Cardiovascular diseases remain top in the world mortality statistics even after significant progress in diagnosing and treating the disease. The recognition of the people at risk prior to the emergence of the critical symptoms is thus an important measure to the minimization of the mortality levels and prevention of the secondary outcomes. Clinical data gathered through the electronic health systems have been applied in a wide range of machine learning techniques to this problem. Classical algorithms such as those based on logistic regression, decision trees, K-nearest neighbours, Naive Bayes, and support vector machines have reported encouraging yet limited performance largely because they are shallow learners and because they rely on engineered features. Advanced algorithms like the Random Forests, XGBoost, and other ensemble learners have been making significant strides but even at the cost of working on independent extracted attributes, with deep learning systems slowly replacing these older model systems. Nonetheless, the widely used architectures, which include multilayer perceptrons, CNNs and hybrid CNN-LSTM based models, are based on spatial or temporal structure built in the data, which is absent in the tabular clinical data. To manage this mismatch, this study presents an attention-based TabTransformer framework, which considers categorical attributes as learnable tokens and captures the dynamics of attributes among them with the multi-head self-attention. Parallel dense projections are used to include numerical attributes and the result is merged with the contextualized token representations to do the final prediction.
P.Ramprakash, Rajeswari Manickam, R. Karthik et al.· Journal of Intelligent Decis...· 0 citations
Nowadays, heart diseases are gaining international attention because they are regarded as a potentially fatal epidemic disease with difficult infection control around the world. Large number of models are proposed in the past. Among them machine learning gained more attention in heart disease prediction. Machine learning (ML) is an intelligent technique that can predict events with reasonable accuracy based on prior experience and learning. Meanwhile, a large number of ML models have been proposed to predict disease cases. As a result, the main challenge of this study requires an evaluation and benchmarking of ML models. Furthermore, no single study has addressed the issue of diagnosis model evaluation and benchmarking. This study, on the other hand, proposed an intelligent methodology to assist health organizations in the selection of heart disease diagnosis system. Benchmarking and evaluating diagnostic models for disease is a time-consuming process. There are numerous criteria to evaluate, some of which are in conflict with one another. Our research is organised as a decision matrix (DM) with ten evaluation criteria and twelve diagnostic models. The multi-criteria decision-making (MCDM) method is used to evaluate and benchmark the various chronic disease diagnostic models in terms of the evaluation criteria. TOPSIS-VIKOR is used for benchmarking and ranking, while Entropy is used to calculate the weights of criteria in an integrated MCDM method. The study's findings revealed that the benchmarking and selection issues associated with heart disease diagnosis models can be effectively addressed by combining Entropy and TOPSIS-VIKOR. Six Machine Learning Algorithms such as Naive Bayes, Artificial Neural Network, SVM, kNN, Logistic Regression, and Decision Tree are discussed for comparative analysis. To evaluate each of the methods used in this study, parameters such as accuracy, F1 score, Recall, Precision are used.The ML algorithms achieve accuracy of 0.862 in Naïve Bayes0.83 in K-Nearest Neighbor, 0.872 in Artificial Neural Network, 0.856 in Support Vector Machine, 0.756 in Decision Tree, 0.79 in Logistic Regression. From the results it is shown that the Artificial Neural Network classifier was chosen as the best diagnosis model for heart disease
P.Ramprakash, Rajeswari Manickam, R. Karthik et al.· Journal of Intelligent Decis...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
