Objectives Artificial intelligence (AI) is reshaping healthcare by improving diagnosis and treatment planning, increasing operational efficiency, and streamlining administrative workflows. This paper integrates findings from an extensive PubMed search (2015–2025) with bibliometric analysis using RStudio and VOSviewer to investigate the comparative applications of AI methods in healthcare, collaborative networks, and emerging trends.
Methods A total of 1,243 records were identified through the PubMed search, and after removal of 143 duplicates, 1,100 records were screened. Following full-text assessment and exclusion of ineligible studies, 986 articles were included in the final bibliometric analysis.
Results The main research areas included robotic-assisted surgery, predictive analytics, diagnostic imaging, and precision medicine, with particular emphasis on the prevalence of machine learning and deep learning in imaging and the increasing application of natural language processing to unstructured medical information.
Conclusion The review emphasizes the need for greater budgetary allocation to scalable and pragmatic AI technologies and for interdisciplinary cooperation among researchers, industry, and healthcare providers. Despite this growth, challenges such as algorithmic bias, data integration, and ethical concerns persist. The paper also highlights the importance of equitable collaboration, accountable AI, and multinational partnerships in ensuring that AI can be used ethically and efficiently in healthcare over the long term to improve patient care and biomedical innovation. It does so by mapping international and regional trends, identifying the most influential authors, institutions, and funding sources, and evaluating methodological approaches.
<sec>
<title>Objectives</title>
<p>This study was designed to analyze biomarker-related publications and patent trends which biobanks could consider in planning biosample collections for biomarker research.</p></sec>
<sec>
<title>Methods</title>
<p>Publications and patents containing the term “biomarker” in the title published between 1998 to 2017 were retrieved using Scopus database and Google Patents search engine.</p></sec>
<sec>
<title>Results</title>
<p>Over the last 20 years there has been a steady increase in biomarker-related publications and patents; however this has slowed for patents over the last few years. Publications in 2017 that contained blood, serum, and plasma search terms in the abstract accounted for 50%, and serum as a search term in the title and abstract was more numerous than those containing blood, plasma, tissue, or urine. Blood-related patents were the most common patent in the last 10 years, and accounted for 110 patents in 2017. Biomarker-related publications since 2010 containing RNA and protein search terms in the title and abstract, were more numerous than those containing DNA and metabolite search terms. More than 27% of biomarker-related publications in 2017 and 21% of biomarker-related patents were associated with cancer.</p></sec>
<sec>
<title>Conclusion</title>
<p>The results of this study will help biobanks establish a biosample collection strategy for clinical application.</p></sec>
Citations
Citations to this article as recorded by
Efficient sampling strategies for assessing the prognostic value of a new biomarker at a temporal endpoint Hani Samawi, Haresh Rochani Research in Statistics.2025;[Epub] CrossRef
Barriers and opportunities in biobank utilization: insights from a 3-year repeated cross-sectional survey of the Female Breast and Genital Disease with Microbiome Biobank Network (FDMNet) in South Korea Sangzin Ahn, Hwa Jin Cho, Mi-Seon Kang, An Na Seo, Lucia Kim, Kyung Un Choi, Mee Sook Roh, Eun-Young Kim Journal of Yeungnam Medical Science.2025; 42: 46. CrossRef
Nanodiamond Solid-Phase Extraction and Triton X-114 Cloud Point Separation for Robust Fractionation and Shotgun Proteomics Analysis of the Human Serum Proteome Hai Lam Khuong, Chein-Hung Chen, Jung-Lee Lin, Trong-Nghia Le, Thi Hue Pham, Thi Bich Thao Le, Xuan Canh Nguyen, Van Chi Phan, Hoang Ha Chu, Wesley Wei-Wen Hsiao, Thi Minh Phuong Nguyen, Dinh Minh Pham Journal of Proteome Research.2022; 21(1): 67. CrossRef
Sampling strategies to evaluate the prognostic value of a new biomarker on a time-to-event end-point Francesca Graziano, Maria Grazia Valsecchi, Paola Rebora BMC Medical Research Methodology.2021;[Epub] CrossRef
MicroRNA-181b Serves as a Circulating Biomarker and Regulates Inflammation in Heart Failure Hongxiao Yang, Lina Shan, Yunan Gao, Lin Li, Guifen Xu, Bin Wang, Xiaoxue Yin, Chengfang Gao, Jiaren Liu, Wei Yang, Gaetano Isola Disease Markers.2021; 2021: 1. CrossRef
Enhancing Cooperation Between Academic Biobanks and Biomedical Industry: Better Mutual Understanding and New Collaborative Models Are Needed Daniel Simeon-Dubach, Michael H. Roehrl, Paul Hofman, Pascal Puchois Biopreservation and Biobanking.2020; 18(2): 144. CrossRef
<sec><title>Objectives</title><p>Cancer has attracted worldwide attention. The incidence and prevalence are increasing, and it is the main cause of death. The purpose of this study was to identify the characteristics of hospitalized cancer patients.</p></sec><sec><title>Methods</title><p>This study is a secondary data study using the Korean National Hospital Discharge In-depth Injury Survey Data conducted annually by the Korea Centers for Disease Control and Prevention. Using these data, we extracted inpatients who principal diagnosis is cancer for nine years from 2005 to 2013.</p></sec><sec><title>Results</title><p>According to the analysis, the annual trend of cancer inpatients is steadily increasing. In 2025, it is expected to increase to about 670,000 inpatients. A cancer diagnosis created a change in medical utilization depending on the characteristics of patients and hospital. Men are more at risk of cancer than women. The number of hospital beds and hospital days were inversely proportional to cancer inpatients. There was also a difference in the equity of medical utilization by region. Other cancer management policies should be based on sex.</p></sec><sec><title>Conclusion</title><p>Populations between the ages of 45 and 64 years should be a priority in cancer policy. Because of the long-term hospitalization of patients with death as the outcome, a terminal cancer patient care facility is needed. These conclusions can provide a basis for various health policies.</p></sec>
Citations
Citations to this article as recorded by
Health-related quality of life by veterans RAND 12 and healthcare resource utilization in cancer patients with sleep disorders: insights from the Medical Expenditure Panel Survey Yuan Kao, Ying-Jia Lin, Shih-Feng Weng, Jhi-Joung Wang, I-Chen Lee, Chien-Cheng Huang Supportive Care in Cancer.2024;[Epub] CrossRef
First
Prev
