Search

One outstanding challenge for machine learning in diagnostic biomedical imaging is algorithm interpretability. A key application is the identification of subtle epileptogenic focal cortical dysplasias (FCDs) from structural MRI. FCDs are difficult to visualize on structural MRI but are often amenable to surgical resection. We aimed to develop an open-source, interpretable, surface-based machine-learning algorithm to automatically identify FCDs on heterogeneous structural MRI data from epilepsy surgery centres worldwide. The Multi-centre Epilepsy Lesion Detection (MELD) Project collated and harmonized a retrospective MRI cohort of 1015 participants, 618 patients with focal FCD-related epilepsy and 397 controls, from 22 epilepsy centres worldwide. We created a neural network for FCD detection based on 33 surface-based features. The network was trained and cross-validated on 50% of the total cohort and tested on the remaining 50% as well as on 2 independent test sites. Multidimensional feature analysis and integrated gradient saliencies were used to interrogate network performance. Our pipeline outputs individual patient reports, which identify the location of predicted lesions, alongside their imaging features and relative saliency to the classifier. On a restricted ‘gold-standard’ subcohort of seizure-free patients with FCD type IIB who had T1 and fluid-attenuated inversion recovery MRI data, the MELD FCD surface-based algorithm had a sensitivity of 85%. Across the entire withheld test cohort the sensitivity was 59% and specificity was 54%. After including a border zone around lesions, to account for uncertainty around the borders of manually delineated lesion masks, the sensitivity was 67%. This multicentre, multinational study with open access protocols and code has developed a robust and interpretable machine-learning algorithm for automated detection of focal cortical dysplasias, giving physicians greater confidence in the identification of subtle M

Objective: Drug-resistant focal epilepsy is often caused by focal cortical dysplasias (FCDs). The distribution of these lesions across the cerebral cortex and the impact of lesion location on clinical presentation and surgical outcome are largely unknown. We created a neuroimaging cohort of patients with individually mapped FCDs to determine factors associated with lesion location and predictors of postsurgical outcome. Methods: The MELD (Multi-centre Epilepsy Lesion Detection) project collated a retrospective cohort of 580 patients with epilepsy attributed to FCD from 20 epilepsy centers worldwide. Magnetic resonance imaging-based maps of individual FCDs with accompanying demographic, clinical, and surgical information were collected. We mapped the distribution of FCDs, examined for associations between clinical factors and lesion location, and developed a predictive model of postsurgical seizure freedom. Results: FCDs were nonuniformly distributed, concentrating in the superior frontal sulcus, frontal pole, and temporal pole. Epilepsy onset was typically before the age of 10 years. Earlier epilepsy onset was associated with lesions in primary sensory areas, whereas later epilepsy onset was associated with lesions in association cortices. Lesions in temporal and occipital lobes tended to be larger than frontal lobe lesions. Seizure freedom rates varied with FCD location, from around 30% in visual, motor, and premotor areas to 75% in superior temporal and frontal gyri. The predictive model of postsurgical seizure freedom had a positive predictive value of 70% and negative predictive value of 61%. Significance: FCD location is an important determinant of its size, the age at epilepsy onset, and the likelihood of seizure freedom postsurgery. Our atlas of lesion locations can be used to guide the radiological search for subtle lesions in individual patients. Our atlas of regional seizure freedom rates and associated predictive model can be used to estimate individua

Preterm newborns with germinal matrix-intraventricular hemorrhage (GM-IVH) are at a higher risk of evidencing neurodevelopmental alterations. Present study aimed to explore the long-term effects that GM-IVH have on hippocampal subfields, and their correlates with memory. The sample consisted of 58 participants, including 36 preterm-born (16 with GM-IVH and 20 without neonatal brain injury), and 22 full-term children aged between 6 and 15 years old. All participants underwent a cognitive assessment and magnetic resonance imaging study. GM-IVH children evidenced lower scores in Full Intelligence Quotient and memory measures compared to their low-risk preterm and full-term peers. High-risk preterm children with GM-IVH evidenced significantly lower total hippocampal volumes bilaterally and hippocampal subfield volumes compared to both low-risk preterm and full-term groups. Finally, significant positive correlations between memory and hippocampal subfield volumes were only found in preterm participants together; memory and the right CA-field correlation remained significant after Bonferroni correction was applied (p = .002). In conclusion, memory alterations and both global and regional volumetric reductions in the hippocampus were found to be specifically related to a preterm sample with GM-IVH. Nevertheless, results also suggest that prematurity per se has a long-lasting impact on the association between the right CA-field volume and memory during childhood

Neuroprostheses aimed to restore lost functions after a limb amputation are based on the interaction with the nervous system by means of neural interfaces. Among the different designs, intraneural electrodes implanted in peripheral nerves represent a good strategy to stimulate nerve fibers to send sensory feedback and to record nerve signals to control the prosthetic limb. However, intraneural electrodes, as any device implanted in the body, induce a foreign body reaction (FBR) that results in the tissue encapsulation of the device. The FBR causes a progressive decline of the electrode functionality over time due to the physical separation between the electrode active sites and the axons to the interface. Modulation of the inflammatory response has arisen as a good strategy to reduce the FBR and maintain electrode functionality. In this paper, transversal intraneural multi-channel electrodes (TIMEs) were implanted in the rat sciatic nerve and tested for three months to evaluate stimulation and recording capabilities under chronic administration of dexamethasone. Dexamethasone treatment significantly reduced the threshold for evoking muscle responses during the follow-up compared to saline-treated animals, without affecting the selectivity of stimulation. However, dexamethasone treatment did not improve the signal-to-noise ratio of the recorded neural signals. Dexamethasone treatment allowed to maintain more working active sites along time than saline treatment. Thus, systemic administration of dexamethasone appears as a useful treatment in chronically implanted animals with neural electrodes as it increases the number of functioning contacts of the implanted TIME and reduces the intensity needed to stimulate the nerve.

Segregation of regenerating motor and sensory axons may be a good strategy to improve selective functionality of regenerative interfaces to provide closed-loop commands. Provided that extracellular matrix components and neurotrophic factors exert guidance effects on different neuronal populations, we assessed in vivo the potential of separating sensory and motor axons regenerating in a bicompartmental Y-type tube, with each branch prefilled with an adequate combination of extracellular matrix and neurotrophic factors. The severed rat sciatic nerve was repaired using a bicompartmental tube filled with a collagen matrix enriched with fibronectin (FN) and brain-derived neurotrophic factor (BDNF) encapsulated in poly-lactic co-glycolic acid microspheres (FN + MP.BDNF) in one compartment to preferentially attract motor axons and collagen enriched with laminin (LM) and nerve growth factor (NGF) and neurotrophin-3 (NT-3) in microspheres (LM + MP.NGF/NT-3) in the other compartment for promoting sensory axons regeneration. Control animals were implanted with the same Y-tube with a collagen matrix with microspheres (MP) containing PBS (Col + MP.PBS). By using retrotracer labelling, we found that LM + MP.NGF/NT-3 did not attract higher number of regenerated sensory axons compared with controls, and no differences were observed in sensory functional recovery. However, FN + MP.BDNF guided a higher number of regenerating motor axons compared with controls, improving also motor recovery. A small proportion of sensory axons with large soma size, likely proprioceptive neurons, was also attracted to the FN + MP.BDNF compartment. These results demonstrate that muscular axonal guidance can be modulated in vivo by the addition of fibronectin and BDNF.