Abstract
BACKGROUND: Ischemic stroke is a leading global cause of death and disability, presenting significant challenges in diagnosis, treatment, and prognosis. Radiomics, an emerging interdisciplinary methodology, employs machine learning to extract high-dimensional features from medical imaging and has demonstrated superior predictive performance in ischemic stroke research. However, the rapidly accumulating publications lack systematic bibliometric synthesis. We therefore conducted a visual bibliometric analysis to map research evolution and emerging trends. METHODS: This study conducted a bibliometric and visual analysis of ischemic stroke radiomics research from 2004 to 2024 using tools like CiteSpace and VOSviewer. The analysis explored publication trends, research hotspots, and technological advancements, identifying collaborations and key advancements in the field. RESULTS: Radiomics research in ischemic stroke has grown exponentially since its inception in 2014, with China and the United States emerging as major contributors. The primary focus has been on AIS, utilizing advanced imaging techniques such as computed tomography (CT) and magnetic resonance imaging (MRI). Machine learning models, particularly deep learning architectures, are being widely applied for lesion segmentation, risk assessment, and functional prognosis prediction. Despite rapid advancements, challenges persist in standardizing imaging protocols, enhancing interdisciplinary collaborations, and ensuring clinical translation. CONCLUSION: Radiomics is transforming ischemic stroke research by enabling detailed imaging analyses and facilitating data-driven clinical decision-making. Future endeavors should prioritize addressing standardization issues, expanding multicenter collaborations, and developing interpretable models that integrate radiomics with clinical and molecular biomarkers. Such efforts will accelerate the translation of radiomics into routine ischemic stroke care and improve patient outcomes.