Automated Estimation of Acute Infarct Volume from Noncontrast Head CT Using Image Intensity Inhomogeneity Correction

Abstract

Identification of early ischemic changes (EIC) on noncontrast head CT scans performed within the first few hours of stroke onset may have important implications for subsequent treatment, though early stroke is poorly delimited on these studies. Lack of sharp lesion boundary delineation in early infarcts precludes manual volume measures, as well as measures using edge-detection or region-filling algorithms. We wished to test a hypothesis that image intensity inhomogeneity correction may provide a sensitive method for identifying the subtle regional hypodensity which is characteristic of early ischemic infarcts. A digital image analysis algorithm was developed using image intensity inhomogeneity correction (IIC) and intensity thresholding. Two different IIC algorithms (FSL and ITK) were compared. The method was evaluated using simulated infarcts and clinical cases. For synthetic infarcts, measured infarct volumes demonstrated strong correlation to the true lesion volume (for 20% decreased density "infarcts," Pearson r = 0.998 for both algorithms); both algorithms demonstrated improved accuracy with increasing lesion size and decreasing lesion density. In clinical cases (41 acute infarcts in 30 patients), calculated infarct volumes using FSL IIC correlated with the ASPECTS scores (Pearson r = 0.680) and the admission NIHSS (Pearson r = 0.544). Calculated infarct volumes were highly correlated with the clinical decision to treat with IV-tPA. Image intensity inhomogeneity correction, when applied to noncontrast head CT, provides a tool for image analysis to aid in detection of EIC, as well as to evaluate and guide improvements in scan quality for optimal detection of EIC.

Figure 1

Illustration of the image processing pipeline. After brain-extraction, the NCCT is passed through IIC to generate the intensity-corrected or “restored” image. The input image is then subtracted from the product image to generate a difference map highlighting areas of reduced radiodensity. After thresholding, the volume of the contralateral hemisphere difference map is subtracted from volume of the ipsilateral hemisphere difference map to yield the infarct volume. An MR diffusion-weighted image of the same case performed the following day is included for reference.

Figure 2

Simulated MCA territory infarcts. The figure illustrates the visual appearance of simulated MCA-territory infarcts of various (known) attenuations (top) and the appearance of each known attenuation on the difference map (bottom). The simulated territory is shown top-left for reference. Infarcts of less than 20% decreased density show poor margin delineation.

Figure 3

Calculated infarct volume plotted against true infarct volume, for (3) different input volumes (122cc, 45cc, and 17cc) and (4) different infarct densities (with density reduced by 5%, 10%, 20%, and 30%), FSL (left), ITK (right).

Figure 4

Time series images. Initial brain extracted image at 2.5 hours after symptom onset (a), at 5 hrs (b), and at 28 hrs (c). Corresponding thresholded difference maps (d–f). Stroke volumes were estimated (FSL) 12cc, 81.4cc, and 117.9cc as the infarct evolved.

Figure 5

Box and whiskers plot of calculated stroke volumes as a function of the IIC program used (FSL or ITK), correlated with dichotomized ASPECTS scores (left) and clinical decision to treat with IV TPA (right).