Search

Individuals with diabetes often experience nerve damage complications due to diabetic neuropathy, with up to 50% of cases being asymptomatic, increasing injury risk. Reducing the risk of long-term complications related to diabetes requires the ability to predict and track the severity of diabetic neuropathy in patients. Diabetes care strategies and management by predicting the severity of diabetic neuropathy can offer insight into disease progression and assist healthcare professionals in managing high-risk patients. The study introduces a novel approach by predicting diabetic neuropathy severity using electromyography signals, moving beyond traditional medical data. It also innovates by employing a hybrid method of Ensemble models with Particle Swarm Optimization for parameter optimization. The study predicts diabetic neuropathy severity by selecting the top 10, 20, and 30 features from 90 extracted electromyography signal features using a correlation matrix, principal component analysis, and recursive feature elimination. Various machine learning models, as well as two ensemble models, boosting and stacking, were employed for prediction. Optimal parameters for each learner were determined using Particle Swarm Optimizer to enhance prediction performance. Among 144 experiment scenarios, the hybrid stacking-particle swarm optimizer model outperformed others, showing 14.29% higher accuracy, 13.93% higher F-1 score, 4.17% higher recall, and 14.58% higher precision compared to other stacking models. The prediction results can be used for early identification of complications in diabetic patients. [ABSTRACT FROM AUTHOR]

In the era of social media and online communication, the surveillance and classification of disease-related information in real-time is crucial. Twitter data on dengue-related are rarely used for classification, especially textbased classification in Indonesian. Even though, the classification of dengue-related news tweets is capable of being utilized for a variety of purposes. This study presents a novel approach to classifying dengue fever-related tweets in the Indonesian context, utilizing the potential of advanced language models and hybrid neural networks. The method proposed incorporates the advantages of two deep learning architectures: Bidirectional encoder representations from transformers indonesian-based (Indo-BERT) and a hybrid convolutional neural network-long short-term memory (CNN-LSTM) model which is still rarely used in this context. Indo-BERT, a pre-trained language model, extracts complex semantic and contextual information from text, enabling a deeper comprehension of dengue-related tweets. The Hybrid CNN-LSTM model processes textual data in tandem, extracting features via convolutional layers while maintaining temporal dependencies. To assure this model's effectiveness in the Indonesian context, a collection of dengue fever-related tweets in Indonesian was created and labeled for supervised learning. Experiments were conducted in multiple scenarios, and the results demonstrated that the combination of the Indo-BERT and Hybrid CNN-LSTM models had superior performance in classifying tweets about dengue illness into five labels, namely infected, awareness, informative, news, and others. The best models deliver an accuracy of 0.91, an F1-score of 0.90, a precision of 0.91, and a recall of 0.89. The hybrid model's efficacy surpasses that of previous approaches with an average difference in accuracy of 0.25, precision of 0.27, and recall of 0.26. It is anticipated that the Health Service can use the results of this classification to improve the dengue surveillance system with the initial data obtained from this classification. [ABSTRACT FROM AUTHOR]

Diphtheria is an infectious disease with a high mortality rate. In Indonesia, the number of diphtheria cases has remained relatively high in recent years, so efforts to prevent and control diphtheria are needed. Forecasting of the number of diphtheria cases was carried out in this study by applying a type-2 fuzzy logic systems method. Forecasting in this study was carried out by involving the variables of the number of diphtheria sufferers, the percentage of immunization coverage comprising four immunization types, and population density. Regions are grouped into three clusters based on the number of cases that have occurred. Each cluster is taken and sampled in the form of one region to acquire a robust model for other regions. The forecasting results for the next 24 periods show that the performance of the type-2 fuzzy logic systems method is quite good, with accuracy values in the Malang area showing an MSE of 8.785 and an SMAPE of 54.91%. In the Surabaya area, the forecasting accuracy results have an MSE value of 14.940 and an SMAPE of 35.51%. In the Sumenep area, the forecasting accuracy results show an MSE value of 2.188 and an SMAPE of 67.63%. The results of the forecasting of the number of cases can be used as a guide in planning and making decisions regarding the prevention and management of diphtheria. [ABSTRACT FROM AUTHOR]

Induced Draft Fans (IDF) and Primary Air Fans (PAF) are critical equipment in steam power plants. Anomaly detection based on machine learning models is an approach that is currently being developed for optimization and increasing the effectiveness of predictive maintenance (PDM) as well as increasing the reliability of thermal power plants. The research aims to develop a data-driven model for diagnostic and prognostic equipment, produce accurate predictions using many sensor data taken in real-time from the SCADA system, and design a PDM management framework using an anomaly detection system. This research proposes to use a combined recurrent neural network (RNN)-autoencoder approach as a "normal" behavior model (NBM) with the Mahalanobis Distance (MD) statistical method for the detection of anomalies in power plant equipment. Based on time-series input sensor data, the RNN - autoencoder is utilized to predict the behavior of the equipment in health circumstances. In contrast, the MD is used to determine the distance between the actual parameters of the equipment and its "normal" behavior prediction to determine the anomaly condition. This study examined the performance of the long short-term memory (LSTM) and gated recurrent unit (GRU) models in modeling normal behavior with hyperparameter optimization. The LSTM with the best hyperparameters had a validation loss of 5,690 x 10-4 and a validation accuracy of 93.36 percent, whereas the GRU with the best hyperparameters had a validation loss of 4.484 x 10-4 and a validation accuracy of 93.47 percent. GRU can outperform the LSTM model. The proposed framework can detect anomaly conditions in various cases of IDF and PAF equipment disturbances, both in early warning of equipment failure and during downtime conditions. [ABSTRACT FROM AUTHOR]

Climate and geography factors significantly influence the spread of dengue fever. It's critical to figure out the specifics of climatic and geographic conditions and the relationships between current patients. For further preventive measures, it is also necessary to identify the transmission source patients. This study aims to improve the understanding of dengue fever patients spreading under climate and geographic locations. Patients are clustered based on climatic and geographical variables, and influential patients are found in the established network using Fuzzy CMeans and Social Network Analysis. The scenario of cluster numbers' alteration and degree of fuzziness with Fuzzy C-Means made the three groups of patient clusters. A total of 52% of the patients are included in the lowland cluster, based on climate conditions and altitude. The patients grouped better with this approach than with the other compared methods. The Calinski Harabasz score has an average difference of 1644.105. The following relationship in the network is constructed once the cluster has been formed. It is given a fuzzy rule representing the distance of residence and the period of illness between patients. According to the Social Network Analysis approach applied to the region and month scenario, the three areas sensitive to the spreading center are Dau, Kepanjen, and Karangploso. The most significant patient data distribution occurred during the rainy season, peaking in January-February. The centrality algorithm shows that patients with male characteristics and the age range of children and adults could be the source of the disease's spread every month. Kepanjen area is the site of residence with the most significant impact with a proportion of 62.5 % in a year, and the initial illness is the fever related to dengue. Analysis results of this study can use by the Public Health Office to plan and manage resources to prevent extensive spread. [ABSTRACT FROM AUTHOR]

Dengue fever is an endemic disease that occurs throughout the year. Forecasting cases of dengue fever based on actual data is needed for monitoring and taking action. Recently, developing countries have faced problems related to the dengue fever surveillance system caused by the data delay factor. On the other hand, availability and access to health-related information on the internet have changed people’s behaviors and habits. However, the effect of internet data usage has not been widely studied, especially in areas with different levels of internet penetration. This study examines the impact of dengue fever case reported data, Google Trends, Twitter, and climate data in areas with many cases and varying levels of internet penetration to forecast dengue fever cases. Split time-series cross-validation (STSCV) and blocked time-series cross-validation (BTSCV) are used to obtain various training and testing results. The hybrid Decomposition-Bidirectional Long Short-Term Memory(D-BiLSTM) method is proposed. D-BiLSTM applied to eight different scenarios across multiple level areas. According to the results of the experiments, the DBiLSTM model with STSCV outperforms the BTSCV. In the high internet penetration area, the average error is 9,517, while in the low internet penetration area, it is 5,188. In areas with high internet penetration, adding the variables Google Trends and Twitter does not significantly reduce the error forecasting. However, in the low penetration area, the inclusion of Google Trends and Twitter significantly decreases errors. In general, the D-BiLSTM model performed well. Then, when compared with other approaches, the D-BiLSTM model as a whole can reduce the average RMSE and the average MAE of the comparison model by 94,120 and 45,132, respectively, in areas of high internet penetration with the best SMAPE model of 0.310. In the low internet penetration area, the average decline in RMSE and MAE was 54,390 and 19,362, with the best SMAPE model performance of 0.183. [ABSTRACT FROM AUTHOR]

Since local human movements can influence dengue spread, a network-based prediction model considers the dynamic relation between dengue case incidences and their location over time. Some approaches often generated the networks in a certain period with a single spanning time until several months or in one year, called static networks. However, one annual-based dengue spread model could have different simulations depending on the selected months to show different dynamicity. Other approaches do not involve any networks for generating the spread models but employ them for validating the models with simulations. Considering the evolution of dengue circumstances that could quickly change between periods, we proposed a Dengue Spread Dynamic Network (DSDN) model with some timespan and location boundaries variants. DSDN includes five network models with nodes representing localities and links showing dengue spread which varied every day depending on infections presence and environmental conditions in a certain period. With our proposed method, daily dengue spread from one location to another can be predicted based on the location-based incidence historical data as outbreaks prevention initiative. We also analyzed how dengue spreads differently in outbreak and non-outbreak periods using Dynamic Network Link Prediction (DNLP) method. From our experiments result, Neighbor Network which modeled that dengue only spreads between neighboring localities produced an accuracy of 92.54% for the entire period. When applied only to outbreak data, there was a performance increase of 3.39 points, which suggested that link prediction performs better when dengue is rapidly spreading. In addition to that, our experiments concluded that dengue potentially spreads to a location with no current infections if local incidence often occurred in the past. [ABSTRACT FROM AUTHOR]