Scholarlink

<p>Peripheral intravenous extravasation poses significant clinical challenges, demanding accurate assessment for timely intervention. This study advances extravasation management by integrating Visual Question Answering (VQA) technology within the Thammasat University eXtravasation Assessment Tool (TUXAT) framework, designed for generating structured clinical reports. The investigation evaluated two distinct VQA approaches—a single large model versus a mixture of models—assessing their capacity to produce reports detailing Findings (Skin Discoloration, Integrity, and Edema), Implications (Severity), and Treatment Plans based on established clinical guidelines. Inter-rater reliability analysis indicated moderate-to-substantial agreement for Discoloration assessment, although inherent subjectivity limited the precision of Integrity and Edema scoring. Statistical analysis confirmed the system’s sensitivity to clinical severity (Severity effect, <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(p=0.013\)</EquationSource> </InlineEquation>) and captured expected expert variability (Annotator effect, <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(p&lt;0.001\)</EquationSource> </InlineEquation>). Notably, while overall annotator assessments varied, the relative differences in their ratings across severity levels remained consistent (Severity<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(\times\)</EquationSource> </InlineEquation>Annotator interaction, <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(p=0.128\)</EquationSource> </InlineEquation>). Crucially, both VQA approaches yielded comparable performance profiles (Model effect, <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(p=0.904\)</EquationSource> </InlineEquation>), demonstrating VQA’s robustness for this application irrespective of architectural choice. These findings highlight VQA’s significant potential to automate and standardize extravasation reporting, offering a promising pathway towards improved clinical decision support.</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

Intravenous extravasation report generation using deep learning, generative artificial intelligence, and visual question answering techniques

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月11日

查看全文
  • Adirek Munthuli,
  • Pitchaporn Liao,
  • Padcha Pongcharoen,
  • Sinee Wetchawalit,
  • Panlop Chakkavittumrong,
  • Borwarnluck Thongthawee,
  • Thitiporn Pathomjaruwat,
  • Charturong Tantibundhit

摘要

Peripheral intravenous extravasation poses significant clinical challenges, demanding accurate assessment for timely intervention. This study advances extravasation management by integrating Visual Question Answering (VQA) technology within the Thammasat University eXtravasation Assessment Tool (TUXAT) framework, designed for generating structured clinical reports. The investigation evaluated two distinct VQA approaches—a single large model versus a mixture of models—assessing their capacity to produce reports detailing Findings (Skin Discoloration, Integrity, and Edema), Implications (Severity), and Treatment Plans based on established clinical guidelines. Inter-rater reliability analysis indicated moderate-to-substantial agreement for Discoloration assessment, although inherent subjectivity limited the precision of Integrity and Edema scoring. Statistical analysis confirmed the system’s sensitivity to clinical severity (Severity effect, \(p=0.013\) ) and captured expected expert variability (Annotator effect, \(p<0.001\) ). Notably, while overall annotator assessments varied, the relative differences in their ratings across severity levels remained consistent (Severity \(\times\) Annotator interaction, \(p=0.128\) ). Crucially, both VQA approaches yielded comparable performance profiles (Model effect, \(p=0.904\) ), demonstrating VQA’s robustness for this application irrespective of architectural choice. These findings highlight VQA’s significant potential to automate and standardize extravasation reporting, offering a promising pathway towards improved clinical decision support.