Sepsis prediction in intensive care unit based on genetic feature optimization and stacked deep ensemble learning

Predecting power transformer health index and life expectation based on digital twins and multitask LSTM-GRU model

Power transformers play a crucial role in enabling the integration of renewable energy sources and improving the overall efficiency and reliability of smart grid systems. They facilitate the conversion, transmission, and distribution of power from various sources and help to balance the load between different parts of the grid. The Transformer Health Index (THI) is one of the most important indicators of ensuring their reliability and preventing unplanned outages. To this end, this study introduces a proposed new architecture called a Smart Electricity Monitoring System based on Fog Computing and Digital Twins (SEMS-FDT) for monitoring the health performance of transformers by measuring the THI rate in real time. The SEMS-FDT is specifically designed to enable the observation of the transformer's health and performance purposes in real time. The study investigates the role of machine learning (ML) models, including traditional and ensemble methods, in predicting THI and LI (Heat Load Index) by exploring the use of the entire set of features and optimized feature subsets for prediction. To improve the forecasting prediction process and achieve optimal performance a novel multitasks LSTM_GRU model is also proposed. The experimental results demonstrate that there is a promising performance of 2.543, 0.13646, 0.0284, and 0.985 for MSE, MAE, MedAE, and R2 scores respectively. Moreover, the framework is extended by incorporating model explanations, which include global explanations, which provide insights based on the entire dataset, and local explanations, which offer instance-specific explanations. The integration of the proposed model and explainability features provides engineers with comprehensive outcomes regarding the model's result.

Early detection of sepsis using machine learning algorithms: a systematic review and network meta-analysis

Background

With machine learning (ML) carving a niche in diverse medical disciplines, its role in sepsis prediction, a condition where the 'golden hour' is critical, is of paramount interest. This study assesses the factors influencing the efficacy of ML models in sepsis prediction, aiming to optimize their use in clinical practice.

Methods

We searched Medline, PubMed, Google Scholar, and CENTRAL for studies published from inception to October 2023. We focused on studies predicting sepsis in real-time settings in adult patients in any hospital settings without language limits. The primary outcome was area under the curve (AUC) of the receiver operating characteristic. This meta-analysis was conducted according to PRISMA-NMA guidelines and Cochrane Handbook recommendations. A Network Meta-Analysis using the CINeMA approach compared ML models against traditional scoring systems, with meta-regression identifying factors affecting model quality.

Results

From 3,953 studies, 73 articles encompassing 457,932 septic patients and 256 models were analyzed. The pooled AUC for ML models was 0.825 and it significantly outperformed traditional scoring systems. Neural Network and Decision Tree models demonstrated the highest AUC metrics. Significant factors influencing AUC included ML model type, dataset type, and prediction window.

Conclusion

This study establishes the superiority of ML models, especially Neural Network and Decision Tree types, in sepsis prediction. It highlights the importance of model type and dataset characteristics for prediction accuracy, emphasizing the necessity for standardized reporting and validation in ML healthcare applications. These findings call for broader clinical implementation to evaluate the effectiveness of these models in diverse patient groups.

Systematic review registration

https://inplasy.com/inplasy-2023-12-0062/, identifier, INPLASY2023120062.

Investigating computational models for diagnosis and prognosis of sepsis based on clinical parameters: Opportunities, challenges, and future research directions

This study investigates the use of computational frameworks for sepsis. We consider two dimensions for investigation - early diagnosis of sepsis (EDS) and mortality prediction rate for sepsis patients (MPS). We concentrate on the clinical parameters on which sepsis diagnosis and prognosis are currently done, including customized treatment plans based on historical data of the patient. We identify the most notable literature that uses computational models to address EDS and MPS based on those clinical parameters. In addition to the review of the computational models built upon the clinical parameters, we also provide details regarding the popular publicly available data sources. We provide brief reviews for each model in terms of prior art and present an analysis of their results, as claimed by the respective authors. With respect to the use of machine learning models, we have provided avenues for model analysis in terms of model selection, model validation, model interpretation, and model comparison. We further present the challenges and limitations of the use of computational models, providing future research directions. This study intends to serve as a benchmark for first-hand impressions on the use of computational models for EDS and MPS of sepsis, along with the details regarding which model has been the most promising to date. We have provided details regarding all the ML models that have been used to date for EDS and MPS of sepsis.

Swin-GA-RF: genetic algorithm-based Swin Transformer and random forest for enhancing cervical cancer classification

Cervical cancer is a prevalent and concerning disease affecting women, with increasing incidence and mortality rates. Early detection plays a crucial role in improving outcomes. Recent advancements in computer vision, particularly the Swin transformer, have shown promising performance in image classification tasks, rivaling or surpassing traditional convolutional neural networks (CNNs). The Swin transformer adopts a hierarchical and efficient approach using shifted windows, enabling the capture of both local and global contextual information in images. In this paper, we propose a novel approach called Swin-GA-RF to enhance the classification performance of cervical cells in Pap smear images. Swin-GA-RF combines the strengths of the Swin transformer, genetic algorithm (GA) feature selection, and the replacement of the softmax layer with a random forest classifier. Our methodology involves extracting feature representations from the Swin transformer, utilizing GA to identify the optimal feature set, and employing random forest as the classification model. Additionally, data augmentation techniques are applied to augment the diversity and quantity of the SIPaKMeD1 cervical cancer image dataset. We compare the performance of the Swin-GA-RF Transformer with pre-trained CNN models using two classes and five classes of cervical cancer classification, employing both Adam and SGD optimizers. The experimental results demonstrate that Swin-GA-RF outperforms other Swin transformers and pre-trained CNN models. When utilizing the Adam optimizer, Swin-GA-RF achieves the highest performance in both binary and five-class classification tasks. Specifically, for binary classification, it achieves an accuracy, precision, recall, and F1-score of 99.012, 99.015, 99.012, and 99.011, respectively. In the five-class classification, it achieves an accuracy, precision, recall, and F1-score of 98.808, 98.812, 98.808, and 98.808, respectively. These results underscore the effectiveness of the Swin-GA-RF approach in cervical cancer classification, demonstrating its potential as a valuable tool for early diagnosis and screening programs.

PSO-XnB: a proposed model for predicting hospital stay of CAD patients

Coronary artery disease poses a significant challenge in decision-making when predicting the length of stay for a hospitalized patient. This study presents a predictive model-a Particle Swarm Optimized-Enhanced NeuroBoost-that combines the deep autoencoder with an eXtreme gradient boosting model optimized using particle swarm optimization. The model uses a fuzzy set of rules to categorize the length of stay into four distinct classes, followed by data preparation and preprocessing. In this study, the dimensionality of the data is reduced using deep neural autoencoders. The reconstructed data obtained from autoencoders is given as input to an eXtreme gradient boosting model. Finally, the model is tuned with particle swarm optimization to obtain optimal hyperparameters. With the proposed technique, the model achieved superior performance with an overall accuracy of 98.8% compared to traditional ensemble models and past research works. The model also scored highest in other metrics such as precision, recall, and particularly F1 scores for all categories of hospital stay. These scores validate the suitability of our proposed model in medical healthcare applications.

Explainable machine learning models based on multimodal time-series data for the early detection of Parkinson's disease

BACKGROUND AND OBJECTIVES

Parkinson's Disease (PD) is a devastating chronic neurological condition. Machine learning (ML) techniques have been used in the early prediction of PD progression. Fusion of heterogeneous data modalities proved its capability to improve the performance of ML models. Time series data fusion supports the tracking of the disease over time. In addition, the trustworthiness of the resulting models is improved by adding model explainability features. The literature on PD has not sufficiently explored these three points.

METHODS

In this work, we proposed an ML pipeline for predicting the progression of PD that is both accurate and explainable. We explore the fusion of different combinations of five time series modalities from the Parkinson's Progression Markers Initiative (PPMI) real-world dataset, including patient characteristics, biosamples, medication history, motor, and non-motor function data. Each patient has six visits. The problem has been formulated in two ways: ❶ a three-class based progression prediction with 953 patients in each time series modality, and ❷ a four-class based progression prediction with 1,060 patients in each time series modality. The statistical features of these six visits were calculated from each modality and diverse feature selection methods were applied to select the most informative feature sets. The extracted features were used to train a set of well-known ML models including Support vector machines (SVM), random forests (RF), extra tree classifier (ETC), light gradient boosting machines (LGBM), and stochastic gradient descent (SGD). We examined a number of data-balancing strategies in the pipeline with different combinations of modalities. ML models have been optimized using the Bayesian optimizer. A comprehensive evaluation of various ML methods has been conducted, and the best models have been extended to provide different explainability features.

RESULTS

We compare the performance of ML models before and after optimization and using and without using feature selection. In the three-class experiment and with various modality fusions, the LGBM model produced the most accurate results with a 10-fold cross-validation (10-CV) accuracy of 90.73% using non-motor function modality. RF produced the best results in the four-class experiment with various modality fusions with a 10-CV accuracy of 94.57% using non-motor modality. With the fused dataset of non-motor and motor function modalities, the LGBM model outperformed the other ML models in both the 3-class and 4-class experiments (i.e., 10-CV accuracy of 94.89% and 93.73%, respectively). Using the Shapely Additive Explanations (SHAP) framework, we employed global and instance-based explanations to explain the behavior of each ML classifier. Moreover, we extended the explainability by implementing the LIME and SHAPASH local explainers. The consistency of these explainers has been explored. The resultant classifiers were accurate, explainable, and thus medically more relevant and applicable.

CONCLUSIONS

The select modalities and feature sets were confirmed by the literature and medical experts. The various explainers suggest that the bradykinesia (NP3BRADY) feature was the most dominant and consistent. By providing thorough insights into the influence of multiple modalities on the disease risk, the suggested approach is expected to help improve the clinical knowledge of PD progression processes.

Predicting CTS Diagnosis and Prognosis Based on Machine Learning Techniques

Carpal tunnel syndrome (CTS) is a clinical disease that occurs due to compression of the median nerve in the carpal tunnel. The determination of the severity of carpal tunnel syndrome is essential to provide appropriate therapeutic interventions. Machine learning (ML)-based modeling can be used to classify diseases, make decisions, and create new therapeutic interventions. It is also used in medical research to implement predictive models. However, despite the growth in medical research based on ML and Deep Learning (DL), CTS research is still relatively scarce. While a few studies have developed models to predict diagnosis of CTS, no ML model has been presented to classify the severity of CTS based on comprehensive clinical data. Therefore, this study developed new classification models for determining CTS severity using ML algorithms. This study included 80 patients with other diseases that have an overlap in symptoms with CTS, such as cervical radiculopathysasas, de quervian tendinopathy, and peripheral neuropathy, and 80 CTS patients who underwent ultrasonography (US)-guided median nerve hydrodissection. CTS severity was classified into mild, moderate, and severe grades. In our study, we aggregated the data from CTS patients and patients with other diseases that have an overlap in symptoms with CTS, such as cervical radiculopathysasas, de quervian tendinopathy, and peripheral neuropathy. The dataset was randomly split into training and test data, at 70% and 30%, respectively. The proposed model achieved promising results of 0.955%, 0.963%, and 0.919% in terms of classification accuracy, precision, and recall, respectively. In addition, we developed a machine learning model that predicts the probability of a patient improving after the hydro-dissection injection process based on the aggregated data after three different months (one, three, and six). The proposed model achieved accuracy after six months of 0.912%, after three months of 0.901%, and after one month 0.877%. The overall performance for predicting the prognosis after six months outperforms the prediction after one and three months. We utilized statistics tests (significance test, Spearman's correlation test, and two-way ANOVA test) to determine the effect of injection process in CTS treatment. Our data-driven decision support tools can be used to help determine which patients to operate on in order to avoid the associated risks and expenses of surgery.

A Holistic Approach to Identify and Classify COVID-19 from Chest Radiographs, ECG, and CT-Scan Images Using ShuffleNet Convolutional Neural Network

Early and precise COVID-19 identification and analysis are pivotal in reducing the spread of COVID-19. Medical imaging techniques, such as chest X-ray or chest radiographs, computed tomography (CT) scan, and electrocardiogram (ECG) trace images are the most widely known for early discovery and analysis of the coronavirus disease (COVID-19). Deep learning (DL) frameworks for identifying COVID-19 positive patients in the literature are limited to one data format, either ECG or chest radiograph images. Moreover, using several data types to recover abnormal patterns caused by COVID-19 could potentially provide more information and restrict the spread of the virus. This study presents an effective COVID-19 detection and classification approach using the Shufflenet CNN by employing three types of images, i.e., chest radiograph, CT-scan, and ECG-trace images. For this purpose, we performed extensive classification experiments with the proposed approach using each type of image. With the chest radiograph dataset, we performed three classification experiments at different levels of granularity, i.e., binary, three-class, and four-class classifications. In addition, we performed a binary classification experiment with the proposed approach by classifying CT-scan images into COVID-positive and normal. Finally, utilizing the ECG-trace images, we conducted three experiments at different levels of granularity, i.e., binary, three-class, and five-class classifications. We evaluated the proposed approach with the baseline COVID-19 Radiography Database, SARS-CoV-2 CT-scan, and ECG images dataset of cardiac and COVID-19 patients. The average accuracy of 99.98% for COVID-19 detection in the three-class classification scheme using chest radiographs, optimal accuracy of 100% for COVID-19 detection using CT scans, and average accuracy of 99.37% for five-class classification scheme using ECG trace images have proved the efficacy of our proposed method over the contemporary methods. The optimal accuracy of 100% for COVID-19 detection using CT scans and the accuracy gain of 1.54% (in the case of five-class classification using ECG trace images) from the previous approach, which utilized ECG images for the first time, has a major contribution to improving the COVID-19 prediction rate in early stages. Experimental findings demonstrate that the proposed framework outperforms contemporary models. For example, the proposed approach outperforms state-of-the-art DL approaches, such as Squeezenet, Alexnet, and Darknet19, by achieving the accuracy of 99.98 (proposed method), 98.29, 98.50, and 99.67, respectively.

Psychological factors of college students’ learning pressure under the online education mode during the epidemic

The emergence of the network environment is the product of the combination of the development of computer technology and the development of network technology. Internet technology is slowly penetrating into all aspects of people’s lives and has had a huge impact and change on people’s lives. With the repeated outbreak of the epidemic in recent years, online education has been increasingly applied to the study and life of college students. The epidemic has lasted for 3 years, while the life of college students is only 4 years. In recent years, most of the campus study and life of college students have been carried out in the online education mode. This not only changed the mode of class, but also changed the mental health of college students. Taking the online education model during the epidemic as the research background, this paper selects the psychological factors of college students’ learning pressure to analyze, combined with the design and implementation of the questionnaire, to understand the impact of online education on college students’ cognition, emotion, willpower, and social interaction. The purpose is to find out the psychological factors of college students’ learning pressure under the online education mode, and to propose effective solutions. The analysis of the psychological factors of college students’ learning pressure in the form of questionnaires is more accurate than other forms of experimental investigation, the efficiency is increased by 32%, and the accuracy is also increased by 18%.

Sentiment Analysis of Users’ Reactions on Social Media During the Pandemic

During the outbreak of the COVID-19 pandemic, social networks became the preeminent medium for communication, social discussion, and entertainment. Social network users are regularly expressing their opinions about the impacts of the coronavirus pandemic. Therefore, social networks serve as a reliable source for studying the topics, emotions, and attitudes of users that have been discussed during the pandemic. In this paper, we investigate the reactions and attitudes of people towards topics raised on social media platforms. We collected data of two large-scale COVID-19 datasets from Twitter and Instagram for six and three months, respectively. This paper analyzes the reaction of social network users in terms of different aspects including sentiment analysis, topic detection, emotions, and the geo-temporal characteristics of our dataset. We show that the dominant sentiment reactions on social media are neutral, while the most discussed topics by social network users are about health issues. This paper examines the countries that attracted a higher number of posts and reactions from people, as well as the distribution of health-related topics discussed in the most mentioned countries. We shed light on the temporal shift of topics over countries. Our results show that posts from the top-mentioned countries influence and attract more reactions worldwide than posts from other parts of the world.

Heterogeneous Ensemble Deep Learning Model for Enhanced Arabic Sentiment Analysis

Sentiment analysis was nominated as a hot research topic a decade ago for its increasing importance in analyzing the people’s opinions extracted from social media platforms. Although the Arabic language has a significant share of the content shared across social media platforms, its content’s sentiment analysis is still limited due to its complex morphological structures and the varieties of dialects. Traditional machine learning and deep neural algorithms have been used in a variety of studies to predict sentiment analysis. Therefore, a need of changing current mechanisms is required to increase the accuracy of sentiment analysis prediction. This paper proposed an optimized heterogeneous stacking ensemble model for enhancing the performance of Arabic sentiment analysis. The proposed model combines three different of pre-trained Deep Learning (DL) models: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) in conjunction with three meta-learners Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM) in order to enhance model’s performance for predicting Arabic sentiment analysis. The performance of the proposed model with RNN, LSTM, GRU, and the five regular ML techniques: Decision Tree (DT), LR, K-Nearest Neighbor (KNN), RF, and Naive Bayes (NB) are compared using three benchmarks Arabic dataset. Parameters of Machine Learning (ML) and DL are optimized using Grid search and KerasTuner, respectively. Accuracy, precision, recall, and f1-score were applied to evaluate the performance of the models and validate the results. The results show that the proposed ensemble model has achieved the best performance for each dataset compared with other models.