This paper proposes an Ensemble Convolutional Neural Network (ECNN) model for the classification of foetal planes in ultrasound images. The model combines multiple pre-trained convolutional neural networks (CNNs) to achieve higher classification accuracy. The dataset used contains 12,400 images of six foetal planes: brain, abdomen, femur, thorax, cervix, and others. The study evaluates the performance of various deep CNN models, with DenseNet-169 achieving an accuracy of 93.6%. The proposed ECNN model uses three pre-trained CNNs as base learners, and their features are used to train a Deep Neural Network (DNN) as a meta-learner. The stacked ensemble model achieved an accuracy of 96%, which is higher than the accuracy of individual models. The study also investigates the impact of hyperparameter tuning on the performance of the baseline models. The results show that the stacked ensemble approach improves classification accuracy and generalization ability, making it effective for unseen data. The study focuses on three main aspects: analyzing foetal planes and generating three pre-trained models, developing a method for classifying foetal images using an ensemble stacked model, and optimizing the best stacked ensemble method through hyperparameter tuning. The performance of the stacked ensemble model is compared with baseline models and similar research. The study highlights the potential of deep learning in improving the accuracy of foetal plane identification and abnormality detection in antenatal care. The results suggest that the proposed ECNN model can be a valuable tool for automating foetal plane identification and improving the outcomes of antenatal care.This paper proposes an Ensemble Convolutional Neural Network (ECNN) model for the classification of foetal planes in ultrasound images. The model combines multiple pre-trained convolutional neural networks (CNNs) to achieve higher classification accuracy. The dataset used contains 12,400 images of six foetal planes: brain, abdomen, femur, thorax, cervix, and others. The study evaluates the performance of various deep CNN models, with DenseNet-169 achieving an accuracy of 93.6%. The proposed ECNN model uses three pre-trained CNNs as base learners, and their features are used to train a Deep Neural Network (DNN) as a meta-learner. The stacked ensemble model achieved an accuracy of 96%, which is higher than the accuracy of individual models. The study also investigates the impact of hyperparameter tuning on the performance of the baseline models. The results show that the stacked ensemble approach improves classification accuracy and generalization ability, making it effective for unseen data. The study focuses on three main aspects: analyzing foetal planes and generating three pre-trained models, developing a method for classifying foetal images using an ensemble stacked model, and optimizing the best stacked ensemble method through hyperparameter tuning. The performance of the stacked ensemble model is compared with baseline models and similar research. The study highlights the potential of deep learning in improving the accuracy of foetal plane identification and abnormality detection in antenatal care. The results suggest that the proposed ECNN model can be a valuable tool for automating foetal plane identification and improving the outcomes of antenatal care.