Dataset Information

Minimum detectable spinal cord atrophy with automatic segmentation: Investigations using an open-access dataset of healthy participants.

ABSTRACT: Spinal cord atrophy is a well-known biomarker in multiple sclerosis (MS) and other diseases. It is measured by segmenting the spinal cord on an MRI image and computing the average cross-sectional area (CSA) over a few slices. Introduced about 25 years ago, this procedure is highly sensitive to the quality of the segmentation and is prone to rater-bias. Recently, fully-automated spinal cord segmentation methods, which remove the rater-bias and enable the automated analysis of large populations, have been introduced. A lingering question related to these automated methods is: How reliable are they at detecting atrophy? In this study, we evaluated the precision and accuracy of automated atrophy measurements by simulating scan-rescan experiments. Spinal cord MRI data from the open-access spine-generic project were used. The dataset aggregates 42 sites worldwide and consists of 260 healthy subjects and includes T1w and T2w contrasts. To simulate atrophy, each volume was globally rescaled at various scaling factors. Moreover, to simulate patient repositioning, random rigid transformations were applied. Using the DeepSeg algorithm from the Spinal Cord Toolbox, the spinal cord was segmented and vertebral levels were identified. Then, the average CSA between C3-C5 vertebral levels was computed for each Monte Carlo sample, allowing us to derive measures of atrophy, intra/inter-subject variability, and sample-size calculations. The minimum sample size required to detect an atrophy of 2% between unpaired study arms, commonly seen in MS studies, was 467 +/- 13.9 using T1w and 467 +/- 3.2 using T2w images. The minimum sample size to detect a longitudinal atrophy (between paired study arms) of 0.8% was 60 +/- 25.1 using T1w and 10 +/- 1.2 using T2w images. At the intra-subject level, the estimated CSA, observed in this study, showed good precision compared to other studies with COVs (across Monte Carlo transformations) of 0.8% for T1w and 0.6% for T2w images. While these sample sizes seem small, we would like to stress that these results correspond to a "best case" scenario, in that the dataset used here was of particularly good quality and the model for simulating atrophy does not encompass all the variability met in real-life datasets. The simulated atrophy and scan-rescan variability may over-simplify the biological reality. The proposed framework is open-source and available at https://csa-atrophy.readthedocs.io/.

SUBMITTER: Bautin P

PROVIDER: S-EPMC8503570 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The spinal cord is frequently affected by atrophy and/or lesions in multiple sclerosis (MS) patients. Segmentation of the spinal cord and lesions from MRI data provides measures of damage, which are key criteria for the diagnosis, prognosis, and longitudinal monitoring in MS. Automating this operation eliminates inter-rater variability and increases the efficiency of large-throughput analysis pipelines. Robust and reliable segmentation across multi-site spinal cord data is challenging because of the large variability related to acquisition parameters and image artifacts. In particular, a precise delineation of lesions is hindered by a broad heterogeneity of lesion contrast, size, location, and shape. The goal of this study was to develop a fully-automatic framework - robust to variability in both image parameters and clinical condition - for segmentation of the spinal cord and intramedullary MS lesions from conventional MRI data of MS and non-MS cases. Scans of 1042 subjects (459 healthy controls, 471 MS patients, and 112 with other spinal pathologies) were included in this multi-site study (n = 30). Data spanned three contrasts (T1-, T2-, and T2∗-weighted) for a total of 1943 vol and featured large heterogeneity in terms of resolution, orientation, coverage, and clinical conditions. The proposed cord and lesion automatic segmentation approach is based on a sequence of two Convolutional Neural Networks (CNNs). To deal with the very small proportion of spinal cord and/or lesion voxels compared to the rest of the volume, a first CNN with 2D dilated convolutions detects the spinal cord centerline, followed by a second CNN with 3D convolutions that segments the spinal cord and/or lesions. CNNs were trained independently with the Dice loss. When compared against manual segmentation, our CNN-based approach showed a median Dice of 95% vs. 88% for PropSeg (p ≤ 0.05), a state-of-the-art spinal cord segmentation method. Regarding lesion segmentation on MS data, our framework provided a Dice of 60%, a relative volume difference of -15%, and a lesion-wise detection sensitivity and precision of 83% and 77%, respectively. In this study, we introduce a robust method to segment the spinal cord and intramedullary MS lesions on a variety of MRI contrasts. The proposed framework is open-source and readily available in the Spinal Cord Toolbox.

Project description:Brain images contain information suitable for automatically sorting subjects into categories such as healthy controls and patients. We sought to identify morphometric criteria for distinguishing controls (n?=?28) from patients with unilateral temporal lobe epilepsy (TLE), 60 with and 20 without hippocampal atrophy (TLE-HA and TLE-N, respectively), and for determining the presumed side of seizure onset. The framework employs multi-atlas segmentation to estimate the volumes of 83 brain structures. A kernel-based separability criterion was then used to identify structures whose volumes discriminate between the groups. Next, we applied support vector machines (SVM) to the selected set for classification on the basis of volumes. We also computed pairwise similarities between all subjects and used spectral analysis to convert these into per-subject features. SVM was again applied to these feature data. After training on a subgroup, all TLE-HA patients were correctly distinguished from controls, achieving an accuracy of 96 ± 2% in both classification schemes. For TLE-N patients, the accuracy was 86 ± 2% based on structural volumes and 91 ± 3% using spectral analysis. Structures discriminating between patients and controls were mainly localized ipsilaterally to the presumed seizure focus. For the TLE-HA group, they were mainly in the temporal lobe; for the TLE-N group they included orbitofrontal regions, as well as the ipsilateral substantia nigra. Correct lateralization of the presumed seizure onset zone was achieved using hippocampi and parahippocampal gyri in all TLE-HA patients using either classification scheme; in the TLE-N patients, lateralization was accurate based on structural volumes in 86 ± 4%, and in 94 ± 4% with the spectral analysis approach. Unilateral TLE has imaging features that can be identified automatically, even when they are invisible to human experts. Such morphometric image features may serve as classification and lateralization criteria. The technique also detects unsuspected distinguishing features like the substantia nigra, warranting further study.

Dataset Information

Minimum detectable spinal cord atrophy with automatic segmentation: Investigations using an open-access dataset of healthy participants.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets