Dual-Stage AI Model for Enhanced CT Imaging: Precision Segmentation of Kidney and Tumors.

Journal: Tomography (Ann Arbor, Mich.)
PMID: 39852683

Abstract

OBJECTIVES: Accurate kidney and tumor segmentation of computed tomography (CT) scans is vital for diagnosis and treatment, but manual methods are time-consuming and inconsistent, highlighting the value of AI automation. This study develops a fully automated AI model using vision transformers (ViTs) and convolutional neural networks (CNNs) to detect and segment kidneys and kidney tumors in Contrast-Enhanced (CECT) scans, with a focus on improving sensitivity for small, indistinct tumors.

Authors

  • Nalan Karunanayake
    Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Lin Lu
    School of Economics and Management, Guangxi Normal University, Guilin, China.
  • Hao Yang
    College of Agricultural Science and Engineering, Hohai University, Nanjing 210098, China.
  • Pengfei Geng
    Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Oguz Akin
    Memorial Sloan Kettering Cancer Center, New York, NY, USA.
  • Helena Furberg
    Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
  • Lawrence H Schwartz
    Department of Radiology, Columbia University College of Physicians and Surgeons, New York, NY, USA.
  • Binsheng Zhao
    Department of Radiology, Columbia University Medical Center, 630 West 168th Street, New York, NY 10032.

Keywords

External Resources

Popular Topics
Recent Journals