Web of Science: 17 cites, Scopus: 20 cites, Google Scholar: cites,
Automated quantification of cerebral edema following hemispheric infarction : Application of a machine-learning algorithm to evaluate CSF shifts on serial head CTs
Chen, Yasheng (Department of Neurology, Washington University, St. Louis, MO 63110, USA)
Dhar, Rajat (Department of Neurology, Washington University, St. Louis, MO 63110, USA)
Heitsch, Laura (Emergency Medicine, Washington University, St. Louis, MO 63110, USA)
Ford, Andria (Department of Neurology, Washington University, St. Louis, MO 63110, USA)
Fernandez-Cadenas, Israel (Hospital Universitari Vall d'Hebron. Institut de Recerca)
Carrera, Caty (Hospital Universitari Vall d'Hebron. Institut de Recerca)
Montaner, Joan (Hospital Universitari Vall d'Hebron. Institut de Recerca)
Lin, Weili (Dept. of Radiology, University of North Carolina, Chapel Hill, NC 27599, USA)
Shen, Dinggang (Department of Brain and Cognitive Engineering, Korea University, Seoul 02841, Republic of Korea)
An, Hongyu (Radiology, Washington University, St. Louis, MO 63110, USA)
Lee, Jin-Moo (Biomedical Engineering, Washington University, St. Louis, MO 63110, USA)
Universitat Autònoma de Barcelona

Data: 2016
Resum: Although cerebral edema is a major cause of death and deterioration following hemispheric stroke, there remains no validated biomarker that captures the full spectrum of this critical complication. We recently demonstrated that reduction in intracranial cerebrospinal fluid (CSF) volume (∆ CSF) on serial computed tomography (CT) scans provides an accurate measure of cerebral edema severity, which may aid in early triaging of stroke patients for craniectomy. However, application of such a volumetric approach would be too cumbersome to perform manually on serial scans in a real-world setting. We developed and validated an automated technique for CSF segmentation via integration of random forest (RF) based machine learning with geodesic active contour (GAC) segmentation. The proposed RF + GAC approach was compared to conventional Hounsfield Unit (HU) thresholding and RF segmentation methods using Dice similarity coefficient (DSC) and the correlation of volumetric measurements, with manual delineation serving as the ground truth. CSF spaces were outlined on scans performed at baseline (< 6 h after stroke onset) and early follow-up (FU) (closest to 24 h) in 38 acute ischemic stroke patients. RF performed significantly better than optimized HU thresholding (p < 10 − 4 in baseline and p < 10 − 5 in FU) and RF + GAC performed significantly better than RF (p < 10 − 3 in baseline and p < 10 − 5 in FU). Pearson correlation coefficients between the automatically detected ∆ CSF and the ground truth were r = 0. 178 (p = 0. 285), r = 0. 876 (p < 10 − 6) and r = 0. 879 (p < 10 − 6) for thresholding, RF and RF + GAC, respectively, with a slope closer to the line of identity in RF + GAC. When we applied the algorithm trained from images of one stroke center to segment CTs from another center, similar findings held. In conclusion, we have developed and validated an accurate automated approach to segment CSF and calculate its shifts on serial CT scans. This algorithm will allow us to efficiently and accurately measure the evolution of cerebral edema in future studies including large multi-site patient populations.
Nota: Altres ajuts: The research was supported in part by the following grants from National Institutes of Health: 1R01HL129241 (to A.F.), 5K23HL129241 (to A.F.), 1R01NS082561-01A1 (to H.A.), R01NS084028 (to J.M.L.),R01NS085419 (to J.M.L.), P50 NS55977 (to J.M.L.) and UL1 TR000448. The Neurovascular Research Laboratory is supported by the Spanish stroke research network (INVICTUS), Pre-Test Stroke Project (PMP15/00022) and the European Stroke Network (EUSTROKE 7FP Health F2-08-202213) and Fondo Europeo de Deasarrollo Regional (ISCIII-FEDER). C-C. is supported by NIH R01 grant, GENISIS project.
Nota: Número d'acord de subvenció ISCIII/CP12-03298
Nota: Número d'acord de subvenció ISCIII/PI15-01978
Nota: Número d'acord de subvenció ISCIII/PMP15-00022
Drets: Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, i la comunicació pública de l'obra, sempre que no sigui amb finalitats comercials, i sempre que es reconegui l'autoria de l'obra original. No es permet la creació d'obres derivades. Creative Commons
Llengua: Anglès
Document: article ; recerca ; publishedVersion
Matèria: Active contour ; Cerebral edema ; CSF segmentation ; Ischemic stroke CT ; Mass effect ; Random forest
Publicat a: NeuroImage, Vol. 12 (september 2016) , p. 673-680, ISSN 2213-1582

DOI: 10.1016/j.nicl.2016.09.018
PMID: 27761398


8 p, 2.3 MB

El registre apareix a les col·leccions:
Articles > Articles de recerca
Articles > Articles publicats

 Registre creat el 2018-02-07, darrera modificació el 2021-02-20



   Favorit i Compartir