Combination of Blood-Flow Numerical Simulation and Machine Learning for Predictive Medicine

Marie Oshima
Institute of Industrial Science
Professor      Interfaculty Initiative in Information Studies    Professor
To predict the risk of excess blood flow after surgery for arterial stenosis as well as stroke, model-driven blood-flow numerical simulation and data-driven machine learning are combined to obtain the post-surgical hemodynamic status and assess the risk of stroke rapidly with simulation-level accuracy.
Uncertainty quantification in the prediction of hyperperfusion syndrome using a surrogate model
Automatic extraction of 3D vessel geometry from medical images using deep learning and data analysis for geometry parameterization and hemodynamic prediction models

Related links

Research collaborators

・The University of Tokyo Hospital
・Shiga University of Medical Science
・Saitama Medical University
・Fujita Health University
・Japan Agency for Medical Research and Development(AMED)

Related publications

・Yuhn, C., Hoshina, K., Miyahara, K., Oshima, M., Computational simulation of flow-induced arterial remodeling of the pancreaticoduodenal arcade associated with celiac artery stenosis, Journal of Biomechanics, Vol.92, pp.146–154, 2019.
・Kobayashi, M., Hoshina, K., Nemoto, Y., Takagi, S., Shojima, M., Hayakawa, M., Yamada, S., Oshima, M., A penalized spline fitting method to optimize geometric parameters of arterial centerlines extracted from medical images, Computerized Medical Imaging and Graphics, Vol.84, pp.101746, 2020.
・Zhang, H., Fujiwara, N., Kobayashi, M., Yamada, S., Liang, F., Takagi, S., Oshima, M., Development of a numerical method for patient-specific cerebral circulation using 1D–0D simulation of the entire cardiovascular system with SPECT data, Annals of Biomedical Engineering, Vol.44, No.8, pp.2351–2363, 2016.
・Kobayashi, M., Hoshina, K., Yamamoto, S., Nemoto, Y., Akai, T., Shigematsu, K., Watanabe, T., Oshima, M., Development of an image-based modeling system to investigate evolutional geometric changes of a stent graft in an abdominal aortic aneurysm, Circulation Journal, Vol.79, No.7, pp.1534-1541, 2015.


  • SDG3 Ensure healthy lives and promote well-being for all at all ages
  • SDG9 Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation


  • Interfaculty Initiative in Information Studies
  • ex. 56205
  • Tel: +81-3-5452-6205
  • Email: olab[at]
Access Map
Kashiwa Campus
Hongo Campus
Komaba Campus