Aymeric Stamm

The DTI Challenge: Toward Standardized Evaluation of Diffusion Tensor Imaging Tractography for Neurosurgery

Diffusion tensor imaging (DTI) tractography reconstruction of white matter pathways can help guide brain tumor resection. However, DTI tracts are complex mathematical objects and the validity of tractography‐derived information in clinical settings has yet to be fully established. To address this issue, we initiated the DTI Challenge, an international working group of clinicians and scientists whose goal was to provide standardized evaluation of tractography methods for neurosurgery. The purpose of this empirical study was to evaluate different tractography techniques in the first DTI Challenge workshop.

Prognostic factors for outcomes after mechanical thrombectomy with solitaire stent

BACKGROUND AND PURPOSE Endovascular mechanical thrombectomy is emerging as a promising therapeutic approach for acute ischemic stroke. This study was aimed at identifying factors influencing outcomes after thrombectomy with a Solitaire stent device. MATERIALS AND METHODS Forty-five consecutive patients treated with thrombectomy using Solitaire FR were retrospectively included. Clinical, imaging and logistic variables were analyzed. A multivariate logistic regression analysis was used to identify variables influencing clinical outcome, based on discharge NIHSS score change and mRS at 3 months. RESULTS Patient mean age and initial NIHSS score was 58 years (range 24-88) and 17 (range 6-32), respectively. An MRI was performed for 80% of patients, showing severe DWI lesion for 28% of patients and associated FLAIR hyperintensity for 58% of patients. Mean time from symptom onset to recanalization was 299min for the 32 ACO and 473min for the 13 PCO. Angiographic efficacy (TICI 2b-3) was achieved for 93% of patients and good clinical outcomes at discharge and at 3 months (mRS≤2) were achieved for 49% and 58% of patients, respectively. Independent prognostic factors for predicting good clinical outcomes at discharge were a short time to recanalization and FLAIR negativity. At 3 months, they were a short time to recanalization and patient age. DWI lesion severity was an associated prognostic factor. CONCLUSION Two main prognostic factors for predicting a good clinical outcome after thrombectomy at 3 months were short time from symptom onset to recanalization and patient age.

Innovations in functional MR imaging of the brain: arterial spin labeling and diffusion

The standard technique for brain activation functional MRI (fMRI) is the BOLD sequence. Two new techniques have emerged: arterial spin labeling (ASL) MRI and diffusion MRI. Both have the theoretical advantage of more accurately directly demonstrating neuronal activation compared to BOLD imaging, resulting in improved spatial and temporal resolution. ASL is a perfusion sequence using labeled arterial protons as an endogenous perfusion agent. In spite of methodological difficulties, quantitative CBF measurements are possible. ASL is less susceptible to venous contamination than BOLD and more reproducible. Diffusion MRI evaluates neuronal activation at the cellular level with the prospect of excellent spatial resolution. The main limitations for both techniques are the technical difficulties in the acquisition and the low SNR. AS such, ASL is not widely used clinically and diffusion remains in the field of research. However, the increasing availability of 3T MR systems coupled with multi-channel surface coils and improved postprocessing techniques should improve the detection of the brain activation signal. It is thus possible that these techniques could become clinically available either in complement to or as a replacement for BOLD imaging.

Adaptive multi-modal particle filtering for probabilistic white matter tractography

Particle filtering has recently been introduced to perform probabilistic tractography in conjunction with DTI and Q-Ball models to estimate the diffusion information. Particle filters are particularly well adapted to the tractography problem as they offer a way to approximate a probability distribution over all paths originated from a specified voxel, given the diffusion information. In practice however, they often fail at consistently capturing the multi-modality of the target distribution. For brain white matter tractography, this means that multiple fiber pathways are unlikely to be tracked over extended volumes. We propose to remedy this issue by formulating the filtering distribution as an adaptive M-component non-parametric mixture model. Such a formulation preserves all the properties of a classical particle filter while improving multi-modality capture. We apply this multi-modal particle filter to both DTI and Q-Ball models and propose to estimate dynamically the number of modes of the filtering distribution. We show on synthetic and real data how this algorithm outperforms the previous versions proposed in the literature.

Non-local Robust Detection of DTI White Matter Differences with Small Databases

Diffusion imaging, through the study of water diffusion, allows for the characterization of brain white matter, both at the population and individual level. In recent years, it has been employed to detect brain abnormalities in patients suffering from a disease, e.g., from multiple sclerosis (MS). State-of-the-art methods usually utilize a database of matched (age, sex, ...) controls, registered onto a template, to test for differences in the patient white matter. Such approaches however suffer from two main drawbacks. First, registration algorithms are prone to local errors, thereby degrading the comparison results. Second, the database needs to be large enough to obtain reliable results. However, in medical imaging, such large databases are hardly available. In this paper, we propose a new method that addresses these two issues. It relies on the search for samples in a local neighborhood of each pixel to increase the size of the database. Then, we propose a new test based on these samples to perform a voxelwise comparison of a patient image with respect to a population of controls. We demonstrate on simulated and real MS patient data how such a framework allows for an improve detection power and a better robustness and reproducibility, even with a small database.