The first trimester of pregnancy is critical for fetal development with the mid sagittal plane (MSP) serving as a standard for prenatal screening to examine fetal growth and anomalies. Accurate identification of the MSP is crucial for effective diagnosis and monitoring but is often hampered by challenges like speckle noise, weak edges, and fuzzy borders in ultrasound images. These limitations can lead to errors, unnecessary medical expenses and parental anxiety. To address this, developed an automated system leveraging the Faster Region Convolutional Neural Network (Faster R-CNN) for accurate MSP detection in fetal ultrasound images. The system was trained on 1500 annotated ultrasound images. The anatomical markers considered in this study are nasal tip, nasal bone, palate, diencephalon, and nuchal translucency. The model achieved a mean average precision (mAP) of 0.9151 with class wise average precision values of 0.9393 for the diencephalon, 0.97 for nuchal translucency, 0.936 for the palate, 0.753 for the nasal tip, and 0.8082 for the nasal bone. The training process demonstrated consistent reduction in RPN and Faster R-CNN losses over 50 epochs affirming model efficiency. By reducing the reliance on manual assessments, it holds significant potential for enhancing clinical applications and mitigating diagnostic variability in early pregnancy.