Computer-aided detection (CAD) system for breast MRI in assessment of local tumor extent, nodal status, and multifocality of invasive breast cancers: Preliminary study

Sung Eun Song, Bo Kyoung Seo, Kyu Ran Cho, Ok Hee Woo, Gil Soo Son, Chulhan Kim, Sung Bum Cho, Soon Sun Kwon

Research output: Contribution to journalArticlepeer-review

20 Citations (Scopus)

Abstract

Background: We aimed to investigate the efficacy of computer-aided detection (CAD) for MRI in the assessment of tumor extent, lymph node status, and multifocality in invasive breast cancers in comparison with other breast imaging modalities. Methods: Two radiologists measured the maximum tumor size, as well as, analyzed lymph node status and multifocality in 86 patients with invasive breast cancers using mammography, ultrasound, CT, MRI with and without CAD, and 18-fludeoxyglucose positron emission tomography (FDG-PET). The assessed data were compared with pathology. Results: For tumor extent, there were no significant differences between pathological size and measured size using mammography, ultrasound, CT, or MRI with and without CAD (P∈>∈0.05). For evaluation of lymph node status, ultrasound had the best kappa coefficients (0.522) for agreement between imaging and pathology, and diagnostic performance with 92.1% specificity and 90.0% positive predictive value. For multifocality, MRI with CAD had the highest area under the receiver operating characteristic curve (AUC∈=∈0.888). Conclusions: CAD for MRI is feasible to assess tumor extent and multifocality in invasive breast cancer patients. However, CAD is not effective in evaluation of nodal status.

Original languageEnglish
Article number1
JournalCancer Imaging
Volume15
Issue number1
DOIs
Publication statusPublished - 2015 Feb 8

Keywords

  • Breast
  • Breast neoplasm
  • Computer-assisted diagnosis
  • Magnetic resonance imaging
  • Tumor staging

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Oncology
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Computer-aided detection (CAD) system for breast MRI in assessment of local tumor extent, nodal status, and multifocality of invasive breast cancers: Preliminary study'. Together they form a unique fingerprint.

Cite this