Brain extraction using the watershed transform from markers

Abstract

Isolation of the brain from other tissue types in magnetic resonance (MR) images is an important step in many types of neuro-imaging research using both humans and animal subjects. The importance of brain extraction is well appreciated-numerous approaches have been published and the benefits of good extraction methods to subsequent processing are well known. We describe a tool-the marker based watershed scalper (MBWSS)-for isolating the brain in T1-weighted MR images built using filtering and segmentation components from the Insight Toolkit (ITK) framework. The key elements of MBWSS-the watershed transform from markers and aggressive filtering with large kernels-are techniques that have rarely been used in neuroimaging segmentation applications. MBWSS is able to reliably isolate the brain without expensive preprocessing steps, such as registration to an atlas, and is therefore useful as the first stage of processing pipelines. It is an informative example of the level of accuracy achievable without using priors in the form of atlases, shape models or libraries of examples. We validate the MBWSS using a publicly available dataset, a paediatric cohort, an adolescent cohort, intra-surgical scans and demonstrate flexibility of the approach by modifying the method to extract macaque brains.

Keywords: Insight Toolkit; brain extraction; human brain extraction; macaque brain extraction; mathematical morphology; scalping; watershed transform from markers.

Figure 1

Reconstruction by erosion of 1D signal from markers. Original signal in dashed black, marker signal in red, reconstructed signal in green. The green signal only has regional minima where the marker signal is zero.

Figure 2

Summary of processing steps used in MBWSS.

Figure 3

Processing steps for generating Stage 1 markers.

Figure 4

Stage 1 segmentation result.

Figure 5

Stage 2 marker generation phases. N_dark and N_bright are overlaid on R_B. N₂, the final stage 2 marker image, is overlaid on R_D.

Figure 6

Segmentation produced by Stage 2—raw (top) and smoothed (bottom).

Figure 7

Processing steps producing stage 1 marker, MN₁.

Figure 8

Stage 1 segmentation for macaque.

Figure 9

Stage 2 marker for macaque on gradient control image (top) and resulting stage 2 segmentation (bottom).

Figure 10

Brain extraction (Stage 1), for two intra-surgical scans with open skull. Top image shows a contrast enhancing tumor. Acquisition on IMRIS surgical scanner.

Figure 11

Brain extraction (Stage 1, in green) and corresponding graph cut segmentation (in cyan), which has a incorrectly removed brain tissue in the vicinity of the open skull.