PuSH - Publikationsserver des Helmholtz Zentrums München

Roque, T.* ; Papiez, B.W.* ; Kersemans, V.* ; Smart, S.* ; Allen, D.* ; Chappell, M.* ; Schnabel, J.A.*

Tumor Growth Estimation via Registration of DCE-MRI Derived Tumor Specific Descriptors.

In: (2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 26 June 2016 - 01 July 2016, Las Vegas, NV, USA). 2016. 507-515 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops)
DOI
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) provides information on changes occurring during tumor growth in the tumor micro-environment and vasculature. In the present paper, tumor voxel-wise estimates of tumor descriptors including total cell number, proliferative cell number, hypoxic cell number, necrotic cell number and oxygen level derived from DCE-MRI data are used to guide the deformable registration of subsequent time points over the tumor growth cycle, evaluating their predictive value for tumor growth. The analysis of three preclinical colon carcinoma longitudinal cases shows that using physiologically meaningful measures of tumor as guidance information can improve non-rigid registration of longitudinal tumor imaging data when compared to a stateof-the-art local correlation coefficient Demons approach. Moreover, using the determinant of the Jacobian of the estimated displacement field as an indicator of volume change allows us to observe a correlation between the tumor descriptor values and tumor growth, especially when maps of hypoxic cells and level of oxygen were used to aid registration. To the best of our knowledge, this work demonstrates for the first time the feasibility of using biologically meaningful tumor descriptors (total cell number, proliferative cell number, hypoxic cell number, necrotic cell number and oxygen level) derived from DCE-MRI to aid non-rigid registration of longitudinal tumor data as well as to estimate tumor growth.
Altmetric
Weitere Metriken?
[➜Einloggen]
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Konferenzbeitrag
Korrespondenzautor
ISSN (print) / ISBN 2160-7508
e-ISSN 2160-7516
Konferenztitel 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Konferzenzdatum 26 June 2016 - 01 July 2016
Konferenzort Las Vegas, NV, USA
Quellenangaben Band: , Heft: , Seiten: 507-515 Artikelnummer: , Supplement: ,
Nichtpatentliteratur Publikationen
Institut(e) Institute for Machine Learning in Biomed Imaging (IML)