Pregnancy is an immensely difficult process and requires constant care. It is crucial that pregnant women have access to all the necessary help they deserve, not only for themselves but also for their babies.
Kenya sits high in the world’s top morality rates, and approximately one-third of Kenya’s maternal and neonatal deaths occur from mothers not having sufficient and prompt supervision. To help with this matter, Pennsylvania State University and Jacaranda Health developed a new tool that could potentially upgrade maternal and neonatal outcomes for those in need.
The tool, called TRIage for Mothers using AI (TRIM-AI) utilizes natural language processing, an AI method that grants machines to comprehend words and text to analyze and recognize any danger. From there, marked messages can be delivered to a small team that could provide help in emergencies. TRIM-AI, when compared to Jacaranda Health’s baseline model, is around 17 percent more precise when transcribing potential emergencies, allowing help-desk agents to have their work reduced by 12 percent.
TRIM-AI is also more structured and precise, providing additional help for Kenyan mothers. “Being able to quickly and accurately automate the classification of messages is a critical step in improving maternal health care,” said PNC Career Development Assistant Professor at Penn State’s College of Information Sciences and Technology Amulya Yadav.
“Our tool improves the system’s ability to identify at-risk cases in real-time, helping mothers connect more promptly with local hospitals and clinics for care.” Jacaranda Health’s baseline model called PROMPTS, a two-way text messaging system that TRIM-AI uses, allows pregnant and new mothers to freely use the platform. It provides crucial information on pregnancy, birth, and postpartum care that mothers need.
“At our current scale on PROMPTS, we increasingly depend on rapid, reliable systems to fast-track mothers in need of lifesaving care. Our partnership with Penn State has helped us improve both the accuracy and speed by which we interact with the 2 million mothers on PROMPTS, ensuring that they get the right care at the right time,” shared co-executive director of Jacaranda Health, Sathyanath Rajasekharan.
PROMPTS has more than 350,000 users across Kenya and sends 1.1 million messages per month, which is far more than what help-desk agents can possibly handle. This is where the new TRIMP-AI jumps in, allowing a much faster and more efficient operation. Jacaranda Health said mothers who called for help using the AI system were contacted by a help desk agent within just one hour and taken into professional care, and approximately 85 percent of cases ended up in the hospital.
Yadav concluded: “With the success of our pilot deployment, we are incentivized to explore the possibility of scaling up our risk-prediction system for improving maternal care in other developing countries around the world.”
The research team will share their research and innovative AI tool at the 37th AAAI Conference on Artificial Intelligence located in Washington, D.C. starting February 7 to 14.