Advance in the use of Artificial Intelligence of Pulmonary nodule: evolution, trends, and future directions.

BACKGROUND

Timely detection and intervention for pulmonary nodules play a vital role in decreasing lung cancer-related deaths. Nevertheless, the precise differentiation between benign and malignant nodules continues to face a major clinical challenge. With the rapid progress of artificial intelligence (AI), significant improvements have been made in the detection, classification, and clinical decision-making related to pulmonary nodules. Although scholarly interest in this domain has surged in recent years, there is still a lack of comprehensive bibliometric studies that systematically map its current landscape and evolution. This study seeks to explore emerging research trends, highlight thematic focus areas, and analyze patterns of collaboration within the field of AI-assisted pulmonary nodule research over the past 20 years.

METHODS

A literature search was conducted in the Web of Science Core Collection to collect relevant studies published from 2005 to 2024 concerning the application of AI in pulmonary nodules. Bibliometric analysis was carried out using tools such as CiteSpace, VOSviewer, and the Online Analysis Platform of Literature Metrology to examine contributions from countries, institutions, authors, journals, keywords, and references.

RESULTS

A total of 1,657 relevant publications were retrieved, reflecting a consistent upward trend in research output over the past two decades, with a marked acceleration observed after 2014. The leading contributors in terms of publication volume were China, the United States, and India. Shanghai Jiao Tong University stood out as the most prolific research institution. Analysis of keyword co-occurrence revealed several prominent thematic clusters, notably centered around Deep Convolutional Neural Network models, major diameter, lung nodule detection, false-positive reduction, cancer diagnosis, quantitative-semantic models, double reading, and clinical utility studies.

CONCLUSIONS

This bibliometric study offers a thorough assessment of the scholarly landscape concerning AI applications in pulmonary nodule research, underscoring major developments and key contributors. The insights gained may serve as a strategic reference for researchers in the medical and AI fields, facilitating informed future directions. Notably, the intersection of AI and pulmonary nodule research is concentrated in the following areas: 1. Application of AI in pulmonary nodule detection and classification; 2. AI in malignancy risk prediction and growth modeling; 3.AI-driven development of drug efficacy evaluation metrics may be a future direction for pulmonary nodule treatment research.