Congenital Heart Disease (CHD) detection has continued to witness a consistent increase in research attention. CHD diseases are vast and also span across diverse demographics, without sparing pregnant women, unborn babies, or newly born babies. The aim of this study is to develop a detection model capable of detecting heart disease among infants with high accuracy and to also suggest solutions to manage it. To achieve this, the CHD types were identified to develop a data model that considered infants. The data model was processed through imputation, feature selection, and transformation using the imputation method and Principal Component Analysis (PCA). After the data processing stage, Artificial Neural Network (ANN) algorithm is adopted and trained with the data to generate the model for the detection of the CHD. Comparative analysis was used to evaluate the performance of the models in comparison with other models adopted in other works, considering metrics that defined the success of the detection models. The results showed that the ANN has the best detection outcome with an accuracy of 97.44%. Although the use of Logistic Regression algorithm attained a high level of performance with an accuracy of 95.00%, but it still falls below the proposed ANN algorithm. Another highly performing algorithm is the use of Extreme Gradient Boosting (XGB) which achieved an accuracy of 92.00%.