On May 16th, the NJ Big Data Alliance hosted its 12th Annual Symposium titled "Empowered by AI: Innovation and the Rise of the Augmented Workforce." During this event, a group of graduate students from Rutgers University presented their paper, "MOMCare: An AI Chatbot for Postpartum Depression Counseling." The accompanying poster received an honorable mention in the graduate student division.
The paper was authored by Rimshah Jawad, Jiatong Yang, Ivania Martinez Zamora, Md Mozammel Hoque, and Xinyi Zhao under the mentorship of Jim Samuel. It addresses postpartum depression (PPD), a condition affecting many women worldwide. According to the World Health Organization (WHO), around 13% of women experience postpartum mental health disorders, with rates rising to nearly 20% in developing countries.
"MOMCare is designed to support mothers dealing with PPD," said Jim Samuel. The chatbot employs a retrieval-augmented architecture that includes data preprocessing and response generation. It uses hybrid classification and a dual embedding system to generate empathetic responses relevant to PPD. The design ensures fast and factual responses while providing context through gpt-3.5-turbo.
The chatbot's performance was evaluated using both automated and human metrics, showing strong results in both areas. This highlights the potential for chatbot interventions in postpartum mental health care. Additionally, MOMCare is built on ethical principles emphasizing safety and fairness.
The manuscript has been accepted for publication in Springer's proceedings as part of the Smart Innovation, Systems and Technologies series at the 10th International Conference on Information and Communication Technology for Intelligent Systems (ICTIS 2025) held on May 23rd in New York.
Jim Samuel added that expanding MOMCare's knowledge base using conversation history with users is also being developed. The final version will be published on SpringerLink.
