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 Concept Videos

Per-Unit Sequence Models01:26

Per-Unit Sequence Models

433
An ideal Y-Y transformer, grounded through neutral impedances, displays per-unit sequence networks akin to those of a single-phase ideal transformer when subjected to balanced positive- or negative-sequence currents. These currents do not produce neutral currents, and their associated voltage drops.
Zero-sequence currents, which are identical in magnitude and phase, generate a neutral current, resulting in voltage drops across the neutral impedance and the low-voltage winding. If the...
433
Cis-regulatory Sequences02:02

Cis-regulatory Sequences

11.6K
Cis-regulatory sequences are short fragments of non-coding DNA that are present on the same chromosomes as the genes that they regulate. These fragments serve as binding sites for transcriptional regulators, proteins that are responsible for controlling gene transcription and differential gene expression across cell types in eukaryotes. Cis-regulatory sequences can be close to the gene of interest or thousands of bases away in the DNA sequence; however, those sequences that are further away are...
11.6K
Cis-regulatory Sequences02:02

Cis-regulatory Sequences

4.0K
4.0K
Sanger Sequencing01:57

Sanger Sequencing

773.5K
DNA sequencing is a fundamental technique that is routinely used in the biological sciences. This method can be applied to a range of questions at different scales - from the sequencing of a cloned DNA fragment or the study of a mutation in a gene up to whole-genome sequencing. However, despite the widespread use of sequencing today, it was not until 1977 that Fredrick Sanger and his collaborators developed the chain-termination method to decode DNA sequences. It relies on the separation of a...
773.5K
Detection of Gross Error: The Q Test01:00

Detection of Gross Error: The Q Test

6.9K
When one or more data points appear far from the rest of the data, there is a need to determine whether they are outliers and whether they should be eliminated from the data set to ensure an accurate representation of the measured value. In many cases, outliers arise from gross errors (or human errors) and do not accurately reflect the underlying phenomenon. In some cases, however, these apparent outliers reflect true phenomenological differences. In these cases, we can use statistical methods...
6.9K
Next-generation Sequencing03:00

Next-generation Sequencing

97.9K
The first human genome sequencing project cost $2.7 billion and was declared complete in 2003, after 15 years of international cooperation and collaboration between several research teams and funding agencies. Today, with the advent of next-generation sequencing technologies, the cost and time of sequencing a human genome have dropped over 100 fold.
Next-Generation Sequencing Methods
Although all next-generation methods use different technologies, they all share a set of standard features....
97.9K

You might also read

Related Articles

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

Sort by
Same author

Vein of Marshall Alcohol Ablation Facilitated by a 10-French Bidirectional Catheter Supported by Glide Advantage Wire.

Pacing and clinical electrophysiology : PACE·2026
Same author

Case Report: Immune checkpoint inhibitor-associated myocarditis, myositis, and myasthenia gravis overlap syndrome with flow cytometric phenotyping before and after treatment in a patient with urothelial carcinoma.

Frontiers in immunology·2026
Same author

Advanced Multimodal AI for Predicting Long-Term Functional Outcomes After Ischemic Stroke Using Only Admission Data.

Research square·2026
Same author

Right Side Accessory Pathway Mediated Cardiomyopathy Treated with Amiodarone: First Adult Case Report.

American journal of therapeutics·2026
Same author

General-purpose large language models outperform specialized clinical AI tools on medical benchmarks.

Nature medicine·2026
Same author

Spontaneous Coronary Artery Dissection in a 49-Year-Old Woman Following Sexual Activity: A Case Report.

Cureus·2026

Related Experiment Video

Updated: Jan 22, 2026

Rare Event Detection Using Error-corrected DNA and RNA Sequencing
10:36

Rare Event Detection Using Error-corrected DNA and RNA Sequencing

Published on: August 3, 2018

12.6K

Detecting insertion, substitution, and deletion errors in radiology reports using neural sequence-to-sequence models.

John Zech1, Jessica Forde2, Joseph J Titano1

  • 1Department of Radiology, Icahn School of Medicine, New York, NY, USA.

Annals of Translational Medicine
|July 19, 2019
PubMed
Summary
This summary is machine-generated.

Neural sequence-to-sequence (seq2seq) models effectively detect errors in radiology reports. Site- and modality-specific training is crucial for optimal performance in identifying real-world errors.

Keywords:
Radiologyartificial intelligencemachine learningnatural language processingneural networks (computer)

More Related Videos

Detection of Copy Number Alterations Using Single Cell Sequencing
09:45

Detection of Copy Number Alterations Using Single Cell Sequencing

Published on: February 17, 2017

12.1K
Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples
07:30

Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples

Published on: June 8, 2020

12.7K

Related Experiment Videos

Last Updated: Jan 22, 2026

Rare Event Detection Using Error-corrected DNA and RNA Sequencing
10:36

Rare Event Detection Using Error-corrected DNA and RNA Sequencing

Published on: August 3, 2018

12.6K
Detection of Copy Number Alterations Using Single Cell Sequencing
09:45

Detection of Copy Number Alterations Using Single Cell Sequencing

Published on: February 17, 2017

12.1K
Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples
07:30

Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples

Published on: June 8, 2020

12.7K

Area of Science:

  • Medical Informatics
  • Natural Language Processing
  • Radiology

Background:

  • Radiology reports commonly contain grammatical, spelling, and usage errors.
  • Automated detection of these errors is essential for improving report quality.

Purpose of the Study:

  • To develop and evaluate a neural sequence-to-sequence (seq2seq) model for automatically detecting word-level errors in radiology reports.
  • To assess the model's performance across different clinical sites and imaging modalities.

Main Methods:

  • Seq2seq models were trained on corrupted radiology report sentences (insertions, deletions, substitutions) to predict original, uncorrupted sentences.
  • Training datasets included head CT and chest radiograph reports from Mount Sinai Hospital, Mount Sinai Queens, and MIMIC-III.
  • Model performance was evaluated on same-site, same-modality test sets, with manual review of a subset of original sentences.

Main Results:

  • Seq2seq models achieved high accuracy in detecting corrupted sentences (90.3% for head CTs, 88.2% for chest radiographs) with high specificity (97.7% and 98.8%).
  • For real-world errors in uncorrupted sentences, the model demonstrated a positive predictive value (PPV) of 0.393 and a negative predictive value (NPV) of 0.986.
  • Estimated sensitivity for detecting real-world errors was 0.389, with a specificity of 0.986.

Conclusions:

  • Seq2seq models show significant potential for identifying word-level errors in radiology reports.
  • High performance necessitates training data specific to the site and imaging modality.
  • Further improvements in detecting real-world errors can be achieved by incorporating additional targeted training data.