EXPLORING MACHINE LEARNING POTENTIALS TO IMPROVE MEDICAL IMAGING SERVICES OF CHILDREN AND ADOLESCENTS IN LOW-RESOURCE SETTINGS

Abstract

The existing body of evidence in literature raises concerns over the growing overuse, underuse, and misuse of medical imaging. These trends worsen among children and adolescents because they present a unique individuality within the medical imaging landscape. Overuse of medical imaging implies giving patients care that they do not need, underuse means failing to provide patients the correct care that they need while, misuse of medical imaging implies making errors that can harm people's health. Giving the correct medical imaging service and the right radiation dose to children and adolescents transcends the boundaries of radiation protection and good medical practice. Several challenges persist in providing quality medical imaging services, radiation protection, and safety for children and adolescents in low-resource settings, especially in Sub-Sahara Africa. However, these challenges are surmountable if machine learning potentials are used to improve medical imaging services for children and adolescents in low-resource settings. Digital literacy and data readiness are indispensable requirements to achieve the desired objective because a machine learning model is only as good as its training data set.