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

Ng et al. BMC Medical Informatics and Decision Making

(2025) 25:236

Page 20 of 24

Table 3 Detailed risk of bias assessment for the various studies using QUADAS-2 tool Study Risk of bias

Applicability concerns

Patient Selection

Index Test

Reference Standard

Flow and Timing

Patient Selection

Index Test Reference Standard

Almario et al., 2015 [13] Almario et al., 2015 [14]

LOW LOW LOW HIGH LOW LOW LOW LOW LOW LOW HIGH LOW HIGH HIGH LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW

LOW LOW LOW LOW LOW LOW LOW LOW HIGH LOW LOW HIGH LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW HIGH LOW

LOW LOW LOW LOW LOW

LOW LOW LOW LOW LOW

LOW LOW LOW HIGH LOW LOW LOW LOW LOW LOW LOW HIGH LOW HIGH LOW LOW LOW LOW HIGH LOW LOW LOW LOW LOW LOW LOW LOW HIGH

LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW

Biro et al., 2025 [32]

Blackley et al., 2020 [21] Balloch et al., 2024 [4] Bundy et al., 2024 [23]

UNCLEAR

UNCLEAR

LOW UNCLEAR

Cao et al., 2024 [24]

LOW LOW

LOW LOW

LOW LOW LOW LOW

Duggan et al., 2025 [33] Goss et al., 2019 [20] Harbele et al., 2024 [25] Happe et al., 2003 [10] Hodgson et al., 2017 [16] Islam et al., 2024 [26] Issenman et al., 2004 [12] Kodish-Wachs et al., 2018 [17] Misurac et al., 2024 [28] Mohr et al., 2003 [11] Moryousef et al., 2025 [35] Owens et al., 2024 [29] Sezgin et al., 2024 [30] Suominen et al., 2015 [15] van Buchem et al., 2024 [31] Van Woensel et al., 2022 [22] Shah et al., 2025 [36] Liu et al., 2024 [27] Lybarger et al., 2018 [18] Ma et al., 2025 [34]

UNCLEAR

UNCLEAR

UNCLEAR

LOW UNCLEAR

LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW HIGH

LOW HIGH LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW

LOW LOW LOW LOW HIGH LOW LOW LOW LOW LOW HIGH LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW LOW

UNCLEAR

UNCLEAR

UNCLEAR

UNCLEAR

UNCLEAR

UNCLEAR

UNCLEAR

Zhou et al., 2018 [19] Zick et al., 2001 [9]

LOW HIGH

Fig. 2 Stacked bar chart displaying risk of bias of the studies reviewed using QUADAS-2 tool

transcription technology in healthcare and highlight cer tain common challenges to overcome in order to advance the field. A broad array of AI models and software systems emerged from the review, from older ASR-based tools, such as Dragon NaturallySpeaking Medical Suite and

Dragon Medical One [9, 12, 15, 16, 20], to more advanced products that incorporate LLMs, including DAX Copi lot and GPT-4–driven systems [31, 33, 34]. The newer studies tend to describe ambient AI scribes, which not only transcribe but also summarize and repurpose clini cal notes. Such systems may overcome some limitations