Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Initial evaluation of a continuous speech recognition program for radiology.

K M Kanal1, N J Hangiandreou, A M Sykes

  • 1Department of Radiology, University of Washington, Seattle 98195-7115, USA.

Journal of Digital Imaging
|April 20, 2001
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

MRgFUS ablation of a recurrent tenosynovial giant cell tumor in the foot using ExAblate 2100 system in combination with patient immobilization device.

Radiography (London, England : 1995)·2024
Same author

Generative Adversarial Networks for Brain MRI Synthesis: Impact of Training Set Size on Clinical Application.

Journal of imaging informatics in medicine·2024
Same author

Risk caused by the propagation of earthquake losses through the economy.

Nature communications·2022
Same author

Practical implementation of robust MR-thermometry during clinical MR-guided microwave ablations in the liver at 1.5 T.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2019
Same author

Does having a rotator cuff repair prior to reverse total shoulder arthroplasty influence the outcome?

The bone & joint journal·2019
Same author

Waveguide effects and implications for cardiac magnetic resonance elastography: A finite element study.

NMR in biomedicine·2018
Same journal

Bayesian Convolutional Neural Networks in Medical Imaging Classification: A Promising Solution for Deep Learning Limits in Data Scarcity Scenarios.

Journal of digital imaging·2023
Same journal

Detecting and Characterizing Inferior Vena Cava Filters on Abdominal Computed Tomography with Data-Driven Computational Frameworks.

Journal of digital imaging·2023
Same journal

DMCA-GAN: Dual Multilevel Constrained Attention GAN for MRI-Based Hippocampus Segmentation.

Journal of digital imaging·2023
Same journal

Left Ventricular Myocardial Dysfunction Evaluation in Thalassemia Patients Using Echocardiographic Radiomic Features and Machine Learning Algorithms.

Journal of digital imaging·2023
Same journal

Public Imaging Datasets of Gastrointestinal Endoscopy for Artificial Intelligence: a Review.

Journal of digital imaging·2023
Same journal

External Validation of Robust Radiomic Signature to Predict 2-Year Overall Survival in Non-Small-Cell Lung Cancer.

Journal of digital imaging·2023
See all related articles

This study found continuous speech recognition software for radiology reports is about 90% accurate. Accuracy was not affected by gender, but nonnative English speakers had higher error rates.

Area of Science:

  • Medical Informatics
  • Speech Recognition Technology

Background:

  • Continuous speech recognition (CSR) software offers potential for transcribing medical reports.
  • Evaluating CSR accuracy and its performance across different user demographics is crucial for clinical adoption.

Purpose of the Study:

  • To measure the accuracy of IBM MedSpeak/Radiology software (version 1.1).
  • To assess the software's accuracy based on speaker gender and native English speaker status.
  • To evaluate the software's suitability for routine transcription of radiology reports.

Main Methods:

  • Six speakers (3 male, 2 nonnative English speakers) dictated 72 radiology reports (neurologic and body imaging).
  • Comparison of dictated and original reports to compute error rates (overall, significant, subtle significant).

Related Experiment Videos

  • Statistical analysis (ttests) to evaluate dependence of error rates on modality, native speaker status, and gender.

    • Main Results:

    • The overall error rate was 10.3% +/- 3.3%, indicating approximately 90% accuracy.
    • No significant difference in accuracy was found between male and female speakers.
    • Significant differences in overall and significant error rates were observed between native and nonnative English speakers (P = .009 and P = .008, respectively).

    Conclusions:

    • Current practical implementation issues, not accuracy, limit widespread routine use of this speech recognition software in radiology.
    • The software shows potential for niche applications, such as in emergency room settings.
    • The described methodology provides a sensitive approach for comparing CSR product accuracy over time.