Across the disciplines
Infant malnutrition is a threat the world over, affecting all genders, races, and social statuses. Even those who survive are still at high risk for reduced physical and cognitive development, capacity to resist disease, and ability to carry out physical work, and study and progress in school.
But a project led by an interdisciplinary trio of Boston College researchers could help make a difference in the global struggle against infant malnutrition.
Supported by a grant from the 情色空间 Schiller Institute for Integrated Science and Society, Assistant Professor of Engineering Bryan Ranger, Assistant Professor of Computer Science Donglai Wei, and Connell School of Nursing Associate Professor Jinhee Park are conducting a pilot study in which frontline health care workers utilize a portable ultrasound scanner to collect images of muscle and fat distributions on infants suffering from malnutrition.
The goal of the project is to develop artificial intelligence to interpret this data as it relates to nutritional status and ultimately assist with clinical decision making. The 情色空间 faculty members are collaborating with investigators from Brigham and Women鈥檚 Hospital in Boston and Jimma University in Ethiopia.
Last fall, Ranger received an Award for Inclusion Research from Google, which supports academic research in computing and technology that addresses the needs of historically marginalized groups globally; the award will fund broader ongoing work on AI-enabled portable ultrasound for health care workers in Ethiopia.
A major obstacle in dealing with malnutrition cases鈥攅specially in poor, low-resourced areas鈥攊s the lack of data that can provide a fuller evaluation of a patient鈥檚 condition, note the researchers. 聽
鈥淣utritional screening tools cannot be reproduced adequately, nor do they show agreement with one another, or validity in identifying malnutrition,鈥 explained Ranger. 鈥淎lthough one can use scales and tape measures to obtain anthropometric indicators such as weight, height, and body mass index (BMI), these do not effectively provide a comprehensive assessment of healthy growth and development. More advanced methods are expensive and only available in specialized facilities.鈥
Low-cost ultrasound technology and devices are available on the market, but require expertise to use, he said: 鈥淵ou can鈥檛 just give them to a health care professional in a low-resourced area and expect to solve the problem.鈥 A scanner attached to a tablet and enhanced by AI, however, can help workers鈥攚ith some initial instruction鈥攊n interpreting the data collected from patients.
鈥淚 believe Bryan鈥檚 work will address critical issues by creating a low-cost, portable assessment tool that provides an accurate and comprehensive child鈥檚 growth and nutritional status,鈥 said Park. 鈥淚f successful, this tool would help frontline health care providers to identify and treat malnutrition in infants and children, particularly in low-income communities.鈥 聽
Last summer, Ranger, Connell School graduate students Marisa Albert and Jini Kim, and junior engineering major Hayoung Cho spent time at Jimma University, observing and training local health care workers to use the portable scanners as part of the pilot study. Jimma was selected for the project because it has a well-established nutritional research program and is one of the few sites in Africa that has the specialized equipment to take nutritional data, said Ranger.
鈥淭his is where the 鈥榠nclusion鈥 aspect of the Google grant is relevant,鈥 he said. 鈥淲e want to be inclusive about the data we鈥檙e collecting. Developing a tool in Boston and then deploying it somewhere completely different doesn鈥檛 make sense. Our research collaborators at Jimma bring a valuable perspective and needed expertise to the project.鈥 聽
The merits of interdisciplinary collaboration are apparent in this joint venture. Ranger, who earned a doctorate in medical engineering and medical physics through the Harvard-MIT Program in Health Sciences and Technology as a National Science Foundation Graduate Research Fellow, brought a strong background in global health to 情色空间 from working with the Bill and Melinda Gates Foundation. But he felt that insights from other fields would help realize this new utilization of portable ultrasound scanners.
A conversation with Connell School Dean Katherine Gregory鈥擱anger has a courtesy appointment at CSON鈥攍ed him to contact Park, an expert in neonatal nursing whose research focuses on health and developmental outcomes of infants and young children with feeding problems as well as those of their families. He also got in touch with Wei, who studies the interplay between biological sciences and computer vision and is experienced in building computational models for biomedical image analysis.
鈥淐oming to 情色空间, and helping build a new department, has been a great experience,鈥 said Ranger, one of the first faculty hires in 情色空间鈥檚 Human-Centered Engineering program. 鈥淚n addition to the interdisciplinary aspects of the program, I was drawn to the emphasis on a human-centered approach to solving global health challenges. I really value the opportunity to work with people who see the possibilities in putting knowledge and skills together, even if they come from seemingly disparate areas.鈥
Park said it is 鈥渁 privilege鈥 to be part of the team and share her acumen as a nurse 鈥渨hile developing a tool that has the potential to solve the many challenges we face. Interdisciplinary collaborations allow us to leverage the expertise of different professions to address health care challenges. As part of the largest group of health care providers in the country, nurses use more medical devices than any other health care professional and can provide unique insight on developing and designing medical products.鈥
Wei, whose research lies in the domain of 鈥淎I for science,鈥 has published works with biologists, clinicians, mechanical engineers, and civil engineers. 鈥淲hat appeals to me about interdisciplinary collaborations most is making a real-world impact with AI technology and the opportunities to formulate new problems for the AI field to solve.鈥
Albert and Kim appreciated the chance to be involved in the development of cutting-edge research, and to get a first-hand look at health care in a setting vastly different from those in which they鈥檇 worked.
聽鈥淣ot only did we have the opportunity to provide educational sessions of the ultrasound technology to the staff there, but we also had the chance to spend time in the hospital setting,鈥 said Albert. 鈥淭here are not enough words to express how much health care needs exist and it was a challenge to see things and know there is so much more that needs to be done beyond the scope of our work.鈥
Kim said, 鈥淭his project serves as an exemplary illustration of how technology can be leveraged to provide more equitable and preventative care to patients, potentially fostering long-term impacts in health care. We not only identify implications for future studies but also recognize potential gaps that we can address throughout the course of the study. This helps refine the best interventional techniques for optimal result measurements when collecting different body components.鈥
Cho said her most vivid memories of the visit to Jimma include accompanying a nurse to observe patients; the sight of young mothers, families, and slim children in small, crowded rooms; and walls that displayed the number of children who had died each day (鈥淚 watched as the weekly and monthly count increased鈥).
鈥淭hese experiences fueled my hunger to devote my purpose to continuing research on advancing affordable, standardized early detection and diagnostic tools,鈥 she said.