This paper introduces an Ensemble Convolution Neural Network (ECNN) model designed to classify foetal planes from ultrasound images, which is crucial for antenatal care. The model combines three pre-trained CNN models (base learners) using transfer learning to classify foetal planes into six categories: brain, abdomen, femur, thorax, cervix, and others. The dataset used contains 12,400 images from 1792 patients. Prior studies have shown that the Densenet-169 model achieved an accuracy of 93.6% on this dataset. The ECNN model, which includes a Deep Neural Network (DNN) meta-learner, achieved an accuracy of 96%, outperforming the individual base models. The study also explores hyperparameter tuning and evaluates the model's performance on unseen data, demonstrating strong generalization capabilities. The research aims to enhance the accuracy of foetal plane classification, aiding in early detection of abnormalities and improving pregnancy outcomes.This paper introduces an Ensemble Convolution Neural Network (ECNN) model designed to classify foetal planes from ultrasound images, which is crucial for antenatal care. The model combines three pre-trained CNN models (base learners) using transfer learning to classify foetal planes into six categories: brain, abdomen, femur, thorax, cervix, and others. The dataset used contains 12,400 images from 1792 patients. Prior studies have shown that the Densenet-169 model achieved an accuracy of 93.6% on this dataset. The ECNN model, which includes a Deep Neural Network (DNN) meta-learner, achieved an accuracy of 96%, outperforming the individual base models. The study also explores hyperparameter tuning and evaluates the model's performance on unseen data, demonstrating strong generalization capabilities. The research aims to enhance the accuracy of foetal plane classification, aiding in early detection of abnormalities and improving pregnancy outcomes.