Health issues during pregnancy have become a serious predicament. The mortality of the fetus can infrequently arise from these issues, which are more prevalent in underdeveloped and emerging nations. Cardiotocography, also known as CTG, is a non-invasive technique frequently utilized by obstetricians to evaluate the physical health of a fetus at any point during pregnancy. CTG provides a visual graphical representation of normal or pathological uterine contractions and fetal heart rate, which avails in identifying newborn’s overall health conditions including any birth defects or abnormalities. To proceed with treatment, a precise examination of the cardiotocograms is required. The evaluation of the fetal state that utilizes the machine learning and deep learning method and incorporates the cardiotocogram data has consequently received substantial attention. Due to its expeditious development, deep learning (DL) is widely utilized in disciplines such as medicine and healthcare to address several ailments. This study examines the results and analysis of a deep-learning classification model for fetal health. Utilizing an open access cardiotocography dataset, the approach was built. Although the dataset is relatively small, it has several eminent values. This data was analyzed and incorporated into numerous Machine Learning models. It is discovered that the proposed dense model employed in this study yields an accuracy of 89% which outperforms previously state-of-the-art techniques in classifying CTG and computational momentum for informed decision-making and quality care.