Scholarlink

Introduction <p>In 2020, an estimated 287,000 women died in pregnancy or childbirth-related causes (70% in Africa), while 2.3&#xa0;million newborns died in the first 28 days of life in 2022 (46% in Africa). The utility of data science and Artificial Intelligence (AI) in health has potential to contribute and accelerate innovation, improve data use for evidence-based decision making, and better planning for policy decision making.</p> Aim <p>To map the current landscape of data science and artificial intelligence applications in maternal, newborn and child health (MNCH) across Africa and identify gaps, challenges, and opportunities for future implementation.</p> Methods <p>A scoping review was conducted following the Arksey and O’Malley framework across five databases (PubMed, Web of Science, EMBASE, SCOPUS, Ovid) and grey literature published before December 2023. Thematic analysis was conducted using previously published seven-domain framework to identify patterns in data science applications, challenges and opportunities. Additionally, projects within the maternal newborn and child health focus on the WHO’s Digital Health Initiatives Atlas were reviewed.</p> Results <p>Of 11,320 articles screened, 52 articles from 31 African countries met inclusion criteria. Most studies (<i>n</i> = 34, 65.4%) were from Eastern Africa. Most studies were research projects (<i>n</i> = 28, 53.8%) and demographic health surveys (<i>n</i> = 22, 43.1%) rather than operational implementation government programmes. Key themes identified included infrastructure challenges, data quality issues, limited workforce capacity, and heavy reliance on external funding. The WHO’s Digital Health Atlas revealed 659 data science initiatives across Africa with the earliest recorded in 2002. 316 (48%) projects were focussed on MNCH and were implemented in 44 African countries.</p> Conclusion <p>The limited application of data science and artificial intelligence in MNCH at the national level highlights a significant gap in Africa. Our review found that 28 studies (53.8%) were research projects compared to operational implementations integrated into routine systems to inform policy. Main barriers include inadequate infrastructure, limited data stewardship capacity, and insufficient government commitment. Opportunities exist through Africa’s youthful demographics, expanding mobile technology, and latecomer advantage in digital innovation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data science and artificial intelligence for maternal, newborn and child health: scoping review and thematic analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月28日

查看全文
  • Joseph Akuze,
  • Grieven P. Otieno,
  • Samson Yohannes Amare,
  • Bancy Ngatia,
  • Phillip Wanduru,
  • Fati Kirakoya-Samadoulougou,
  • Rornald Muhumuza Kananura,
  • Agbessi Amouzou,
  • Abiy Seifu Estifanos,
  • Eric O. Ohuma

摘要

Introduction

In 2020, an estimated 287,000 women died in pregnancy or childbirth-related causes (70% in Africa), while 2.3 million newborns died in the first 28 days of life in 2022 (46% in Africa). The utility of data science and Artificial Intelligence (AI) in health has potential to contribute and accelerate innovation, improve data use for evidence-based decision making, and better planning for policy decision making.

Aim

To map the current landscape of data science and artificial intelligence applications in maternal, newborn and child health (MNCH) across Africa and identify gaps, challenges, and opportunities for future implementation.

Methods

A scoping review was conducted following the Arksey and O’Malley framework across five databases (PubMed, Web of Science, EMBASE, SCOPUS, Ovid) and grey literature published before December 2023. Thematic analysis was conducted using previously published seven-domain framework to identify patterns in data science applications, challenges and opportunities. Additionally, projects within the maternal newborn and child health focus on the WHO’s Digital Health Initiatives Atlas were reviewed.

Results

Of 11,320 articles screened, 52 articles from 31 African countries met inclusion criteria. Most studies (n = 34, 65.4%) were from Eastern Africa. Most studies were research projects (n = 28, 53.8%) and demographic health surveys (n = 22, 43.1%) rather than operational implementation government programmes. Key themes identified included infrastructure challenges, data quality issues, limited workforce capacity, and heavy reliance on external funding. The WHO’s Digital Health Atlas revealed 659 data science initiatives across Africa with the earliest recorded in 2002. 316 (48%) projects were focussed on MNCH and were implemented in 44 African countries.

Conclusion

The limited application of data science and artificial intelligence in MNCH at the national level highlights a significant gap in Africa. Our review found that 28 studies (53.8%) were research projects compared to operational implementations integrated into routine systems to inform policy. Main barriers include inadequate infrastructure, limited data stewardship capacity, and insufficient government commitment. Opportunities exist through Africa’s youthful demographics, expanding mobile technology, and latecomer advantage in digital innovation.