Hand, foot, and mouth disease (HFMD) is a highly contagious disease with several outbreaks in Asian-Pacific countries, including Thailand. With such epidemic characteristics and potential economic impact, HFMD is a significant public health issue in Thailand. Generally, contagious/infectious diseases’ transmission dynamics vary across geolocations due to different socioeconomic situations, demography, and lifestyles. Hence, a nationwide comprehensive model of the disease’s epidemic dynamics can provide information to understand better and predict a potential outbreak of this disease and efficiently and effectively manage its impact. However, there is no nationwide and comprehensive (i.e., the inclusion of reinfections in the model) model of HFDM dynamics for Thailand. This paper has endeavoured to promote nationwide comprehensive modelling of HFMD’s epidemic dynamics and comprehend the reinfection cases. We have formulated the SEIRS epidemiological model with dynamic vitals, including reinfections, to explore this disease’s prevalence. We also introduced periodic seasonality to reproduce the seasonal effect. The pattern of spread of this disease is uneven across the provinces in Thailand, so we used K -means clustering algorithm to cluster those provinces into three groups (i.e., highly, moderately, and least affected levels). We also analysed health records collected from district hospitals, which suggest significant reinfection cases. For example, we found that 11% (approximately) of infectious patients return for repeat treatment within the study period. We also performed sensitivity analysis which indicates that the basic reproduction number ( R 0 ) is sensitive to the rate of transmission ( β ) and the rate at which infected people recover ( γ ). By fitting the model with HFMD confirmed data for the provinces in each cluster, the basic reproduction number ( R 0 ) was estimated to be 2.643, 1.91, and 3.246 which are greater than 1. Based on this high R 0 , this study recommends that this disease will persist in the coming years under identical cultural and environmental conditions.

Hand, Foot and Mouth Disease (HFMD) is a highly contagious paediatric disease showing up symptoms like fever, diarrhoea, oral ulcers and rashes on the hands and foot, and even in the mouth. This disease has become an epidemic with several outbreaks in many Asian-Pacific countries with the basic reproduction number R 0 > 1. HFMD's diagnosis is very challenging as its lesion pattern may appear quite similar to other skin diseases such as herpangina, aseptic meningitis, and poliomyelitis. Therefore, clinical symptoms are essential besides skin lesion's pattern and position for precise diagnose of this disease. A deep learning-based HFMD detection system can play a significant role in the digital diagnosis of this disease. Various machine learning and deep learning architectures have been proposed for skin disease diagnosis and classification. However, these models are limited to the image classification problem. The diagnosis of similar appearing skin diseases using the image classification approach may result in misclassification or misdiagnosis of the disease. Parallel integration of clinical symptoms and images can improve disease diagnosis and classification performance. However, no deep learning architecture has been developed to diagnose HFMD disease from images and clinical data. This paper has proposed a novel Hybrid Deep Neural Networks integrating Multi-Layer Perceptron (MLP) network and Convolutional Neural Network into a single framework for the diagnosis of HFMD using the integrated features from clinical and image data. The proposed Hybrid Deep Neural Networks is particularly a multi branched model comprising of Multi-Layer Perceptron (MLP) network in the first branch to extract the clinical features and the modified pre-trained CNN architecture: MobileNet or NasNetMobile in the second branch to extract the features from skin disease lesion images. The features learnt from both the branches are merged to form an integrated feature from clinical data and images, which is fed to the subsequent classification network. We conducted several experiments employing image data only, clinical data only and both sources of data. The analyses compared and evaluated the performance of a typical MLP model and CNN model with our proposed Hybrid Deep Neural Networks. The novel approach promotes the existing image classification model and clinical symptoms based disease classification model, particularly the MLP model. From the cross-validated experiments, the results reveal that the proposed Hybrid Deep Neural Networks can diagnose the disease 99%-100% accurately.

Background Hand Foot and Mouth Disease (HFMD) is a highly contagious disease and has become an epidemic in many Asian-Pacific countries, including Thailand. With such epidemic characteristics and potential economic impact, HFMD is a significant public health issue. Comprehensive modelling of HFMD’s epidemic dynamics can be useful in understanding and predicting any potential outbreak of it, and manage its impact efficiently and effectively. Generally, the transmission dynamics of infectious diseases vary across geolocations due to different socio-economic situations, demography, and people’s lifestyles. However, there is no nation-wide and comprehensive (i.e., the inclusion of reinfections in the model) modelling of HFDM dynamics in Thailand. We aim to develop a nation-wide comprehensive modelling of HFMD’s epidemic dynamics and understand the reinfection cases in Thailand.Methods We have formulated Susceptible - Exposed - Infectious - Recovered - Susceptible (SEIRS) epidemiological model with dynamic vitals, including reinfections, to investigate the transmission of this disease in Thailand. We also introduced periodic seasonality to model the seasonal effect. According to the model, the spread of this disease is uneven throughout the provinces in Thailand. So, we have grouped the provinces into three clusters (i.e., highly, moderately and least affected provinces) using K-means unsupervised machine learning algorithm for better estimation of the parameters and fitting the model. We collected data from three local hospitals in Thailand to analyze the reinfection cases.Results The result from the analysis of HFMD recorded cases from three hospital (years 2012 to 2016) shows that 11% (approximately) are reinfections. By fitting the model with HFMD confirmed cases (years 2011 to 2019) and considering the reinfections, the basic reproductive number (R0) was estimated to be 2.643, 1.91 and 3.246 for three clustered provinces.Conclusion In a conclusion, it is found that HFMD is re-infectious disease in Thailand. It is also found that the spread of HFMD is not uniform across the provinces in Thailand. The basic reproductive number R 0 was estimated to be greater than 1 for all the three clusters. This indicates that under the same social and environmental condition, this disease will persist in coming years.