Cardiovascular disease detection from high utility rare rule mining.

We propose a method to search rare cardiovascular disease symptom rules from historical health examination records according to its hazard ratio utility and further detect the disease given new medical record data. Further, we aim to assist both medical experts and patients by alerting the current symptoms and preparing the early treatments. In general, the proposed method first deals with the uncertainty of age and other continuous features using a fuzzy set. Next, we define the hazard ratio utility of each item set to assist the mining process. Based on the utility, we discover the rare cardiovascular disease patterns employing High Utility Rare Itemset Mining. At last, we add a prediction step to check the given health record data whether diagnosed cardiovascular. Subsequently, we can obtain rare symptoms of cardiovascular disease, which are later applied to detect the new related record data. The rare symptoms that are confirmed by their utility risk for cardiovascular disease can assist the medical experts' decision better than the common symptoms as it is often hard to be recognized at a glance. The proposed method evaluated on a public cardiovascular dataset. The experimental results showed that the generated rare cardiovascular disease patterns successfully applied to detect the cardiovascular given the symptoms data.

Exploration of Black Boxes of Supervised Machine Learning Models: A Demonstration on Development of Predictive Heart Risk Score.

Machine learning (ML) often provides applicable high-performance models to facilitate decision-makers in various fields. However, this high performance is achieved at the expense of the interpretability of these models, which has been criticized by practitioners and has become a significant hindrance in their application. Therefore, in highly sensitive decisions, black boxes of ML models are not recommended. We proposed a novel methodology that uses complex supervised ML models and transforms them into simple, interpretable, transparent statistical models. This methodology is like stacking ensemble ML in which the best ML models are used as a base learner to compute relative feature weights. The index of these weights is further used as a single covariate in the simple logistic regression model to estimate the likelihood of an event. We tested this methodology on the primary dataset related to cardiovascular diseases (CVDs), the leading cause of mortalities in recent times. Therefore, early risk assessment is an important dimension that can potentially reduce the burden of CVDs and their related mortality through accurate but interpretable risk prediction models. We developed an artificial neural network and support vector machines based on ML models and transformed them into a simple statistical model and heart risk scores. These simplified models were found transparent, reliable, valid, interpretable, and approximate in predictions. The findings of this study suggest that complex supervised ML models can be efficiently transformed into simple statistical models that can also be validated.

Development of Nonlaboratory-Based Risk Prediction Models for Cardiovascular Diseases Using Conventional and Machine Learning Approaches.

Criticism of the implementation of existing risk prediction models (RPMs) for cardiovascular diseases (CVDs) in new populations motivates researchers to develop regional models. The predominant usage of laboratory features in these RPMs is also causing reproducibility issues in low-middle-income countries (LMICs). Further, conventional logistic regression analysis (LRA) does not consider non-linear associations and interaction terms in developing these RPMs, which might oversimplify the phenomenon. This study aims to develop alternative machine learning (ML)-based RPMs that may perform better at predicting CVD status using nonlaboratory features in comparison to conventional RPMs. The data was based on a case-control study conducted at the Punjab Institute of Cardiology, Pakistan. Data from 460 subjects, aged between 30 and 76 years, with (1:1) gender-based matching, was collected. We tested various ML models to identify the best model/models considering LRA as a baseline RPM. An artificial neural network and a linear support vector machine outperformed the conventional RPM in the majority of performance matrices. The predictive accuracies of the best performed ML-based RPMs were between 80.86 and 81.09% and were found to be higher than 79.56% for the baseline RPM. The discriminating capabilities of the ML-based RPMs were also comparable to baseline RPMs. Further, ML-based RPMs identified substantially different orders of features as compared to baseline RPM. This study concludes that nonlaboratory feature-based RPMs can be a good choice for early risk assessment of CVDs in LMICs. ML-based RPMs can identify better order of features as compared to the conventional approach, which subsequently provided models with improved prognostic capabilities.

Simulation of COVID-19 outbreaks via Graphical User Interface (GUI).

BACKGROUND: This research aimed to model the outbreak of COVID-19 in Malaysia and develop a GUI-based model. DESIGN AND METHODS: The model is an improvement of the susceptible, infected, recovery, and death (SIRD) compartmental model.  The epidemiological parameters of the infection, recovery, and death rates were formulated as time dependent piecewise functions by incorporating the control measures of lockdown, social distancing, quarantine, lockdown lifting time and the percentage of people who abide by the rules. An improved SIRD model was solved via the 4th order Runge-Kutta (RK4) method and 14 unknown parameters were estimated by using Nelder-Mead algorithm and pattern-search technique. The publicly available data for COVID-19 outbreak in Malaysia was used to validate the performance of the model. The GUI-based SIRD model was developed to simulate the number of active cases of COVID-19 over time by considering movement control order (MCO) lifted date and the percentage of people who abide the rules. RESULTS: The simulator showed that the improved SIRD model adequately fitted Malaysia COVID-19 data indicated by low values of root mean square error (RMSE) as compared to other existing models. The higher the percentage of people following the SOP, the lower the spread of disease. Another key point is that the later the lifting time after the lockdown, the lower the spread of disease. CONCLUSIONS: These findings highlight the importance of the society to obey the intervention measures in preventing the spread of the COVID-19 disease.

Nonclinical Features in Predictive Modeling of Cardiovascular Diseases: A Machine Learning Approach.

BACKGROUND: In the broader healthcare domain, the prediction bears more value than an explanation considering the cost of delays in its services. There are various risk prediction models for cardiovascular diseases (CVDs) in the literature for early risk assessment. However, the substantial increase in CVDs-related mortality is challenging global health systems, especially in developing countries. This situation allows researchers to improve CVDs prediction models using new features and risk computing methods. This study aims to assess nonclinical features that can be easily available in any healthcare systems, in predicting CVDs using advanced and flexible machine learning (ML) algorithms. METHODS: A gender-matched case-control study was conducted in the largest public sector cardiac hospital of Pakistan, and the data of 460 subjects were collected. The dataset comprised of eight nonclinical features. Four supervised ML algorithms were used to train and test the models to predict the CVDs status by considering traditional logistic regression (LR) as the baseline model. The models were validated through the train-test split (70:30) and tenfold cross-validation approaches. RESULTS: Random forest (RF), a nonlinear ML algorithm, performed better than other ML algorithms and LR. The area under the curve (AUC) of RF was 0.851 and 0.853 in the train-test split and tenfold cross-validation approach, respectively. The nonclinical features yielded an admissible accuracy (minimum 71%) through the LR and ML models, exhibiting its predictive capability in risk estimation. CONCLUSION: The satisfactory performance of nonclinical features reveals that these features and flexible computational methodologies can reinforce the existing risk prediction models for better healthcare services.

Modifiable risk factors and overall cardiovascular mortality: Moderation of urbanization.

Background: Modifiable risk factors are associated with cardiovascular mortality (CVM) which is a leading form of global mortality. However, diverse nature of urbanization and its objective measurement can modify their relationship. This study aims to investigate the moderating role of urbanization in the relationship of combined exposure (CE) of modifiable risk factors and CVM. Design and Methods: This is the first comprehensive study which considers different forms of urbanization to gauge its manifold impact. Therefore, in addition to existing original quantitative form and traditional two categories of urbanization, a new form consisted of four levels of urbanization was duly introduced. This study used data of 129 countries mainly retrieved from a WHO report, Non-Communicable Diseases Country Profile 2014. Factor scores obtained through confirmatory factor analysis were used to compute the CE. Age-income adjusted regression model for CVM was tested as a baseline with three bootstrap regression models developed for the three forms of urbanization. Results: Results revealed that the CE and CVM baseline relationship was significantly moderated through the original quantitative form of urbanization. Contrarily, the two traditional categories of urbanization could not capture the moderating impact. However, the four levels of urbanization were objectively estimated the urbanization impact and subsequently indicated that the CE was more alarming in causing the CVM in levels 2 and 3 urbanized countries, mainly from low-middle-income countries. Conclusion: This study concluded that the urbanization is a strong moderator and it could be gauged effectively through four levels whereas sufficiency of two traditional categories of urbanization is questionable.