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

Ng et al. BMC Medical Informatics and Decision Making

(2025) 25:236

Page 8 of 24

Key Findings Novel Features

The authors recommended using templates and macros to insert standard examination findings, as this can reduce transcription time and the likelihood of

transcription errors; this could also ensure that only clinically relevant portions of the exam are included, making documenta tion more efficient and accurate​.

Charts dictated using the voice recog nition program were considerably less costly than the manually transcribed

charts; computer voice recognition is nearly as accurate as traditional transcription, has a much shorter

turnaround time and is less expensive than traditional transcription. Recom mended its use as a tool for physician charting in the ED.

AI Transcription

Proficiency (paper

specific outcomes)

NR Difference between

voice recognition and

transcription (95% CI) 1.2 (0.8–1.5) 1.3 (0.67–1.88)

0.12 (–0.34–0.58)

35.95 (24.59–47.31) 40.4 (34.4–46.39)

Metric (F1 score,

Precision, Recall, WER)

Standard Care - Accuracy (%)

- Average no.

errors/chart

- Average dicta

tion and correc

tion time (min)

- Average turn

around time for

receipt of a com

pleted document (min)

- Throughput words/min

Comparator Type Subcategories Performance

Study Reference Zick et Table 2 Key study findings for studies reviewed Standard al., 2001 [9] Voice recogni tion chart was used as a basis for the tradi tional transcrip tion chart