A New Index for Intelligent Classification of Early Syndromic of Cardiovascular (CVD) Diseases Based on Electrocardiogram (ECG) (Published)
Most disease that affects the heart or blood vessels is referred to as cardiovascular disease(CVD). The main aim of this work is to build a system capable of modeling and predicting early syndromic cardiovascular diseases (CVD) based on electrocardiogram (ECG). The study considers the implementation of computationally intelligent system for detecting and classifying early syndromic assessment of CVD. The clinical and ECG recordings of patients diagnosed with pulmonary hypertension at the University of Uyo Teaching Hospital (UUTH) were obtained. The datasets were segmented into Demographic and ECG datasets. A quantitative research approach was used for the study with examination of several segments based on recommended framework. Three (3) classifier models were adopted to detect cardiac related problems using specified datasets. The classifiers such as; Random Forest Ensemble (RFE), Support Vector (SVM) Classifier and Artificial Neural Network (ANN) was employed for Machine Learning process. The models were implemented using a robust programming languages (Python and Jupyter notebook). The datasets were further segmented into two categories: training sets and testing in the ratio of 80:20 respectively. The test data reflects; precision, recall and sensitivity: Results show Radom Forest Model: 0.50 (50%) accuracy, 0.48 (48%) precision score and 0.65 (65%) recall sensitivity score (RSS), SVM classifier indicated 0.70 (70%) accuracy score, 0.47 (47) % precision score as well as 0.52 (52 %) sensitivity score. The ANN model illustrates 0.50 (50%) score for accuracy, precision and recall. Research Findings demonstrated that, RFE, SVM, ANN illustrate 100% accuracy in precision and recall sensitivity. The interaction effects of the various clinical factors influencing the CVD of patient was appraised and performance evaluation were further done using standard data science measures; Confusion Matrix (CM), MAP, MAPE, RMSE was deployed. The final results obtained shows that RFE, SVM, ANN models support satisfactorily the assessment and classification of early syndromic conditions of CVD.
Keywords: ANN, Accuracy, ECG, RFE, SVM Classifier, UUTH, precision score and recall sensitivity score (RSS).
Analysis on Deep Learning Performance with Low Complexity (Published)
In this article, presenting deep learning to the LTE-A uplink channel estimation system. This work involved creating two SC-FDMA databases, one for training and one for testing, based on three different channel propagation models. The first part of this work consists in applying artificial neural networks to estimate the channel of the SC-FDMA link. Neural network training is an iterative process consisting of adapting the values (weights and biases) of its parameters. After training, the neural network is tested and implemented in the recipient. The second part of this work addresses the same experiment, but uses deep learning, not traditional neural networks. The simulation results show that deep learning has improved significantly compared to traditional methods for bit error rate and processing speed. The third part of this work is devoted to complexity research. Deep learning has been shown to provide better performance than less complex MMSE estimators.
Keywords: ANN, Artificial Intelligence, LTE-A, SC-FDMA, and portable, channel estimation, communications systems, deep learning, mobile
Application of Expert System for Diagnosing Medical Conditions: A Methodological Review (Published)
Naturally, human diseases should be treated on time; otherwise the patients might die if there is delay in attending to such patient or scarcity of medical practitioners’ or experts. Several attempts have been made through studies to design and built software based medical expert systems for probing and prognosis of several medical conditions using artificial and non-artificial based approaches for patients and medical facilities. This paper represents a comprehensive methodological review of existing medical expert systems used for diagnosis of various diseases based on the increasing demand of expert systems to support the human experts. The study provides a concise evaluation of the various techniques used such as rule-based, fuzzy, artificial neural networks and intelligent hybrid models. The rule-based techniques is not too efficient based on its inability to learn and require powerful search strategies for its knowledge-base; while the fuzzy or ANN models are less efficient when compared to the hybrid models that can give a more accurate results.
Keywords: AI, ANN, Expert System, Fuzzy Logic, Intelligent hybrid model, Rule-based