While advances in medical care have reduced mortality rates across the globe, perinatal mortality has decreased at a slower pace. This is particularly true of low- and middle-income countries, which contribute by far the largest proportion of perinatal deaths annually. Key reasons for this include lack of systematic screening, lack of early health advice-seeking and robust referral. To address these issues we are developing a scalable mHealth referral system embedded into an existing network of practising traditional birth attendants. Two off-the-shelf sensors are bundled into the intervention, a 1D Doppler foetal heart monitor (cost $35) and a pulse oximeter, to address key clinical perinatal complications: the detection of foetal distress and intrauterine growth restriction by automatic algorithmic analysis of the Doppler ultrasound signal. The devices directly interface with the smartphone for data transfer and to enable on-the-spot assessment. For the development of our system we partnered with a local healthcare NGO working in rural Guatemala, one of the poorest regions in Latin America. For the first time, the systemic barriers to maternal-child care posed by the acute lack of access to basic diagnostic technology, decision support, and reliable linkages between lay birth attendants and higher levels of care will be addressed. An initial usability study has been conducted to assess the feasibility of introducing the proposed system for use by frontline health care workers. The high rate of successful signal recordings and the positive feedback on system use and utility are encouraging indicators that, by bringing together engineers, physicians, public health workers, anthropologists, and local NGOs, we are developing an intuitive system capable in assisting practicing birth attendants to elevate the level of services they provide.