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

Ng et al. BMC Medical Informatics and Decision Making

(2025) 25:236

Page 15 of 24

Key Findings Novel Features

Some clinicians provided positive feed back, sharing that they could spend more time consulting and dealing with their patient’s issues rather than bearing the

burden of documentation. However, some clinicans preferred human transcription

ists who could process what was spoken, providing a summarised transcript that did not require additional time to proofread for errors and misheard words.

The study documented the use of SR in context of creating notes in the electronic health record. However, newer systems

can be directly utilised in the exam room or a mobile device to capture the clinical encounter.

Even with SR experience, participants encountered frequent SR errors, par

ticularly with medical terms, names, and abbreviations.

Dictation required extensive correction ef forts, highlighting the need for improved error detection and correction mecha

nisms to alleviate the editing burden. This improvement would support clinicians

who may be less adept at real-time editing while dictating. Although SR-produced notes were more

comprehensive, they sometimes included redundant information, suggesting a need to optimize SR systems for conciseness without losing detail.

69% of respondents used SR for 75–100% of their patients. Higher satis faction was linked to greater efficiency

and fewer errors. Odds of satisfaction increased as user efficiency increased and as the number of errors and edit ing time decreased. Future studies should examine the

influence of other factors (e.g., accent) on SR usability and accuracy.

No evidence was found that SR is significantly more efficient or accurate for creating clinical notes despite clinicians feeling that SR saves time, increases efficiency and accuracy. SR generated documents had errors that had to be manually corrected.

AI Transcription

Proficiency (paper

specific outcomes)

Efficiency and satisfaction NR Percentage of user agreement 77.1

NR Dictation mean 1.5 4.1

4 m 23s 7.7

Typing mean 2.9 33.9*

5 m 18s 6.6*

Metric (F1 score,

Precision, Recall, WER)

errors

- Corrected errors

- Documentation time

- Quality score†

Comparator Type Subcategories Performance assessment of SR accuracy, efficiency, and satisfaction

al., 2019 [20] NR Survey-based self

Blinded comparison - Uncorrected

Standard

Annotated

and analysed

recordings that

captured infor

mation about

the documenta tion process, errors and

corrections.

Table 2 (continued)

Study Reference Goss et

Blackley et al.,

2020

[21]