[Paper Review] What We Know So Far: Artificial Intelligence in African Healthcare

Naome A. Etori, Ebasa Temesgen+1 more|arXiv (Cornell University)|May 10, 2023

Artificial Intelligence in Healthcare and EducationMedicine3 citations

TL;DR

This paper reviews the application of artificial intelligence (AI) in African healthcare, focusing on AI's role in improving diagnostics, disease monitoring, and healthcare access in low-resource settings. It highlights successful AI applications in predicting malaria, Ebola, cancer, and diabetes using machine learning models like Random Forest and Neural Networks, while identifying infrastructure, data, and regulatory challenges to implementation.

ABSTRACT

Healthcare in Africa is a complex issue influenced by many factors including poverty, lack of infrastructure, and inadequate funding. However, Artificial intelligence (AI) applied to healthcare, has the potential to transform healthcare in Africa by improving the accuracy and efficiency of diagnosis, enabling earlier detection of diseases, and supporting the delivery of personalized medicine. This paper reviews the current state of how AI Algorithms can be used to improve diagnostics, treatment, and disease monitoring, as well as how AI can be used to improve access to healthcare in Africa as a low-resource setting and discusses some of the critical challenges and opportunities for its adoption. As such, there is a need for a well-coordinated effort by the governments, private sector, healthcare providers, and international organizations to create sustainable AI solutions that meet the unique needs of the African healthcare system.

Motivation & Objective

• To examine the current state of AI integration in African healthcare systems, particularly in low-resource settings.
• To identify how AI algorithms such as machine learning, deep learning, and NLP are applied to improve diagnostics and disease monitoring.
• To investigate the impact of resource limitations—such as poor infrastructure and data scarcity—on AI implementation in Africa.
• To highlight opportunities for sustainable AI adoption through coordinated efforts among governments, private sector, and international partners.
• To address the lack of diverse, representative datasets in AI models for African populations, especially low-resource languages and regional health data.

Proposed method

• Systematic review of peer-reviewed literature and case studies on AI applications in African healthcare.
• Analysis of machine learning models including Random Forest, Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Convolutional Neural Networks (CNN) in disease prediction.
• Evaluation of natural language processing (NLP) for analyzing electronic health records and social media data to detect health trends.
• Use of interpretable models like decision trees and Random Forest to improve transparency and clinical trust in AI predictions.
• Application of Bayesian and ensemble learning models for predicting outcomes in infectious diseases (e.g., Ebola, malaria) and chronic conditions (e.g., diabetes, cancer).
• Assessment of AI performance using standard metrics such as AUC-ROC, accuracy, sensitivity, and specificity across diverse African datasets.

Experimental results

Research questions

• RQ1 What role do AI algorithms play in African healthcare systems, particularly in diagnostics and disease monitoring?
• RQ2 How do resource constraints—such as limited data, infrastructure, and funding—affect the implementation and performance of AI in African healthcare?
• RQ3 What are the most effective AI models for predicting and managing diseases like malaria, Ebola, cancer, and diabetes in African populations?
• RQ4 How can AI be adapted to work effectively with low-resource, low-bandwidth, and low-literacy healthcare environments in Africa?
• RQ5 What policy, data, and infrastructure improvements are needed to ensure sustainable and equitable AI adoption in African healthcare?

Key findings

• Random Forest models achieved 98.33% accuracy, 100% sensitivity, and 96.55% specificity in predicting breast cancer risk among Nigerian women using selected features.
• An Artificial Neural Network (ANN) model achieved an AUC-ROC of 87% for colorectal cancer recurrence and 82% for survival prediction in South Africa.
• A Random Forest model using climate and ecological data predicted malaria vector biting rates in rural Burkina Faso with 99% accuracy.
• Ensemble learning models predicted cervical cancer with 87.21% accuracy based on risk factors in African populations.
• AI models such as ANN, ANFIS, and SVM demonstrated high accuracy in forecasting daily COVID-19 case trends across African regions.
• AI-based systems using NLP and social media data successfully identified diabetes prevalence factors in Nigeria, demonstrating potential for public health surveillance.