Ultrasound for Ugandan midwives
2010 - 2012
UX researcher and designer on a Bill & Melinda Gates-sponsored project focused on building a low-cost ultrasound device for midwives in developing regions, specifically for women working in rural environments.

Context and problem

Maternal morality rates in sub-Saharan Africa are unacceptably high. Many factors contribute to this, including the near absence of ultrasound technology, which can potentially decrease deaths by identifying common pregnancy complications.

In 2010 Seattle radiologist Dr. Rob Nathan began donating GE portable ultrasound machines costing $20,000 to small, rural villages in Uganda. This price was unsustainable and the midwives had to train on these devices in Kampala, taking them away for long periods from their desperately needed work in village clinics. I, along with Alexis Hope, Wayne Gerard, Pratik Prasad, and Dr. Beth Kolko, decided to help Dr. Rob Nathan design a low cost portable ultrasound device.

We were asked to build the device for under $5,000 and to create a simplified UI focused on three common prenatal complications:

1) Multiple gestations (twins)

2) placenta previa

3) breech presentation

Surveys, classes, and field research

The team first mailed surveys to midwives working on the current GE machines in Kampala and in rural southern Uganda. We also conducted contextual inquires at local Seattle hospitals to learn about ultrasound technology.

We then traveled to Uganda and visited an ultrasound institute in Kampala, where we attended ultrasound classes with midwives. We also visited several clinics, interviewing doctors, midwives, and mothers to understand the health care system in Uganda.

We then traveled south and conducted fieldwork in the Isingiro District, meeting with midwives who often worked 14 to 18 hours a day without reliable electricity. The donated GE machines were very complicated, and when minor technical issues occurred, the midwives put the device in a corner or closet without further use.

Actions taken based on this research

Based on research and resource constraints, we built a simplified ultrasound application that utilized an Interson USB probe attached to a netbook. The most basic interactions were prioritized first, such as filling in the patient record and performing a simple scan. We iterated until radiologists, sonographers, and midwives both in the U.S. and Uganda verified that the device could successfully diagnose the three most common complications.

We were able to get the cost of the device to $3,500 using a modular off-the-shelf approach rather than an all-in-one system. We also added an integrated contextual help feature that helped supplement the limited sonography training received by the midwives, helping to answer diagnostic questions when a radiologist isn't available.

uganda field work

Scanning fetus head

scanning fetus head

Browsing ultrasound topics

browing topics

Learning topic about sonography

learning sonography

Comparing ultrasound images

comparing images

Output and impact

I co-authored three academic papers:

  1. W. Brunette; W. Gerard; M. Hicks; A. Hope; R. Anderson; G. Borriello; B. Kolko; R. Nathan. "Portable Antenatal Ultrasound Platform for Village Midwives." ACM Computing for Development (DEV), December 2010.
  2. A. Hope; M. Hicks; W. Gerard; P. Prasad; K. Saville; W. Brunette; L. Ham; J. Keh; R. Anderson; G. Borriello; B. Kolko; R. Nathan. "Designing an Intelligent Medical Assistant for Diagnostic Ultrasound." International Conference on Intelligent User Interfaces (IUI), February 2011.
  3. W. Brunette; M. Hicks; A. Hope; G. Ruddy; R. Anderson; B. Kolko. "Reducing Maternal Mortality: An Ultrasound System for Village Midwives." IEEE Global Humanitarian Technology Conference (GHTC), October 2011.

Ultimately, our work broadened the conversation in the medical community about the needs for simplified lower cost medical technology. GE, Philips, and other medical device companies took notice. In addition, our HCI professor started a company called Shift Labs focusing on developing low cost medical technology.

For more about this work, please watch this TEDx presentation from HCI professor and now Shift Labs CEO Beth Kolko: