When Maryam Mustafa walked into a small, crowded clinic in Pakistan she carefully scanned the space—a shared lobby that doubled as an examination room. Dozens of women waited quietly as doctors moved from patient to patient. Each woman had only moments to explain why she was there before the physician moved on. Mustafa observed the interactions closely.
“[Doctors] have three minutes per patient,” she says. “They only have the bandwidth to ask, ‘Why are you here today?’”
For many Pakistani women who are pregnant, this kind of visit is their first and only prenatal medical appointment, which puts them at increased risk of life-threatening conditions like preeclampsia, anemia and gestational diabetes. Early detection can save lives, but the lack of affordable and accessible care, health clinic workers’ limited time with each patient, and lack of privacy in under resourced clinics can lead to a deadly gap in care for mothers and their unborn children.
“This happens so often, where women come in for a first visit and then you don’t see them again until they land in the emergency room, often when it’s too late,” Mustafa says.
She was in the clinic that day for an important reason—one that could help transform maternal health care in low-resource settings. As a computer scientist dedicated to accelerating progress in health and development, she was developing accessible, cheap technology to help health workers detect pregnancy risks early and save the lives of mothers and babies. In Pakistan, more than 150 women die for every 100,000 live births—each one a mother who might have been saved with timely care.
A path where human needs and technology meet
In Mustafa’s Pakistani family, education came first. “My parents were an anomaly in how they raised their daughters,” she says. They insisted that all six of their children—their daughters as well as their sons—have access to the very best educational opportunities. But for Mustafa, choosing a path wasn’t simple.
At Lahore University of Management Sciences (LUMS), one of Pakistan’s top universities, she majored in computer science, albeit reluctantly. She excelled as an undergraduate but felt that something was missing. “There’s always been this inherent friction in my life,” she says. “I was always interested in anthropology and sociology. I wanted to understand people’s lives.”
Though trained as a computer scientist at top universities in Pakistan, the U.S., and Germany, Mustafa was drawn to the human side of technology. Her passion for anthropology and design led her to specialize in human-computer interaction—a field that blends people, psychology, and innovation.
A maternal health solution powered by artificial intelligence
Mustafa’s work to improve maternal health care began with years of on-the-ground research. She visited clinics and hospitals and spoke to countless women, seeking to understand why so many were dying from preventable conditions like high blood pressure, which leads to eclampsia—one of the leading causes of maternal mortality in Pakistan.
The answer became clear: Critical symptoms were being missed due to rushed consultations, overworked doctors, and lack of access to trained health care workers which led to delayed or missed diagnoses.
To help solve this, Mustafa and her team developed Awaaz-e-Sehat—“Voice of Health” in Urdu—a voice-assisted app designed specifically for frontline health workers. Using simple smartphones, the app guides providers through structured questions in local languages, helping them identify warning signs of conditions like pre-eclampsia or gestational diabetes during brief consultations. By capturing patient responses in real time and flagging potential risks, the tool supports more attentive, informed care—even in the most resource-strapped settings.
As the project evolved, Mustafa and her team also developed a simpler voice-based tool for patients themselves—delivered entirely over WhatsApp. It allows women to answer key health questions in their own language before seeing a doctor, generating basic medical records and helping ensure critical symptoms aren’t missed.
Backed by a Grand Challenges grant from the Gates Foundation, her team partnered with one of Pakistan’s largest maternal health hospitals to scale the app and improve recordkeeping.
Promising results in early detection of dangerous conditions
Awaaz-e-Sehat is still in the pilot phase of development, but Mustafa and her team are already seeing success with the technology. The app is flagging symptoms that could lead to a variety of dangerous prenatal conditions far more accurately than doctors in traditional medical settings. For example, early data shows that while doctors in those settings are detecting symptoms of gestational diabetes and hypertension just 7% of the time, the app is flagging those same symptoms 40% of the time.
Mustafa’s goal in the next phase of the study is to scale up to multiple hospitals and clinics in Punjab and support a wider variety of health care workers including midwives, doctors, and nurses. Now a professor at LUMS and Director of the Interactive Media Lab and the Saida Waheed Gender and Technology cluster, Mustafa is expanding Awaaz-e-Sehat to clinics across Punjab. Early results are promising: the app flags dangerous conditions like gestational diabetes far more reliably than doctors working without it.
“Why do you have women dying from high blood pressure? That’s the third-leading cause of maternal deaths in this country,” she says. These kinds of alarming statistics keep her laser focused on expanding her sample size and gathering more data to help save more lives through early detection.
For Mustafa, this is about more than just building a tool—it’s about employing new technologies to help ensure that more mothers survive childbirth, more babies are born healthy, and fewer lives are lost to preventable causes.
Read next
