Dataset Information

A machine learning approach for early identification of patients with severe imported malaria.

ABSTRACT:

Background

The aim of this study is to design ad hoc malaria learning (ML) approaches to predict clinical outcome in all patients with imported malaria and, therefore, to identify the best clinical setting.

Methods

This is a single-centre cross-sectional study, patients with confirmed malaria, consecutively hospitalized to the Lazzaro Spallanzani National Institute for Infectious Diseases, Rome, Italy from January 2007 to December 2020, were recruited. Different ML approaches were used to perform the analysis of this dataset: support vector machines, random forests, feature selection approaches and clustering analysis.

Results

A total of 259 patients with malaria were enrolled, 89.5% patients were male with a median age of 39 y/o. In 78.3% cases, Plasmodium falciparum was found. The patients were classified as severe malaria in 111 cases. From ML analyses, four parameters, AST, platelet count, total bilirubin and parasitaemia, are associated to a negative outcome. Interestingly, two of them, aminotransferase and platelet are not included in the current list of World Health Organization (WHO) criteria for defining severe malaria.

Conclusion

In conclusion, the application of ML algorithms as a decision support tool could enable the clinicians to predict the clinical outcome of patients with malaria and consequently to optimize and personalize clinical allocation and treatment.

SUBMITTER: D'Abramo A

PROVIDER: S-EPMC10865572 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A machine learning approach for early identification of patients with severe imported malaria.

D'Abramo Alessandra A Rinaldi Francesco F Vita Serena S Mazzieri Riccardo R Corpolongo Angela A Palazzolo Claudia C Ascoli Bartoli Tommaso T Faraglia Francesca F Giancola Maria Letizia ML Girardi Enrico E Nicastri Emanuele E

Malaria journal 20240213 1

<h4>Background</h4>The aim of this study is to design ad hoc malaria learning (ML) approaches to predict clinical outcome in all patients with imported malaria and, therefore, to identify the best clinical setting.<h4>Methods</h4>This is a single-centre cross-sectional study, patients with confirmed malaria, consecutively hospitalized to the Lazzaro Spallanzani National Institute for Infectious Diseases, Rome, Italy from January 2007 to December 2020, were recruited. Different ML approaches were ...[more]

PMID: 38351021

Similar Datasets

Project description:Background/Objectives: Studies attempting to predict the development of severe respiratory failure in patients with a COVID-19 infection using machine learning algorithms have yielded different results due to differences in variable selection. We aimed to predict the development of severe respiratory failure, defined as the need for high-flow oxygen support, continuous positive airway pressure, or mechanical ventilation, in patients with COVID-19, using machine learning algorithms to identify the most important variables in achieving this prediction. Methods: This retrospective, cross-sectional study included COVID-19 patients with mild respiratory failure (mostly receiving oxygen through a mask or nasal cannula). We used XGBoost, support vector machines, multi-layer perceptron, k-nearest neighbor, random forests, decision trees, logistic regression, and naïve Bayes methods to accurately predict severe respiratory failure in these patients. Results: A total of 320 patients (62.1% male; average age, 54.67 ± 15.82 years) were included in this study. During the follow-ups of these cases, 114 patients (35.6%) required high-level oxygen support, 67 (20.9%) required intensive care unit admission, and 43 (13.4%) died. The machine learning algorithms with the highest accuracy values were XGBoost, support vector machines, k-nearest neighbor, logistic regression, and multi-layer perceptron (0.7395, 0.7395, 0.7291, 0.7187, and 0.75, respectively). The method that obtained the highest ROC-AUC value was logistic regression (ROC-AUC = 0.7274). The best predictors of severe respiratory failure were a low lymphocyte count, a high computed tomography score in the right and left upper lung zones, an elevated neutrophil count, a small decrease in CRP levels on the third day of admission, a high Charlson comorbidity index score, and a high serum procalcitonin level. Conclusions: The development of severe respiratory failure in patients with COVID-19 could be successfully predicted using machine learning methods, especially logistic regression, and the best predictors of severe respiratory failure were the lymphocyte count and the degree of upper lung zone involvement.

Project description:BackgroundImported malaria continues to be reported in Sri Lanka after it was eliminated in 2012, and a few progress to life-threatening severe malaria.MethodsData on imported malaria cases reported in Sri Lanka from 2013 to 2023 were extracted from the national malaria database maintained by the Anti Malaria Campaign (AMC) of Sri Lanka. Case data of severe malaria as defined by the World Health Organization were analysed with regard to patients' general characteristics and their health-seeking behaviour, and the latter compared with that of uncomplicated malaria patients. Details of the last three cases of severe malaria in 2023 are presented.Results532 imported malaria cases were diagnosed over 11 years (2013-2023); 46 (8.6%) were severe malaria, of which 45 were Plasmodium falciparum and one Plasmodium vivax. Most severe malaria infections were acquired in Africa. All but one were males, and a majority (87%) were 26-60 years of age. They were mainly Sri Lankan nationals (82.6%). Just over half (56.5%) were treated at government hospitals. The average time between arrival of the person in Sri Lanka and onset of illness was 4 days. 29 cases of severe malaria were compared with 165 uncomplicated malaria cases reported from 2015 to 2023. On average both severe and uncomplicated malaria patients consulted a physician equally early (mean = 1 day) with 93.3% of severe malaria doing so within 3 days. However, the time from the point of consulting a physician to diagnosis of malaria was significantly longer (median 4 days) in severe malaria patients compared to uncomplicated patients (median 1 day) (p = 0.012) as was the time from onset of illness to diagnosis (p = 0.042). All severe patients recovered without sequelae except for one who died.ConclusionsThe risk of severe malaria among imported cases increases significantly beyond 5 days from the onset of symptoms. Although patients consult a physician early, malaria diagnosis tends to be delayed by physicians because it is now a rare disease. Good access to expert clinical care has maintained case fatality rates of severe malaria at par with those reported elsewhere.

Dataset Information

A machine learning approach for early identification of patients with severe imported malaria.

Background

Methods

Results

Conclusion

Publications

A machine learning approach for early identification of patients with severe imported malaria.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets