xRead - Incorporating Artificial Intelligence into Clinical Practice (March 2026)

FIGURE 1 | Summary flowchart of the outcomes of the identification and screening process. Flow chart format has been adapted from the PRISMA 2020 flow diagram for new systematic reviews [11]. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

majority of studies were not classifiable as per Oxford level of evidence ( n = 134).

[60, 61, 88, 91, 107, 109, 115, 124, 144, 155] and Gemini (Google DeepMind, London, UK) [28, 36, 49, 52, 84, 97, 146, 148, 158, 168] were each included in 10 studies, Copilot (Microsoft, Redmond, WA, USA) in 5 [82, 97, 108, 146, 158], and Claude (Anthropic, San Francisco, CA, USA) [47, 49] and Llama (Meta, Menlo Park, CA, USA) in 2 [90, 118]. 3.2 | Level of Evidence Level of evidence was found as follows: Level 1 ( n = 1), Level 2 ( n =2), Level 3 ( n =2), Level 4 ( n =25), and Level 5 ( n = 1). The

3.3 | Clinical Uses 3.3.1 | Patient Education and Monitoring

There were 42 studies investigating NLP tools for pa tient education [31, 34, 35, 39, 40, 42, 44, 51, 56, 62–64, 70, 76, 77, 94, 95, 97–101, 104, 106, 111, 113, 122, 123, 126, 128, 146, 153, 157, 159–161, 165, 168, 169, 174, 176, 178].

3051