Reducing Maternal Mortality: An Ultrasound System for Village Midwives Waylon Brunette 1 , Matthew Hicks 2 , Alexis Hope 3 , Ginger Ruddy 4 , Ruth E. Anderson 1 , Beth Kolko 3 1 Dept. of Computer Science & Engineering University of Washington Seattle, WA USA {wrb, rea}@cse.uw.edu 2 The Information School University of Washington Seattle, WA USA mattheau@uw.edu 3 Dept. of Human Centered Design & Engineering University of Washington Seattle, WA USA {alexishg, bkolko}@uw.edu 4 School of Public Health University of Washington Seattle, WA USA ginger.ruddy@gmail.com Abstract—On average a woman dies in childbirth approximately every 90 seconds, and the majority of these deaths take place in Africa and Asia [1]. While ultrasound imaging is an effective tool for identifying maternal mortality risk factors, it is nearly absent in many rural healthcare facilities in developing regions due to the high costs of both equipment and required training. To leverage existing healthcare systems commonly found in these contexts, we have focused on increasing the diagnostic capabilities of village midwives—often central medical figures in rural and low- income communities. We have developed a low-cost, portable, easy-to-use ultrasound system designed specifically to enable local midwives to identify high-risk conditions for referral to a well-equipped health care facility. Our focus has been on designing an appropriate system for our context. Specifically, we had to: simplify our user interface, support a solitary work environment, balance cost and features, and create an integrated teaching help system. This paper describes technical, socio-technical, and socio-cultural factors, drawn from our collaboration with the University of Washington Department of Radiology and our field experiences with midwives in Uganda, that have influenced our design. Keywords-ultrasound, Uganda, ICTD, healthcare technologies, maternal health, design, appropriate technology I. INTRODUCTION Globally, a woman dies in childbirth on average every ~90 seconds; over ninety percent of women who die are in Africa and Asia [1]. In Uganda, for example, the maternal mortality rate is 430 per 100,000 live births. Countries that are more politically unstable, such as Chad and Somalia, have rates of 1,200 maternal deaths per 100,000 live births. On the other hand, rates of maternal mortality in the United States are 24/100,000 [1]. However, small changes, such as identifying high risk patients, can potentially reduce preventable deaths in developing countries. Most of these preventable deaths occur because of severe bleeding, infections, ecclampsia, obstructed labor and the consequences of unsafe abortions [2]. These conditions can often be prevented if a woman is diagnosed early enough to be treated in a well-equipped healthcare facility by trained medical personnel. Three of these five conditions, specifically causes of hemorrhage, risk of obstructed labor, and dangerous sequelae of abortion, can be predicted or diagnosed using ultrasound technology, as can three of the top five causes of neonatal mortality [3]: birth asphyxia, injuries, and defects. However, the effect of ultrasound on maternal health outcomes is still inconclusive as researchers have argued that longitudinal studies must be performed to assess the ability of ultrasound to lessen maternal deaths [4]. To create a viable longitudinal study to answer this question, we need to have a low-cost, portable ultrasound system that can be easily deployed in rural developing regions. Cost is traditionally seen as the major barrier for widespread deployment of medical technology, but the situation is more nuanced. Longitudinal studies must integrate an ultrasound system that is appropriate to the populations in which they are introduced. The problem space of maternal ultrasound and health outcomes, then, is twofold. While lower cost commercial systems are increasingly available, those systems lack some characteristics that would allow them to be fully integrated into maternal health practice outside of better-equipped hospitals as appropriate technology. The suitability of ultrasound for low-resource settings has been addressed by researchers Harris and Marks, and drawing on UNESCO technology evaluation guidelines, they highlight the criteria of feasibility, appropriateness, and impact as key [4]. Harris and Marks discuss issues of appropriateness at length, and their considered analysis of how a technology can usefully be deployed—without replicating neocolonial patterns or disrupting valuable forms of local knowledge—is a welcome addition to technology and development considerations. Our own work considers a balance of these three criteria, and our efforts are focused on developing a low cost, easy-to-use ultrasound system that will respond specifically to the need for feasible and appropriate technologies within low-resource settings. To help lessen maternal and neonatal mortality rates, we have developed a portable ultrasound platform targeted at health practitioners with limited training. Specifically, our project aims to create an ultrasound system for midwives in Uganda that will enable them to identify potential pregnancy complications. Once a condition is identified the midwife then refers the mother with high-risk conditions to a well- equipped medical facility for delivery. In our prior work [6], we discussed the feasibility of our system, demonstrating that the combination of older ultrasound technology and a