Preoperative kidney tumor risk estimation with AI: From logistic regression to transformer.

Journal: PloS one
Published Date:

Abstract

We consider the problem of renal mass risk classification to support doctors in adjuvant treatment decisions following nephrectomy. Recommendation of adjuvant therapy based on the mass appearance poses two major challenges: first, morphologic patterns may sometimes overlap across subtypes of varying risks. Second, interobserver variability is large. These complexities encourage the use of computational models as accurate noninvasive tools to find relevant relationships between individual perioperative renal mass characteristics and patient risk. In addition, recent evidence highlights the importance of clinical context as a promising direction to inform treatment decisions post-nephrectomy. In this work, we aim to identify relevant clinical markers that can be predictive of renal cancer prognosis. As a starting point, we perform a clinical feature ablation study by training a logistic regression baseline model to predict renal cancer patients' eligibility for adjuvant therapy. The training dataset consisted of medical records of 300 individuals with renal tumors who underwent partial or radical nephrectomy between 2011 and 2020. In addition, we evaluate the same task using a transformer-based model pretrained on a much larger dataset of over 300,000 clinical records of individuals from the UK Biobank. Our findings demonstrate the pretrained model's efficacy in knowledge transfer across different populations, with radiographic data from preoperative cross-sectional imaging playing an important role in informing renal risk and treatment decisions.

Authors

  • Vesna Barros
    From the AI for Accelerated Healthcare & Life Sciences Discovery, IBM R&D Laboratories, University of Haifa Campus, Mount Carmel, Haifa 3498825, Israel (V.B., T.T., E.B., E.H., M.R.Z.); The Hebrew University of Jerusalem, Ein Kerem Campus, Jerusalem, Israel (V.B., M.R.Z.); IBM Watson Health, Cambridge, Mass (D.G.); RadPartners, Jefferson Radiology, East Hartford, Conn (D.G.); Department of Imaging, Assuta Medical Center, Tel Aviv, Israel (M.G.); and Ben-Gurion University Medical School, Be'er Sheva, Israel (M.G.).
  • Nour Abdallah
    University of Connecticut, Hartford, CT, USA.
  • Michal Ozery-Flato
    IBM Research-Haifa, Haifa, Israel.
  • Avihu Dekel
    IBM Research Israel, Haifa, Israel.
  • Moshiko Raboh
    IBM Research Israel, Haifa, Israel.
  • Nicholas Heller
    Department of Computer Science and Engineering, University of Minnesota - Twin Cities, Minneapolis, Minnesota.
  • Simona Rabinovici-Cohen
    IBM Research Haifa, Haifa University Campus, Mount Carmel, Haifa, Israel.
  • Alex Golts
    IBM Research Israel, Haifa, Israel.
  • Amilcare Gentili
    San Diego VA Health Care System, San Diego, CA, USA.
  • Daniel Lang
    Institute of Radiation Medicine, Helmholtz Zentrum München, Ingolstädter Landstraße 1, 85764, Neuherberg, Germany.
  • Suman Chaudhary
    College of Information and Computer, Taiyuan University of Technology, Taiyuan, China.
  • Varsha Satish
    Indian Institute of Technology Bombay, Bombay, India.
  • Resha Tejpaul
    Department of Computer Science and Engineering, University of Minnesota, Minneapolis, Minnesota, USA.
  • Ivan Eggel
    University of Applied Sciences Western Switzerland (HES-SO), Sierre, Switzerland.
  • Itai Guez
    IBM Research Israel, Haifa, Israel.
  • Ella Barkan
    From the Department of Healthcare Informatics, IBM Research, IBM R&D Labs, University of Haifa Campus, Mount Carmel, Haifa 3498825, Israel (A.A.B., M.C., Y.S., A.S., A.H., R.M., E.B., S.N., E.K., Y.G., M.R.Z.); MaccabiTech, MKM, Maccabi Healthcare Services, Tel Aviv, Israel (E.H., G.K., V.S.); and Department of Imaging, Assuta Medical Centers, Tel Aviv, Israel (M.G.).
  • Henning Muller
  • Efrat Hexter
    IBM Research-Haifa, Haifa, Israel.
  • Michal Rosen-Zvi
    IBM Research-Haifa, Haifa, Israel.
  • Christopher Weight
    Department of Urology, University of Minnesota, Minneapolis, Minnesota, USA.