Multiparametric Iterative Self-Organizing MR Imaging Data Analysis Technique for Assessment of Tissue Viability in Acute Cerebral Ischemia
Panayiotis D. Mitsiasa,
James R. Ewinga,b,
Mei Luc,
Mohammed M. Khalighia,b,
Mamatha Pasnoord,
Hassan B. Ebadiana,b,
Qingming Zhaoa,b,
Sunitha Santhakumard,
Michael A. Jacobse,
Nikolaos Papamitsakisa,
Hamid Soltanian-Zadehb,
David Hearshenb,
Suresh C. Patelb and
Michael Choppa
a Department of Neurology, Henry Ford Health Sciences Center, Detroit, MI
b Department of Radiology, Henry Ford Health Sciences Center, Detroit, MI
c Department of Biostatistics and Research Epidemiology, Henry Ford Health Sciences Center, Detroit, MI
d Department of Neurology, Kansas University Medical Center, Kansas, KS
e Department of Radiology, Johns Hopkins University, Baltimore, MD

View larger version (62K):
[in a new window]
|
FIG 1. Standardized ISODATA image reveals a right hemispheric ischemic lesion. The ischemic lesion is inhomogeneous and contains two different tissue signatures (4 and 5). The MR imaging study was performed 12 hours after symptom onset.
| |

View larger version (31K):
[in a new window]
|
FIG 2. Calculation of tissue recovery based on acute phase ISODATA image and outcome-phase T2-weighted image.
A, Acute phase ISODATA image.
B, Acute phase diffusion-weighted image.
C, Outcome (3-month) phase T2-weighted image.
| |