Artificial intelligence in diagnostic medical imaging: A systematic review

Introduction: There is a significant change in global communities due to evolution of artificial intelligence (AI). Healthcare centers are not left behind in AI especially in radiology departments. AI Deployment in radiology, mainly in multidimensional imaging, mammography and ultrasound, transformed through its ability in detection, analyze, interpret, characterize tumor and compare images across modalities.

Objectives: This paper aims to clarify the importance of AI in radiology as an aiding tool rather than a threat, analyze the success of AI in detection, characterization and image interpretation in multidimensional imaging and mammography, discuss optimal utilization in these areas, and examine future prospects and challenges, particularly in developing countries along with strategies to overcome them.

Methods: Relevant publications on AI, radiology, multi-dimensional imaging and mammography across PubMed, Science Direct and Google Scholar over three months. Online videos from conferences, seminars and webinars about AI applications, results, challenges and further research areas were also reviewed. A total of 20 papers were published between 2019 and 2024 were included in the study.

Results: AI improves early tumor detection and characterization of tumors, images analyses and interpretation and increased workflow. However, it faces limitations such as integration into existing systems, data privacy concerns, legal regulations, and inadequate personnel training, along with challenges like staff shortages, high installation costs, and limited PACS availability, particularly in developing countries like Nigeria.

Conclusion: Evidence indicates that AI integration improves diagnostic accuracy and may reduce medical expenditures. Despite technical and regulatory challenges, it functions as an essential aiding tool.