Predicting the necessity of oxygen therapy in the early stage of COVID-19 using machine learning

Medical oxygen is a critical element in the treatment process of COVID-19 patients which its shortage impacts the treatment process adversely. This study aims to apply machine learning (ML) to predict the requirement for oxygen-based treatment for hospitalized COVID-19 patients. In the first phase, demographic information, symptoms, and patient's background were extracted from the databases of two local hospitals in Iran, and preprocessing actions were applied. In the second step, the related features were selected. Lastly, five ML models including logistic regression (LR), random forest (RF), XGBoost, C5.0, and neural networks (NNs) were implemented and compared based on their accuracy and capability. Among the variables related to the patient's background, consuming opium due to the high rate of opium users in Iran was considered in the models. Of the 398 patients included in the study, 112 (28.14%) received oxygen-based treatment. Shortness of breath (71.42%), fever (62.5%), and cough (59.82%) had the highest frequency in patients with oxygen requirements. The most important variables for prediction were shortness of breath, cough, age, and fever. For opioid-addicted patients, in addition to the high mortality rate (23.07%), the rate of oxygen-based treatment was twice as high as non-addicted patients. XGBoost and LR obtained the highest area under the curve with values of 88.7% and 88.3%, respectively. For accuracy, LR and NNs achieved the best and same accuracy (86.42%). This approach provides a tool that accurately predicts the need for oxygen in the treatment process of COVID-19 patients and helps hospital resource management.

Using machine learning in prediction of ICU admission, mortality, and length of stay in the early stage of admission of COVID-19 patients

The recent COVID-19 pandemic has affected health systems across the world. Especially, Intensive Care Units (ICUs) have played a pivotal role in the treatment of critically-ill patients. At the same time however, the increasing number of admissions due to the vast prevalence of the virus have caused several problems for ICU wards such as overburdening of staff and shortages of medical resources. These issues might have affected the quality of healthcare services provided directly impacting a patient's survival. The objective of this research is to leverage Machine Learning (ML) on hospital data in order to support hospital managers and practitioners with the treatment of COVID-19 patients. This is accomplished by providing more detailed inference about a patient's likelihood of ICU admission, mortality and in case of hospitalization the length of stay (LOS). In this pursuit, the outcome variables are in three separate models predicted by five different ML algorithms: eXtreme Gradient Boosting (XGB), K-Nearest Neighbor (KNN), Random Forest (RF), bagged-CART (b-CART), and LogitBoost (LB). With the exception of KNN, the studied models show good predictive capabilities when evaluating relevant accuracy scores, such as area under the curve. By implementing an ensemble stacking approach (either a Neural Net or a General Linear Model) on top of the aforementioned ML algorithms the performance is further boosted. Ultimately, for the prediction of admission to the ICU, the ensemble stacking via a Neural Net achieved the best result with an accuracy of over 95%. For mortality at the ICU, the vanilla XGB performed slightly better (1% difference with the meta-model). To predict large length of stays both ensemble stacking approaches yield comparable results. Besides it direct implications for managing COVID-19 patients, the approach presented serves as an example how data can be employed in future pandemics or crises.

A simulation optimization approach for resource allocation in an emergency department

The emergency department (ED) is a primary health care unit and one of the main entrances to the hospital system where appropriate, timely and good performance can save lives. Lack of sufficient resources, such as beds and qualified health care professionals, are major stumbling blocks to providing timely and suitable services; but resources availability and moving towards the ideal situation without attention to budget restrictions is neither practical nor achievable. In this study, simulation optimization is used to finding the best configuration in ED resources (e.g., Bed, Nurse, and GP) that affects a patient's length of stay, subject to budget constraints. Simulation is used to analyze the system and estimate target function an optimization model is then solved under different budget constraints. By considering the current budget, the new configuration of 20 inpatient beds, 3 nurses and 1 GP, with 554.4 minutes of a patient's length of stay shows 8.1% length of stay (LOS) improvement. Whilst with a maximum 35.5 budget units allocation of 20 inpatient beds, 4 nurses and 3 GPs a 9.5% decrease in LOS is proposed.

Predicting the dynamical behavior of COVID-19 epidemic and the effect of control strategies

This paper uses transformed subsystem of ordinary differential equation s e i r s model, with vital dynamics of birth and death rates, and temporary immunity (of infectious individuals or vaccinated susceptible) to evaluate the disease-free D F E X ¯ D F E , and endemic E E X ¯ E E equilibrium points, using the Jacobian matrix eigenvalues λ i of both disease-free equilibriumX ¯ D F E , and endemic equilibriumX ¯ E E for COVID-19 infectious disease to show S, E, I, and R ratios to the population in time-series. In order to obtain the disease-free equilibrium point, globally asymptotically stable (R 0 ≤ 1 ), the effect of control strategies has been added to the model (in order to decrease transmission rateβ , and reinforce susceptible to recovered flow), to determine how much they are effective, in a mass immunization program. The effect of transmission rates β (from S to E) and α (from R to S) varies, and when vaccination effectρ , is added to the model, disease-free equilibriumX ¯ D F E is globally asymptotically stable, and the endemic equilibrium pointX ¯ E E , is locally unstable. The initial conditions for the decrease in transmission rates of β and α , reached the corresponding disease-free equilibriumX ¯ D F E locally unstable, and globally asymptotically stable for endemic equilibriumX ¯ E E . The initial conditions for the decrease in transmission rate s β andα , and increase in ρ , reached the corresponding disease-free equilibriumX ¯ D F E globally asymptotically stable, and locally unstable in endemic equilibriumX ¯ E E .

A multi objective volleyball premier league algorithm for green scheduling identical parallel machines with splitting jobs

Parallel machine scheduling is one of the most common studied problems in recent years, however, this classic optimization problem has to achieve two conflicting objectives, i.e. minimizing the total tardiness and minimizing the total wastes, if the scheduling is done in the context of plastic injection industry where jobs are splitting and molds are important constraints. This paper proposes a mathematical model for scheduling parallel machines with splitting jobs and resource constraints. Two minimization objectives - the total tardiness and the number of waste - are considered, simultaneously. The obtained model is a bi-objective integer linear programming model that is shown to be of NP-hard class optimization problems. In this paper, a novel Multi-Objective Volleyball Premier League (MOVPL) algorithm is presented for solving the aforementioned problem. This algorithm uses the crowding distance concept used in NSGA-II as an extension of the Volleyball Premier League (VPL) that we recently introduced. Furthermore, the results are compared with six multi-objective metaheuristic algorithms of MOPSO, NSGA-II, MOGWO, MOALO, MOEA/D, and SPEA2. Using five standard metrics and ten test problems, the performance of the Pareto-based algorithms was investigated. The results demonstrate that in general, the proposed algorithm has supremacy than the other four algorithms.

Applications of different machine learning approaches in prediction of breast cancer diagnosis delay

Background

The increasing rate of breast cancer (BC) incidence and mortality in Iran has turned this disease into a challenge. A delay in diagnosis leads to more advanced stages of BC and a lower chance of survival, which makes this cancer even more fatal.

Objectives

The present study was aimed at identifying the predicting factors for delayed BC diagnosis in women in Iran.

Methods

In this study, four machine learning methods, including extreme gradient boosting (XGBoost), random forest (RF), neural networks (NNs), and logistic regression (LR), were applied to analyze the data of 630 women with confirmed BC. Also, different statistical methods, including chi-square, p-value, sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC), were utilized in different steps of the survey.

Results

Thirty percent of patients had a delayed BC diagnosis. Of all the patients with delayed diagnoses, 88.5% were married, 72.1% had an urban residency, and 84.8% had health insurance. The top three important factors in the RF model were urban residency (12.04), breast disease history (11.58), and other comorbidities (10.72). In the XGBoost, urban residency (17.54), having other comorbidities (17.14), and age at first childbirth (>30) (13.13) were the top factors; in the LR model, having other comorbidities (49.41), older age at first childbirth (82.57), and being nulliparous (44.19) were the top factors. Finally, in the NN, it was found that being married (50.05), having a marriage age above 30 (18.03), and having other breast disease history (15.83) were the main predicting factors for a delayed BC diagnosis.

Conclusion

Machine learning techniques suggest that women with an urban residency who got married or had their first child at an age older than 30 and those without children are at a higher risk of diagnosis delay. It is necessary to educate them about BC risk factors, symptoms, and self-breast examination to shorten the delay in diagnosis.

Analysis of non-pharmaceutical interventions impacts on COVID-19 pandemic in Iran

The COVID-19 pandemic shows to have a huge impact on people's health and countries' infrastructures around the globe. Iran was one of the first countries that experienced the vast prevalence of the coronavirus outbreak. The Iranian authorities applied various non-pharmaceutical interventions to eradicate the epidemic in different periods. This study aims to investigate the effectiveness of non-pharmaceutical interventions in managing the current Coronavirus pandemic and to predict the next wave of infection in Iran. To achieve the research objective, the number of cases and deaths before and after the interventions was studied and the effective reproduction number of the infection was analyzed under various scenarios. The SEIR generic model was applied to capture the dynamic of the pandemic in Iran. To capture the effects of different interventions, the corresponding reproduction number was considered. Depending on how people are responsive to interventions, the effectiveness of each intervention has been investigated. Results show that the maximum number of the total of infected individuals will occur around the end of May and the start of June 2021. It is concluded that the outbreak could be smoothed if full lockdown and strict quarantine continue. The proposed modeling could be used as an assessment tool to evaluate the effects of different interventions in new outbreaks.

Machine learning algorithms to uncover risk factors of breast cancer: insights from a large case-control study

Introduction

This large case-control study explored the application of machine learning models to identify risk factors for primary invasive incident breast cancer (BC) in the Iranian population. This study serves as a bridge toward improved BC prevention, early detection, and management through the identification of modifiable and unmodifiable risk factors.

Methods

The dataset includes 1,009 cases and 1,009 controls, with comprehensive data on lifestyle, health-behavior, reproductive and sociodemographic factors. Different machine learning models, namely Random Forest (RF), Neural Networks (NN), Bootstrap Aggregating Classification and Regression Trees (Bagged CART), and Extreme Gradient Boosting Tree (XGBoost), were employed to analyze the data.

Results

The findings highlight the significance of a chest X-ray history, deliberate weight loss, abortion history, and post-menopausal status as predictors. Factors such as second-hand smoking, lower education, menarche age (>14), occupation (employed), first delivery age (18-23), and breastfeeding duration (>42 months) were also identified as important predictors in multiple models. The RF model exhibited the highest Area Under the Curve (AUC) value of 0.9, as indicated by the Receiver Operating Characteristic (ROC) curve. Following closely was the Bagged CART model with an AUC of 0.89, while the XGBoost model achieved a slightly lower AUC of 0.78. In contrast, the NN model demonstrated the lowest AUC of 0.74. On the other hand, the RF model achieved an accuracy of 83.9% and a Kappa coefficient of 67.8% and the XGBoost, achieved a lower accuracy of 82.5% and a lower Kappa coefficient of 0.6.

Conclusion

This study could be beneficial for targeted preventive measures according to the main risk factors for BC among high-risk women.

Application of job shop scheduling approach in green patient flow optimization using a hybrid swarm intelligence

With the increasing demand for hospital services amidst the COVID-19 pandemic, allocation of limited public resources and management of healthcare services are of paramount importance. In the field of patient flow scheduling, previous research primarily focused on classical-based objective functions, while ignoring environmental-based objective functions. This study presents a flexible job shop scheduling problem to optimize patient flow and, thereby, minimize the total carbon footprint, as the sustainability-based objective function. Since flexible job shop scheduling is an NP-hard problem, a metaheuristic optimization algorithm, called Chaotic Salp Swarm Algorithm Enhanced with Opposition-Based Learning and Sine Cosine (CSSAOS), was developed. The proposed algorithm integrates the Salp Swarm Algorithm (SSA) with chaotic maps to update the position of followers, the sine cosine algorithm to update the leader position, and opposition-based learning for a better exploration of the search space. generating more accurate solutions. The proposed method was successfully applied in a real-world case study and demonstrated better performance than other well-known metaheuristic algorithms, including differential evolution, genetic algorithm, grasshopper optimization algorithm, SSA based on opposition-based learning, quantum evolutionary SSA, and whale optimization algorithm. In addition, it was found that the proposed method is scalable to different sizes and complexities.