Cross-site Validation of AI Segmentation and Harmonization in Breast MRI.

This work aims to perform a cross-site validation of automated segmentation for breast cancers in MRI and to compare the performance to radiologists. A three-dimensional (3D) U-Net was trained to segment cancers in dynamic contrast-enhanced axial MRIs using a large dataset from Site 1 (n = 15,266; 4...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Imaging Informatics in Medicine Vol. 38; no. 3; pp. 1642 - 1653
Autores principales: Huang, Yu, Leotta, Nicholas J., Hirsch, Lukas, Gullo, Roberto Lo, Hughes, Mary, Reiner, Jeffrey, Saphier, Nicole B., Myers, Kelly S., Panigrahi, Babita, Ambinder, Emily, Di Carlo, Philip, Grimm, Lars J., Lowell, Dorothy, Yoon, Sora, Ghate, Sujata V., Parra, Lucas C., Sutton, Elizabeth J.
Formato: diagnostic images research tables/charts Journal Article
Publicado: Springer Nature Jun2025
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=185280511&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 185280511
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        29482925
        NR3A
      jtl: Journal of Imaging Informatics in Medicine
      issn: 29482925
      maglogo: N
    pubinfo:
      dt: Jun2025
      vid: 38
      iid: 3
      pid: 237
      pub: Springer Nature
      place: New York, New York
    artinfo:
      ui:
        185280511
        185280511
        189894284
        185280511
        10.1007/s10278-024-01266-9
        185280511
      ppf: 1642
      ppct: 11
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
      tig:
        atl: Cross-site Validation of AI Segmentation and Harmonization in Breast MRI.
      aug:
        au:
          Huang, Yu
          Leotta, Nicholas J.
          Hirsch, Lukas
          Gullo, Roberto Lo
          Hughes, Mary
          Reiner, Jeffrey
          Saphier, Nicole B.
          Myers, Kelly S.
          Panigrahi, Babita
          Ambinder, Emily
          Di Carlo, Philip
          Grimm, Lars J.
          Lowell, Dorothy
          Yoon, Sora
          Ghate, Sujata V.
          Parra, Lucas C.
          Sutton, Elizabeth J.
        affil: https://ror.org/00wmhkr98 Department of Biomedical Engineering, The City College of the City University of New York, 160 Convent Ave, 10031, New York, NY, USA
      sug:
        subj:
          Breast Neoplasms Diagnosis
          Magnetic Resonance Imaging
          Automation
          Artificial Intelligence
          Validity Evaluation
          Radiologists
          Deep Learning
          Image Enhancement
          Image Interpretation, Computer Assisted
          Human
          Female
          United States
          Retrospective Design
          Comparative Studies
          Nonparametric Statistics
          Descriptive Statistics
          Imaging, Three-Dimensional
          Funding Source
          Female
      ab: This work aims to perform a cross-site validation of automated segmentation for breast cancers in MRI and to compare the performance to radiologists. A three-dimensional (3D) U-Net was trained to segment cancers in dynamic contrast-enhanced axial MRIs using a large dataset from Site 1 (n = 15,266; 449 malignant and 14,817 benign). Performance was validated on site-specific test data from this and two additional sites, and common publicly available testing data. Four radiologists from each of the three clinical sites provided two-dimensional (2D) segmentations as ground truth. Segmentation performance did not differ between the network and radiologists on the test data from Sites 1 and 2 or the common public data (median Dice score Site 1, network 0.86 vs. radiologist 0.85, n = 114; Site 2, 0.91 vs. 0.91, n = 50; common: 0.93 vs. 0.90). For Site 3, an affine input layer was fine-tuned using segmentation labels, resulting in comparable performance between the network and radiologist (0.88 vs. 0.89, n = 42). Radiologist performance differed on the common test data, and the network numerically outperformed 11 of the 12 radiologists (median Dice: 0.85–0.94, n = 20). In conclusion, a deep network with a novel supervised harmonization technique matches radiologists' performance in MRI tumor segmentation across clinical sites. We make code and weights publicly available to promote reproducible AI in radiology.
      pubtype: Academic Journal
      doctype:
        diagnostic images
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N