AI solutions improving sexual and reproductive health across the Global South
Innovation hubs across the Global South are developing solutions that realize the promise of responsible artificial intelligence (AI) for better sexual and reproductive health. These hubs are funded by IDRC and will soon receive additional financial support through a new partnership with the United Kingdom’s Foreign, Commonwealth and Development Office (FCDO).
Thirty-one projects across four hubs in Latin America and the Caribbean, sub-Saharan Africa, Asia, and the Middle East and North Africa are responding to local sexual and reproductive health (SRH) needs through AI-based innovations that are ethical, inclusive and sustainable in their design and implementation. The focus on these tenets of responsible AI is central to these hubs, which are focused on closing specific regional gaps in health equity and redressing root drivers of poor access to sexual health and reproductive rights.
Each of the four hubs provides technical and financial support, as well as mentorship and networking opportunities, to their grantees. In addition, the hubs are connected across regions and develop cross-regional reflections and syntheses as part of the AI for Global Health initiative and the broader AI for Development program at IDRC.
Africa
The Hub for Artificial Intelligence in Maternal Sexual and Reproductive Health (HASH) is supporting 10 projects across seven countries (Ghana, Ethiopia, Kenya, Namibia, Nigeria, Tanzania and Uganda). Each initiative addresses one or more of the priority SRH needs identified for sub-Saharan Africa: maternal health, sexually transmitted infections (STIs), HIV and adolescent SRH.
The HASH network, which is grounded by these innovative projects, is a dynamic platform fostering community, collaboration and learning. With over 100 members and ongoing growth, the network aims to bring together various stakeholders who are interested in the rapidly developing field of AI and who want to explore its application in SRH.
A HASH project in Nigeria, called mDoc: Harnessing the power of Artificial Intelligence to augment patients' knowledge, understanding and behaviours with Sexually Transmitted Infections, has used its funding to modify a rules-based AI chatbot model (Kem) to one that uses large language models. This shift enhances Kem’s ability to deliver personalized responses to STI questions, considering the user’s gender and age. The new version includes content in Pidgin, Yoruba, Igbo and Hausa, in addition to English.
HASH is a multi-disciplinary consortium between the Infectious Diseases Institute and AI Lab at Makerere University and Sunbird AI in Uganda.
Latin America and the Caribbean
The Centro de Inteligencia Artificial y Salud para América Latina y El Caribe (CLIAS) is supporting seven projects to improve SRH outcomes for vulnerable populations across Latin America and the Caribbean. The projects, based in Colombia, Guatemala and Argentina, address regionally specific SRH priorities such as unintended adolescent pregnancy, preeclampsia, maternal morbidity, SRH and cervical cancer. Specific underserved groups such as adolescents and persons with disabilities are central to these initiatives.
For example, the Faculty of Exact and Natural Sciences at the University of Buenos Aires in Argentina is developing the first feasibility analysis for using AI tools in the diagnosis of cervical cancer from national data. The team is creating an extensive bank of digitized images of biological samples based on retrospective digitization of samples already available in physical format, which will serve to train the learning model. Examined alongside other relevant patient information, including age and previous controls, the aim is to improve health outcomes.
CLIAS has produced technical documents on issues such as ethical principles for AI in health care, review of experiences using AI for SRH in the region and the challenges and opportunities of using AI for sexual and reproductive health and rights in the region. CLIAS is led by the Center for Implementation and Innovation in Health Policies of the Institute of Clinical and Health Effectiveness, Argentina.
Asia
AI-Sarosh (AI for Sexual, Reproductive and Maternal Health in South Asia) is supporting nine projects in four countries: Bangladesh, Nepal, Pakistan and Sri Lanka. These projects, individually and collectively, address the regionally specific SRH needs of maternal health, family planning and male partner involvement, adolescent SRH, maternal mental health, and improving counselling and services for STIs and contraception.
Functioning as a pivotal networking and innovation hub, AI-Sarosh assumes responsibility for coordinating activities, identifying opportunities, overseeing projects and fostering knowledge-sharing among countries in the region. The focus on mental health within this hub is a noteworthy feature, with three of the nine projects choosing to focus on this growing area of concern in SRH.
Mental health among pregnant and new mothers is a growing concern in South Asia. In Pakistan, the prevalence of postnatal depression is reported to be 37%, which is among the highest in Asian countries. This significant and often silenced issue among women requires evidence-based solutions for prevention and treatment. Zahid Memon from Aga Khan University in Pakistan is leading a project on bridging the gap in early detection and intervention through development and evaluation of an AI-based predictive model for postnatal depression.
AI-Sarosh is a partnership co-led by PHC Global in Pakistan and the GTA Foundation in Nepal.
Middle East and North Africa
The Global Health and Artificial Intelligence Network in the Middle East and North Africa region (GHAIN MENA) is supporting six projects across Lebanon, Morocco, the West Bank and Gaza, Qatar, Turkey and Jordan. These projects address regional SRH priorities such as family planning, fostering partner involvement, adolescent SRH, strengthening the health workforce and gender-based violence, focusing on primary health care settings. In addition to offering sub-grants to other grantees, GHAIN MENA is positioned to actively contribute to the responsible adoption of AI and support the creation and strengthening of a network of researchers, who are interested or working in the realm of AI in global health.
The GHAIN MENA project in Morocco, entitled Saving Mothers’ Lives using Artificial Intelligence in Rural Areas in Marrakech, is developing an AI-based application that can accurately diagnose and prevent high-risk maternal health outcomes across 12 rural primary health-care centres. The project's goal is to reduce the maternal mortality ratio by enhancing accurate clinical diagnosis and facilitating timely referrals for pregnant women using the employed AI model.
GHAIN MENA is implemented by a consortium led by the Global Health Institute at the American University of Beirut, in partnership with the Jordan University of Science and Technology and the Eastern Mediterranean Public Health Network.
What does the future hold?
These four Global South-led innovation hubs are blazing new trails to improve equitable SRH outcomes for all through responsible AI solutions. These hubs were designed with the intention of improving health outcomes through their support of innovation, mentorship, peer-learning and exchange, alongside providing robust and accessible evidence. Connections within and across the hubs lay the foundation for accelerating the achievements of SDGs 3 (on health) and 5 (on gender equality) by demonstrating how best to develop, deploy, govern and evaluate responsible AI solutions in a way that is meaningful to and respectful of the diversity of contexts across the Global South.
Learn more:
Contributors: Chaitali Sinha, senior program specialist, IDRC; Cintia Cejas, coordinator, Institute for Clinical Effectiveness and Health Policy (Argentina); Elizabeth Oseku, project coordinator, Makerere University (Uganda); Nour El Arnaout, manager, American University of Beirut (Lebanon); and Mohammad Imram, executive director, PHC Global (Pakistan).