Exploratory Data Analysis and Artificial Neural Network for Prediction of Leptospirosis Occurrence in Seremban, Malaysia Based on Meteorological Data

Fariq Rahmat,Asnor Juraiza Ishak,Zed Zulkafli,Samsul Bahari Mohd Noor,Afiqah Masrani,Hazlina Yahaya

doi:10.3389/feart.2020.00377

Fariq Rahmat, Asnor Juraiza Ishak + Show 4 more

Open Access

https://doi.org/10.3389/feart.2020.00377

Copy DOI

Abstract

Leptospirosis outbreaks in various parts of the world have been linked to changes in the weather. Furthermore, the effects have been shown to occur at different lags of up to 10 months, affecting the performance of simulation models that predict leptospirosis occurrence. In Malaysia, the link between different weather parameters, at different time lags, has yet to be established despite an increasing number of cases in recent years. In this study, a combination of data mining and machine learning is used to analyse, capture, and predict the relation between leptospirosis occurrence and temperature, rainfall, and relative humidity using the Seremban district in Malaysia as a case study. First, the optimal time lags for rainfall were determined using graphical exploratory data analysis (EDA) while non-graphical EDA was used for temperature. Then, an artificial neural network model (ANN) is developed to classify the combination of selected features into disease occurrence and non-occurrence using back-propagation training, optimizing the number of hidden layers and hidden nodes. The success is measured using accuracy, sensitivity and specificity of each model. EDA has shown that leptospirosis occurrence in Seremban is highly correlated with weekly average temperature at lag 16 weeks and weekly rainfall amount at lag 12 to 20 weeks. Using these selected features, the ANN model achieved the highest accuracy, sensitivity, and specificity at 84.00\%, 86.44\% and 79.33\% respectively. Overall, the EDA approach has increased the accuracy of the predictive model by 13.30\% to 31.26\% from the baseline models.

Highlights

This section is divided into three parts, which are discussion of the baseline models, exploratory data analysis (EDA)-based time lag selection for the meteorological input, and discussion on the performance of the proposed predictive model by using the selected time lagged predictors
These baseline models take in meteorological data without any time lags as input
A similar trend can be observed for Seremban City, where the accuracy increases from 40 to 62.30% as the number of neurons increases to 9; the accuracy starts to decrease to 60.01% when the number of neurons increases to 10

Summary

Introduction

Leptospirosis is a zoonotic disease caused by bacterial infection, i.e., by Leptospira with clinical symptoms, such as fever, headaches, muscle pains, and meningitis (Slack, 2010). The bacteria can be found in animals, such as rodents, dogs, pigs, and cattle (Mgode et al, 2015). More than 300 serovars have been identified worldwide and over 250 serotypes have been classified as pathogenic, which can cause diseases in people in varying severity (Levett, 2001; Lehmann et al, 2014). Human infection of leptospirosis occurs through direct contact with the product of infected animals, such as urine. Infection can occur indirectly through contact with contaminated water or soil that contain pathogenic Leptospira species, while human to human transmission is considered rare (Chadsuthi et al, 2012)

Methods

Results

Conclusion