background and rationale
the use of ai is showing promise in transforming how health systems are planned and health services are delivered across low- and middle-income countries (lmics) today. responsibly designed and implemented ai has the potential to contribute to improved global health outcomes. in the area of srmh outcomes, ai applications have the potential to strengthen point-of-care services, inform diagnoses and provide personalized information based on real-time analysis of relevant data. for example, ai models and applications have been developed to support the early detection of pre-eclampsia and improve accuracy of hiv testing.
in lower-income countries, indicators for sexual, reproductive and maternal health — referred to by some more generally as sexual and reproductive health rights (srhr) — are not on track to achieve sustainable development goal (sdg) 3: good health and well-being. inadequate srmh service provision has been linked with unintended pregnancies — including teen pregnancies — early and forced marriage, complications related to unsafe abortions, gender-based violence and increases in sexually transmitted infections. this is particularly pronounced for vulnerable groups such as adolescents, people living with disabilities, refugees and internally displaced populations. the covid-19 pandemic has had a devastating impact on srmh outcomes such as access to essential services and support for survivors of gender-based violence. gender inequality, the focus of sdg5, is a central challenge for women and girls to overcome, especially in matters related to sexuality and reproduction.
despite the promise of ai to improve srmh outcomes, there are important ethical, legal and social risks that need to be appropriately managed, mitigated and governed. for example, harmful biases can be integrated into algorithms, which can translate into biased public health messaging, diagnoses and treatment protocols. furthermore, much of the data required to develop ai models to tackle srmh challenges are non-representative or inaccessible. the current increase in demand for health-focused ai solutions in lmics is not commensurate with the investments being made to strengthen health systems, credible data, skilled individuals and requisite computing infrastructure.
key objectives
through this call, up to three hubs will be set up and managed, one each in mena, lac and asia. these hubs will be managed by a regionally based organization or consortium of organizations. each of the hubs will be tasked with establishing, managing and supporting implementation research networks in their respective region. setting up the network should consist of running an open call to select implementation research projects in the corresponding region working on or researching ai innovations for srmh, with a typical approach being to select six to ten projects. it is expected that the selected projects will represent linguistic, gender and geographic diversity across the respective region; diversity across relevant areas of application of ai to srmh; and diversity of relevant involved stakeholders (e.g., university researchers, start-ups, ministries of health, research-oriented think-tanks, consultancies, labs or community groups). the selected innovation research projects will receive funding and support as sub-grants for an implementation research project for a duration of at least 18 months.
the general objective of each of the three innovation research networks is to advance srmh in mena, lac and asia through implementation research promoting responsible development and deployment of ai innovations. by responsible, we are referring to ai innovations that are ethical, respect human rights, inclusive and contribute to environmental sustainability.
specific objectives are to:
- strengthen the body of multidisciplinary evidence emerging from lmic-based researchers on how to develop and scale responsible ai innovations for improving srmh in mena, lac or asia.
- build innovation research capacities in mena, lac or asia to develop, deploy and scale responsible ai applications in “real world” settings to improve srmh outcomes.
- deepen understanding and informed practices to enhance gender and social inclusion and limit related biases in the design, implementation and use of ai solutions for srmh.
- influence ai and srmh policies, practices and efforts to scale up and/or commercialize responsible ai solutions.
the primary responsibilities of the hubs include:
- developing and managing an open call for proposals process to select innovation research projects focused on using ai to improve srmh outcomes;
- strengthening how gender equality and inclusion (gei) and intersectional analysis are addressed in the innovation research projects, and ensure each research project team has the appropriate multidisciplinary experience required;
- supporting networking among selected research teams to strengthen knowledge exchange;
- harvesting and synthesizing outcomes related to health outcomes, innovation processes, strengthening gei and scaling strategies;
- facilitating equitable engagement of women in research and leadership roles.
for more information, please contact jill sherman, international research facilitator, at intl.research@lakeheadu.ca.