Computational Method of Predicting Down Syndrome on Foetus by Utilizing First Trimester Ultrasound Scan

Down syndrome(DS) is a common chromosomal abnormality that affects about 1 in 800 live births. It is a genetic condition that occurs due to the presence of an extra copy of chromosome 21. Early detection of Down syndrome during pregnancy is crucial for appropriate prenatal care and planning. Ultrasound imaging is a widely used method for detecting fetal abnormalities, including Down syndrome. In recent years, deep Convolutional Neural Networks (CNNs) have shown great promise in analyzing medical images for disease diagnosis. In this study, we propose a deep CNN-based approach for Down syndrome prediction from ultrasound images. Our approach consists of a pre-processing step for image normalization and a CNN-based model for feature extraction and classification. We trained and evaluated our model on a publicly available dataset of ultrasound images with both normal and Down syndrome cases. Our experimental results show that our proposed CNN-based approach achieved an accuracy of 98.7% for Down syndrome prediction, outperforming traditional machine learning algorithms and other deep learning models. Our proposed approach has the potential to assist clinicians in the early detection of Down syndrome and improving prenatal care for affected individuals.