Deep Learning Segmentation of the Nucleus Basalis of Meynert on 3T MRI.

BACKGROUND AND PURPOSE: The nucleus basalis of Meynert is a key subcortical structure that is important in arousal and cognition and has been explored as a deep brain stimulation target but is difficult to study due to its small size, variability among patients, and lack of contrast on 3T MR imaging. Thus, our goal was to establish and evaluate a deep learning network for automatic, accurate, and patient-specific segmentations with 3T MR imaging. MATERIALS AND METHODS: Patient-specific segmentations can be produced manually; however, the nucleus basalis of Meynert is difficult to accurately segment on 3T MR imaging, with 7T being preferred. Thus, paired 3T and 7T MR imaging data sets of 21 healthy subjects were obtained. A test data set of 6 subjects was completely withheld. The nucleus was expertly segmented on 7T, providing accurate labels for the paired 3T MR imaging. An external data set of 14 patients with temporal lobe epilepsy was used to test the model on brains with neurologic disorders. A 3D-Unet convolutional neural network was constructed, and a 5-fold cross-validation was performed. RESULTS: The novel segmentation model demonstrated significantly improved Dice coefficients over the standard probabilistic atlas for both healthy subjects (mean, 0.68 [SD, 0.10] versus 0.45 [SD, 0.11], P = .002, t test) and patients (0.64 [SD, 0.10] versus 0.37 [SD, 0.22], P < .001). Additionally, the model demonstrated significantly decreased centroid distance in patients (1.18 [SD, 0.43] mm, 3.09 [SD, 2.56] mm, P = .007). CONCLUSIONS: We developed the first model, to our knowledge, for automatic and accurate patient-specific segmentation of the nucleus basalis of Meynert. This model may enable further study into the nucleus, impacting new treatments such as deep brain stimulation.

Targeting Epilepsy Through the Foremen Ovale: How Many Helical Needles are Needed?

Laser ablation of the hippocampus offers medically refractory epilepsy patients an alternative to invasive surgeries. Emerging commercial solutions deliver the ablator through a burr hole in the back of the head. We recently introduced a new access path through the foremen ovale, using a helical needle, which minimizes the amount of healthy brain tissue the needle must pass through on its way to the hippocampus, and also enables the needle to follow the medial axis of the hippocampus more closely. In this paper, we investigate whether helical needles should be designed and fabricated on a patient-specific basis as we had previously proposed, or whether a small collection of pre-defined needle shapes can apply across many patients. We propose a new optimization strategy to determine this needle set using patient data, and investigate the accuracy with which these needles can reach the the medial axis of the hippocampus. We find that three basic tube shapes (mirrored as necessary for left vs. right hippocampi) are all that is required, across 20 patient datasets (obtained from 10 patient CT scans), to reduce worst-case maximum error below 2 mm.

Lead cap localization using ultrasound in deep brain stimulation surgery: technical note.

In deep brain stimulation (DBS) surgery, after intracranial lead implantation, lead caps are tunneled into the subgaleal space for later connection to internal pulse generator (IPG) extension wires. In the subsequent IPG implantation procedure, the lead cap must be localized by palpation in order to plan an incision in the scalp to complete this connection. However, if the IPG implantation is done the same day as the intracranial lead implantation, palpation of the lead cap may be challenging in a thick or postoperatively edematous scalp. Manufacturers suggest using fluoroscopy in these instances, but fluoroscopy provides poor soft tissue visualization, requires further unnecessary radiation exposure to both the patient and the surgical team, and can be cumbersome. Portable ultrasound (US) machines are readily available in many operating rooms, and can be used to easily and accurately localize the lead cap prior to IPG implantation.

Social brain networks: Resting-state and task-based connectivity in youth with and without epilepsy.

Individuals with epilepsy often experience social difficulties and deficits in social cognition. It remains unknown how disruptions to neural networks underlying such skills may contribute to this clinical phenotype. The current study compared the organization of relevant brain circuits-the "mentalizing network" and a salience-related network centered on the amygdala-in youth with and without epilepsy. Functional connectivity between the nodes of these networks was assessed, both at rest and during engagement in a social cognitive task (facial emotion recognition), using functional magnetic resonance imaging. There were no group differences in resting-state connectivity within either neural network. In contrast, youth with epilepsy showed comparatively lower connectivity between the left posterior superior temporal sulcus and the medial prefrontal cortex-but greater connectivity within the left temporal lobe-when viewing faces in the task. These findings suggest that the organization of a mentalizing network underpinning social cognition may be disrupted in youth with epilepsy, though differences in connectivity within this circuit may shift depending on task demands. Our results highlight the importance of considering functional task-based engagement of neural systems in characterizations of network dysfunction in epilepsy.

Epilepsy surgery trends in the United States, 1990-2008.

OBJECTIVE: To examine national time trends of resective surgery for the treatment of medically refractory epilepsy before and after Class I evidence demonstrating its efficacy and subsequent practice guidelines recommending early surgical evaluation. METHODS: We performed a population-based cohort study with time trends of patients admitted to US hospitals for medically refractory focal epilepsy between 1990 and 2008 who did or did not undergo lobectomy, as reported in the Nationwide Inpatient Sample. RESULTS: Weighted data revealed 112,026 hospitalizations for medically refractory focal epilepsy and 6,653 resective surgeries (lobectomies and partial lobectomies) from 1990 to 2008. A trend of increasing hospitalizations over time was not accompanied by an increase in surgeries, producing an overall trend of decreasing surgery rates (F = 13.6, p < 0.01). Factors associated with this trend included a decrease in epilepsy hospitalizations at the highest-volume epilepsy centers, and increased hospitalizations to lower-volume hospitals that were found to be less likely to perform surgery. White patients were more likely to have surgery than racial minorities (relative risk [RR], 1.13; 95% confidence interval [CI], 1.10-1.17), and privately insured individuals were more likely to receive lobectomy than those with Medicaid or Medicare (RR, 1.28; 95% CI, 1.25-1.30). CONCLUSION: Despite Class I evidence and subsequent practice guidelines, the utilization of lobectomy has not increased from 1990 to 2008. Surgery continues to be heavily underutilized as a treatment for epilepsy, with significant disparities by race and insurance coverage. Patients who are medically refractory after failing 2 antiepileptic medications should be referred to a comprehensive epilepsy center for surgical evaluation.