Oslo Universitetssykehus Hf

Kyrre Eeg Emblem

Atle Bjørnerud

The invention provides a method, an image analysis software product, and a system for medical imaging analysis for estimating contrast agent extravasation in contrast agent based perfusion imaging such as MRI dynamic contrast enhanced (DCE) imaging, and in particular correction, compensation, or visualization of extravascular leakage of contrast agent in tumors. According to the invention, the effect of extravasation is directly manifested in the tail part of an observed, apparent residue function, R'(t), obtained directly by de-convoluting the expression C(t)=R&#x2032;(t)C&#x2032;(t) with the arterial input function (AIF). A leakage rate or extravasation constant is determined directly from the tail part of the determined apparent residue function. The invention also relates to distinguishing between T-dominant and T*-dominant extravasation effects in perfusion imaging to from the sign of the tail part of the determined apparent residue function and to an automated method for DSC-MRI involving correction for contrast agent extravasation and partial volume effects.

Estimating and correcting for contrast agent extravasation in tissue perfusion imaging

The invention relates to segmenting blood vessels in perfusion MR images, more particularly, the invention relates to automated vessel removal in identified tumor regions prior to tumor grading. Pixels from a perfusion related map from dynamic contrast enhancement (DCE) MR images are clustered into arterial pixels and venous pixels by e.g. a k-means class cluster analysis. The analysis applies parameters representing the degree to which the tissue entirely consists of blood (such as relative blood volume (rBV), peak enhancement (&#x394;R) and/or post first-pass enhancement level (&#x394;R)) and parameters representing contrast arrival time at the tissue (such as first moment of the area (fmAUC) and/or contrast arrival time (T)). The artery and venous pixels are used to form a vessel mask. The invention also relates to a computer aided diagnostic (CAD) system for tumor grading, comprising automated tumor segmentation, vessel segmentation, and tumor grading.

Vessel segmentation in DCE MR imaging

An embodiment of the invention is to make possible a non-invasive grading of a tumor based on parameters determined from a frequency distribution (histogram) of values in a map representing cerebral blood volume (CBV) or cellular metabolism in the tumor. The method is especially applicable to brain tumors such as gliomas where histological grading is difficult. The invention provides a precise and consistent grading since it relies on values selected from the whole tumor (not just from hot spots); since it takes the diversity or heterogeneity of the vascularization into account by analyzing the frequency distribution (not just a mean value); and since it involves and allows for a more automated procedure wherein any subjective contributions from human operators is not critical to the resulting grading. CBV maps may be obtained by perfusion imaging using MRI or CT scanning. Cellular metabolism maps may be obtained from a glucose metabolism map obtained by positron emission tomography (PET).

Growing community of inventors

Cambridge, MA, United States of America

Kyrre Eeg Emblem