Head CT deep learning model is highly accurate for early infarct estimation.

Journal: Scientific reports
PMID:

Abstract

Non-contrast head CT (NCCT) is extremely insensitive for early (< 3-6 h) acute infarct identification. We developed a deep learning model that detects and delineates suspected early acute infarcts on NCCT, using diffusion MRI as ground truth (3566 NCCT/MRI training patient pairs). The model substantially outperformed 3 expert neuroradiologists on a test set of 150 CT scans of patients who were potential candidates for thrombectomy (60 stroke-negative, 90 stroke-positive middle cerebral artery territory only infarcts), with sensitivity 96% (specificity 72%) for the model versus 61-66% (specificity 90-92%) for the experts; model infarct volume estimates also strongly correlated with those of diffusion MRI (r > 0.98). When this 150 CT test set was expanded to include a total of 364 CT scans with a more heterogeneous distribution of infarct locations (94 stroke-negative, 270 stroke-positive mixed territory infarcts), model sensitivity was 97%, specificity 99%, for detection of infarcts larger than the 70 mL volume threshold used for patient selection in several major randomized controlled trials of thrombectomy treatment.

Authors

  • Romane Gauriau
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Bernardo C Bizzo
    Department of Radiology, Massachusetts General Hospital, Boston, MA, United States; Harvard Medical School, Boston, MA, United States; Mass General Brigham Data Science Office (DSO), Boston, MA, United States.
  • Donnella S Comeau
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • James M Hillis
    Digital Clinical Research Organization, Data Science Office, Mass General Brigham, Boston, Massachusetts.
  • Christopher P Bridge
    Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA; Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, USA.
  • John K Chin
    From the Department of Radiology, Brigham and Women's Hospital (BWH), Harvard Medical School, 75 Francis St, Boston, MA 02115 (C.R.W., D.I.G., K.P.A., B.D., W.W.M-S.); Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, Mass (J.P., C.P.B., J. Kalpathy-Cramer); Health Sciences and Technology Department, Massachusetts Institute of Technology, Cambridge, Mass (J.P.); Department of Radiology, Massachusetts General Hospital (MGH), Harvard Medical School, Boston, Mass (B.C.B., K.D.); and MGH & BWH Center for Clinical Data Science, Boston, Mass (B.C.B., C.P.B., K.P.A., J. K. Chin, K.D., J. Kalpathy-Cramer).
  • Jayashri Pawar
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Ali Pourvaziri
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Ivana Sesic
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Elshaimaa Sharaf
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Jinjin Cao
    Department of Radiology, Massachusetts General Hospital, White 270, 55 Fruit Street, Boston, MA, 02114, USA.
  • Flavia T C Noro
    Data Science Office, Mass General Brigham, 100 Cambridge St, Suite 1303, Boston, MA, 02114, USA.
  • Walter F Wiggins
    Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass (W.F.W., M.T.C., K.M., S.A.G., E.G., M.H.R., G.C.G., K.P.A.); and MGH & BWH Center for Clinical Data Science, Boston, Mass (W.F.W., M.T.C., K.M., K.P.A.).
  • M Travis Caton
    Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass (W.F.W., M.T.C., K.M., S.A.G., E.G., M.H.R., G.C.G., K.P.A.); and MGH & BWH Center for Clinical Data Science, Boston, Mass (W.F.W., M.T.C., K.M., K.P.A.).
  • Felipe Kitamura
    Diagnosticos da America SA (Dasa), Barueri, SP, Brazil.
  • Keith J Dreyer
    Department of Radiology, Massachusetts General Hospital, Boston, MA, United States; Harvard Medical School, Boston, MA, United States; Mass General Brigham Data Science Office (DSO), Boston, MA, United States.
  • John F Kalafut
    GE Healthcare, Chicago, IL, USA.
  • Katherine P Andriole
    Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass (W.F.W., M.T.C., K.M., S.A.G., E.G., M.H.R., G.C.G., K.P.A.); and MGH & BWH Center for Clinical Data Science, Boston, Mass (W.F.W., M.T.C., K.M., K.P.A.).
  • Stuart R Pomerantz
    Department of Radiology, Massachusetts General Hospital, Boston, MA, USA.
  • Ramon G Gonzalez
    Department of Radiology, Massachusetts General Hospital, Boston, MA, USA.
  • Michael H Lev
    Department of Radiology, Massachusetts General Hospital, Boston, MA, USA.