Mathematical Biosciences and Engineering最新文献

The global impact of the COVID-19 pandemic is widely recognized as a significant concern, with human flow playing a crucial role in its propagation. Consequently, recent research has focused on identifying and analyzing factors that can effectively regulate human flow. However, among the multiple factors that are expected to have an effect, few studies have investigated those that are particularly associated with human flow during the COVID-19 pandemic. In addition, few studies have investigated how regional characteristics and the number of vaccinations for these factors affect human flow. Furthermore, increasing the number of verified cases in countries and regions with insufficient reports is important to generalize conclusions. Therefore, in this study, a group-level analysis was conducted for Narashino City, Chiba Prefecture, Japan, using a human flow prediction model based on machine learning. High-importance groups were subdivided by regional characteristics and the number of vaccinations, and visual and correlation analyses were conducted at the factor level. The findings indicated that tree-based models, especially LightGBM, performed better in terms of prediction. In addition, the cumulative number of vaccinated individuals and the number of newly infected individuals are likely explanatory factors for changes in human flow. The analyses suggested a tendency to move with respect to the number of newly infected individuals in Japan or Tokyo, rather than the number of new infections in the area where they lived when vaccination had not started. With the implementation of vaccination, attention to the number of newly infected individuals in their residential areas may increase. However, after the spread of vaccination, the perception of infection risk may decrease. These findings can contribute to the proposal of new measures for efficiently controlling human flows and determining when to mitigate or reinforce specific measures.

It is important to classify electroencephalography (EEG) signals automatically for the diagnosis and treatment of epilepsy. Currently, the dominant single-modal feature extraction methods cannot cover the information of different modalities, resulting in poor classification performance of existing methods, especially the multi-classification problem. We proposed a multi-modal feature fusion (MMFF) method for epileptic EEG signals. First, the time domain features were extracted by kernel principal component analysis, the frequency domain features were extracted by short-time Fourier extracted transform, and the nonlinear dynamic features were extracted by calculating sample entropy. On this basis, the features of these three modalities were interactively learned through the multi-head self-attention mechanism, and the attention weights were trained simultaneously. The fused features were obtained by combining the value vectors of feature representations, while the time, frequency, and nonlinear dynamics information were retained to screen out more representative epileptic features and improve the accuracy of feature extraction. Finally, the feature fusion method was applied to epileptic EEG signal classifications. The experimental results demonstrated that the proposed method achieves a classification accuracy of 92.76 ± 1.64% across the five-category classification task for epileptic EEG signals. The multi-head self-attention mechanism promotes the fusion of multi-modal features and offers an efficient and novel approach for diagnosing and treating epilepsy.

In this paper, our main objective was to investigate the central pattern generator (CPG) neural network model for quadruped gait with time delay. First, we computed the normal form of the model on the center manifold, the bifurcation direction, and stability conditions of the bifurcating periodic solutions. Second, we applied the CPG model for quadruped gait to obtain reference models for goat's diagonal trotting gait on the flat ground and walking gait on the 18 degree slope through the trust region inversion algorithm. Finally, we performed numerical simulations to support theoretical analysis.

Chemical heterogeneity significantly influences the dynamics of phytoplankton and zooplankton interactions through its effects on phytoplankton carrying capacity and zooplankton ingestion rates. Our central objective of this study was to develop and examine a nonautonomous model of phytoplankton-zooplankton growth, which incorporates season-driven variations in light intensity and chemical heterogeneity. The dynamics of the system is characterized by positive invariance, dissipativity, boundary dynamics, and internal dynamics. Subsequently, numerical simulations were conducted to validate the theoretical findings and to elucidate the effects of seasonal light intensity, nutrient availability, and zooplankton loss rates on phytoplankton dynamics. The outcomes of our model and analysis offer a potential explanation for seasonal phytoplankton blooms.

Convolutional Neural Networks (CNNs) have received substantial attention as a highly effective tool for analyzing medical images, notably in interpreting endoscopic images, due to their capacity to provide results equivalent to or exceeding those of medical specialists. This capability is particularly crucial in the realm of gastrointestinal disorders, where even experienced gastroenterologists find the automatic diagnosis of such conditions using endoscopic pictures to be a challenging endeavor. Currently, gastrointestinal findings in medical diagnosis are primarily determined by manual inspection by competent gastrointestinal endoscopists. This evaluation procedure is labor-intensive, time-consuming, and frequently results in high variability between laboratories. To address these challenges, we introduced a specialized CNN-based architecture called GastroFuse-Net, designed to recognize human gastrointestinal diseases from endoscopic images. GastroFuse-Net was developed by combining features extracted from two different CNN models with different numbers of layers, integrating shallow and deep representations to capture diverse aspects of the abnormalities. The Kvasir dataset was used to thoroughly test the proposed deep learning model. This dataset contained images that were classified according to structures (cecum, z-line, pylorus), diseases (ulcerative colitis, esophagitis, polyps), or surgical operations (dyed resection margins, dyed lifted polyps). The proposed model was evaluated using various measures, including specificity, recall, precision, F1-score, Mathew's Correlation Coefficient (MCC), and accuracy. The proposed model GastroFuse-Net exhibited exceptional performance, achieving a precision of 0.985, recall of 0.985, specificity of 0.984, F1-score of 0.997, MCC of 0.982, and an accuracy of 98.5%.

The COVID-19 pandemic continues to pose significant challenges to global public health, necessitating the development of effective vaccination strategies to mitigate disease transmission. In Thailand, the COVID-19 epidemic has undergone multiple waves, prompting the implementation of various control measures, including vaccination campaigns. Understanding the dynamics of disease transmission and the impact of vaccination strategies is crucial for guiding public health interventions and optimizing epidemic control efforts. In this study, we developed a comprehensive mathematical model, termed $ S{S}_{v}I{H}_{1}C{H}_{2}RD $, to elucidate the dynamics of the COVID-19 epidemic in Thailand. The model incorporates key epidemiological parameters, vaccination rates, and disease progression stages to assess the effectiveness of different vaccination strategies in curbing disease transmission. Parameter estimation and model fitting were conducted using real-world data from COVID-19 patients in Thailand, enabling the simulation of epidemic scenarios and the exploration of optimal vaccination rates. Our results showed that optimizing vaccination strategies, particularly by administering approximately 119,625 doses per day, can significantly reduce the basic reproduction number ($ {R}_{0} $) below 1, thereby accelerating epidemic control. Simulation results demonstrated that the optimal vaccination rate led to a substantial decrease in the number of infections, with the epidemic projected to be completely eradicated from the population by June 19, 2022. These findings underscore the importance of targeted vaccination efforts and proactive public health interventions in mitigating the spread of COVID-19 and minimizing the burden on healthcare systems. Our study provides valuable insights into the optimization of vaccination strategies for epidemic control, offering guidance for policymakers and healthcare authorities in Thailand and beyond. By leveraging mathematical modeling techniques and real-world data, stakeholders can develop evidence-based strategies to combat the COVID-19 pandemic and safeguard public health.

Vaccine hesitancy threatens to reverse the progress in tackling vaccine-preventable diseases. We used an $ SIS $ model with a game theory model for vaccination and parameters from the COVID-19 pandemic to study how vaccine hesitancy impacts epidemic dynamics. The system showed three asymptotic behaviors: total rejection of vaccinations, complete acceptance, and oscillations. With increasing fear of infection, stable endemic states become periodic oscillations. Our results suggest that managing fear of infection relative to vaccination is vital to successful mass vaccinations.

Incorporating the self-regulatory mechanism with time delay to a plant-sulfide feedback system for intertidal salt marshes, we proposed and studied a functional reaction-diffusion model. We analyzed the stability of the positive steady state of the system, and derived the sufficient conditions for the occurrence of Hopf bifurcations. By deriving the normal form on the center manifold, we obtained the formulas determining the properties of the Hopf bifurcations. Our analysis showed that there is a critical value of time delay. When the time delay is greater than the critical value, the system will show asymptotical temporal periodic patterns while the system will display asymptotical spatial homogeneous patterns when the time delay is smaller than the critical value. Our numerical study showed that there are transient fairy circles for any time delay while there are different types of fairy circles and rings in the system. Our results enhance the concept that transient fairy circle patterns in intertidal salt marshes can infer the underlying ecological mechanisms and provide a measure of ecological resilience when the self-regulatory mechanism with time delay is considered.

Heart rate variability (HRV) is an important metric in cardiovascular health monitoring. Spectral analysis of HRV provides essential insights into the functioning of the cardiac autonomic nervous system. However, data artefacts could degrade signal quality, potentially leading to unreliable assessments of cardiac activities. In this study, we introduced a novel approach for estimating uncertainties in HRV spectrum based on matrix completion. The proposed method utilises the low-rank characteristic of HRV spectrum matrix to efficiently estimate data uncertainties. In addition, we developed a refined matrix completion technique to enhance the estimation accuracy and computational cost. Benchmarking on five public datasets, our model shows effectiveness and reliability in estimating uncertainties in HRV spectrum, and has superior performance against five deep learning models. The results underscore the potential of our developed matrix completion-based statistical machine learning model in providing reliable HRV spectrum uncertainty estimation.