Mental health during pregnancy plays a critical role in fetal development and maternal well-being. However, psychological conditions such as depression, stress, and anxiety in pregnant women often go undetected, especially in primary healthcare settings. This research aims to design and develop a web-based expert system capable of diagnosing the mental health conditions of pregnant women using Forward Chaining and Backward Chaining inference techniques. Forward Chaining is applied to infer possible conditions based on reported symptoms, while Backward Chaining is used to validate hypotheses by tracing required supporting symptoms. The system was developed using patient data collected from three health centers in Lhokseumawe City, totaling 500 records with parameters including name, age, gestational age, number of children, and reported complaints. It incorporates 30 symptoms and 9 diagnostic rules to classify the mental condition and its severity.The results indicate that 179 women were diagnosed with depression (mild 107, moderate 33, severe 39), 150 with anxiety (mild 24, moderate 91, severe 35), and 171 with stress (mild 82, moderate 50, severe 39). The system also demonstrates diagnostic probability (e.g., 66.67% in a specific case). Validation using 20 test cases yielded an accuracy of 85%, showing the system performs reliably in aligning symptoms with diagnostic outcomes. This study makes two significant contributions. Practically, it offers a decision-support tool for midwives and general practitioners to perform early mental health screening of pregnant women, especially in regions lacking access to psychiatric specialists. Scientifically, it demonstrates the effectiveness of a hybrid reasoning approach in handling overlapping psychological symptoms and in assessing severity levels, thereby enriching the development of domain-specific expert systems in maternal mental health. In conclusion, this system provides a practical and accessible solution to support early detection and intervention in maternal mental health, ultimately contributing to improved health outcomes for both mothers and their babies.