Deep learning accurately classifies elbow joint effusion in adult and pediatric radiographs.

Journal: Scientific reports
Published Date:

Abstract

Joint effusion due to elbow fractures are common among adults and children. Radiography is the most commonly used imaging procedure to diagnose elbow injuries. The purpose of the study was to investigate the diagnostic accuracy of deep convolutional neural network algorithms in joint effusion classification in pediatric and adult elbow radiographs. This retrospective study consisted of a total of 4423 radiographs in a 3-year period from 2017 to 2020. Data was randomly separated into training (n = 2672), validation (n = 892) and test set (n = 859). Two models using VGG16 as the base architecture were trained with either only lateral projection or with four projections (AP, LAT and Obliques). Three radiologists evaluated joint effusion separately on the test set. Accuracy, precision, recall, specificity, F1 measure, Cohen's kappa, and two-sided 95% confidence intervals were calculated. Mean patient age was 34.4 years (1-98) and 47% were male patients. Trained deep learning framework showed an AUC of 0.951 (95% CI 0.946-0.955) and 0.906 (95% CI 0.89-0.91) for the lateral and four projection elbow joint images in the test set, respectively. Adult and pediatric patient groups separately showed an AUC of 0.966 and 0.924, respectively. Radiologists showed an average accuracy, sensitivity, specificity, precision, F1 score, and AUC of 92.8%, 91.7%, 93.6%, 91.07%, 91.4%, and 92.6%. There were no statistically significant differences between AUC's of the deep learning model and the radiologists (p value > 0.05). The model on the lateral dataset resulted in higher AUC compared to the model with four projection datasets. Using deep learning it is possible to achieve expert level diagnostic accuracy in elbow joint effusion classification in pediatric and adult radiographs. Deep learning used in this study can classify joint effusion in radiographs and can be used in image interpretation as an aid for radiologists.

Authors

  • Jarno T Huhtanen
    Faculty of Health and Well-Being, Turku University of Applied Sciences, Turku, Finland. jarno.huhtanen@turkuamk.fi.
  • Mikko Nyman
    Department of Radiology, University of Turku and Turku University Hospital, Turku, Finland.
  • Dorin Doncenco
    Faculty of Engineering and Business, Turku University of Applied Sciences, Turku, Finland.
  • Maral Hamedian
    Faculty of Engineering and Business, Turku University of Applied Sciences, Turku, Finland.
  • Davis Kawalya
    Faculty of Engineering and Business, Turku University of Applied Sciences, Turku, Finland.
  • Leena Salminen
    Department of Nursing Science, University of Turku and Director of Nursing (Part-Time) Turku University Hospital, Turku, Finland.
  • Roberto Blanco Sequeiros
    Department of Radiology, University of Turku and Turku University Hospital, Turku, Finland.
  • Seppo K Koskinen
    Department of Diagnostic Radiology, Karolinska University Hospital, 171 76, Stockholm, Sweden.
  • Tomi K Pudas
    Terveystalo Inc, Jaakonkatu 3, Helsinki, Finland.
  • Sami Kajander
    Department of Radiology, University of Turku, Turku, Finland.
  • Pekka Niemi
    Department of Radiology, University of Turku, Turku, Finland.
  • Jussi Hirvonen
    Department of Radiology, Turku University Hospital & University of Turku, Kiinamyllynkatu 4-8, 20521 Turku, Finland.
  • Hannu J Aronen
    Department of Diagnostic Radiology, University of Turku, Turku, Finland.
  • Mojtaba Jafaritadi
    Faculty of Engineering and Business, Turku University of Applied Sciences, Turku, Finland.