Analyzing the impact of feature selection methods on machine learning algorithms for heart disease prediction.

The present study examines the role of feature selection methods in optimizing machine learning algorithms for predicting heart disease. The Cleveland Heart disease dataset with sixteen feature selection techniques in three categories of filter, wrapper, and evolutionary were used. Then seven algorithms Bayes net, Naïve Bayes (BN), multivariate linear model (MLM), Support Vector Machine (SVM), logit boost, j48, and Random Forest were applied to identify the best models for heart disease prediction. Precision, F-measure, Specificity, Accuracy, Sensitivity, ROC area, and PRC were measured to compare feature selection methods' effect on prediction algorithms. The results demonstrate that feature selection resulted in significant improvements in model performance in some methods (e.g., j48), whereas it led to a decrease in model performance in other models (e.g. MLP, RF). SVM-based filtering methods have a best-fit accuracy of 85.5. In fact, in a best-case scenario, filtering methods result in + 2.3 model accuracy. SVM-CFS/information gain/Symmetrical uncertainty methods have the highest improvement in this index. The filter feature selection methods with the highest number of features selected outperformed other methods in terms of models' ACC, Precision, and F-measures. However, wrapper-based and evolutionary algorithms improved models' performance from sensitivity and specificity points of view.

Comparing artificial neural network training algorithms to predict length of stay in hospitalized patients with COVID-19.

BACKGROUND: The exponential spread of coronavirus disease 2019 (COVID-19) causes unexpected economic burdens to worldwide health systems with severe shortages in hospital resources (beds, staff, equipment). Managing patients' length of stay (LOS) to optimize clinical care and utilization of hospital resources is very challenging. Projecting the future demand requires reliable prediction of patients' LOS, which can be beneficial for taking appropriate actions. Therefore, the purpose of this research is to develop and validate models using a multilayer perceptron-artificial neural network (MLP-ANN) algorithm based on the best training algorithm for predicting COVID-19 patients' hospital LOS. METHODS: Using a single-center registry, the records of 1225 laboratory-confirmed COVID-19 hospitalized cases from February 9, 2020 to December 20, 2020 were analyzed. In this study, first, the correlation coefficient technique was developed to determine the most significant variables as the input of the ANN models. Only variables with a correlation coefficient at a P-value < 0.2 were used in model construction. Then, the prediction models were developed based on 12 training algorithms according to full and selected feature datasets (90% of the training, with 10% used for model validation). Afterward, the root mean square error (RMSE) was used to assess the models' performance in order to select the best ANN training algorithm. Finally, a total of 343 patients were used for the external validation of the models. RESULTS: After implementing feature selection, a total of 20 variables were determined as the contributing factors to COVID-19 patients' LOS in order to build the models. The conducted experiments indicated that the best performance belongs to a neural network with 20 and 10 neurons in the hidden layer of the Bayesian regularization (BR) training algorithm for whole and selected features with an RMSE of 1.6213 and 2.2332, respectively. CONCLUSIONS: MLP-ANN-based models can reliably predict LOS in hospitalized patients with COVID-19 using readily available data at the time of admission. In this regard, the models developed in our study can help health systems to optimally allocate limited hospital resources and make informed evidence-based decisions.

Cardiopulmonary resuscitation of infants at birth: predictable or unpredictable?

BACKGROUND: Anticipating the need for at-birth cardiopulmonary resuscitation (CPR) in neonates is very important and complex. Timely identification and rapid CPR for neonates in the delivery room significantly reduce mortality and other neurological disabilities. The aim of this study was to create a prediction system for identifying the need for at-birth CPR in neonates based on Machine Learning (ML) algorithms. METHODS: In this study, 3,882 neonatal medical records were retrospectively reviewed. A total of 60 risk factors was extracted, and five ML algorithms of J48, Naïve Bayesian, multilayer perceptron, support vector machine (SVM), and random forest were compared to predict the need for at-birth CPR in neonates. Two types of resuscitation were considered: basic and advanced CPR. Using five feature selection algorithms, features were ranked based on importance, and important risk factors were identified using the ML algorithms. RESULTS: To predict the need for at-birth CPR in neonates, SVM using all risk factors reached 88.43% accuracy and F-measure of 88.4%, while J48 using only the four first important features reached 90.89% accuracy and F-measure of 90.9%. The most important risk factors were gestational age, delivery type, presentation, and mother's addiction. CONCLUSIONS: The proposed system can be useful in predicting the need for CPR in neonates in the delivery room.

A new COVID-19 intubation prediction strategy using an intelligent feature selection and K-NN method.

BACKGROUND: Predicting severe respiratory failure due to COVID-19 can help triage patients to higher levels of care, resource allocation and decrease morbidity and mortality. The need for this research derives from the increasing demand for innovative technologies to overcome complex data analysis and decision-making tasks in critical care units. Hence the aim of our paper is to present a new algorithm for selecting the best features from the dataset and developing Machine Learning(ML) based models to predict the intubation risk of hospitalized COVID-19 patients. METHODS: In this retrospective single-center study, the data of 1225 COVID-19 patients from February 9, 2020, to July 20, 2021, were analyzed by several ML algorithms which included, Decision Tree(DT), Support Vector Machine (SVM), Multilayer perceptron (MLP), and K-Nearest Neighbors(K-NN). First, the most important predictors were identified using the Horse herd Optimization Algorithm (HOA). Then, by comparing the ML algorithms' performance using some evaluation criteria, the best performing one was identified. RESULTS: Predictive models were trained using 12 validated features. Also, it found that proposed DT-based predictive model enables a reasonable level of accuracy (=93%) in predicting the risk of intubation among hospitalized COVID-19 patients. CONCLUSIONS: The experimental results demonstrate the effectiveness of the proposed meta-heuristic feature selection technique in combining with DT model in predicting intubation risk for hospitalized patients with COVID-19. The proposed model have the potential to inform frontline clinicians with quantitative and non-invasive tool to assess illness severity and to identify high risk patients.

Performance evaluation of selected machine learning algorithms for COVID-19 prediction using routine clinical data: With versus Without CT scan features.

BACKGROUND: Given coronavirus disease (COVID-19's) unknown nature, diagnosis, and treatment is very complex up to the present time. Thus, it is essential to have a framework for an early prediction of the disease. In this regard, machines learning (ML) could be crucial to extract concealed patterns from mining of huge raw datasets then it establishes high-quality predictive models. At this juncture, we aimed to apply different ML techniques to develop clinical predictive models and select the best performance of them. MATERIALS AND METHODS: The dataset of Ayatollah Talleghani hospital, COVID-19 focal center affiliated to Abadan University of Medical Sciences have been taken into consideration. The dataset used in this study consists of 501 case records with two classes (COVID-19 and non COVID-19) and 32 columns for the diagnostic features. ML algorithms such as Naïve Bayesian, Bayesian Net, random forest (RF), multilayer perceptron, K-star, C4.5, and support vector machine were developed. Then, the recital of selected ML models was assessed by the comparison of some performance indices such as accuracy, sensitivity, specificity, precision, F-score, and receiver operating characteristic (ROC). RESULTS: The experimental results indicate that RF algorithm with the accuracy of 92.42%, specificity of 75.70%, precision of 92.30%, sensitivity of 92.40%, F-measure of 92.00%, and ROC of 97.15% has the best capability for COVID-19 diagnosis and screening. CONCLUSION: The empirical results reveal that RF model yielded higher performance as compared to other six classification models. It is promising to the implementation of RF model in the health-care settings to increase the accuracy and speed of disease diagnosis for primary prevention, screening, surveillance, and early treatment.

Teleoncology for children with cancer: A scoping review on applications and outcomes.

BACKGROUND AND OBJECTIVES: Teleoncology can be used to reduce the limitations due to the lack of access to specialists, inadequate resources and training, and reducing unnecessary travels and arising of the costs. The purpose of this study was to review the literatures to identify and classify the areas of application and outcomes of using teleoncology in diagnosis, management, and treatment of children with cancer. METHODS: This scoping review of the published literatures was conducted by searching the Web of Science, PubMed/Medline, Scopus, and Cochrane Library databases in October 2019. Studies investigated telemedicine in diagnosis, management, and treatment of cancer in children were also included. We identified and classified different applications and the reported outcomes of this technology. RESULTS: In this study, 1834 articles were retrieved, and after removing the unrelated and duplicated articles, 20 articles were reviewed ultimately. We found that, teleoncology services were provided to the patients with cancer, their parents, and nurses in various clinical fields such as telepathology, telemental care (telepsychology), teleneurology, teledermatology, telehematology, and teleophthalmology. The findings also showed that, the outcomes of using telemedicine in children with cancer can be classified into six general categories (five primary and 14 secondary outcomes). Primary outcomes including diagnosis accuracy, reduced costs as well as mortality and secondary outcomes consist of improved relationship and training, better care management, satisfaction, and workload. CONCLUSION: The use of telemedicine for children with cancer is growing, and there is a tendency for using this technology for families and clinical staff. Providing teleoncology services to children with cancer may improve diagnosis accuracy and reduce the cost and mortality rate. Also, better care management, appropriate relationships and training, increased satisfaction, and decreased workload may be achieved.

Clinical decision support system to predict chronic kidney disease: A fuzzy expert system approach.

BACKGROUND AND OBJECTIVES: Diagnosis and early intervention of chronic kidney disease are essential to prevent loss of kidney function and a large amount of financial resources. To this end, we developed a fuzzy logic-based expert system for diagnosis and prediction of chronic kidney disease and evaluate its robustness against noisy data. METHODS: At first, we identified the diagnostic parameters and risk factors through a literature review and a survey of 18 nephrologists. Depending on the features selected, a set of fuzzy rules for the prediction of chronic kidney disease was determined by reviewing the literature, guidelines and consulting with nephrologists. Fuzzy expert system was developed using MATLAB software and Mamdani Inference System. Finally, the fuzzy expert system was evaluated using data extracted from 216 randomly selected medical records of patients with and without chronic kidney disease. We added noisy data to our dataset and compare the performance of the system on original and noisy datasets. RESULTS: We selected 16 parameters for the prediction of chronic kidney disease. The accuracy, sensitivity, and specificity of the final system were 92.13 %, 95.37 %, and 88.88 %, respectively. The area under the curve was 0.92 and the Kappa coefficient was 0.84, indicating a very high correlation between the system diagnosis and the final diagnosis recorded in the medical records. The performance of the system on noisy input variables indicated that in the worse scenario, the accuracy, sensitivity, and specificity of the system decreased only by 4.43 %, 7.48 %, and 5.41 %, respectively. CONCLUSION: Considering the desirable performance of the proposed expert system, the system can be useful in the prediction of chronic kidney disease.

A Clinical Decision Support System for Predicting the Early Complications of One-Anastomosis Gastric Bypass Surgery.

BACKGROUND/OBJECTIVE: One of the most effective treatments for patients with obesity, albeit with some complications, is obesity surgery. The aim of this study was to develop a clinical decision support system (CDSS) to predict the early complications of one-anastomosis gastric bypass (OAGB) surgery. SUBJECTS/METHODS: This study was conducted in Tehran, Iran on patients who underwent OAGB surgery in 2011-2014 in five hospitals. Initially, variables affecting the OAGB early complications were identified using the literature review. Patients' data were extracted from an existing database of obesity surgery. Then, different artificial neural networks (ANNs) (multilayer perceptron (MLP) network) were developed and evaluated for prediction of 10-day, 1-month, and 3-month complications. RESULTS: Factors including age, BMI, smoking status, intra-operative complications, comorbidities, laboratory tests, sonography results, and endoscopy results were considered important factors for predicting early complications of OAGB. A CDSS was developed with these variables. The accuracy, specificity, and sensitivity of the 10-day prediction system in the test data were 98.4%, 98.6%, and 98.3%, respectively. These figures for 1-month system were 96%, 93%, and 98.4% and for the 3-month system were 89.3%, 86.6%, and 91.5%, respectively. CONCLUSIONS: Using the CDSS designed, we could accurately predict the early complications of OAGB surgery.

Effectiveness of Anonymization Methods in Preserving Patients' Privacy: A Systematic Literature Review.

BACKGROUND: An ever growing for application of electronic health records (EHRs) has improved healthcare providers' communications, access to data for secondary use and promoted the quality of services. Patient's privacy has been changed to a great issue today since there are large loads of critical information in EHRs. Therefore, many privacy preservation techniques have been proposed and anonymization is a common one. OBJECTIVES: This study aimed to investigate the effectiveness of anonymization in preserving patients' privacy. METHODS: The articles published in the 2005-2016 were included. Pubmed, Cochrane, IEEE and ScienceDirect were searched with a variety of related keywords. Finally, 18 articles were included. RESULTS: In the present study, the relevant anonymization issues were investigated in four categories: secondary use of anonymized data, re-identification risk, anonymization effect on information extraction and inadequacy of current methods for different document types. CONCLUSION: The results revealed that though anonymization cannot reduce the risk of re-identification to zero, if implemented correctly, can manage to help preserve patient's privacy.

A novel method for fuzzy diagnostic system design.

Background: In recent years, liver disorders have been continuously increased. Proper performance of data mining techniques in decision-making and forecasting caused to use them commonly in designing of automatic medical diagnostic systems. The main aim of this paper is to introduce a classifier for diagnosis of liver disease that not only has high precision but also is understandable and has been created without expert knowledge. Methods: In regards to this purpose, fuzzy association rules have been extracted from dataset according to fuzzy membership functions which determined by fuzzy C-means clustering method; while each time, extracting fuzzy association rules, one of the five quality measures including confidence, coverage, reliability, comprehensibility and interestingness is used and five fuzzy rule-bases extracted based on them. Then, five fuzzy inference systems are designed on the basis of obtained rule-bases and evaluated in order to choose the best model in terms of diagnostic accuracy. Results: The proposed diagnostic method was examined using data set of Indian liver patients available at UCI repository. Results showed that among considered quality measures, interestingness, reliability and truth outperformed respectively, and yielded precision, sensitivity, specificity and accuracy of more than 90%. Conclusion: In this paper, a classification method was developed to predict liver disease which in addition to high classification accuracy, it has been created without expert knowledge and provided an understandable explanation of data. This method is convenient, user friendly, efficient and requires no expertise.

A fuzzy rule-based expert system for diagnosing cystic fibrosis.

BACKGROUND: Finding a valid diagnosis is mostly a prolonged process. Current advances in the sector of artificial intelligence have led to the appearance of expert systems that enrich the experiences and capabilities of doctors for making decisions for their patients. OBJECTIVE: The objective of this research was developing a fuzzy expert system for diagnosing Cystic Fibrosis (CF). METHODS: Defining the risk factors and then, designing the fuzzy expert system for diagnosis of CF were carried out in this cross-sectional study. To evaluate the performance of the proposed system, a dataset that corresponded to 70 patients with respiratory disease who were serially admitted to the CF Clinic in the Pediatric Respiratory Diseases Center, Masih Daneshvari Hospital in Tehran, Iran during August 2016 to January 2017 was considered. Whole procedures of system construction were implemented in a MATLAB environment. RESULTS: Results showed that the suggested system can be used as a strong diagnostic tool with 93.02% precision, 89.29% specificity, 95.24% sensitivity and 92.86% accuracy for diagnosing CF. There was also a good relationship between the user and the system through the appealing user interface. CONCLUSION: The system is equipped with information, knowledge, and expertise from certified specialists; hence, as a training tool it can be useful for new physicians. It is worth mentioning that the accomplishment of this project depends on advocacy of decision making in CF diagnosis. Nevertheless, it is expected that the system will reduce the number of false positives and false negatives in unusual cases.