Denoising diffusion-based MRI to CT image translation enables automated spinal segmentation.

BACKGROUND: Automated segmentation of spinal magnetic resonance imaging (MRI) plays a vital role both scientifically and clinically. However, accurately delineating posterior spine structures is challenging. METHODS: This retrospective study, approved by the ethical committee, involved translating T1-weighted and T2-weighted images into computed tomography (CT) images in a total of 263 pairs of CT/MR series. Landmark-based registration was performed to align image pairs. We compared two-dimensional (2D) paired - Pix2Pix, denoising diffusion implicit models (DDIM) image mode, DDIM noise mode - and unpaired (SynDiff, contrastive unpaired translation) image-to-image translation using "peak signal-to-noise ratio" as quality measure. A publicly available segmentation network segmented the synthesized CT datasets, and Dice similarity coefficients (DSC) were evaluated on in-house test sets and the "MRSpineSeg Challenge" volumes. The 2D findings were extended to three-dimensional (3D) Pix2Pix and DDIM. RESULTS: 2D paired methods and SynDiff exhibited similar translation performance and DCS on paired data. DDIM image mode achieved the highest image quality. SynDiff, Pix2Pix, and DDIM image mode demonstrated similar DSC (0.77). For craniocaudal axis rotations, at least two landmarks per vertebra were required for registration. The 3D translation outperformed the 2D approach, resulting in improved DSC (0.80) and anatomically accurate segmentations with higher spatial resolution than that of the original MRI series. CONCLUSIONS: Two landmarks per vertebra registration enabled paired image-to-image translation from MRI to CT and outperformed all unpaired approaches. The 3D techniques provided anatomically correct segmentations, avoiding underprediction of small structures like the spinous process. RELEVANCE STATEMENT: This study addresses the unresolved issue of translating spinal MRI to CT, making CT-based tools usable for MRI data. It generates whole spine segmentation, previously unavailable in MRI, a prerequisite for biomechanical modeling and feature extraction for clinical applications. KEY POINTS: • Unpaired image translation lacks in converting spine MRI to CT effectively. • Paired translation needs registration with two landmarks per vertebra at least. • Paired image-to-image enables segmentation transfer to other domains. • 3D translation enables super resolution from MRI to CT. • 3D translation prevents underprediction of small structures.

Resective epilepsy surgery in patients aged 50years and older - a retrospective study regarding seizure outcome, memory performance, and psychopathology.

PURPOSE: To assess clinical and demographic characteristics in two cohorts of elderly patients with drug-resistant focal epilepsy, undergoing resective epilepsy surgery (RES). Further, to determine seizure, neuropsychological, and mental health outcomes after RES and evaluate possible influencing factors. METHODS: Consecutive patients aged ≥50 years with temporal lobe epilepsy (TLE) who underwent curative RES in the Hamburg epilepsy surgery program (2004-2017) were identified. Data were retrospectively analyzed. Seizure outcome was classified according to ILAE and Engel outcome scales in patients with first-time surgeries and with reoperations. Previously reported predictors of the seizure outcome were evaluated using regression analyses. Changes in verbal memory were assessed for patients with complete pre- and postoperative datasets (n=30) using repeated-measures analysis of variance. For evaluation of possible predictors of psychopathologic changes after RES a regression analysis was conducted. RESULTS: Fifty-one elderly patients underwent RES of the temporal lobe, including twelve aged ≥60 years, and five with reoperations. After one year, 65% of the patients with first-time surgeries were seizure free and 91% had a favorable outcome. At last follow-up, 49% were seizure free since surgery. Three reoperated patients had an Engel I outcome. Seizure outcome was not dependent on age at surgery, duration of epilepsy, or other evaluated variables. There was no significant decline in the memory performance after surgery. Significant improvements in mental health were found. CONCLUSION: RES for drug-resistant TLE is safe, effective, and improves mental health also in patients aged ≥ 50 years. Thus, it should be evaluated as the treatment of choice also in this age group.

Preparation of thick silica coatings on carbon fibers with fine-structured silica nanotubes induced by a self-assembly process.

A facile method to coat carbon fibers with a silica shell is presented in this work. By immobilizing linear polyamines on the carbon fiber surface, the high catalytic activity of polyamines in the sol-gel-processing of silica precursors is used to deposit a silica coating directly on the fiber's surface. The surface localization of the catalyst is achieved either by attaching short-chain polyamines (e.g., tetraethylenepentamine) via covalent bonds to the carbon fiber surface or by depositing long-chain polyamines (e.g., linear poly(ethylenimine)) on the carbon fiber by weak non-covalent bonding. The long-chain polyamine self-assembles onto the carbon fiber substrate in the form of nanoscopic crystallites, which serve as a template for the subsequent silica deposition. The silicification at close to neutral pH is spatially restricted to the localized polyamine and consequently to the fiber surface. In case of the linear poly(ethylenimine), silica shells of several micrometers in thickness can be obtained and their morphology is easily controlled by a considerable number of synthesis parameters. A unique feature is the hierarchical biomimetic structure of the silica coating which surrounds the embedded carbon fiber by fibrillar and interconnected silica fine-structures. The high surface area of the nanostructured composite fiber may be exploited for catalytic applications and adsorption purposes.