InfarctSizer: computing infarct volume from brain images of a stroke animal model

School of Computer Science and Engineering, Inha University, Incheon, South Korea.
Computer Methods in Biomechanics and Biomedical Engineering (Impact Factor: 1.79). 06/2011; 14(6):497-504. DOI: 10.1080/10255842.2010.482528
Source: PubMed

ABSTRACT Many computational methods for determining the infarct volume from the image of 2,3,5-triphenyltetrazolium chloride-stained brain slices rely on the discretion of the user to determine the infarct region by visual inspection. Once the user determines the infarct boundary by visual inspection, the methods compute the area within the boundary with the assumption that all the spots within the boundary have been infarcted at the same level. However, in the same brain image, partially infarcted spots often tend to appear pinkish whereas fully or severely infarcted spots appear white. We developed a program called InfarctSizer, which automatically detects the infarct region and computes the infarct volume proportional to infarction levels. Comparison of InfarctSizer with other methods shows that InfarctSizer computes the infarct volume more accurately and efficiently than other methods. InfarctSizer and sample brain images are available at

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Ethyl pyruvate (EP) is a pyruvate derivative that has been reported recently to prevent lethality in mice with established lethal sepsis and systemic inflammation. In this study, we examined the neuroprotective effect of EP in a rat cerebral ischemia model of middle cerebral artery occlusion (MCAO). Male Sprague-Dawley rats were subjected to 1 hour of MCAO, and EP was administered at various time points before or after MCAO. The changes in the brain infarction, neurological deficits, microglia activation, and proinflammatory cytokine expression were evaluated. BV2 microglial cells were also used to access the anti-inflammatory effect of EP. The administration of EP intraperitoneally at 30 minutes before or at 4 or 12 hours after MCAO reduced the infarct volume to 10.3+/-3.4% (n=6; P<0.05), 21.5+/-2.7% (n=6; P<0.05), and 44.3+/-4.0% (n=6; P<0.05), respectively, of that of the control group. The significant reduction in infarct volume was accompanied by the suppression of the clinical manifestations associated with cerebral ischemia, including motor impairment and neurological deficits, microglial activation, and proinflammatory cytokine expression. The neuroprotective effect of EP was yet evident when it was administered as late as 24 hours after MCAO/reperfusion (76.5+/-4.70%; n=6; P<0.05). EP suppressed lipopolysaccharide induced activation of BV2 cells, as was evidenced by a reduction in NO release and the accompanying induction of proinflammatory cytokines. These results suggest that EP affords the strong protection of the delayed cerebral ischemic injury with a wide therapeutic window.
    Stroke 11/2005; 36(10):2238-43. DOI:10.1161/01.STR.0000181779.83472.35 · 6.02 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Triphenyltetrazolium chloride (TTC) has been used to detect experimental brain ischemia but its accuracy has not been established fully, especially in models of early brain ischaemia. We have developed a technique that combines TTC-staining and perfusion-fixation with formalin in the same specimen. The left middle cerebral artery of 12 rats, anaesthetized with halothane were exposed via a subtemporal approach and occluded for 4 h. The animals were perfused transcardially with TTC (2%) and formalin (10%). The forebrain was sliced and the size of the ischaemic area, delineated by TTC, was measured. The brain slices were then processed for light microscopy, and the amount of ischaemic damage determined. The ischaemic volumes of the hemisphere and cortex assessed by TTC (127.4 +/- 21.7 mm3 and 18.4 +/- 3.0 mm3) were larger than the volumes measured by microscopy (104.3 +/- 16.6 mm3 and 15.2 +/- 3.6 mm3) but there was no statistical difference between them. Correlation of the ischaemic volumes in hemisphere and cortex between the two methods was good (r2 = 0.9221, P less than 0.01 and r2 = 0.9243, P less than 0.001), but the correlation of the ischaemic areas at specific coronal planes was poor. The ischaemic volume of caudate nucleus measured by TTC (15.2 +/- 3.6 mm3) was smaller than the volume assessed by microscopy (18.4 +/- 3.0 mm3) and the correlation was poor (r2 = 0.6650) It is concluded that TTC-staining provides only an approximation of the amount of early ischaemic brain damage subsequently identified by conventional light microscopy.
    Neuropathology and Applied Neurobiology 08/1988; 14(4):289-98. DOI:10.1111/j.1365-2990.1988.tb00889.x · 4.97 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Tetrazolium salts, histochemical indicators of mitochondrial respiratory enzymes, have been used by some pathologists to detect infarcts in myocardium. We explored the utility of this technique in detecting experimental brain infarcts and report our findings. Infarcts were produced in cats, gerbils, and rats by unilateral temporal and permanent cerebral vessel occlusion. After various time periods the animals were killed, and their brains were reacted with 2,3,5, triphenyl, 2H-tetrazolium chloride (TTC). The experimental and contralateral hemispheres were examined by light and electron microscopy. The TTC-stained tissue was correlated with histology. In some situations the histological condition of the tissue correlated well with the TTC staining results. Brain regions supplied by temporarily occluded vessels and judged infarcted by light and electron microscopy did not stain. In these regions less than 6% of the mitochondria were intact. In brain tissue from animals with permanent vessel occlusion (no reflow) mitochondria were intact despite the fact that other cellular organelles, such as nuclei, were destroyed. TTC stained such mitochondria and as a result could not distinguish infarcted brain in complete ischemia situations (no reflow). Another draw back to this staining procedure was 36 h after infarction macrophages with intact mitochondria would replace damage neurons and be stained. Under ideal conditions though this technique can detect irreversibly damaged brain as early as 2.5 h after artery occlusion.
    Acta Neuropathologica 02/1984; 65(2):150-7. DOI:10.1007/BF00690469 · 9.78 Impact Factor