When Ellen Kaphamtengo felt a sharp pain in her lower abdomen, she thought she might be in labour. It was the ninth month of her first pregnancy and she wasn’t taking any chances. With the help of her mother, the 18-year-old climbed on to a motorcycle taxi and rushed to a hospital in Malawi’s capital, Lilongwe, a 20-minute ride away.
At the Area 25 health centre, they told her it was a false alarm and took her to the maternity ward. But things escalated quickly when a routine ultrasound revealed that her baby was much smaller than expected for her pregnancy stage, which can cause asphyxia – a condition that limits blood flow and oxygen to the baby.
In Malawi, about 19 out of 1,000 babies die during delivery or in the first month of life. Birth asphyxia is a leading cause of neonatal mortality in the country, and can mean newborns suffering brain damage, with long-term effects including developmental delays and cerebral palsy.
Doctors reclassified Kaphamtengo, who had been anticipating a normal delivery, as a high-risk patient. Using AI-enabled foetal monitoring software, further testing found that the baby’s heart rate was dropping. A stress test showed that the baby would not survive labour.
The hospital’s head of maternal care, Chikondi Chiweza, knew she had less than 30 minutes to deliver Kaphamtengo’s baby by caesarean section. Having delivered thousands of babies at some of the busiest public hospitals in the city, she was familiar with how quickly a baby’s odds of survival can change during labour.
Chiweza, who delivered Kaphamtengo’s baby in good health, says the foetal monitoring programme has been a gamechanger for deliveries at the hospital.
“[In Kaphamtengo’s case], we would have only discovered what we did either later on, or with the baby as a stillbirth,” she says.
The software, donated by the childbirth safety technology company PeriGen through a partnership with Malawi’s health ministry and Texas children’s hospital, tracks the baby’s vital signs during labour, giving clinicians early warning of any abnormalities. Since they began using it three years ago, the number of stillbirths and neonatal deaths at the centre has fallen by 82%. It is the only hospital in the country using the technology.
“The time around delivery is the most dangerous for mother and baby,” says Jeffrey Wilkinson, an obstetrician with Texas children’s hospital, who is leading the programme. “You can prevent most deaths by making sure the baby is safe during the delivery process.”
The AI monitoring system needs less time, equipment and fewer skilled staff than traditional foetal monitoring methods, which is critical in hospitals in low-income countries such as Malawi, which face severe shortages of health workers. Regular foetal observation often relies on doctors performing periodic checks, meaning that critical information can be missed during intervals, while AI-supported programs do continuous, real-time monitoring. Traditional checks also require physicians to interpret raw data from various devices, which can be time consuming and subject to error.
Area 25’s maternity ward handles about 8,000 deliveries a year with a staff of 80 physicians. While only about 10% are trained to perform traditional electronic monitoring, most can use the AI software to detect anomalies, so doctors are aware of any riskier or more complex births. Hospital staff also say that using AI has standardised important aspects of maternity care at the clinic, such as interpretations on foetal wellbeing and decisions on when to intervene.
Kaphamtengo, who is excited to be a new mother, believes the doctor’s interventions may have saved her baby’s life. “They were able to discover that my baby was distressed early enough to act,” she says, holding her son, Justice.
Doctors at the hospital hope to see the technology introduced in other hospitals in Malawi, and across Africa.
“AI technology is being used in many fields, and saving babies’ lives should not be an exception,” says Chiweza. “It can really bridge the gap in the quality of care that underserved populations can access.”
