Post By bderenzi in  Projects 

South King County encompasses several cities in Washington State, including Tukwila, Sea-Tac and Federal Way. Its residents experience some of the worst health outcomes in the country. For example, diabetes prevalence in South King County is more than twice the rate of King County, while rates of obesity and heart disease are also disproportionately high. The incredibly diverse ethnic and linguistic makeup of the area also leads to residents’ reduced access to healthcare facilities and services.

In partnership with Global to Local, we are adapting the video-creation model developed by Digital Green to establish a community-based program for creating culturally-appropriate video that demonstrate healthy behaviors. By working with Global to Local and its Community Health Promoters (CHPs), the project works to both build capacity within Global to Local and the community for creating and disseminating videos.

The first set of videos have been filmed and set to be shared at community forums in September/October 2012. These videos target the Eritrean, Latino, and high school communities in South King County.

Sep 13, 2012 | 0 Comments 

Post By bderenzi in  Projects 

A job aid is typically a one- or two-page instruction sheet or poster that walks the user through a series of steps to perform certain tasks. Job aids range from covering simple step-wise processes to more complex decision-making or conducting calculations. In the public health context in developing countries and other low-resource settings, job aids are designed to help health practitioners who often just one brief training session on a particular procedure. In these cases, job aids are particularly important in guiding workers through procedures they rarely perform or that are complex to conduct.

For years, paper-based job aids have been widely used, and in many situations they serve their purpose well: they provide a comprehensive overview of the procedure, clearly worded instructions, and visual depictions to guide a user through a procedure. However, traditional paper-based job aids have limitations: they can get lost; they might be on a poster in a room separate from a treatment room; they make it difficult for multiple people in multiple locations to use them; they are time-intensive to update when best practices change. Finally, the procedure might be complex and require calculations or decision-making that are difficult to follow on paper, particularly when a health worker is juggling other tasks and interacting with one or more patients.

A smartphone-based job aid application can offer benefits that improve upon paper-based job aids. An app can help users with calculations and decision-making. Animated images can more fully illustrate the procedure. Job aid apps could also automatically be updated with new information when the user connects to the internet. In addition, smart phones are portable and have the capacity to store numerous job aids–serving as an instructional manual of sorts that can be easily accessed with a click or swipe.

Sep 13, 2012 | 0 Comments 

Post By bderenzi in  Projects 

Government, social and health organizations working in developing countries use large-scale data collection to measure their impact and control the quality of the services they provide. Many of these organizations rely on paper forms to perform this data collection. Paper forms are a well-understood and trusted medium, and the low cost and ease-of-use of paper forms suggest that paper will continue to be utilized in developing communities for many years to come. However, the potential benefits of digitizing data from paper forms for statistical analysis and aggregation are significant, and there is thus a need for tools that bridge the gap between the paper and digital worlds, allowing organizations to continue to use familiar, cheap paper forms while also facilitating the collection of digital data.

We built an Android application that automatically captures digital data from paper forms. The phone’s camera is used to photograph the form and computer vision algorithms running on the phone automatically extract digital data from the image. Our design utilizes a lightweight form description language to facilitate the processing of existing paper forms without the need to redesign or add coded marks to the form. We field tested our initial prototype by using it to digitize “bubble” forms that collect vaccine statistics in clinics in Mozambique, and were able to digitize one month’s vaccine statistics for one clinic in 30 seconds with over 99% accuracy.

Although our initial prototype focused on digitizing multiple-choice or “bubble” forms, our latest prototype is capable of reading check-boxes and assisting with the entry of handwritten form fields. We have also integrated the application with the Open Data Kit platform and are looking for NGO partners and collaborators interested in pilot testing the application with their paper-based workflow.

Publications

Nicola Dell, Nathan Breit, Timóteo Chaluco, Jessica Crawford, Gaetano Borriello. 2012. Digitizing paper forms with mobile imaging technologies. In Proceedings of the 2nd ACM Symposium on Computing for Development. ACM, New York, NY, USA, 2:1–2:10. PDF

Nicola Dell, Nathan Breit, Gaetano Borriello. 2012. Digitizing Paper Forms with Mobile Imaging Technologies. In 1st International Workshop on Mobile Data Collection in the Developing World (DataDev 2012). IEEE, Bangalore, India, .

Sep 13, 2012 | 0 Comments 

Post By bderenzi in  Projects 

Remote health monitoring and disease detection in the developing world are hampered by a lack of accurate, convenient and affordable diagnostic tests. Many of the tests routinely administered in well-equipped clinical laboratories are inappropriate for the settings encountered at the point of care, where low-income patients may be best served. To address this problem, medical researchers have developed innovative rapid diagnostic tests (RDTs) that are capable of detecting diseases at the point of care within a single patient visit to a clinic. However, for these new diagnostic technologies to be effective, tools must be developed to support the health workers who will be responsible for administering the tests and interpreting their results.

Broadly, this work seeks to answer the following research question: How can we support health workers as they are required to make increasingly complex diagnostic decisions for an increasing number of diseases and medical conditions? To address this challenge, we designed ODK Diagnostics, a smartphone application that supports health workers in three ways: (1) by facilitating the creation of digital job aids to guide users step-by-step through the process of administering each test correctly, (2) by automatically interpreting the test results using computer vision algorithms running locally on the phone and delivering the diagnosis to health workers, and (3) by automating the process of collecting data regarding the type and outcome of the test administered, which will alleviate some of the reporting burden placed on health workers and allow them to spend more time focusing on patients and less time doing paperwork and record keeping.

Thus far, we have completed the technical implementation and initial evaluation of Diagnostics, which represent only the first steps towards building a effective point-of-care diagnostic system. Our results suggest that the system is ready to be field tested with health workers, and our next steps will involve more focused field studies that rigorously evaluate both the digital job aids and the algorithm for automatically interpreting the test results to ensure that they are usable and appropriate for point-of-care settings in developing countries.

Publications

Nicola Lee Dell, Sugandhan Venkatachalam, Dean Stevens, Paul Yager, Gaetano Borriello. 2011. Towards a point-of-care diagnostic system: automated analysis of immunoassay test data on a cell phone. In Proceedings of the 5th ACM workshop on Networked systems for developing regions. ACM, New York, NY, USA, 3–8. PDF

Sep 13, 2012 | 0 Comments 

Post By bderenzi in  Projects 

It is estimated that in resource-poor areas of the world, 3.3 million neonatal deaths occur within the first four weeks of life every year. According to the Lancet Child Survival Series, 13% of deaths of children under the age of five could be prevented by breastfeeding alone. Availability of breast milk to vulnerable infants can be increased significantly by establishing human milk banks. However, providing safe breast milk to infants in developing regions continues to be a challenge. Commercial-grade pasteurizers are too expensive and beyond the reach of most organizations. Affordable, low-tech pasteurization methods lack the appropriate quality control mechanisms, which prevents implementation at a large-scale.

In this project we have leveraged mobile and sensing technologies to create an alternate system to safely pasteurize human breast milk. The system, called FoneAstra, enables low-level sensors like temperature probes, to be connected to mobile phones. It ensures that milk is pasteurized correctly by providing appropriate audiovisual feedback to guide users performing the procedure. At the end of the procedure, users are able to print a pasteurization report and labels for pasteurized milk bottles using a Bluetooth-enabled printer. The system automatically uploads temperature curves of procedures to a server, which enables supervisors to remotely monitor facilities where procedures are performed.

The first trial of the system started in Durban, South Africa, in May 2012. We have deployed it at two locations. First, it is being used to perform routine pasteurizations at a milk bank located in the neonate ward of a district-level hospital. Second, our in-country partners in the pediatrics department of the University of KwaZulu-Natal, are using it to validate the efficacy of the system by doing microbial activity tests on pre- and post-pasteurized milk samples. The results received so far have been encouraging – while pre-pasteurized milk samples showed microbial activity, none of the post-pasteurized samples showed any microbial activity. The system will be installed at two new milk banks in Durban in the coming months. Our in-country partners are also promoting the system with the South African Dept. of Health, who are in the process of scaling up milk banking across the country, and are looking for alternate, affordable methods to safely pasteurize human breast milk.

Publications

Rohit Chaudhri, Darivanh Vlachos, Jabili Kaza, Joy Palludan, Nathan Bilbao, Troy Martin, Gaetano Borriello, Beth Kolko, Kiersten Israel-Ballard. 2011. A system for safe flash-heat pasteurization of human breast milk. In Proceedings of the 5th ACM workshop on Networked systems for developing regions. ACM, New York, NY, USA, 9–14. PDF

Sep 12, 2012 | 0 Comments 

Post By bderenzi in  Projects 

Community health workers (CHWs) have the potential to reduce neonatal and maternal mortality rates when used effectively. However, community health programs are difficult to setup and run. Mobile applications, like CommCare, that are used by CHWs to collect information during routine home visits, provide opportunities for increased supervision, accountability, and feedback.

In this project, we explore completing the feedback loop with CHWs by sending back performance feedback in the form of rich graphs. Once a week a CHW will receive an SMS message with a link to her specific performance graph. The project will be evaluated with a randomized controlled trial in India using three groups: (1) a control group, (2) a treatment group that only receives information about individual performance, and (3) a group that receives information about individual performance, as well as how it ranks with a set of her peers.

The specific graphs to be used are currently being piloted and the software being developed. The study will launch before the end of 2012 in India with 120 CHWs.

Sep 12, 2012 | 0 Comments 

Post By bderenzi in  Projects 

Market penetration of smartphones as a computing and communications platform has increased significantly in recent years. This has created an opportunity to integrate mobile consumer devices (e.g., smartphones, tablets) with external sensors to enable the collection of sensed data directly onto mobile devices. The work is part of the larger Open Data Kit (ODK) project that couples mobile devices with cloud services to create mobile information systems that magnify human resources through appropriately designed technology. By lowering the barriers to add external sensing components, we hope to expand mobile data collection applications to include an even richer set of data types. Even though capturing sensor data directly eliminates many of the errors that plague traditional data collection techniques, such as manual form-filling, it is still not widely used in developing regions because of the high level of technical expertise required to develop a mobile sensing application. The technical challenges include managing the details of different physical communication channels, processing sensor-specific data, developing a user interface, and designing application control logic.  Because of these complications, we hypothesize that including sensors in mobile data collection poses several technical barriers that, if reduced, would enable more applications to leverage sensors for data collection across varied domains. To address this, we created an application-level driver framework that enables convenient reuse of sensor-specific code between applications by logically separating the high-level application from the underlying sensor driver.

The focus of ODK Sensors is on enabling the integration of data from a variety of sensors over both wired and wireless communication channels. It simplifies application development by creating a single interface that can control virtually any kind of sensor (both external and built-in) and reduces the amount of code needed to access a sensor. Having a single interface is appropriate for lightly trained technical workers because it hides a large number of the details involved in developing sensing applications. Applications that leverage the framework to communicate with external sensors can be implemented in fewer lines of code by removing sensor communication code. Additionally, the sensor framework automatically multiplexes the communication channels allowing different types of sensors to be used simultaneously by an application. For example, an application can easily use two USB and three Bluetooth sensors simultaneously to record several phenomena at once. The ODK Sensors framework is designed to flexibly meet any application’s needs regardless of data type, data collection rate and size, sensor configuration requirements, or communication channel. The framework also provides a simple, high-performing, and flexible abstraction on which to develop and deploy user-level device drivers on Android. While a device driver abstraction is a standard concept, the framework includes features that make development of device drivers easier by handling sensor state (e.g., connection, buffered data, threading) and only requiring driver developers to implement sensor-specific commands and data processing.

Unlike traditional PC devices, the new class of mobile consumer devices are often locked by service providers or manufacturers, and most end-users do not have the administrative rights, technical ability, or organizational capacity to modify or customize the operating system. As a result, it’s impractical to rely on conventional in-kernel device driver frameworks to integrate external sensors with consumer smartphones. Our project explores ways to package software so that non-technical users can access external sensors from a locked mobile device running a stock version of the Android operating system. The framework assumes the consumer device is ‘locked down’ and an end-user only has the skills to install applications from a standard app marketplace such as Google Play (Google’s Android app store). Using standard Android app distribution channels will make it easy for users to download functionality enhancements (application-level device drivers) to their unmodified Android OS. This simple method of deployment will hopefully lead to the creation of new sensing-based mobile data collection applications that improve information services in under-resourced contexts that typically lack a rich technology infrastructure (both physically and in terms of expertise). ODK Sensors increases the variation of input data types possible by simplifying access to sensing resources through the creation of a single interface that makes external sensors as easy to integrate as built-in sensors. By creating a framework designed to follow ODK’s philosophy of modular components, we aim to expand the tool suite to allow end-users to easily augment their Android device with external sensing options. The component enables easy reuse of sensor drivers that will hopefully lead to an ecosystem of drivers further promoting the creation of novel mobile sensing applications.

Publications

Waylon Brunette, Rita Sodt, Rohit Chaudhri, Mayank Goel, Michael Falcone, Jaylen Van Orden, Gaetano Borriello. 2012. Open data kit sensors: a sensor integration framework for android at the application-level. In Proceedings of the 10th international conference on Mobile systems, applications, and services. ACM, New York, NY, USA, 351–364. PDF

Rohit Chaudhri, Waylon Brunette, Mayank Goel, Rita Sodt, Jaylen Van Orden, Michael Falcone, Gaetano Borriello. 2012. Open data kit sensors: mobile data collection with wired and wireless sensors. In Proceedings of the 2nd ACM Symposium on Computing for Development. ACM, New York, NY, USA, 9:1–9:10. PDF

Sep 12, 2012 | 0 Comments 

Post By bderenzi in  Projects 

In Kenya, only 40% of women deliver in a health facility, despite the dangers for home-based delivery. Similarly, few women complete the three recommended antenatal visits at the facility and few receive proper counseling in the postnatal period. Most projects in maternal health target specific timelines of as opposed to the entire continuum of care.

With the Mobile WaCH project, we are sending targeted (both timeline and health challenges) SMS messages to pregnant women during the entire continuum of care to achieve behavior change around the uptake of government health services. This will be evaluated with a randomized control trial with three groups: (1) a control group, (2) a treatment group with one-way messages, and (3) a treatment group with two-messages that request a reply.

We hypothesize that a woman’s reply can be used as a proxy for her engagement with the health system and eventual uptake of health services.

The exact messages to be used are currently being piloted and iterated on. The study will begin by the end of 2012 at two sites in Kenya: a clinic in a dense urban setting in Nairobi and a clinic in a more rural setting in western Kenya.

Sep 12, 2012 | 0 Comments 

Post By bderenzi in  Projects 

In the Tukwila–SeaTac area, compared to King County averages, life expectancy is seven years shorter and the number of households below the poverty line is 76% higher. Currently there are over 70 languages spoken in SeaTac and Tukwila schools and households. By helping people overcome language barriers perhaps we can improve outcomes.

The  Interpreter Connect project seeks to make mobile phone interpretation services available for free to anyone in the SeaTac-Tuckwila community. By texting “spanish” to the Interpreter Connect phone number, people can request interpretation from a number of volunteers in their community. Volunteer interpreters are given Android phones preloaded with an app for accepting or rejecting interpretation requests, and scheduling times when they will be available to interpret.

There is a pilot that has been running since March, and will continue until the free phone service donated by T-Mobile runs out this August. There are 11 interpreters currently registered in the system and a total of 8 phone calls have been placed through it.

Sep 12, 2012 | 0 Comments