Optimization of Radiodiagnostics Based on Artificial Intelligence to Enhance Clinical Accuracy and Efficiency

The integration of Artificial Intelligence (AI) into radiodiagnostics has emerged as a transformative innovation in medical imaging, offering substantial improvements in diagnostic accuracy, workflow efficiency, and clinical decision-making. Over the past five years, AI-driven technologies such as deep learning, convolutional neural networks (CNNs), and radiomics have shown remarkable potential in automating image interpretation, reducing human error, and supporting early disease detection. According to Liu et al. (2021) and Thrall et al. (2022), AI has demonstrated superior performance in identifying complex imaging patterns in modalities such as CT, MRI, and PET scans, surpassing traditional manual analysis in several diagnostic domains. In the Indonesian context, the adoption of AI in radiodiagnostic practices is still in its early stages, with challenges including limited digital infrastructure, lack of technical expertise, and regulatory uncertainty (Suryaningrum et al., 2023). However, initiatives by the Indonesian Radiology Society and academic institutions have begun to explore pilot implementations and collaborative research with global AI developers (Utama & Raharjo, 2024). These efforts highlight the growing recognition of AI’s capacity to optimize clinical workflow efficiency and improve diagnostic precision. This literature review aims to analyze recent advances in AI-based radiodiagnostics, evaluate their implications for clinical accuracy and operational efficiency, and identify the opportunities and challenges specific to the Indonesian healthcare system. Through a comprehensive synthesis of studies published between 2020 and 2025, this paper discusses key trends, algorithmic approaches, ethical considerations, and policy perspectives necessary to guide the responsible integration of AI into clinical radiology practice.