Recurrent Neural Network and Reinforcement Learning Model for COVID-19 Prediction

Deep learning in public health: Comparative predictive models for COVID-19 case forecasting

The COVID-19 pandemic has had a significant impact on both the United Arab Emirates (UAE) and Malaysia, emphasizing the importance of developing accurate and reliable forecasting mechanisms to guide public health responses and policies. In this study, we compared several cutting-edge deep learning models, including Long Short-Term Memory (LSTM), bidirectional LSTM, Convolutional Neural Networks (CNN), hybrid CNN-LSTM, Multilayer Perceptron's, and Recurrent Neural Networks (RNN), to project COVID-19 cases in the aforementioned regions. These models were calibrated and evaluated using a comprehensive dataset that includes confirmed case counts, demographic data, and relevant socioeconomic factors. To enhance the performance of these models, Bayesian optimization techniques were employed. Subsequently, the models were re-evaluated to compare their effectiveness. Analytic approaches, both predictive and retrospective in nature, were used to interpret the data. Our primary objective was to determine the most effective model for predicting COVID-19 cases in the United Arab Emirates (UAE) and Malaysia. The findings indicate that the selected deep learning algorithms were proficient in forecasting COVID-19 cases, although their efficacy varied across different models. After a thorough evaluation, the model architectures most suitable for the specific conditions in the UAE and Malaysia were identified. Our study contributes significantly to the ongoing efforts to combat the COVID-19 pandemic, providing crucial insights into the application of sophisticated deep learning algorithms for the precise and timely forecasting of COVID-19 cases. These insights hold substantial value for shaping public health strategies, enabling authorities to develop targeted and evidence-based interventions to manage the virus spread and its impact on the populations of the UAE and Malaysia. The study confirms the usefulness of deep learning methodologies in efficiently processing complex datasets and generating reliable projections, a skill of great importance in healthcare and professional settings.

Innovative applications of artificial intelligence during the COVID-19 pandemic

The COVID-19 pandemic has created unprecedented challenges worldwide. Artificial intelligence (AI) technologies hold tremendous potential for tackling key aspects of pandemic management and response. In the present review, we discuss the tremendous possibilities of AI technology in addressing the global challenges posed by the COVID-19 pandemic. First, we outline the multiple impacts of the current pandemic on public health, the economy, and society. Next, we focus on the innovative applications of advanced AI technologies in key areas such as COVID-19 prediction, detection, control, and drug discovery for treatment. Specifically, AI-based predictive analytics models can use clinical, epidemiological, and omics data to forecast disease spread and patient outcomes. Additionally, deep neural networks enable rapid diagnosis through medical imaging. Intelligent systems can support risk assessment, decision-making, and social sensing, thereby improving epidemic control and public health policies. Furthermore, high-throughput virtual screening enables AI to accelerate the identification of therapeutic drug candidates and opportunities for drug repurposing. Finally, we discuss future research directions for AI technology in combating COVID-19, emphasizing the importance of interdisciplinary collaboration. Though promising, barriers related to model generalization, data quality, infrastructure readiness, and ethical risks must be addressed to fully translate these innovations into real-world impacts. Multidisciplinary collaboration engaging diverse expertise and stakeholders is imperative for developing robust, responsible, and human-centered AI solutions against COVID-19 and future public health emergencies.

Predicting COVID-19 new cases in California with Google Trends data and a machine learning approach

BACKGROUND

Google Trends data can be a valuable source of information for health-related issues such as predicting infectious disease trends.

OBJECTIVES

To evaluate the accuracy of predicting COVID-19 new cases in California using Google Trends data, we develop and use a GMDH-type neural network model and compare its performance with a LTSM model.

METHODS

We predicted COVID-19 new cases using Google query data over three periods. Our first period covered March 1, 2020, to July 31, 2020, including the first peak of infection. We also estimated a model from October 1, 2020, to January 7, 2021, including the second wave of COVID-19 and avoiding possible biases from public interest in searching about the new pandemic. In addition, we extended our forecasting period from May 20, 2020, to January 31, 2021, to cover an extended period of time.

RESULTS

Our findings show that Google relative search volume (RSV) can be used to accurately predict new COVID-19 cases.  We find that among our Google relative search volume terms, "Fever," "COVID Testing," "Signs of COVID," "COVID Treatment," and "Shortness of Breath" increase model predictive accuracy.

CONCLUSIONS

Our findings highlight the value of using data sources providing near real-time data, e.g., Google Trends, to detect trends in COVID-19 cases, in order to supplement and extend existing epidemiological models.

Optimal adaptive nonpharmaceutical interventions to mitigate the outbreak of respiratory infections following the COVID-19 pandemic: a deep reinforcement learning study in Hong Kong, China

BACKGROUND

Long-lasting nonpharmaceutical interventions (NPIs) suppressed the infection of COVID-19 but came at a substantial economic cost and the elevated risk of the outbreak of respiratory infectious diseases (RIDs) following the pandemic. Policymakers need data-driven evidence to guide the relaxation with adaptive NPIs that consider the risk of both COVID-19 and other RIDs outbreaks, as well as the available healthcare resources.

METHODS

Combining the COVID-19 data of the sixth wave in Hong Kong between May 31, 2022 and August 28, 2022, 6-year epidemic data of other RIDs (2014-2019), and the healthcare resources data, we constructed compartment models to predict the epidemic curves of RIDs after the COVID-19-targeted NPIs. A deep reinforcement learning (DRL) model was developed to learn the optimal adaptive NPIs strategies to mitigate the outbreak of RIDs after COVID-19-targeted NPIs are lifted with minimal health and economic cost. The performance was validated by simulations of 1000 days starting August 29, 2022. We also extended the model to Beijing context.

FINDINGS

Without any NPIs, Hong Kong experienced a major COVID-19 resurgence far exceeding the hospital bed capacity. Simulation results showed that the proposed DRL-based adaptive NPIs successfully suppressed the outbreak of COVID-19 and other RIDs to lower than capacity. DRL carefully controlled the epidemic curve to be close to the full capacity so that herd immunity can be reached in a relatively short period with minimal cost. DRL derived more stringent adaptive NPIs in Beijing.

INTERPRETATION

DRL is a feasible method to identify the optimal adaptive NPIs that lead to minimal health and economic cost by facilitating gradual herd immunity of COVID-19 and mitigating the other RIDs outbreaks without overwhelming the hospitals. The insights can be extended to other countries/regions.

Predicting medical device failure: a promise to reduce healthcare facilities cost through smart healthcare management

Background

The advancement of biomedical research generates myriad healthcare-relevant data, including medical records and medical device maintenance information. The COVID-19 pandemic significantly affects the global mortality rate, creating an enormous demand for medical devices. As information technology has advanced, the concept of intelligent healthcare has steadily gained prominence. Smart healthcare utilises a new generation of information technologies, such as the Internet of Things (loT), big data, cloud computing, and artificial intelligence, to completely transform the traditional medical system. With the intention of presenting the concept of smart healthcare, a predictive model is proposed to predict medical device failure for intelligent management of healthcare services.

Methods

Present healthcare device management can be improved by proposing a predictive machine learning model that prognosticates the tendency of medical device failures toward smart healthcare. The predictive model is developed based on 8,294 critical medical devices from 44 different types of equipment extracted from 15 healthcare facilities in Malaysia. The model classifies the device into three classes; (i) class 1, where the device is unlikely to fail within the first 3 years of purchase, (ii) class 2, where the device is likely to fail within 3 years from purchase date, and (iii) class 3 where the device is likely to fail more than 3 years after purchase. The goal is to establish a precise maintenance schedule and reduce maintenance and resource costs based on the time to the first failure event. A machine learning and deep learning technique were compared, and the best robust model for smart healthcare was proposed.

Results

This study compares five algorithms in machine learning and three optimizers in deep learning techniques. The best optimized predictive model is based on ensemble classifier and SGDM optimizer, respectively. An ensemble classifier model produces 77.90%, 87.60%, and 75.39% for accuracy, specificity, and precision compared to 70.30%, 83.71%, and 67.15% for deep learning models. The ensemble classifier model improves to 79.50%, 88.36%, and 77.43% for accuracy, specificity, and precision after significant features are identified. The result concludes although machine learning has better accuracy than deep learning, more training time is required, which is 11.49 min instead of 1 min 5 s when deep learning is applied. The model accuracy shall be improved by introducing unstructured data from maintenance notes and is considered the author's future work because dealing with text data is time-consuming. The proposed model has proven to improve the devices' maintenance strategy with a Malaysian Ringgit (MYR) cost reduction of approximately MYR 326,330.88 per year. Therefore, the maintenance cost would drastically decrease if this smart predictive model is included in the healthcare management system.

Global analysis and prediction scenario of infectious outbreaks by recurrent dynamic model and machine learning models: A case study on COVID-19

It is essential to evaluate patient outcomes at an early stage when dealing with a pandemic to provide optimal clinical care and resource management. Many methods have been proposed to provide a roadmap against different pandemics, including the recent pandemic disease COVID-19. Due to recurrent epidemic waves of COVID-19, which have been observed in many countries, mathematical modeling and forecasting of COVID-19 are still necessary as long as the world continues to battle against the pandemic. Modeling may aid in determining which interventions to try or predict future growth patterns. In this article, we design a combined approach for analyzing any pandemic in two separate parts. In the first part of the paper, we develop a recurrent SEIRS compartmental model to predict recurrent outbreak patterns of diseases. Due to its time-varying parameters, our model is able to reflect the dynamics of infectious diseases, and to measure the effectiveness of the restrictive measures. We discuss the stable solutions of the corresponding autonomous system with frozen parameters. We focus on the regime shifts and tipping points; then we investigate tipping phenomena due to parameter drifts in our time-varying parameters model that exhibits a bifurcation in the frozen-in case. Furthermore, we propose an optimal numerical design for estimating the system’s parameters. In the second part, we introduce machine learning models to strengthen the methodology of our paper in data analysis, particularly for prediction scenarios. We use MLP, RBF, LSTM, ANFIS, and GRNN for training and evaluation of COVID-19. Then, we compare the results with the recurrent dynamical system in the fitting process and prediction scenario. We also confirm results by implementing our methods on the released data on COVID-19 by WHO for Italy, Germany, Iran, and South Africa between 1/22/2020 and 7/24/2021, when people were engaged with different variants including Alpha, Beta, Gamma, and Delta. The results of this article show that the dynamic model is adequate for long-term analysis and data fitting, as well as obtaining parameters affecting the epidemic. However, it is ineffective in providing a long-term forecast. In contrast machine learning methods effectively provide disease prediction, although they do not provide analysis such as dynamic models. Finally, some metrics, including RMSE, R-Squared, and accuracy, are used to evaluate the machine learning models. These metrics confirm that ANFIS and RBF perform better than other methods in training and testing zones.

A comparative study of three models to analyze the impact of air pollutants on the number of pulmonary tuberculosis cases in Urumqi, Xinjiang

In this paper, we separately constructed ARIMA, ARIMAX, and RNN models to determine whether there exists an impact of the air pollutants (such as PM2.5, PM10, CO, O3, NO2, and SO2) on the number of pulmonary tuberculosis cases from January 2014 to December 2018 in Urumqi, Xinjiang. In addition, by using a new comprehensive evaluation index DISO to compare the performance of three models, it was demonstrated that ARIMAX (1,1,2) × (0,1,1)12 + PM2.5 (lag = 12) model was the optimal one, which was applied to predict the number of pulmonary tuberculosis cases in Urumqi from January 2019 to December 2019. The predicting results were in good agreement with the actual pulmonary tuberculosis cases and shown that pulmonary tuberculosis cases obviously declined, which indicated that the policies of environmental protection and universal health checkups in Urumqi have been very effective in recent years.

A Hybrid Model for Coronavirus Disease 2019 Forecasting Based on Ensemble Empirical Mode Decomposition and Deep Learning

The novel coronavirus pneumonia that began to spread in 2019 is still raging and has placed a burden on medical systems and governments in various countries. For policymaking and medical resource decisions, a good prediction model is necessary to monitor and evaluate the trends of the epidemic. We used a long short-term memory (LSTM) model and the improved hybrid model based on ensemble empirical mode decomposition (EEMD) to predict COVID-19 trends; Methods: The data were collected from the Harvard Dataverse. Epidemic data from 21 January 2020 to 25 April 2021 for California, the most severely affected state in the United States, were used to develop an LSTM model and an EEMD-LSTM hybrid model, which is an LSTM model combined with ensemble empirical mode decomposition. In this study, ninety percent of the data were adopted to fit the models as a training set, while the subsequent 10% were used to test the prediction effect of the models. The mean absolute percentage error, mean absolute error, and root mean square error were used to evaluate the prediction performances of the models; Results: The results indicated that the number of confirmed cases in California was increasing as of 25 April 2021, with no obvious evidence of a sharp decline. On 25 April 2021, the LSTM model predicted 3666418 confirmed cases, whereas the EEMD-LSTM predicted 3681150. The mean absolute percentage errors for the LSTM and EEMD-LSTM models were 0.0151 and 0.0051, respectively. The mean absolute and root mean square errors were 5.58 × 10⁴ and 5.63 × 10⁴ for the LSTM model and 1.9 × 10⁴ and 2.43 × 10⁴ for the EEMD-LSTM model, respectively; Conclusions: The results showed the advantage of an EEMD-LSTM model over a single LSTM model, and the established EEMD-LSTM model may be suitable for monitoring and evaluating the epidemic situation and providing quantitative analysis evidence for epidemic prevention and control.

Prediction of Omicron Virus Using Combined Extended Convolutional and Recurrent Neural Networks Technique on CT-Scan Images

COVID-19 has sparked a global pandemic, with a variety of inflamed instances and deaths increasing on an everyday basis. Researchers are actively increasing and improving distinct mathematical and ML algorithms to forecast the infection. The prediction and detection of the Omicron variant of COVID-19 brought new issues for the health fraternity due to its ubiquity in human beings. In this research work, two learning algorithms, namely, deep learning (DL) and machine learning (ML), were developed to forecast the Omicron virus infections. Automatic disease prediction and detection have become crucial issues in medical science due to rapid population growth. In this research study, a combined Extended CNN-RNN research model was developed on a chest CT-scan image dataset to predict the number of +ve and -ve cases of Omicron virus infections. The proposed research model was evaluated and compared against the existing system utilizing a dataset of 16,733-sample training and testing CT-scan images collected from the Kaggle repository. This research article aims to introduce a combined ML and DL technique based on the combination of an Extended Convolutional Neural Network (ECNN) and an Extended Recurrent Neural Network (ERNN) to diagnose and predict Omicron virus-infected cases automatically using chest CT-scan images. To overcome the drawbacks of the existing system, this research proposes a combined research model that is ECNN-ERNN, where ECNN is used for the extraction of deep features and ERNN is used for exploration using extracted features. A dataset of 16,733 Omicron computer tomography images was used as a pilot assessment for this proposed prototype. The investigational experiment results show that the projected prototype provides 97.50% accuracy, 98.10% specificity, 98.80% of AUC, and 97.70% of F1-score. To the last, the study outlines the advantages being offered by the proposed model with respect to other existing models by comparing different parameters of validation such as accuracy, error rate, data size, time complexity, and execution time.

Quantifying the dynamic transmission of COVID-19 asymptomatic and symptomatic infections: Evidence from four Chinese regions

The dynamic transmission of asymptomatic and symptomatic COVID-19 infections is difficult to quantify because asymptomatic infections are not readily recognized or self-identified. To address this issue, we collected data on asymptomatic and symptomatic infections from four Chinese regions (Beijing, Dalian, Xinjiang, and Guangzhou). These data were considered reliable because the government had implemented large-scale multiple testing during the outbreak in the four regions. We modified the classical susceptible-exposure-infection-recovery model and combined it with mathematical tools to quantitatively analyze the number of infections caused by asymptomatic and symptomatic infections during dynamic transmission, respectively. The results indicated that the ratios of the total number of asymptomatic to symptomatic infections were 0.13:1, 0.48:1, 0.29:1, and 0.15:1, respectively, in the four regions. However, the ratio of the total number of infections caused by asymptomatic and symptomatic infections were 4.64:1, 6.21:1, 1.49:1, and 1.76:1, respectively. Furthermore, the present study describes the daily number of healthy people infected by symptomatic and asymptomatic transmission and the dynamic transmission process. Although there were fewer asymptomatic infections in the four aforementioned regions, their infectivity was found to be significantly higher, implying a greater need for timely screening and control of infections, particularly asymptomatic ones, to contain the spread of COVID-19.

Real-time trajectory privacy protection based on improved differential privacy method and deep learning model

Accurate and real-time trajectory data publishing plays an important role in providing users with the latest traffic and road condition information to help in rationally planning travel time and routes. However, the improper publishing of location information and reverse analysis and reasoning can easily leak users’ personal information, which may threaten users’ privacy and lives. Owing to the inclusion of differential privacy model noise, privacy protection introduces inaccuracies in data publishing and validity. To improve the accuracy and usability of published data, we propose a data publishing method based on deep learning and differential privacy models for securing spatiotemporal trajectory data publishing. The method divides the trajectory data into two-dimensional grid regions, counts the density of trajectories at grids, performs a top-down recursive division of regions, and formulates rules for privacy budget allocation from multiple perspectives as recurrence depth increases. Furthermore, the method integrates spatiotemporal sequence data according to temporal order. Subsequently, it extracts temporal and spatial features of the data by the temporal graph convolutional network model for budget matrix prediction, adds Laplace noise to the regions, and evaluates the effect of differential privacy protection with the original data to protect trajectory data privacy. Experiments demonstrate that under the premise of satisfying ε-difference privacy, the query error and Jensen–Shannon divergence are smaller, the Kendall coefficient is more consistent, and the upper and lower limit values are more stable. Hence, the top-down division method achieves better results than those of the two traditional region division methods of the uniform grid and adaptive grid. The proposed method can be used to allocate the privacy budget more reasonably and achieve privacy protection of trajectories, which can be applied to a large amount of spatiotemporal trajectory data.

Voice spoofing countermeasure for voice replay attacks using deep learning

In our everyday lives, we communicate with each other using several means and channels of communication, as communication is crucial in the lives of humans. Listening and speaking are the primary forms of communication. For listening and speaking, the human voice is indispensable. Voice communication is the simplest type of communication. The Automatic Speaker Verification (ASV) system verifies users with their voices. These systems are susceptible to voice spoofing attacks - logical and physical access attacks. Recently, there has been a notable development in the detection of these attacks. Attackers use enhanced gadgets to record users’ voices, replay them for the ASV system, and be granted access for harmful purposes. In this work, we propose a secure voice spoofing countermeasure to detect voice replay attacks. We enhanced the ASV system security by building a spoofing countermeasure dependent on the decomposed signals that consist of prominent information. We used two main features— the Gammatone Cepstral Coefficients and Mel-Frequency Cepstral Coefficients— for the audio representation. For the classification of the features, we used Bi-directional Long-Short Term Memory Network in the cloud, a deep learning classifier. We investigated numerous audio features and examined each feature’s capability to obtain the most vital details from the audio for it to be labelled genuine or a spoof speech. Furthermore, we use various machine learning algorithms to illustrate the superiority of our system compared to the traditional classifiers. The results of the experiments were classified according to the parameters of accuracy, precision rate, recall, F1-score, and Equal Error Rate (EER). The results were 97%, 100%, 90.19% and 94.84%, and 2.95%, respectively.

Stacking Ensemble Method for Gestational Diabetes Mellitus Prediction in Chinese Pregnant Women: A Prospective Cohort Study

Gestational diabetes mellitus (GDM) is closely related to adverse pregnancy outcomes and other diseases. Early intervention in pregnant women who are at high risk of developing GDM could help prevent adverse health consequences. The study aims to develop a simple model using the stacking ensemble method to predict GDM for women in the first trimester based on easily available factors. We used the data from the Chinese Pregnant Women Cohort Study from July 2017 to November 2018. A total of 6,848 pregnant women in the first trimester were included in the analysis. Logistic regression (LR), random forest (RF), and extreme gradient boosting (XGBoost) were considered as base learners. Optimal feature subsets for each learner were chosen by using recursive feature elimination cross-validation. Then, we built a pipeline to process imbalance data, tune hyperparameters, and evaluate model performance. The learners with the best hyperparameters were employed in the first layer of the proposed stacking method. Their predictions were obtained using optimal feature subsets and served as meta-learner's inputs. Another LR was used as a meta-learner to obtain the final prediction results. Accuracy, specificity, error rate, and other metrics were calculated to evaluate the performance of the models. A paired samples t-test was performed to compare the model performance. In total, 967 (14.12%) women developed GDM. For base learners, the RF model had the highest accuracy (0.638 (95% confidence interval (CI) 0.628–0.648)) and specificity (0.683 (0.669–0.698)) and lowest error rate (0.362 (0.352–0.372)). The stacking method effectively improved the accuracy (0.666 (95% CI 0.663–0.670)) and specificity (0.725 (0.721–0.729)) and decreased the error rate (0.333 (0.330–0.337)). The differences in the performance between the stacking method and RF were statistically significant. Our proposed stacking method based on easily available factors has better performance than other learners such as RF.

Symptom-Based COVID-19 Prognosis through AI-Based IoT: A Bioinformatics Approach

Objective

Internet of Things (IoT) integrates several technologies where devices learn from the experience of each other thereby reducing human-intervened likely errors. Modern technologies like IoT and machine learning enable the conventional to patient-specific approach transition in healthcare. In conventional approach, the biggest challenge faced by healthcare professionals is to predict a disease by observing the symptoms, monitoring the remote area patient, and also attending to the patient all the time after being hospitalised. IoT provides real-time data, makes decision-making smarter, and provides far superior analytics, and all these to help improve the quality of healthcare. The main objective of the work was to create an IoT-based automated system using machine learning models for symptom-based COVID-19 prognosis.

Methods

Comparative analysis of predictive microbiology of COVID-19 from case symptoms using various machine learning classifiers like logistics regression, k-nearest neighbor, support vector machine, random forest, decision trees, Naïve Bayes, and gradient booster is reported here. For the sake of the validation and verification of the models, performance of each model based on the retrieved cloud-stored data was measured for accuracy.

Results

From the accuracy plot, it was concluded that k-NN was more accurate (97.97%) followed by decision tree (97.79), support vector machine (97.42), logistics regression (96.50), random forest (90.66), gradient boosting classifier (87.77), and Naïve Bayes (73.50) in COVID-19 prognosis.

Conclusion

The paper presents a health monitoring IoT framework having high clinical significance in real-time and remote healthcare monitoring. The findings reported here and the lessons learnt shall enable the healthcare system worldwide to counter not only this ongoing COVID but many other such global pandemics the humanity may suffer from time to come.

Daily surgery caseload prediction: towards improving operating theatre efficiency

BACKGROUND

In many hospitals, operating theatres are not used to their full potential due to the dynamic nature of demand and the complexity of theatre scheduling. Theatre inefficiencies may lead to access block and delays in treating patients requiring critical care. This study aims to employ operating theatre data to provide decision support for improved theatre management.

METHOD

Historical observations are used to predict long-term daily surgery caseload in various levels of granularity, from emergency versus elective surgeries to clinical specialty-level demands. A statistical modelling and a machine learning-based approach are developed to estimate daily surgery demand. The statistical model predicts daily demands based on historical observations through weekly rolling windows and calendar variables. The machine learning approach, based on regression algorithms, learns from a combination of temporal and sequential features. A de-identified data extract of elective and emergency surgeries at a major 783-bed metropolitan hospital over four years was used. The first three years of data were used as historical observations for training the models. The models were then evaluated on the final year of data.

RESULTS

Daily counts of overall surgery at a hospital-level could be predicted with approximately 90% accuracy, though smaller subgroups of daily demands by medical specialty are less predictable. Predictions were generated on a daily basis a year in advance with consistent predictive performance across the forecast horizon.

CONCLUSION

Predicting operating theatre demand is a viable component in theatre management, enabling hospitals to provide services as efficiently and effectively as possible to obtain the best health outcomes. Due to its consistent predictive performance over various forecasting ranges, this approach can inform both short-term staffing choices as well as long-term strategic planning.

COVID-19 Isolation Control Proposal via UAV and UGV for Crowded Indoor Environments: Assistive Robots in the Shopping Malls

Artificial intelligence researchers conducted different studies to reduce the spread of COVID-19. Unlike other studies, this paper isn't for early infection diagnosis, but for preventing the transmission of COVID-19 in social environments. Among the studies on this is regarding social distancing, as this method is proven to prevent COVID-19 to be transmitted from one to another. In the study, Robot Operating System (ROS) simulates a shopping mall using Gazebo, and customers are monitored by Turtlebot and Unmanned Aerial Vehicle (UAV, DJI Tello). Through frames analysis captured by Turtlebot, a particular person is identified and followed at the shopping mall. Turtlebot is a wheeled robot that follows people without contact and is used as a shopping cart. Therefore, a customer doesn't touch the shopping cart that someone else comes into contact with, and also makes his/her shopping easier. The UAV detects people from above and determines the distance between people. In this way, a warning system can be created by detecting places where social distance is neglected. Histogram of Oriented-Gradients (HOG)-Support Vector Machine (SVM) is applied by Turtlebot to detect humans, and Kalman-Filter is used for human tracking. SegNet is performed for semantically detecting people and measuring distance via UAV. This paper proposes a new robotic study to prevent the infection and proved that this system is feasible.

A Visualized Dynamic Prediction Model for Overall Survival in Elderly Patients With Pancreatic Cancer for Smart Medical Services

Background

Pancreatic cancer (PC) is a highly malignant tumor of the digestive system. The number of elderly patients with PC is increasing, and older age is related to a worse prognosis. Accurate prognostication is crucial in treatment decisions made for people diagnosed with PC. However, an accurate predictive model for the prognosis of these patients is still lacking. We aimed to construct nomograms for predicting the overall survival (OS) of elderly patients with PC.

Methods

Patients with PC, older than 65 years old from 2010 to 2015 in the Surveillance, Epidemiology, and End Results database, were selected and randomly divided into training cohort (n = 4,586) and validation cohort (n = 1,966). Data of patients in 2016-2018 (n = 1,761) were used for external validation. Univariable and forward stepwise multivariable Cox analysis was used to determine the independent prognostic factors. We used significant variables in the training set to construct nomograms predicting prognosis. The performance of the models was evaluated for their discrimination and calibration power based on the concordance index (C-index), calibration curve, and the decision curve analysis (DCA).

Results

Age, insurance, grade, surgery, radiation, chemotherapy, T, N, and American Joint Commission on Cancer were independent predictors for OS and thus were included in our nomogram. In the training cohort and validation cohort, the C-indices of our nomogram were 0.725 (95%CI: 0.715-0.735) and 0.711 (95%CI: 0.695-0.727), respectively. The 1-, 3-, and 5-year areas under receiver operating characteristic curves showed similar results. The calibration curves showed a high consensus between observations and predictions. In the external validation cohort, C-index (0.797, 95%CI: 0.778-0.816) and calibration curves also revealed high consistency between observations and predictions. The nomogram-related DCA curves showed better clinical utility compared to tumor-node-metastasis staging. In addition, we have developed an online prediction tool for OS.

Conclusions

A web-based prediction model for OS in elderly patients with PC was constructed and validated, which may be useful for prognostic assessment, treatment strategy selection, and follow-up management of these patients.

FedSGDCOVID: Federated SGD COVID-19 Detection under Local Differential Privacy Using Chest X-ray Images and Symptom Information

Coronavirus (COVID-19) has created an unprecedented global crisis because of its detrimental effect on the global economy and health. COVID-19 cases have been rapidly increasing, with no sign of stopping. As a result, test kits and accurate detection models are in short supply. Early identification of COVID-19 patients will help decrease the infection rate. Thus, developing an automatic algorithm that enables the early detection of COVID-19 is essential. Moreover, patient data are sensitive, and they must be protected to prevent malicious attackers from revealing information through model updates and reconstruction. In this study, we presented a higher privacy-preserving federated learning system for COVID-19 detection without sharing data among data owners. First, we constructed a federated learning system using chest X-ray images and symptom information. The purpose is to develop a decentralized model across multiple hospitals without sharing data. We found that adding the spatial pyramid pooling to a 2D convolutional neural network improves the accuracy of chest X-ray images. Second, we explored that the accuracy of federated learning for COVID-19 identification reduces significantly for non-independent and identically distributed (Non-IID) data. We then proposed a strategy to improve the model's accuracy on Non-IID data by increasing the total number of clients, parallelism (client-fraction), and computation per client. Finally, for our federated learning model, we applied a differential privacy stochastic gradient descent (DP-SGD) to improve the privacy of patient data. We also proposed a strategy to maintain the robustness of federated learning to ensure the security and accuracy of the model.

Development and Validation of a Competitive Risk Model in Elderly Patients With Chromophobe Cell Renal Carcinoma: A Population-Based Study

Background

Renal cell carcinoma (RCC) is the most common renal malignancy in adults, and chromophobe renal cell carcinoma (chRCC) is the third most common subtype of RCC. We aimed to construct a competitive risk model to predict cancer-specific survival (CSS) in elderly patients with chRCC.

Methods

The clinicopathological information of the patients was downloaded from the SEER database, and the patients were randomly divided into the training and validation cohorts. Patients' risk factors for cancer-specific death (CSM) were analyzed using proportional subdistribution hazard (SH). We constructed a competitive risk model to predict the CSS of elderly chRCC patients. Consistency index (C-index), the area under receiver operating curve (AUC), and a calibration curve were used to validate the model's accuracy. Decision curve analysis (DCA) was used to test the clinical value of the model.

Results

A total of 3,522 elderly patients with chRCC were included in the analysis. Patients were randomly assigned to either the training cohort (N = 2,474) or the validation cohort (N = 1,048). SH analysis found that age, race, T, N, and M stage, tumor size, and surgery were risk factors for CSM. We constructed a competitive risk model to predict patients' CSS. In the training set, the model predicted patients' 1-, 3-, and 5-year CSS with C-indices of 82.2, 80.8, and 78.2, respectively. The model predicted patient 1-, 3-, and 5-year CSS in the validation cohort with C-indices of 84.7, 83.4, and 76.9, respectively. The calibration curve showed that the model's predicted value is almost consistent with the observed value, which indicated that the model has good accuracy. The AUC of the training set and validation queue also suggested that the model has good discrimination. The clinical utility of the DCA model in predicting patients' CSS is higher than that of traditional TNM staging.

Conclusions

We constructed a competitive risk model to predict CSS in elderly patients with chRCC. The model has good accuracy and reliability, which can help doctors and patients to make clinical decisions and follow-up strategies.

Automated System for Identifying COVID-19 Infections in Computed Tomography Images Using Deep Learning Models

Coronavirus disease 2019 (COVID-19) is a novel disease that affects healthcare on a global scale and cannot be ignored because of its high fatality rate. Computed tomography (CT) images are presently being employed to assist doctors in detecting COVID-19 in its early stages. In several scenarios, a combination of epidemiological criteria (contact during the incubation period), the existence of clinical symptoms, laboratory tests (nucleic acid amplification tests), and clinical imaging-based tests are used to diagnose COVID-19. This method can miss patients and cause more complications. Deep learning is one of the techniques that has been proven to be prominent and reliable in several diagnostic domains involving medical imaging. This study utilizes a convolutional neural network (CNN), stacked autoencoder, and deep neural network to develop a COVID-19 diagnostic system. In this system, classification undergoes some modification before applying the three CT image techniques to determine normal and COVID-19 cases. A large-scale and challenging CT image dataset was used in the training process of the employed deep learning model and reporting their final performance. Experimental outcomes show that the highest accuracy rate was achieved using the CNN model with an accuracy of 88.30%, a sensitivity of 87.65%, and a specificity of 87.97%. Furthermore, the proposed system has outperformed the current existing state-of-the-art models in detecting the COVID-19 virus using CT images.

Development and Validation of a Nomogram to Predict Cancer-Specific Survival for Middle-Aged Patients With Early-Stage Hepatocellular Carcinoma

Background

Hepatocellular carcinoma is a common cause of death in middle-aged patients. We aimed to construct a new nomogram to predict cancer-specific survival (CSS) in middle-aged patients with hepatocellular carcinoma at an early stage.

Method

We collected clinicopathological information on early middle-aged patients with hepatocellular carcinoma from the SEER database. Univariate and multivariate Cox regression models were used to screen the independent risk factors for prognosis. These risk factors were used to construct predictions of CSS in patients with hepatocellular carcinoma. Consistency index (C- index), calibration curve, area under the receiver operating curve (AUC) were used. A decision analysis curve (DCA) was used to evaluate the clinical utility of the predictive model.

Results

A total of 6,286 patients with hepatocellular carcinoma in early middle age were enrolled. Univariate and multivariate Cox regression analysis showed that sex, marriage, race, histological tumor grade, T stage, surgery, chemotherapy, AFP, and tumor size were independent risk factors for prognosis. All independent risk factors were included in the nomogram to predict CSS at 1-, 3-, and 5-years in early middle age patients with hepatocellular carcinoma. In the training cohort and validation cohort, the C-index of the prediction model was 0.728 (95%CI: 0.716–0.740) and 0.733 (95%CI: 0.715–0.751), respectively. The calibration curve showed that the predicted value of the prediction model is highly consistent with the observed value. AUC also suggested that the model has good discrimination. DCA suggested that the nomogram had better predictive power than T staging.

Conclusion

We constructed a new nomogram to predict CSS in middle-aged patients with early-stage hepatocellular carcinoma. This prediction model has good accuracy and reliability, which can help patients and doctors to judge prognosis and make clinical decisions.

Deep recurrent reinforced learning model to compare the efficacy of targeted local versus national measures on the spread of COVID-19 in the UK

OBJECTIVES

To prevent the emergence of new waves of COVID-19 caseload and associated mortalities, it is imperative to understand better the efficacy of various control measures on the national and local development of this pandemic in space-time, characterise hotspot regions of high risk, quantify the impact of under-reported measures such as international travel and project the likely effect of control measures in the coming weeks.

METHODS

We applied a deep recurrent reinforced learning based model to evaluate and predict the spatiotemporal effect of a combination of control measures on COVID-19 cases and mortality at the local authority (LA) and national scale in England, using data from week 5 to 46 of 2020, including an expert curated control measure matrix, official statistics/government data and a secure web dashboard to vary magnitude of control measures.

RESULTS

Model predictions of the number of cases and mortality of COVID-19 in the upcoming 5 weeks closely matched the actual values (cases: root mean squared error (RMSE): 700.88, mean absolute error (MAE): 453.05, mean absolute percentage error (MAPE): 0.46, correlation coefficient 0.42; mortality: RMSE 14.91, MAE 10.05, MAPE 0.39, correlation coefficient 0.68). Local lockdown with social distancing (LD_SD) (overall rank 3) was found to be ineffective in preventing outbreak rebound following lockdown easing compared with national lockdown (overall rank 2), based on prediction using simulated control measures. The ranking of the effectiveness of adjunctive measures for LD_SD were found to be consistent across hotspot and non-hotspot regions. Adjunctive measures found to be most effective were international travel and quarantine restrictions.

CONCLUSIONS

This study highlights the importance of using adjunctive measures in addition to LD_SD following lockdown easing and suggests the potential importance of controlling international travel and applying travel quarantines. Further work is required to assess the effect of variant strains and vaccination measures.

An Empirical Model to Predict the Diabetic Positive Using Stacked Ensemble Approach

Today, disease detection automation is widespread in healthcare systems. The diabetic disease is a significant problem that has spread widely all over the world. It is a genetic disease that causes trouble for human life throughout the lifespan. Every year the number of people with diabetes rises by millions, and this affects children too. The disease identification involves manual checking so far, and automation is a current trend in the medical field. Existing methods use a single algorithm for the prediction of diabetes. For complex problems, a single model is not enough because it may not be suitable for the input data or the parameters used in the approach. To solve complex problems, multiple algorithms are used. These multiple algorithms follow a homogeneous model or heterogeneous model. The homogeneous model means the same algorithm, but the model has been used multiple times. In the heterogeneous model, different algorithms are used. This paper adopts a heterogeneous ensemble model called the stacked ensemble model to predict whether a person has diabetes positively or negatively. This stacked ensemble model is advantageous in the prediction. Compared to other existing models such as logistic regression Naïve Bayes (72), (74.4), and LDA (81%), the proposed stacked ensemble model has achieved 93.1% accuracy in predicting blood sugar disease.

Machine Learning-Based Prediction Models for Depression Symptoms Among Chinese Healthcare Workers During the Early COVID-19 Outbreak in 2020: A Cross-Sectional Study

BACKGROUND

The 2019 novel coronavirus (COVID-19)-related depression symptoms of healthcare workers have received worldwide recognition. Although many studies identified risk exposures associated with depression symptoms among healthcare workers, few have focused on a predictive model using machine learning methods. As a society, governments, and organizations are concerned about the need for immediate interventions and alert systems for healthcare workers who are mentally at-risk. This study aims to develop and validate machine learning-based models for predicting depression symptoms using survey data collected during the COVID-19 outbreak in China.

METHOD

Surveys were conducted of 2,574 healthcare workers in hospitals designated to care for COVID-19 patients between 20 January and 11 February 2020. The patient health questionnaire (PHQ)-9 was used to measure the depression symptoms and quantify the severity, a score of ≥5 on the PHQ-9 represented depression symptoms positive, respectively. Four machine learning approaches were trained (75% of data) and tested (25% of data). Cross-validation with 100 repetitions was applied to the training dataset for hyperparameter tuning. Finally, all models were compared to evaluate their predictive performances and screening utility: decision tree, logistics regression with least absolute shrinkage and selection operator (LASSO), random forest, and gradient-boosting tree.

RESULTS

Important risk predictors identified and ranked by the machine learning models were highly consistent: self-perceived health status factors always occupied the top five most important predictors, followed by worried about infection, working on the frontline, a very high level of uncertainty, having received any form of psychological support material and having COVID-19-like symptoms. The area under the curve [95% CI] of machine learning models were as follows: LASSO model, 0.824 [0.792-0.856]; random forest, 0.828 [0.797-0.859]; gradient-boosting tree, 0.829 [0.798-0.861]; and decision tree, 0.785 [0.752-0.819]. The calibration plot indicated that the LASSO model, random forest, and gradient-boosting tree fit the data well. Decision curve analysis showed that all models obtained net benefits for predicting depression symptoms.

CONCLUSIONS

This study shows that machine learning prediction models are suitable for making predictions about mentally at-risk healthcare workers predictions in a public health emergency setting. The application of multidimensional machine learning models could support hospitals' and healthcare workers' decision-making on possible psychological interventions and proper mental health management.