KARACHI: The Sindh Health Department, Lahore University of Management Sciences (LUMS) and Aga Khan University (AKU) have signed a three-year collaboration to integrate a national artificial intelligence hub for maternal, newborn and child health into the province’s public health system.
The initiative, supported by a multi-year grant from the Gates Foundation, will focus on deploying AI-enabled tools at primary health care facilities to support frontline workers and improve care for women and children.
The agreement was signed on Wednesday, Sept 2, at the office of Sindh Health and Population Welfare Minister Dr. Azra Fazal Pechuho. It was signed by Sindh Health Secretary Tahir Hussain, LUMS Associate Professor Dr. Maryam Mustafa, who is the hub’s director and principal investigator, and AKU Medical College Dean Dr. Karim Damji.
Dr. Anita Zaidi, president of Gender Equality at the Gates Foundation, attended the signing, along with AKU Associate Professor of Paediatrics and Child Health Dr. Zahra Hoodbhoy, who has worked on AI models for point-of-care ultrasound.
Under the agreement, the three institutions will jointly develop and deploy risk-stratification models, clinical decision-support systems and voice-assisted tools for frontline health care workers providing maternal and child health services.
Pechuho said the initiative would focus on primary health care facilities, with the aim of improving outcomes for mothers and newborns.
“I would like to see healthier babies, and prematurity rates go down,” she said, adding that the technology could also help identify congenital conditions and strengthen paramedical services.
The AI hub builds on a multi-year collaboration involving computer scientists, human-centered design specialists and clinical epidemiologists.
Its clinical research component is led by AKU’s Dr. Fyezah Jehan, Chair of the Department of Paediatrics and Child Health. Her clinical trial networks, maternal and newborn cohort datasets and field-based epidemiological research will provide clinical data for validating predictive models and assessing their performance in primary health care settings.
“We have brilliant faculty and that’s the magic,” Damji said after the signing. “I think it’s fantastic because we are serving people better, in this case maternal, newborn, child health, and looking at ways to understand what value we can offer for safer deliveries for the mothers, but also for the children.”
“We have to embrace [AI] and try to make it as cost effective and efficient as possible,” he added.
AI for frontline health workers
The hub’s technology will build on Awaaz-e-Sehat, a voice-enabled electronic record management platform developed by Mustafa’s team with initial support from a Grand Challenges grant.
The platform uses localized speech recognition and natural-language technologies designed for low-literacy environments, allowing frontline health workers to interact with digital health tools in local languages.
“One of the things that worries me the most as a computer scientist working in AI and health is that Pakistan is far, far behind other countries when it comes to adopting AI tools,” Mustafa said.
“Part of the reason is that we don’t have a formalized data infrastructure that can help scale AI tools. The tech part is actually easier to build. It is how do you embed it in existing infrastructures and get governments to adopt and scale,” she added.
The initiative aims to extend access to AI-enabled health tools beyond major urban centers, including remote and underserved areas.
Roles of the partners
The Sindh Health and Population Welfare Department will provide strategic oversight, facilitate integration into provincial health facilities and support public-sector deployment.
LUMS will serve as the technical partner, leading AI model development, algorithm validation, speech technologies and technical safeguards.
AKU will serve as the clinical, research and implementation partner, overseeing clinical safety, real-world evaluation and the integration of the tools into primary health care settings.
The partners said the collaboration would follow national regulatory requirements, data privacy standards and ethical safeguards. The AI systems are intended to support health care professionals rather than replace clinical judgment.