Effect of Training Data Volume on Performance of Convolutional Neural Network Pneumothorax Classifiers.

Large datasets with high-quality labels required to train deep neural networks are challenging to obtain in the radiology domain. This work investigates the effect of training dataset size on the performance of deep learning classifiers, focusing on chest radiograph pneumothorax detection as a proxy...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Digital Imaging Vol. 35; no. 4; pp. 881 - 893
Autores principales: Thian, Yee Liang, Ng, Dian Wen, Hallinan, James Thomas Patrick Decourcy, Jagmohan, Pooja, Sia, Soon Yiew, Mohamed, Jalila Sayed Adnan, Quek, Swee Tian, Feng, Mengling
Formato: diagnostic images research tables/charts Journal Article
Publicado: Springer Nature Aug2022
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=159195613&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 159195613
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        08971889
        DOQ
      jtl: Journal of Digital Imaging
      issn: 08971889
      maglogo: N
    pubinfo:
      dt: Aug2022
      vid: 35
      iid: 4
      pid: 237
      pub: Springer Nature
      place: New York, New York
    artinfo:
      ui:
        159195613
        155542238
        159195613
        159195613
        10.1007/s10278-022-00594-y
        159195613
      ppf: 881
      ppct: 12
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
      tig:
        atl: Effect of Training Data Volume on Performance of Convolutional Neural Network Pneumothorax Classifiers.
      aug:
        au:
          Thian, Yee Liang
          Ng, Dian Wen
          Hallinan, James Thomas Patrick Decourcy
          Jagmohan, Pooja
          Sia, Soon Yiew
          Mohamed, Jalila Sayed Adnan
          Quek, Swee Tian
          Feng, Mengling
        affil: Department of Diagnostic Imaging, National University Hospital, 5 Lower Kent Ridge Rd, 119074, Queenstown, Singapore
      sug:
        subj:
          Pneumothorax
          Neural Networks (Computer)
          Deep Learning
          Digital Imaging
          Image Processing, Computer Assisted Methods
          Human
          Descriptive Statistics
          Data Analysis Software
          Learning
      ab: Large datasets with high-quality labels required to train deep neural networks are challenging to obtain in the radiology domain. This work investigates the effect of training dataset size on the performance of deep learning classifiers, focusing on chest radiograph pneumothorax detection as a proxy visual task in the radiology domain. Two open-source datasets (ChestX-ray14 and CheXpert) comprising 291,454 images were merged and convolutional neural networks trained with stepwise increase in training dataset sizes. Model iterations at each dataset volume were evaluated on an external test set of 525 emergency department chest radiographs. Learning curve analysis was performed to fit the observed AUCs for all models generated. For all three network architectures tested, model AUCs and accuracy increased rapidly from 2 × 103 to 20 × 103 training samples, with more gradual increase until the maximum training dataset size of 291 × 103 images. AUCs for models trained with the maximum tested dataset size of 291 × 103 images were significantly higher than models trained with 20 × 103 images: ResNet-50: AUC20k = 0.86, AUC291k = 0.95, p < 0.001; DenseNet-121 AUC20k = 0.85, AUC291k = 0.93, p < 0.001; EfficientNet AUC20k = 0.92, AUC 291 k = 0.98, p < 0.001. Our study established learning curves describing the relationship between dataset training size and model performance of deep learning convolutional neural networks applied to a typical radiology binary classification task. These curves suggest a point of diminishing performance returns for increasing training data volumes, which algorithm developers should consider given the high costs of obtaining and labelling radiology data.
      pubtype: Academic Journal
      doctype:
        diagnostic images
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N