Alexander J. Towbin · Automatic Detection of Inadequate Pediatric Lateral Neck Radiographs of the Airway and Soft Tissues using Deep Learning
Follow
Article1 min read·Published Sep 2020Older

Automatic Detection of Inadequate Pediatric Lateral Neck Radiographs of the Airway and Soft Tissues using Deep Learning

pubmed.ncbi.nlm.nih.gov shows its articles on its own site.

Read the article on pubmed.ncbi.nlm.nih.gov ↗

Article · Sep 2020 · 1 min read

In brief

In brief

This study developed a ResNet-50 deep learning algorithm to automatically identify inadequate lateral airway radiographs in pediatric emergency patients, achieving 90% sensitivity and outperforming both technologists and radiologists in interrater agreement. The AI classifier demonstrated superior consistency in quality assessment compared to human raters, suggesting potential for real-time quality control in pediatric airway imaging.

Written by the GCMD Library team from the article.

Alexander J. Towbin

Purpose: To develop and validate a deep learning (DL) algorithm to identify poor-quality lateral airway radiographs. Materials And Methods: A total of 1200 lateral airway radiographs obtained in emergency department patients between January 1, 2000, and July 1, 2019, were retrospectively queried from the picture archiving and communication system. Two radiologists classified each radiograph as adequate or inadequate. Disagreements were adjudicated by a third radiologist. The radiographs were used to train and test the DL classifiers. Three technologists and three different radiologists classified the images in the test dataset, and their performance was compared with that of the DL classifiers. Results: The training set had 961 radiographs and the test set had 239. The best DL classifier (ResNet-50) achieved sensitivity, specificity, and area under the receiver operating characteristic curve of 0.90 (95% confidence interval [CI]: 0.86, 0.94), 0.82 (95% CI: 0.76, 0.90), and 0.86 (95% CI: 0.81, 0.91), respectively. Interrater agreement for technologists was fair (Fleiss κ, 0.36 [95% CI: 0.29, 0.43]), while that for radiologists was moderate (Fleiss κ, 0.59 [95% CI: 0.52, 0.66]). Cohen κ value comparing the consensus rating of ResNet-50 iterations from fivefold cross-validation, consensus technologists' rating, and consensus radiologists' rating to the ground truth were 0.76 (95% CI: 0.63, 0.89), 0.49 (95% CI: 0.37, 0.61), and 0.66 (95% CI: 0.54, 0.78), respectively. Conclusion:

Statements in this article

  1. A total of 1200 lateral airway radiographs obtained in emergency department patients between January 1, 2000, and July 1, 2019, were retrospectively queried and classified by radiologists as adequate or inadequate.

    ResearchMaterials And Methods
  2. The training set contained 961 radiographs and the test set contained 239 radiographs.

    ResearchResults
  3. The ResNet-50 deep learning classifier achieved sensitivity of 0.90 (95% CI: 0.86, 0.94), specificity of 0.82 (95% CI: 0.76, 0.90), and area under the receiver operating characteristic curve of 0.86 (95% CI: 0.81, 0.91) for detecting inadequate lateral airway radiographs.

    ResearchResults
  4. Interrater agreement for technologists in classifying radiograph quality was fair with Fleiss κ of 0.36 (95% CI: 0.29, 0.43).

    ResearchResults
  5. Interrater agreement for radiologists in classifying radiograph quality was moderate with Fleiss κ of 0.59 (95% CI: 0.52, 0.66).

    ResearchResults
  6. Cohen κ comparing ResNet-50 consensus rating to ground truth was 0.76 (95% CI: 0.63, 0.89), compared to 0.49 (95% CI: 0.37, 0.61) for technologists and 0.66 (95% CI: 0.54, 0.78) for radiologists.

    ResearchResults

Read it at the source ↗

Try
Intelligent Search· scoped to this article · not medical adviceSearch the whole library →
GCMD Library — educational content from recorded physician discussions, not medical advice.