Garcia Gonzalez-Moral Sonia, Pennock Erin, Ewedairo Olushola, Green Elizabeth, Elgey James, Mkwashi Andrew
NIHR Innovation Observatory, Population Health Sciences Institute, Newcastle University, The Catalyst Room 3.12, 3 Science Square Newcastle Helix, Newcastle upon Tyne, NE4 5TG, United Kingdom, 44 0191 2082262.
Center for Cancer, Population Health Sciences Insitute, Newcastle University, Newcastle upon Tyne, United Kingdom.
Interact J Med Res. 2025 Sep 11;14:e70323. doi: 10.2196/70323.
Patents are an early sign of innovation, yet their role in horizon scanning for health care remains unclear.
This study investigates the role of, and methods for, patent analysis in advancing health care technology innovation in a sector that is characterized by diverse health care technologies and significant research investment. Patents are critical early indicators of innovation, supporting horizon scanning and weak signal detection. The study aimed to identify intellectual property sources, evaluate methods for patent retrieval and analysis, and outline objectives for using patent data to anticipate trends and inform health care strategies.
A rapid scoping review was conducted following Cochrane Rapid Review Methods recommendations and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, with a preregistered protocol on the Open Science Framework. Searches in Embase, IEEE Xplore, and Web of Science targeted records published 2020 onward to capture the most recent sources, methods, and tools. Three independent reviewers screened studies using Rayyan (Qatar Computing Research Institute). We included any study type published since 2020 that provided patent sources data, methods, and tools applied to the study of health care technologies. Our data extraction included bibliographic details, study characteristics, and methodological information. Risk of bias assessments were not undertaken. Narrative and tabular methods, supplemented by visual charts, were used to synthesize findings.
Our searches identified 1741 studies, of which 124 were included after title, abstract, and full-text screening, with 54% being original research, 43.5% reviews, and the remainder being conference abstracts (2.5%). Most studies (68%) relied solely on patent databases, while others searched the gray and published literature. Research objectives of the included studies were grouped into 10 themes, with trend analysis (50%) and the provision of recommendations for future research, policy, and strategy development (20%) being the most common. Our review identified up to 47 patent databases, with 27% of studies using multiple sources. Whenever time limits were reported, the mean time horizon for patent searches was 24.6 years, ranging from 1900 to 2019. Automated approaches, used in 33% (n=43) of studies, frequently used tools such as Gephi (Gephi Consortium) for network visualization. Disease mapping based on National Institute for Health and Care Excellence classification indicated that cancer (19%) and respiratory conditions (16%), particularly COVID-19, were key areas.
Patent data are valuable for identifying technological trends and informing policy and research strategies. While patents provide crucial insights into emerging technologies, inconsistent deduplication practices across studies pose the risk of data inflation, accentuating the need for transparency and rigor. Finally, this review emphasized the importance of data transformation and visualization in detecting emerging trends, with Python and R being the most commonly used programming languages for developing custom tools.
专利是创新的早期标志，但其在医疗保健领域的前瞻性扫描中的作用仍不明确。
本研究调查专利分析在推动医疗保健技术创新中的作用和方法，该领域以多样化的医疗保健技术和大量研究投资为特征。专利是创新的关键早期指标，有助于前瞻性扫描和微弱信号检测。该研究旨在识别知识产权来源，评估专利检索和分析方法，并概述利用专利数据预测趋势和为医疗保健战略提供信息的目标。
按照Cochrane快速综述方法建议和PRISMA（系统综述和Meta分析的首选报告项目）指南进行快速范围综述，并在开放科学框架上预先注册了方案。在Embase、IEEE Xplore和科学网中进行检索，目标是2020年以后发表的记录，以获取最新的来源、方法和工具。三名独立评审员使用Rayyan（卡塔尔计算研究所）筛选研究。我们纳入了自2020年以来发表的任何研究类型，这些研究提供了专利来源数据、应用于医疗保健技术研究的方法和工具。我们的数据提取包括文献细节、研究特征和方法学信息。未进行偏倚风险评估。采用叙述性和表格方法，并辅以可视化图表来综合研究结果。
我们的检索共识别出1741项研究，其中124项在标题、摘要和全文筛选后被纳入，其中54%为原创研究，43.5%为综述，其余为会议摘要（2.5%）。大多数研究（68%）仅依赖专利数据库，而其他研究则搜索灰色文献和已发表文献。纳入研究的研究目标分为10个主题，趋势分析（50%）以及为未来研究、政策和战略制定提供建议（20%）最为常见。我们的综述识别出多达47个专利数据库，27%的研究使用多个来源。每当报告时间限制时，专利检索的平均时间跨度为24.6年，范围从1900年到2019年。33%（n = 43）的研究使用自动化方法，经常使用诸如Gephi（Gephi联盟）等工具进行网络可视化。基于英国国家卫生与临床优化研究所分类的疾病图谱表明，癌症（19%）和呼吸道疾病（16%），尤其是COVID-19，是关键领域。
专利数据对于识别技术趋势以及为政策和研究战略提供信息很有价值。虽然专利为新兴技术提供了关键见解，但各研究中重复数据删除做法不一致带来了数据膨胀的风险，凸显了透明度和严谨性的必要性。最后，本综述强调了数据转换和可视化在检测新兴趋势中的重要性，Python和R是开发定制工具最常用的编程语言。
