Forecast UTI: aplicativo para previsão de leitos de unidades de terapia intensiva no contexto da pandemia de COVID-19

Neto, Luiz Leduíno de Salles; Martins, Camila Bertini; Chaves, Antônio Augusto; Konstantyner, Thais Cláudia Roma de Oliveira; Yanasse, Horácio Hideki; Campos, Cláudia Barbosa Ladeira de; Bellini, Ana Júlia de Oliveira; Butkeraites, Renan Brito; Correia, Leonardo; Magro, Igor Luciano; Soares, Fernando dos Santos

doi:10.1590/s1679-49742020000400023

Cited by 5 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They started with a low-dose computed tomography (LDCT) scan of each patient, and estimates of body composition were made. As a result, it was possible to identify that the relationship between waist circumference and paravertebral muscle circumference (FMR), in addition to age, were significant predictors of the need for treatment in the ICU of mentioned patients Salles-Neto et al [13] developed an open application named Forecast UTI, which uses polynomial regression models adjusted to the fifth degree to forecast the number of ICU beds occupied by COVID-19 patients. They calculated the root-mean-square of forecast error (RMSE) for each model and chose the model with the lowest RMSE.…”

Section: State-of-the-artmentioning

confidence: 99%

Predicting Intensive Care Unit Admission of COVID-19 Patients with Open Data: Analysis of the First Wave in Colombia

Acosta-Velasquez¹,

Fajardo-Moreno²,

Espinosa-Leal³

2022

Preprint

View full text Add to dashboard Cite

Optimizing intensive care resources using predicting modeling is paramount for fighting the COVID-19 pandemic. In this paper, we model the admission of COVID-19 patients in intensive care units (ICU) in Colombia using openly available data gathered from 18 March 2020 to 14 October 2020. After an intensive preprocessing of the data, we trained four different machine learning models using four different strategies for handling the imbalanced features. Our findings show that our best model (XGBoost) effectively predicts an Area Under the Curve (AUC-ROC) of 0.94, in line with the state-of-the-art results obtained in other predictive models obtained with medical data.

show abstract

Section: State-of-the-artmentioning

confidence: 99%

Predicting Intensive Care Unit Admission of COVID-19 Patients with Open Data: Analysis of the First Wave in Colombia

Acosta-Velasquez¹,

Fajardo-Moreno²,

Espinosa-Leal³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…A ECA2 está relacionada com a função cardíaca e no desenvolvimento de hipertensão e diabetes mellitus e, além disso, essa enzima também foi identificada como um receptor funcional para o SARS-CoV. Como é mostrado na Figura 2.1, o vírus se liga através da proteína spike da superfície viral ao receptor da ECA2 após a ativação da proteína spike pela serina protease transmembrana-2 (TMPRSS2), e como consequência dessa ligação do vírus acaba gerando uma desregulação da ECA2 aumentando a angiotensina II, levando a efeitos deletérios da ativação do sistema reninaangiotensina-aldosterona, causando principalmente alterações cardiovasculares (NETO et al, 2020).…”

Section: Autoresunclassified

“…Nesse contexto, umas das principais indagações de gestores de hospitais do Brasil é prever o número de leitos de Unidade de Terapia Intensiva (UTI) necessário para atender os casos graves da doença, considerando a dinâmica temporal da epidemia, a distribuição desigual de leitos hospitalares no território nacional e o fato da epidemia de COVID-19 encontrar-se em diferentes fases, segundo os municípios e os estados brasileiros (SALLES NETO et al, 2020).…”

Section: Conclusãounclassified

“…A maioria dos casos são assintomáticos ou oligossintomáticos, todavia, relatos sugerem que até 20% dos indivíduos infectados requerem hospitalização, e desses, até 25% necessitam de cuidados em Unidade de Terapia Intensiva (UTI) (NASCIMENTO et al, 2021). Pacientes internados na UTI podem obter administração de respiração assistida por ventilador e apresentam uma alta taxa de mortalidade, mas, essa varia conforme faixa etária e comorbidades que o indivíduo possui (SALLES NETO et al, 2020), sendo as de maior risco: diabetes, hipertensão e doenças cardiovasculares (CHAKRABORTY et al, 2020).…”

Section: Sinais E Sintomasunclassified

See 1 more Smart Citation

Covid-19 - O Vírus Que Movimentou a Ciência - Volume 1

Almeida¹,

Araújo²,

Brandão³

et al. 2021

View full text Add to dashboard Cite

A pandemia do COVID-19, deflagrada no início de 2020, foi acompanhada por rápida disseminação, taxas de mortalidade sem barreiras sociais e uma corrida científica sem precedentes.Neste cenário mundial o Brasil tem lugar de destaque, pois é o 11º país que mais produziu artigos sobre o assunto. Este e-book tenho o objetivo de expor as pesquisas epidemiológicas, revisões científicas, estudos experimentais pré-clínicos, observacionais e clínicos sobre o assunto COVID-19 -O VÍRUS QUE MOVIMENTOU A CIÊNCIA.Esperamos que tenha uma leitura agradável e possa desfrutar ao máximo o conhecimento transmitido por nossos autores.

show abstract

Control Centre for Intensive Care as a Tool for Effective Coordination, Real-Time Monitoring, and Strategic Planning During the COVID-19 Pandemic

Komenda¹,

Černý²,

Šnajdárek³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

In the Czech Republic, the strategic data-based and organizational support for individual regions and for providers of acute care at the nationwide level is coordinated by the Ministry of Health. At the beginning of the COVID-19 pandemic, the country needed to very quickly implement a system for the monitoring, reporting, and overall management of hospital capacities. The aim of this viewpoint is to describe the purpose and basic functions of a web-based application named “Control Centre for Intensive Care,” which was developed and made available to meet the needs of systematic online technical support for the management of intensive inpatient care across the Czech Republic during the first wave of the pandemic in spring 2020. Two tools of key importance are described in the context of national methodology: one module for regular online updates and overall monitoring of currently free capacities of intensive care in real time, and a second module for online entering and overall record-keeping of requirements on medications for COVID-19 patients. A total of 134 intensive care providers and 927 users from hospitals across all 14 regions of the Czech Republic were registered in the central Control Centre for Intensive Care database as of March 31, 2021. This web-based application enabled continuous monitoring and decision-making during the mass surge of critical care from autumn 2020 to spring 2021. The Control Center for Intensive Care has become an indispensable part of a set of online tools that are employed on a regular basis for crisis management at the time of the COVID-19 pandemic.

show abstract

Forecast UTI: aplicativo para previsão de leitos de unidades de terapia intensiva no contexto da pandemia de COVID-19

Cited by 5 publications

References 5 publications

Predicting Intensive Care Unit Admission of COVID-19 Patients with Open Data: Analysis of the First Wave in Colombia

Predicting Intensive Care Unit Admission of COVID-19 Patients with Open Data: Analysis of the First Wave in Colombia

Covid-19 - O Vírus Que Movimentou a Ciência - Volume 1

Control Centre for Intensive Care as a Tool for Effective Coordination, Real-Time Monitoring, and Strategic Planning During the COVID-19 Pandemic

Contact Info

Product

Resources

About