Eye-selective fMRI activity in human primary visual cortex: Comparison between 3 ​T and 9.4 ​T, and effects across cortical depth

Radiofrequency antenna concepts for human cardiac MR at 14.0 T

OBJECTIVE

To examine the feasibility of human cardiac MR (CMR) at 14.0 T using high-density radiofrequency (RF) dipole transceiver arrays in conjunction with static and dynamic parallel transmission (pTx).

MATERIALS AND METHODS

RF arrays comprised of self-grounded bow-tie (SGBT) antennas, bow-tie (BT) antennas, or fractionated dipole (FD) antennas were used in this simulation study. Static and dynamic pTx were applied to enhance transmission field (B₁⁺) uniformity and efficiency in the heart of the human voxel model. B₁⁺ distribution and maximum specific absorption rate averaged over 10 g tissue (SAR_10g) were examined at 7.0 T and 14.0 T.

RESULTS

At 14.0 T static pTx revealed a minimum B₁⁺_ROI efficiency of 0.91 μT/√kW (SGBT), 0.73 μT/√kW (BT), and 0.56 μT/√kW (FD) and maximum SAR_10g of 4.24 W/kg, 1.45 W/kg, and 2.04 W/kg. Dynamic pTx with 8 kT points indicate a balance between B₁⁺_ROI homogeneity (coefficient of variation < 14%) and efficiency (minimum B₁⁺_ROI > 1.11 µT/√kW) at 14.0 T with a maximum SAR_10g < 5.25 W/kg.

DISCUSSION

MRI of the human heart at 14.0 T is feasible from an electrodynamic and theoretical standpoint, provided that multi-channel high-density antennas are arranged accordingly. These findings provide a technical foundation for further explorations into CMR at 14.0 T.

Fast and functionally specific cortical thickness changes induced by visual stimulation

Structural characteristics of the human brain serve as important markers of brain development, aging, disease progression, and neural plasticity. They are considered stable properties, changing slowly over time. Multiple recent studies reported that structural brain changes measured with magnetic resonance imaging (MRI) may occur much faster than previously thought, within hours or even minutes. The mechanisms behind such fast changes remain unclear, with hemodynamics as one possible explanation. Here we investigated the functional specificity of cortical thickness changes induced by a flickering checkerboard and compared them to blood oxygenation level-dependent (BOLD) functional MRI activity. We found that checkerboard stimulation led to a significant thickness increase, which was driven by an expansion at the gray-white matter boundary, functionally specific to V1, confined to the retinotopic representation of the checkerboard stimulus, and amounted to 1.3% or 0.022 mm. Although functional specificity and the effect size of these changes were comparable to those of the BOLD signal in V1, thickness effects were substantially weaker in V3. Furthermore, a comparison of predicted and measured thickness changes for different stimulus timings suggested a slow increase of thickness over time, speaking against a hemodynamic explanation. Altogether, our findings suggest that visual stimulation can induce structural gray matter enlargement measurable with MRI.

Reinforcement learning with associative or discriminative generalization across states and actions: fMRI at 3 T and 7 T

The model-free algorithms of "reinforcement learning" (RL) have gained clout across disciplines, but so too have model-based alternatives. The present study emphasizes other dimensions of this model space in consideration of associative or discriminative generalization across states and actions. This "generalized reinforcement learning" (GRL) model, a frugal extension of RL, parsimoniously retains the single reward-prediction error (RPE), but the scope of learning goes beyond the experienced state and action. Instead, the generalized RPE is efficiently relayed for bidirectional counterfactual updating of value estimates for other representations. Aided by structural information but as an implicit rather than explicit cognitive map, GRL provided the most precise account of human behavior and individual differences in a reversal-learning task with hierarchical structure that encouraged inverse generalization across both states and actions. Reflecting inference that could be true, false (i.e., overgeneralization), or absent (i.e., undergeneralization), state generalization distinguished those who learned well more so than action generalization. With high-resolution high-field fMRI targeting the dopaminergic midbrain, the GRL model's RPE signals (alongside value and decision signals) were localized within not only the striatum but also the substantia nigra and the ventral tegmental area, including specific effects of generalization that also extend to the hippocampus. Factoring in generalization as a multidimensional process in value-based learning, these findings shed light on complexities that, while challenging classic RL, can still be resolved within the bounds of its core computations.

Modelling the depth-dependent VASO and BOLD responses in human primary visual cortex

Functional magnetic resonance imaging (fMRI) using a blood-oxygenation-level-dependent (BOLD) contrast is a common method for studying human brain function noninvasively. Gradient-echo (GRE) BOLD is highly sensitive to the blood oxygenation change in blood vessels; however, the spatial signal specificity can be degraded due to signal leakage from activated lower layers to superficial layers in depth-dependent (also called laminar or layer-specific) fMRI. Alternatively, physiological variables such as cerebral blood volume using the VAscular-Space-Occupancy (VASO) contrast have shown higher spatial specificity compared to BOLD. To better understand the physiological mechanisms such as blood volume and oxygenation changes and to interpret the measured depth-dependent responses, models are needed which reflect vascular properties at this scale. For this purpose, we extended and modified the "cortical vascular model" previously developed to predict layer-specific BOLD signal changes in human primary visual cortex to also predict a layer-specific VASO response. To evaluate the model, we compared the predictions with experimental results of simultaneous VASO and BOLD measurements in a group of healthy participants. Fitting the model to our experimental data provided an estimate of CBV change in different vascular compartments upon neural activity. We found that stimulus-evoked CBV change mainly occurs in small arterioles, capillaries, and intracortical arteries and that the contribution from venules and ICVs is smaller. Our results confirm that VASO is less susceptible to large vessel effects compared to BOLD, as blood volume changes in intracortical arteries did not substantially affect the resulting depth-dependent VASO profiles, whereas depth-dependent BOLD profiles showed a bias towards signal contributions from intracortical veins.

Pre-processing of Sub-millimeter GE-BOLD fMRI Data for Laminar Applications

Over the past 30 years, brain function has primarily been evaluated non-invasively using functional magnetic resonance imaging (fMRI) with gradient-echo (GE) sequences to measure blood-oxygen-level-dependent (BOLD) signals. Despite the multiple advantages of GE sequences, e.g., higher signal-to-noise ratio, faster acquisitions, etc., their relatively inferior spatial localization compromises the routine use of GE-BOLD in laminar applications. Here, in an attempt to rescue the benefits of GE sequences, we evaluated the effect of existing pre-processing methods on the spatial localization of signals obtained with EPIK, a GE sequence that affords voxel volumes of 0.25 mm3 with near whole-brain coverage. The methods assessed here apply to both task and resting-state fMRI data assuming the availability of reconstructed magnitude and phase images.

Zooming-in on higher-level vision: High-resolution fMRI for understanding visual perception and awareness

One of the central questions in visual neuroscience is how the sparse retinal signals leaving our eyes are transformed into a rich subjective visual experience of the world. Invasive physiology studies, which offers the highest spatial resolution, have revealed many facts about the processing of simple visual features like contrast, color, and orientation, focusing on the early visual areas. At the same time, standard human fMRI studies with comparably coarser spatial resolution have revealed more complex, functionally specialized, and category-selective responses in higher visual areas. Although the visual system is the best understood among the sensory modalities, these two areas of research remain largely segregated. High-resolution fMRI opens up a possibility for linking them. On the one hand, it allows studying how the higher-level visual functions affect the fine-scale activity in early visual areas. On the other hand, it allows discovering the fine-scale functional organization of higher visual areas and exploring their functional connectivity with visual areas lower in the hierarchy. In this review, I will discuss examples of successful work undertaken in these directions using high-resolution fMRI and discuss where this method could be applied in the future to advance our understanding of the complexity of higher-level visual processing.

Ultra-high field fMRI reveals origins of feedforward and feedback activity within laminae of human ocular dominance columns

Ultra-high field MRI can functionally image the cerebral cortex of human subjects at the submillimeter scale of cortical columns and laminae. Here, we investigate both in concert, by imaging ocular dominance columns (ODCs) in primary visual cortex (V1) across different cortical depths. We ensured that putative ODC patterns in V1 (a) are stable across runs, sessions, and scanners located in different continents, (b) have a width (~1.3 mm) expected from post-mortem and animal work and (c) are absent at the retinotopic location of the blind spot. We then dissociated the effects of bottom-up thalamo-cortical input and attentional feedback processes on activity in V1 across cortical depth. Importantly, the separation of bottom-up information flows into ODCs allowed us to validly compare attentional conditions while keeping the stimulus identical throughout the experiment. We find that, when correcting for draining vein effects and using both model-based and model-free approaches, the effect of monocular stimulation is largest at deep and middle cortical depths. Conversely, spatial attention influences BOLD activity exclusively near the pial surface. Our findings show that simultaneous interrogation of columnar and laminar dimensions of the cortical fold can dissociate thalamocortical inputs from top-down processing, and allow the investigation of their interactions without any stimulus manipulation.