive Care Units

García Terriza, L.; Risco-Martín, J.; Ayala, J. and Roselló, G. BIOINFORMATICS 2023, ISBN 978-989-758-631-6; ISSN 2184-4305, pages 131-138. DOI: 10.5220/0011621000003414

Abstract: This work presents an integrated recommendation system capable of providing support in healthcare critical environments such as Intensive Care Units or Stroke Care Units using Machine Learning techniques. The system can manage several patients by reading monitoring hemodynamic data in real-time, presenting current death risk probability, and showing recommendations that would reduce such probability and, in some cases, avoid death. This system introduces a novel method to produce recommendations based on genetic models and supervised machine learning. The interface is built upon a web application where clinicians can evaluate recommendations and straightforwardly provide feedback.

Funding: This research has been funded by Instituto de Salud Carlos III (RICORS-RD21/0006/0009) and cofinanced with FEDER Funds and/or from the European funds of the Recovery, Transformation and Resilience Plan and by NextGenerationEU. This work is also supported by Spanish Ministry of Science and Innovation under project PID2019-110866RB-I00.

N-Terminal Pro-Brain Natriuretic Peptide Levels Are Associated with Post-Stroke In-Hospital Complications

Ruiz-Franco ML, Guevara-Sánchez E, Amaya-Pascasio L et al. J Pers Med. 2023 Mar 5;13(3):474. doi: 10.3390/jpm13030474. PMID: 36983656

Abstract: Previous studies have shown the relationship between N-terminal pro-brain natriuretic peptide (NT-proBNP) with stroke mortality and functional outcome after an acute ischemic stroke (AIS). Knowledge of its association with systemic and neurological in-hospital complications is scarce. Our objective is to analyze this. We performed an observational, retrospective study that included consecutive AIS patients during a 1-year period (2020). A multivariate analysis was performed to identify if NT-proBNP levels were independently associated with in-hospital complications. 308 patients were included, of whom 96 (31.1%) developed systemic and 62 (20.12%) neurological in-hospital complications. Patients with any complication (39.3%) showed higher NT-proBNP levels than those without (median (IQR): 864 (2556) vs. 142 (623) pg/dL, p < 0.001). The receiver operating characteristic curve (ROC) pointed to 326 pg/dL of NT-proBNP as the optimal cutoff level for developing in-hospital systemic complications (63.6% sensitivity and 64.7% specificity for any complication; 66.7% and 62.7% for systemic; and 62.9% and 57.7% for neurological complications). Multivariate analyses showed that NT-proBNP > 326 pg/dL was associated with systemic complications (OR 2.336, 95% CI: 1.259–4.335), adjusted for confounders. This did not reach statistical significance for neurological complications. NT-proBNP could be a predictor of in-hospital systemic complications in AIS patients. Further studies are needed.

Funding: This study is part of the Spanish Health Outcomes-Oriented Cooperative Research Networks (RICORTS-ICTUS), Instituto de Salud Carlos III (Carlos III Health Institute), Ministerio de Ciencia e Innovación (Ministry of Science and Innovation), RD21/0006/0010.

Treatment of Vascular Parkinsonism: A Systematic Review

Del Toro-Pérez C, Guevara-Sánchez E, Martínez-Sánchez P. Brain Sci. 2023 Mar 14;13(3):489. doi: 10.3390/brainsci13030489.PMID: 36979299

Abstract: Background and aims: Although the distinction between vascular parkinsonism (VP) and idiopathic Parkinson’s disease (IPD) is widely described, it is not uncommon to find parkinsonisms with overlapping clinical and neuroimaging features even in response to levodopa treatment. In addition, several treatments have been described as possible adjuvants in VP. This study aims to update and analyze the different treatments and their efficacy in VP. Methods: A literature search was performed in PubMed, Scopus and Web of Science for studies published in the last 15 years until April 2022. A systematic review was performed. No meta-analysis was performed as no new studies on response to levodopa in VP were found since the last systematic review and meta-analysis in 2017, and insufficient studies on other treatments were located to conduct it in another treatment subgroup. Results: Databases and other sources yielded 59 publications after eliminating duplicates, and a total of 12 original studies were finally included in the systematic review. The treatments evaluated included levodopa, vitamin D, repetitive transcranial magnetic stimulation (rTMS) and intracerebral transcatheter laser photobiomodulation therapy (PBMT). The response to levodopa was lower in patients with VP with respect to IPD. Despite this, there has been described a subgroup of patients with good response, it being possible to identify them by means of neuroimaging techniques and the olfactory identification test. Other therapies showed encouraging results in studies with some risk of bias. Conclusions: The response of VP to different therapeutic strategies is modest. However, there is evidence that a subgroup of patients can be identified as more responsive to L-dopa based on clinical and neuroimaging criteria. This subgroup should be treated with L-dopa at appropriate doses. New therapies such as vitamin D, rTMS and PBMT warrant further studies to demonstrate their efficacy.

Funding: This study is part of the Spanish Health Outcomes-Oriented Cooperative Research Networks (RICORS-ICTUS), Instituto de Salud Carlos III (Carlos III Health Institute), Ministerio de Ciencia e Innovación (Ministry of Science and Innovation), RD21/0006/0010.