DOI: 10.32322/jhsm.1987220 ISSN: 2636-8579

Paradigm shifts in non-small cell lung cancer imaging: a bibliometric and thematic analysis of the top 100 highly cited articles (2016–2026)

Abdullah Enes Ataş, Abdurrahman Koç, Hasan Eryeşil
Show PDF Cite
Aims: To conduct a bibliometric and thematic analysis of the 100 most-cited articles published over the last decade concerning the intersection of non-small cell lung cancer (NSCLC) and radiology, highlighting paradigm shifts and current research trends.Methods: The Web of Science Core Collection was queried for literature on NSCLC and radiology/imaging published between 2016 and 2026. The top 100 most-cited original articles and reviews were extracted. Data were categorized thematically and mapped bibliometrically using VOSviewer to analyze keyword co-occurrence and chronological evolution.Results: The thematic analysis revealed that “Artificial Intelligence (AI) & Deep Learning” (25 articles, 9711 citations) and “treatment response & Prognosis” (26 articles, 6271 citations) dominated the high-impact literature. Timeline mapping demonstrated a notable shift from traditional anatomical staging toward predictive radiomics, deep learning models, and immunotherapy evaluation. The Lancet Oncology and Harvard University emerged as the leading journal and institution, respectively. Despite high citation rates, a lack of methodological standardization in image processing remains a significant barrier to clinical translation.Conclusion: NSCLC radiology has evolved from a qualitative morphological evaluation into a quantitative, AI-driven discipline capable of predicting genetic mutations and treatment outcomes. Overcoming standardization bottlenecks and developing multimodal systems are crucial for integrating these robust models into clinical workflows.

More from our Archive