Segmentation of nerve bundles and ganglia in spine MRI using particle filters

Parallel and distributed encoding of speech across human auditory cortex

Speech perception is thought to rely on a cortical feedforward serial transformation of acoustic into linguistic representations. Using intracranial recordings across the entire human auditory cortex, electrocortical stimulation, and surgical ablation, we show that cortical processing across areas is not consistent with a serial hierarchical organization. Instead, response latency and receptive field analyses demonstrate parallel and distinct information processing in the primary and nonprimary auditory cortices. This functional dissociation was also observed where stimulation of the primary auditory cortex evokes auditory hallucination but does not distort or interfere with speech perception. Opposite effects were observed during stimulation of nonprimary cortex in superior temporal gyrus. Ablation of the primary auditory cortex does not affect speech perception. These results establish a distributed functional organization of parallel information processing throughout the human auditory cortex and demonstrate an essential independent role for nonprimary auditory cortex in speech processing.

DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes

We present DeepVesselNet, an architecture tailored to the challenges faced when extracting vessel trees and networks and corresponding features in 3-D angiographic volumes using deep learning. We discuss the problems of low execution speed and high memory requirements associated with full 3-D networks, high-class imbalance arising from the low percentage (<3%) of vessel voxels, and unavailability of accurately annotated 3-D training data-and offer solutions as the building blocks of DeepVesselNet. First, we formulate 2-D orthogonal cross-hair filters which make use of 3-D context information at a reduced computational burden. Second, we introduce a class balancing cross-entropy loss function with false-positive rate correction to handle the high-class imbalance and high false positive rate problems associated with existing loss functions. Finally, we generate a synthetic dataset using a computational angiogenesis model capable of simulating vascular tree growth under physiological constraints on local network structure and topology and use these data for transfer learning. We demonstrate the performance on a range of angiographic volumes at different spatial scales including clinical MRA data of the human brain, as well as CTA microscopy scans of the rat brain. Our results show that cross-hair filters achieve over 23% improvement in speed, lower memory footprint, lower network complexity which prevents overfitting and comparable accuracy that does not differ from full 3-D filters. Our class balancing metric is crucial for training the network, and transfer learning with synthetic data is an efficient, robust, and very generalizable approach leading to a network that excels in a variety of angiography segmentation tasks. We observe that sub-sampling and max pooling layers may lead to a drop in performance in tasks that involve voxel-sized structures. To this end, the DeepVesselNet architecture does not use any form of sub-sampling layer and works well for vessel segmentation, centerline prediction, and bifurcation detection. We make our synthetic training data publicly available, fostering future research, and serving as one of the first public datasets for brain vessel tree segmentation and analysis.

Comparison of wideband steady-state free precession and T₂-weighted fast spin echo in spine disorder assessment at 1.5 and 3 T

Wideband steady-state free precession (WB-SSFP) is a modification of balanced steady-state free precession utilizing alternating repetition times to reduce susceptibility-induced balanced steady-state free precession limitations, allowing its use for high-resolution myelographic-contrast spinal imaging. Intertissue contrast and spatial resolution of complete-spine-coverage 3D WB-SSFP were compared with those of 2D T₂-weighted fast spin echo, currently the standard for spine T₂-imaging. Six normal subjects were imaged at 1.5 and 3 T. The signal-to-noise ratio efficiency (SNR per unit-time and unit-volume) of several tissues was measured, along with four intertissue contrast-to-noise ratios; nerve-ganglia:fat, intradural-nerves:cerebrospinal fluid, nerve-ganglia:muscle, and muscle:fat. Patients with degenerative and traumatic spine disorders were imaged at both MRI fields to demonstrate WB-SSFP clinical advantages and disadvantages. At 3 T, WB-SSFP provided spinal contrast-to-noise ratios 3.7-5.2 times that of fast spin echo. At 1.5 T, WB-SSFP contrast-to-noise ratio was 3-3.5 times that of fast spin echo, excluding a 1.7 ratio for intradural-nerves:cerebrospinal fluid. WB-SSFP signal-to-noise ratio efficiency was also higher. Three-dimensional WB-SSFP disadvantages relative to 2D fast spin echo are reduced edema hyperintensity, reduced muscle signal, and higher motion sensitivity. WB-SSFP's high resolution and contrast-to-noise ratio improved visualization of intradural nerve bundles, foraminal nerve roots, and extradural nerve bundles, improving detection of nerve compression in radiculopathy and spinal-stenosis. WB-SSFP's high resolution permitted reformatting into orthogonal planes, providing distinct advantages in gauging fine spine pathology.