Validation of the Scrub Typhus Encephalitis Assessment Tool for the Management of Acute Encephalitis Syndrome.
Аннотация
INTRODUCTION
Acute encephalitis syndrome (AES) is one of the important causes of mortality among children in India. Active management of the cases, followed by addressing the cause of AES, is the key strategy for preventing mortality. Lack of laboratory facility and difficulty of sampling blood and cerebrospinal fluid (CSF) for assessing causes is one of the important barriers to early initiation of treatment. The main objective of the study is to validate the Scrub Typhus Encephalitis Assessment Tool (SEAT) for the management of AES.
METHODS
The study is a continuation of a study conducted in a tertiary care hospital in Eastern Uttar Pradesh. A machine learning (LightGBM) model was built to predict the probability of scrub typhus diagnosis among patients with acute encephalitis. Three models were built: one with sociodemographic characters, the second with Model 1 variables and blood parameters, and the third with Model 2 variables and CSF parameters.
RESULTS
The sensitivity of diagnosing the scrub typhus case was 71%, 77.5%, and 83% in Model 1, Model 2, and Model 3, respectively, and specificity was 61.5%, 75.5%, and 76.3%, respectively, in the models. In Model 1 fever duration, in Models 2 and 3, neutrophil/lymphocyte ratio was the most important predictor for differentiating the scrub and nonscrub cases.
CONCLUSION
With the available sensitivity and specificity of the tool, the SEAT can be a valuable tool for the prediction of scrub typhus as a cause of AES cases in remote areas.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.