Quantitative voxel-based analysis of qualitative images

Abnormalities in the brain can manifest on MRI as changes in shape (morphometry) or changes in the nature of the tissue (signal intensity). Quantitative assessment across the whole brain is possible.

Voxel Based Morphometry (VBM) is a whole brain quantitative way of assessing morphometric changes, and Voxel Based Relaxometry (VBR) directly assesses signal intensity changes in quantitative maps of T2 relaxation time. Unfortunately signal intensity analysis is not straightforward with conventional clinical imaging. This is because typical clinical images, such as T2-weighted MRI, are qualitative in nature: the values can vary from scan-to-scan and assessment of a scan is typically undertaken by a radiologist viewing an image by eye and relying on relative contrast within the image. Whilst quantitative imaging is possible, it usually requires specialised multiple-echo acquisition sequences that can take longer to acquire and are not often available at clinical sites.

To allow quantitative analysis of clinical images, we have developed an objective voxel-based statistical method for evaluation of signal intensity in groups of routinely acquired qualitative images. We call the method Voxel-Based Iterative Sensitivity (VBIS) analysis. It adaptively optimises the relative global scaling of images in an effort to standardise the overall intensity of different images to maximise sensitivity to regional effects. It also includes a final iteration to avoid potentially significant voxels contaminating the scaling factor.

We applied and validated this method of analysis for T2-weighted images of the human brain, by directly comparing with Voxel Based Relaxometry. We studied a group of patients with left hemisphere hippocampal sclerosis and a group of healthy controls. Expected signal abnormalities in the patients were detectable with VBIS-T2, confirming the feasibility of the technique. This opens the door to use of a voxel-based analysis approach for group studies of clinical T2-weighted MRI. More generally, when a quantitative modality is not available, VBIS can be an effective way to quantify differences between groups: for example, the method may assist quantitative analysis of other qualitative modalities such as T1-weighted MRI, SPECT and CT.