Scholarlink

<p>Artificial intelligence (AI) is rapidly transforming surgical practice, including neurosurgery, plastic surgery, and general surgery. However, despite promising applications across surgical specialties, the scope of AI integration in peripheral nerve surgery remains unclear. This systematic review is registered and publicly available in PROSPERO (ID: 1061046) and aimed to explore, evaluate, and summarize current applications of AI in peripheral nerve surgery. Google Scholar, PubMed, Scopus, and IEEE Xplore were searched for peer-reviewed studies on AI in peripheral nerve surgery and related specialties. Eligible studies included original research using AI for imaging, diagnosis, treatment planning, or outcome prediction in this field. Non-English studies, abstracts, conference proceedings, and non-AI-related articles were excluded. Bias and quality were assessed using the Mixed Methods Appraisal Tool (MMAT 2018), A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR-2), Scale for the Assessment of Narrative Review Articles (SANRA), Joanna Briggs Institute (JBI) Critical Appraisal Checklist, and Prediction model Risk Of Bias ASsessment Tool (PROBAST). No meta-analysis was conducted, and findings were synthesized narratively with descriptive statistics where applicable. A total of 32 studies met inclusion criteria.</p><p>Thirty-two studies published between 2009 and 2025 were included in this review. Most were retrospective or prospective human studies (28%). AI models demonstrated high performance (median accuracy 0.93, median sensitivity 0.96), with AI outperforming human experts (71.4%) in four head-to-head studies. Seven studies reported favorable clinical outcomes, including enhanced diagnostic accuracy and improved recovery trajectories. Quality assessments indicated generally high methodological rigor among included studies. This review explores AI’s potential to improve diagnostic precision and streamline workflows in peripheral nerve surgery. Limitations include high heterogeneity, absence of pooled data synthesis, and potential language and publication biases. Future research should focus on multi-center, prospective outcome-based studies and standardized reporting to guide clinical integration.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Applications of artificial intelligence in peripheral neurosurgery: a systematic review

  • 期刊论文

  • 发表日期：

    2026年3月26日

查看全文
  • Destiny L. Green,
  • B. Parker Layton,
  • Justin Knapp,
  • Ammon Driggs,
  • Robert J. Spinner

摘要

Artificial intelligence (AI) is rapidly transforming surgical practice, including neurosurgery, plastic surgery, and general surgery. However, despite promising applications across surgical specialties, the scope of AI integration in peripheral nerve surgery remains unclear. This systematic review is registered and publicly available in PROSPERO (ID: 1061046) and aimed to explore, evaluate, and summarize current applications of AI in peripheral nerve surgery. Google Scholar, PubMed, Scopus, and IEEE Xplore were searched for peer-reviewed studies on AI in peripheral nerve surgery and related specialties. Eligible studies included original research using AI for imaging, diagnosis, treatment planning, or outcome prediction in this field. Non-English studies, abstracts, conference proceedings, and non-AI-related articles were excluded. Bias and quality were assessed using the Mixed Methods Appraisal Tool (MMAT 2018), A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR-2), Scale for the Assessment of Narrative Review Articles (SANRA), Joanna Briggs Institute (JBI) Critical Appraisal Checklist, and Prediction model Risk Of Bias ASsessment Tool (PROBAST). No meta-analysis was conducted, and findings were synthesized narratively with descriptive statistics where applicable. A total of 32 studies met inclusion criteria.

Thirty-two studies published between 2009 and 2025 were included in this review. Most were retrospective or prospective human studies (28%). AI models demonstrated high performance (median accuracy 0.93, median sensitivity 0.96), with AI outperforming human experts (71.4%) in four head-to-head studies. Seven studies reported favorable clinical outcomes, including enhanced diagnostic accuracy and improved recovery trajectories. Quality assessments indicated generally high methodological rigor among included studies. This review explores AI’s potential to improve diagnostic precision and streamline workflows in peripheral nerve surgery. Limitations include high heterogeneity, absence of pooled data synthesis, and potential language and publication biases. Future research should focus on multi-center, prospective outcome-based studies and standardized reporting to guide clinical integration.