人工智慧演算法與標準篩檢方法於老年人口咽部吞嚥困難心理計量及可行性之比較研究：範疇性評論

// A Comparative Study of the Psychometrics and Feasibility of Artificial Intelligence Algorithms Versus the Standard Screening Methods for Oropharyngeal Dysphagia in Older Adults: A Scoping Review

本研究比較臨床與人工智慧（AI）篩檢口咽部吞嚥困難（OD）於老年住院患者的效能｜臨床篩檢敏感性和特異性不穩定且耗時長（每人約5.6至10分鐘），易導致漏診｜AI篩檢，尤其是AIMS-OD，準確性相當且篩檢速度快（每人約0.7秒），具完全自動化特性｜結果顯示AI篩檢為早期發現OD的可靠且可擴展方案，有助改善患者預後。

// This study reviewed the effectiveness and feasibility of clinical versus AI-based screening methods for oropharyngeal dysphagia (OD) in older hospitalized patients. Clinical methods showed variable sensitivity and specificity but were time-consuming (5.6-10 minutes per patient), contributing to underdiagnosis. AI tests, especially the AIMS-OD, demonstrated comparable accuracy with faster screening times (0.7 seconds per patient) and full automation. The findings suggest AI-based screening is a reliable, rapid, and scalable solution for early OD detection, potentially improving patient outcomes.

Published 2026年7月15日