Domain-agnostic segmentation of thalamic nuclei from joint structural and diffusion MRI.

The human thalamus is a highly connected subcortical grey-matter structure within the brain. It comprises dozens of nuclei with different function and connectivity, which are affected differently by disease. For this reason, there is growing interest in studying the thalamic nuclei in vivo with MRI. Tools are available to segment the thalamus from 1 mm T1 scans, but the contrast of the lateral and internal boundaries is too faint to produce reliable segmentations. Some tools have attempted to incorporate information from diffusion MRI in the segmentation to refine these boundaries, but do not generalise well across diffusion MRI acquisitions. Here we present the first CNN that can segment thalamic nuclei from T1 and diffusion data of any resolution without retraining or fine tuning. Our method builds on a public histological atlas of the thalamic nuclei and silver standard segmentations on high-quality diffusion data obtained with a recent Bayesian adaptive segmentation tool. We combine these with an approximate degradation model for fast domain randomisation during training. Our CNN produces a segmentation at 0.7 mm isotropic resolution, irrespective of the resolution of the input. Moreover, it uses a parsimonious model of the diffusion signal at each voxel (fractional anisotropy and principal eigenvector) that is compatible with virtually any set of directions and b-values, including huge amounts of legacy data. We show results of our proposed method on three heterogeneous datasets acquired on dozens of different scanners. An implementation of the method is publicly available at https://freesurfer.net/fswiki/ThalamicNucleiDTI.

Accurate Bayesian segmentation of thalamic nuclei using diffusion MRI and an improved histological atlas.

The human thalamus is a highly connected brain structure, which is key for the control of numerous functions and is involved in several neurological disorders. Recently, neuroimaging studies have increasingly focused on the volume and connectivity of the specific nuclei comprising this structure, rather than looking at the thalamus as a whole. However, accurate identification of cytoarchitectonically designed histological nuclei on standard in vivo structural MRI is hampered by the lack of image contrast that can be used to distinguish nuclei from each other and from surrounding white matter tracts. While diffusion MRI may offer such contrast, it has lower resolution and lacks some boundaries visible in structural imaging. In this work, we present a Bayesian segmentation algorithm for the thalamus. This algorithm combines prior information from a probabilistic atlas with likelihood models for both structural and diffusion MRI, allowing segmentation of 25 thalamic labels per hemisphere informed by both modalities. We present an improved probabilistic atlas, incorporating thalamic nuclei identified from histology and 45 white matter tracts surrounding the thalamus identified in ultra-high gradient strength diffusion imaging. We present a family of likelihood models for diffusion tensor imaging, ensuring compatibility with the vast majority of neuroimaging datasets that include diffusion MRI data. The use of these diffusion likelihood models greatly improves identification of nuclear groups versus segmentation based solely on structural MRI. Dice comparison of 5 manually identifiable groups of nuclei to ground truth segmentations show improvements of up to 10 percentage points. Additionally, our chosen model shows a high degree of reliability, with median test-retest Dice scores above 0.85 for four out of five nuclei groups, whilst also offering improved detection of differential thalamic involvement in Alzheimer's disease (AUROC 81.98%). The probabilistic atlas and segmentation tool will be made publicly available as part of the neuroimaging package FreeSurfer (https://freesurfer.net/fswiki/ThalamicNucleiDTI).

Effect of transauricular nerve stimulation on perioperative pain: a single-blind, analyser-masked, randomised controlled trial.

BACKGROUND: Activation of central autonomic pathways, including those regulating the arterial baroreflex, might reduce acute pain. We tested the hypothesis that transcutaneous auricular nerve stimulation (TAN) reduces pain after orthopaedic trauma surgery through autonomic modulation. METHODS: A total of 86 participants aged >18 yr were randomly assigned to 50 min of either sham or active bilateral TAN, undertaken before, and again 24 h after, surgery for orthopaedic trauma. The primary outcome was absolute change in pain 24 h postoperatively, comparing the 100 mm visual analogue scale (VAS) before and after TAN. Secondary outcomes included the minimal clinically important difference in pain (>10 mm increase or reduction in VAS) before/after surgery, using intention-to-treat analysis. Holter monitoring, the analysis of which was masked to allocation, quantified autonomic modulation of heart rate. RESULTS: From June 22, 2021 to July 7, 2022, 79/86 participants (49 yr; 45% female) completed TAN before and after surgery. For the primary outcome, the mean reduction in VAS was 19 mm (95% confidence interval [CI]: 12-26) after active TAN (n=40), vs 10 mm (95% CI: 3-17) after sham TAN (n=39; P=0.023). A minimally clinically important reduction in postoperative pain occurred in 31/40 (78%) participants after active TAN, compared with 15/39 (38%) allocated to sham TAN (odds ratio 5.51 [95% CI: 2.06-14.73]; P=0.001). Only active TAN increased heart rate variability (log low-frequency power increased by 0.19 ms2 [0.01-0.37 ms2]). Prespecified adverse events (auricular skin irritation) occurred in six participants receiving active TAN, compared with two receiving sham TAN. CONCLUSION: Bilateral TAN reduces perioperative pain through autonomic modulation. These proof-of-concept data support a non-pharmacological, generalisable approach to improve perioperative analgesia.

Deep active learning for suggestive segmentation of biomedical image stacks via optimisation of Dice scores and traced boundary length.

Manual segmentation of stacks of 2D biomedical images (e.g., histology) is a time-consuming task which can be sped up with semi-automated techniques. In this article, we present a suggestive deep active learning framework that seeks to minimise the annotation effort required to achieve a certain level of accuracy when labelling such a stack. The framework suggests, at every iteration, a specific region of interest (ROI) in one of the images for manual delineation. Using a deep segmentation neural network and a mixed cross-entropy loss function, we propose a principled strategy to estimate class probabilities for the whole stack, conditioned on heterogeneous partial segmentations of the 2D images, as well as on weak supervision in the form of image indices that bound each ROI. Using the estimated probabilities, we propose a novel active learning criterion based on predictions for the estimated segmentation performance and delineation effort, measured with average Dice scores and total delineated boundary length, respectively, rather than common surrogates such as entropy. The query strategy suggests the ROI that is expected to maximise the ratio between performance and effort, while considering the adjacency of structures that may have already been labelled - which decrease the length of the boundary to trace. We provide quantitative results on synthetically deformed MRI scans and real histological data, showing that our framework can reduce labelling effort by up to 60-70% without compromising accuracy.

Bi-allelic JAM2 Variants Lead to Early-Onset Recessive Primary Familial Brain Calcification.

Primary familial brain calcification (PFBC) is a rare neurodegenerative disorder characterized by a combination of neurological, psychiatric, and cognitive decline associated with calcium deposition on brain imaging. To date, mutations in five genes have been linked to PFBC. However, more than 50% of individuals affected by PFBC have no molecular diagnosis. We report four unrelated families presenting with initial learning difficulties and seizures and later psychiatric symptoms, cerebellar ataxia, extrapyramidal signs, and extensive calcifications on brain imaging. Through a combination of homozygosity mapping and exome sequencing, we mapped this phenotype to chromosome 21q21.3 and identified bi-allelic variants in JAM2. JAM2 encodes for the junctional-adhesion-molecule-2, a key tight-junction protein in blood-brain-barrier permeability. We show that JAM2 variants lead to reduction of JAM2 mRNA expression and absence of JAM2 protein in patient's fibroblasts, consistent with a loss-of-function mechanism. We show that the human phenotype is replicated in the jam2 complete knockout mouse (jam2 KO). Furthermore, neuropathology of jam2 KO mouse showed prominent vacuolation in the cerebral cortex, thalamus, and cerebellum and particularly widespread vacuolation in the midbrain with reactive astrogliosis and neuronal density reduction. The regions of the human brain affected on neuroimaging are similar to the affected brain areas in the myorg PFBC null mouse. Along with JAM3 and OCLN, JAM2 is the third tight-junction gene in which bi-allelic variants are associated with brain calcification, suggesting that defective cell-to-cell adhesion and dysfunction of the movement of solutes through the paracellular spaces in the neurovascular unit is a key mechanism in CNS calcification.