Natural Language Processing Technologies for Public Health in Africa: A Scoping Review (Preprint)

Опубликована: Ноя. 12, 2024

Natural language processing (NLP) has the potential to promote public health. However, applying these technologies in African health systems faces challenges, including limited digital and computational resources support continent’s diverse languages needs. This scoping review maps evidence on NLP for Africa, addressing following research questions: (1) What needs are being addressed by what unmet remain? (2) factors influence availability of across countries languages? (3) stages deployment have reached, extent they been integrated into systems? (4) measurable impact had outcomes, where such data available? (5) recommendations proposed enhance quality, cost, accessibility health-related Africa? includes academic studies published between January 1, 2013, October 3, 2024. A systematic search was conducted databases, MEDLINE via PubMed, ACL Anthology, Scopus, IEEE Xplore, ACM Digital Library, supplemented gray literature searches. Data were extracted technology functions mapped World Health Organization’s list essential United Nations’ sustainable development goals (SDGs). The analyzed identify trends, gaps, areas future research. follows PRISMA-ScR (Preferred Reporting Items Systematic Reviews Meta-Analyses Extension Scoping Reviews) reporting guidelines, its protocol is publicly available. Of 2186 citations screened, 54 included. While existing a subset SDGs, coverage remains uneven, with widely spoken languages, as Kiswahili, Yoruba, Igbo, Zulu, no most Africa’s >2000 languages. Most prototyping phases, only one fully deployed chatbot vaccine hesitancy. Evidence limited, 15% (8/54) attempting evaluations 4% (2/54) demonstrating positive improved participants’ mood increased intentions. Recommendations include expanding coverage, targeting local needs, enhancing trust, integrating solutions systems, adopting participatory design approaches. reveals industry- nongovernmental organizations–led projects focused deployable applications. tend few major specific use cases, indicating narrower scope than Despite growth health, gaps remain deployment, linguistic inclusivity, outcome evaluation. Future should prioritize cross-sectoral needs-based approaches that engage communities, align incorporate rigorous outcomes. RR2-doi:10.1101/2024.07.02.24309815

Язык: Английский

Natural Language Processing Technologies for Public Health in Africa: A Scoping Review (Preprint)

Journal of Medical Internet Research, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 12, 2024

Natural language processing (NLP) has the potential to promote public health. However, applying these technologies in African health systems faces challenges, including limited digital and computational resources support continent's diverse languages needs. This scoping review maps evidence on NLP for Africa, addressing following research questions: (1) What needs are being addressed by what unmet remain? (2) factors influence availability of across countries languages? (3) stages deployment have reached, extent they been integrated into systems? (4) measurable impact had outcomes, where such data available? (5) recommendations proposed enhance quality, cost, accessibility health-related Africa? includes academic studies published between January 1, 2013, October 3, 2024. A systematic search was conducted databases, MEDLINE via PubMed, ACL Anthology, Scopus, IEEE Xplore, ACM Digital Library, supplemented gray literature searches. Data were extracted technology functions mapped World Health Organization's list essential United Nations' sustainable development goals (SDGs). The analyzed identify trends, gaps, areas future research. follows PRISMA-ScR (Preferred Reporting Items Systematic Reviews Meta-Analyses Extension Scoping Reviews) reporting guidelines, its protocol is publicly available. Of 2186 citations screened, 54 included. While existing a subset SDGs, coverage remains uneven, with widely spoken languages, as Kiswahili, Yoruba, Igbo, Zulu, no most Africa's >2000 languages. Most prototyping phases, only one fully deployed chatbot vaccine hesitancy. Evidence limited, 15% (8/54) attempting evaluations 4% (2/54) demonstrating positive improved participants' mood increased intentions. Recommendations include expanding coverage, targeting local needs, enhancing trust, integrating solutions systems, adopting participatory design approaches. reveals industry- nongovernmental organizations-led projects focused deployable applications. tend few major specific use cases, indicating narrower scope than Despite growth health, gaps remain deployment, linguistic inclusivity, outcome evaluation. Future should prioritize cross-sectoral needs-based approaches that engage communities, align incorporate rigorous outcomes. RR2-doi:10.1101/2024.07.02.24309815.

Язык: Английский