Deep Learning Pre-training Strategy for Mammogram Image Classification: an Evaluation Study.

In this work, we assess how pre-training strategy affects deep learning performance for the task of distinguishing false-recall from malignancy and normal (benign) findings in digital mammography images. A cohort of 1303 breast cancer screening patients (4935 digital mammogram images in total) was r...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Digital Imaging Vol. 33; no. 5; pp. 1257 - 1266
Autores principales: Clancy, Kadie, Aboutalib, Sarah, Mohamed, Aly, Sumkin, Jules, Wu, Shandong
Formato: research tables/charts Journal Article
Publicado: Springer Nature Oct2020
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=146532225&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 146532225
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        08971889
        DOQ
      jtl: Journal of Digital Imaging
      issn: 08971889
      maglogo: N
    pubinfo:
      dt: Oct2020
      vid: 33
      iid: 5
      pid: 237
      pub: Springer Nature
      place: New York, New York
    artinfo:
      ui:
        146532225
        144449944
        146532225
        146532225
        10.1007/s10278-020-00369-3
        146532225
      ppf: 1257
      ppct: 9
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
      tig:
        atl: Deep Learning Pre-training Strategy for Mammogram Image Classification: an Evaluation Study.
      aug:
        au:
          Clancy, Kadie
          Aboutalib, Sarah
          Mohamed, Aly
          Sumkin, Jules
          Wu, Shandong
        affil: Department of Computer Science, University of Pittsburgh, 3240 Craft Place, 15213, Pittsburgh, PA, USA
      sug:
        subj:
          Deep Learning
          Mammography
          Breast Neoplasms Diagnosis
          Cancer Screening
          Human
          Evaluation Research
          Neural Networks (Computer)
          ROC Curve
      ab: In this work, we assess how pre-training strategy affects deep learning performance for the task of distinguishing false-recall from malignancy and normal (benign) findings in digital mammography images. A cohort of 1303 breast cancer screening patients (4935 digital mammogram images in total) was retrospectively analyzed as the target dataset for this study. We assessed six different convolutional neural network model structures utilizing four different imaging datasets (total > 1.4 million images (including ImageNet); medical images different in terms of scale, modality, organ, and source) for pre-training on six classification tasks to assess how the performance of CNN models varies based on training strategy. Representative pre-training strategies included transfer learning with medical and non-medical datasets, layer freezing, varied network structure, and multi-view input for both binary and triple-class classification of mammogram images. The area under the receiver operating characteristic curve (AUC) was used as the model performance metric. The best performing model out of all experimental settings was an AlexNet model incrementally pre-trained on ImageNet and a large Breast Density dataset. The AUC for the six classification tasks using this model ranged from 0.68 to 0.77. In the case of distinguishing recalled-benign mammograms from others, four out of five pre-training strategies tested produced significant performance differences from the baseline model. This study suggests that pre-training strategy influences significant performance differences, especially in the case of distinguishing recalled- benign from malignant and benign screening patients.
      pubtype: Academic Journal
      doctype:
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N