Robust thalamic nuclei segmentation method based on local diffusion magnetic resonance properties

Abstract

The thalamus is an essential relay station in the cortical-subcortical connections. It is characterized by a complex anatomical architecture composed of numerous small nuclei, which mediate the involvement of the thalamus in a wide range of neurological functions. We present a novel framework for segmenting the thalamic nuclei, which explores the orientation distribution functions (ODFs) from diffusion magnetic resonance images at 3 T. The differentiation of the complex intra-thalamic microstructure is improved by using the spherical harmonic (SH) representation of the ODFs, which provides full angular characterization of the diffusion process in each voxel. The clustering was performed using the k-means algorithm initialized in a data-driven manner. The method was tested on 35 healthy volunteers and our results show a robust, reproducible and accurate segmentation of the thalamus in seven nuclei groups. Six of them closely matched the anatomy and were labeled as anterior, ventral anterior, medio-dorsal, ventral latero-ventral, ventral latero-dorsal and pulvinar, while the seventh cluster included the centro-lateral and the latero-posterior nuclei. Results were evaluated both qualitatively, by comparing the segmented nuclei to the histological atlas of Morel, and quantitatively, by measuring the clusters' extent and the clusters' spatial distribution across subjects and hemispheres. We also showed the robustness of our approach across different sequences and scanners, as well as intra-subject reproducibility of the segmented clusters using additional two scan-rescan datasets. We also observed an overlap between the path of the main long-connection tracts passing through the thalamus and the spatial distribution of the nuclei identified with our clustering algorithm. Our approach, based on SH representations of the ODFs, outperforms the one based on angular differences between the principle diffusion directions, which is considered so far as state-of-the-art method. Our findings show an anatomically reliable segmentation of the main groups of thalamic nuclei that could be of potential use in many clinical applications.

Keywords: Orientation distribution function; Segmentation; Spherical harmonics; Thalamic nuclei.

Fig. 1

Outline of the main pre-processing steps for accurate thalamus extraction

Fig. 2

Visualization of the ODFs in a slice of the thalamus. The yellow contour in a delineates the thalamus, while b provides a close-up view of the ODFs shapes inside the thalamic area identified by the light-blue box

Fig. 3

Schematic overview of the clustering framework. Segmentation of the seven thalamic nuclei has been performed using a k-means clustering algorithm with two equally weighted features: the spatial position of the voxels inside the thalamus (x, y, z) and the mean ODF coefficients (C _i , i ∈ [1, 28]) expressed in the SH basis of maximum order 6. k-means is initialized in a data-driven fashion

Fig. 4

Rendering of the weighted mean clustering map by majority voting. The map is superposed on a T1-weighted image in the Montreal Neurological Institute (MNI) space in sagittal (a) and transversal (b) views. Panel c represents the mean ODF characteristic for each cluster. Each averaged ODFs were reconstructed on a representative subject and superposed on the weighted mean clustering map. Thalamic nuclei are color-coded as follows: brown for the anterior group (A), maroon for the ventral anterior group (VA), light pink for the medio-dorsal group (MD), red for the ventral latero-ventral group (VLV), blue for the ventral latero-dorsal group (VLD), green for the pulvinar (Pu), and cyan for the cluster representing the central lateral nucleus, the lateral posterior and a portion of the medial part of the pulvinar (CL–LP–PuM)

Fig. 5

Individual results of the thalamic nuclei segmentation. Spatial distribution of the segmented nuclei are shown in axial view for five different cases and superposed on each subject’s MPRAGE image

Fig. 6

Comparison of the results of our clustering algorithm with the Morel’s histological atlas. a shows a sagittal view of the Morel atlas. b–d show instead the spatial distribution of the thalamic nuclei segmented with our framework in the same sagittal slice for three different cases in the Talairach space. Each color gives the anatomical correspondence of each group of nuclei

Fig. 7

Resulting clustering from the scan–rescan analysis compared with two different axial slices from the Morel’s atlas (D 4.5 and D 10.8 top and bottom row, respectively)

Fig. 8

Reconstruction of thalamic long connections. Sagittal (a) and coronal (b) 3D views of the motor fiber tracts passing through the cluster VLV (in red). Probabilistic tracts (in white) were reconstructed using the whole thalamus mask and the following seed regions (in yellow): left precentral gyrus, left red nucleus, and right superior cerebellar peduncle

Fig. 9

Quantitative measures of overlap between the corresponding clusters in scan–rescan analysis: ODF- versus AD-based segmentation