Federico De Martino

Combining multivariate voxel selection and support vector machines for mapping and classification of fMRI spatial patterns

In functional brain mapping, pattern recognition methods allow detecting multivoxel patterns of brain activation which are informative with respect to a subjects perceptual or cognitive state. The sensitivity of these methods, however, is greatly reduced when the proportion of voxels that convey the discriminative information is small compared to the total number of measured voxels. To reduce this dimensionality problem, previous studies employed univariate voxel selection or region-of-interest-based strategies as a preceding step to the application of machine learning algorithms. Here we employ a strategy for classifying functional imaging data based on a multivariate feature selection algorithm, Recursive Feature Elimination (RFE) that uses the training algorithm (support vector machine) recursively to eliminate irrelevant voxels and estimate informative spatial patterns. Generalization performances on test data increases while features/voxels are pruned based on their discrimination ability. In this article we evaluate RFE in terms of sensitivity of discriminative maps (Receiver Operative Characteristic analysis) and generalization performances and compare it to previously used univariate voxel selection strategies based on activation and discrimination measures. Using simulated fMRI data, we show that the recursive approach is suitable for mapping discriminative patterns and that the combination of an initial univariate activation-based (F-test) reduction of voxels and multivariate recursive feature elimination produces the best results, especially when differences between conditions have a low contrast-to-noise ratio. Furthermore, we apply our method to high resolution (2 x 2 x 2 mm(3)) data from an auditory fMRI experiment in which subjects were stimulated with sounds from four different categories. With these real data, our recursive algorithm proves able to detect and accurately classify multivoxel spatial patterns, highlighting the role of the superior temporal gyrus in encoding the information of sound categories. In line with the simulation results, our method outperforms univariate statistical analysis and statistical learning without feature selection.

Who Is Saying "What"? Brain-Based Decoding of Human Voice and Speech

Can we decipher speech content (“what” is being said) and speaker identity (“who” is saying it) from observations of brain activity of a listener? Here, we combine functional magnetic resonance imaging with a data-mining algorithm and retrieve what and whom a person is listening to from the neural fingerprints that speech and voice signals elicit in the listeners auditory cortex. These cortical fingerprints are spatially distributed and insensitive to acoustic variations of the input so as to permit the brain-based recognition of learned speech from unknown speakers and of learned voices from previously unheard utterances. Our findings unravel the detailed cortical layout and computational properties of the neural populations at the basis of human speech recognition and speaker identification.

Sound Categories Are Represented as Distributed Patterns in the Human Auditory Cortex

The ability to recognize sounds allows humans and animals to efficiently detect behaviorally relevant events, even in the absence of visual information. Sound recognition in the human brain has been assumed to proceed through several functionally specialized areas, culminating in cortical modules where category-specific processing is carried out. In the present high-resolution fMRI experiment, we challenged this model by using well-controlled natural auditory stimuli and by employing an advanced analysis strategy based on an iterative machine-learning algorithm that allows modeling of spatially distributed, as well as localized, response patterns. Sounds of cats, female singers, acoustic guitars, and tones were controlled for their time-varying spectral characteristics and presented to subjects at three different pitch levels. Sound category information--not detectable with conventional contrast-based methods analysis--could be detected with multivoxel pattern analyses and attributed to spatially distributed areas over the supratemporal cortices. A more localized pattern was observed for processing of pitch laterally to primary auditory areas. Our findings indicate that distributed neuronal populations within the human auditory cortices, including areas conventionally associated with lower-level auditory processing, entail categorical representations of sounds beyond their physical properties.

Comprehensive in vivo mapping of the human basal ganglia and thalamic connectome in individuals using 7T MRI.

Basal ganglia circuits are affected in neurological disorders such as Parkinsons disease (PD), essential tremor, dystonia and Tourette syndrome. Understanding the structural and functional connectivity of these circuits is critical for elucidating the mechanisms of the movement and neuropsychiatric disorders, and is vital for developing new therapeutic strategies such as deep brain stimulation (DBS). Knowledge about the connectivity of the human basal ganglia and thalamus has rapidly evolved over recent years through non-invasive imaging techniques, but has remained incomplete because of insufficient resolution and sensitivity of these techniques. Here, we present an imaging and computational protocol designed to generate a comprehensive in vivo and subject-specific, three-dimensional model of the structure and connections of the human basal ganglia. High-resolution structural and functional magnetic resonance images were acquired with a 7-Tesla magnet. Capitalizing on the enhanced signal-to-noise ratio (SNR) and enriched contrast obtained at high-field MRI, detailed structural and connectivity representations of the human basal ganglia and thalamus were achieved. This unique combination of multiple imaging modalities enabled the in-vivo visualization of the individual human basal ganglia and thalamic nuclei, the reconstruction of seven white-matter pathways and their connectivity probability that, to date, have only been reported in animal studies, histologically, or group-averaged MRI population studies. Also described are subject-specific parcellations of the basal ganglia and thalamus into sub-territories based on their distinct connectivity patterns. These anatomical connectivity findings are supported by functional connectivity data derived from resting-state functional MRI (R-fMRI). This work demonstrates new capabilities for studying basal ganglia circuitry, and opens new avenues of investigation into the movement and neuropsychiatric disorders, in individual human subjects.

Mapping the organization of axis of motion selective features in human area mt using high-field fmri

Functional magnetic resonance imaging (fMRI) at high magnetic fields has made it possible to investigate the columnar organization of the human brain in vivo with high degrees of accuracy and sensitivity. Until now, these results have been limited to the organization principles of early visual cortex (V1). While the middle temporal area (MT) has been the first identified extra-striate visual area shown to exhibit a columnar organization in monkeys, evidence of MTs columnar response properties and topographic layout in humans has remained elusive. Research using various approaches suggests similar response properties as in monkeys but failed to provide direct evidence for direction or axis of motion selectivity in human area MT. By combining state of the art pulse sequence design, high spatial resolution in all three dimensions (0.8 mm isotropic), optimized coil design, ultrahigh field magnets (7 Tesla) and novel high resolution cortical grid sampling analysis tools, we provide the first direct evidence for large-scale axis of motion selective feature organization in human area MT closely matching predictions from topographic columnar-level simulations.

Contextual Feedback to Superficial Layers of V1.

Summary Neuronal cortical circuitry comprises feedforward, lateral, and feedback projections, each of which terminates in distinct cortical layers [1–3]. In sensory systems, feedforward processing transmits signals from the external world into the cortex, whereas feedback pathways signal the brain’s inference of the world [4–11]. However, the integration of feedforward, lateral, and feedback inputs within each cortical area impedes the investigation of feedback, and to date, no technique has isolated the feedback of visual scene information in distinct layers of healthy human cortex. We masked feedforward input to a region of V1 cortex and studied the remaining internal processing. Using high-resolution functional brain imaging (0.8 mm3) and multivoxel pattern information techniques, we demonstrate that during normal visual stimulation scene information peaks in mid-layers. Conversely, we found that contextual feedback information peaks in outer, superficial layers. Further, we found that shifting the position of the visual scene surrounding the mask parametrically modulates feedback in superficial layers of V1. Our results reveal the layered cortical organization of external versus internal visual processing streams during perception in healthy human subjects. We provide empirical support for theoretical feedback models such as predictive coding [10, 12] and coherent infomax [13] and reveal the potential of high-resolution fMRI to access internal processing in sub-millimeter human cortex.

Processing of Natural Sounds in Human Auditory Cortex: Tonotopy, Spectral Tuning, and Relation to Voice Sensitivity

Auditory cortical processing of complex meaningful sounds entails the transformation of sensory (tonotopic) representations of incoming acoustic waveforms into higher-level sound representations (e.g., their category). However, the precise neural mechanisms enabling such transformations remain largely unknown. In the present study, we use functional magnetic resonance imaging (fMRI) and natural sounds stimulation to examine these two levels of sound representation (and their relation) in the human auditory cortex. In a first experiment, we derive cortical maps of frequency preference (tonotopy) and selectivity (tuning width) by mathematical modeling of fMRI responses to natural sounds. The tuning width maps highlight a region of narrow tuning that follows the main axis of Heschls gyrus and is flanked by regions of broader tuning. The narrowly tuned portion on Heschls gyrus contains two mirror-symmetric frequency gradients, presumably defining two distinct primary auditory areas. In addition, our analysis indicates that spectral preference and selectivity (and their topographical organization) extend well beyond the primary regions and also cover higher-order and category-selective auditory regions. In particular, regions with preferential responses to human voice and speech occupy the low-frequency portions of the tonotopic map. We confirm this observation in a second experiment, where we find that speech/voice selective regions exhibit a response bias toward the low frequencies characteristic of human voice and speech, even when responding to simple tones. We propose that this frequency bias reflects the selective amplification of relevant and category-characteristic spectral bands, a useful processing step for transforming a sensory (tonotopic) sound image into higher level neural representations.

Cortical Depth Dependent Functional Responses in Humans at 7T: Improved Specificity with 3D GRASE

Ultra high fields (7T and above) allow functional imaging with high contrast-to-noise ratios and improved spatial resolution. This, along with improved hardware and imaging techniques, allow investigating columnar and laminar functional responses. Using gradient-echo (GE) (T2* weighted) based sequences, layer specific responses have been recorded from human (and animal) primary visual areas. However, their increased sensitivity to large surface veins potentially clouds detecting and interpreting layer specific responses. Conversely, spin-echo (SE) (T2 weighted) sequences are less sensitive to large veins and have been used to map cortical columns in humans. T2 weighted 3D GRASE with inner volume selection provides high isotropic resolution over extended volumes, overcoming some of the many technical limitations of conventional 2D SE-EPI, whereby making layer specific investigations feasible. Further, the demonstration of columnar level specificity with 3D GRASE, despite contributions from both stimulated echoes and conventional T2 contrast, has made it an attractive alternative over 2D SE-EPI. Here, we assess the spatial specificity of cortical depth dependent 3D GRASE functional responses in human V1 and hMT by comparing it to GE responses. In doing so we demonstrate that 3D GRASE is less sensitive to contributions from large veins in superficial layers, while showing increased specificity (functional tuning) throughout the cortex compared to GE.

An anatomical and functional topography of human auditory cortical areas.

While advances in magnetic resonance imaging (MRI) throughout the last decades have enabled the detailed anatomical and functional inspection of the human brain non-invasively, to date there is no consensus regarding the precise subdivision and topography of the areas forming the human auditory cortex. Here, we propose a topography of the human auditory areas based on insights on the anatomical and functional properties of human auditory areas as revealed by studies of cyto- and myelo-architecture and fMRI investigations at ultra-high magnetic field (7 Tesla). Importantly, we illustrate that—whereas a group-based approach to analyze functional (tonotopic) maps is appropriate to highlight the main tonotopic axis—the examination of tonotopic maps at single subject level is required to detail the topography of primary and non-primary areas that may be more variable across subjects. Furthermore, we show that considering multiple maps indicative of anatomical (i.e., myelination) as well as of functional properties (e.g., broadness of frequency tuning) is helpful in identifying auditory cortical areas in individual human brains. We propose and discuss a topography of areas that is consistent with old and recent anatomical post-mortem characterizations of the human auditory cortex and that may serve as a working model for neuroscience studies of auditory functions.

Encoding of natural sounds at multiple spectral and temporal resolutions in the human auditory cortex.

Functional neuroimaging research provides detailed observations of the response patterns that natural sounds (e.g. human voices and speech, animal cries, environmental sounds) evoke in the human brain. The computational and representational mechanisms underlying these observations, however, remain largely unknown. Here we combine high spatial resolution (3 and 7 Tesla) functional magnetic resonance imaging (fMRI) with computational modeling to reveal how natural sounds are represented in the human brain. We compare competing models of sound representations and select the model that most accurately predicts fMRI response patterns to natural sounds. Our results show that the cortical encoding of natural sounds entails the formation of multiple representations of sound spectrograms with different degrees of spectral and temporal resolution. The cortex derives these multi-resolution representations through frequency-specific neural processing channels and through the combined analysis of the spectral and temporal modulations in the spectrogram. Furthermore, our findings suggest that a spectral-temporal resolution trade-off may govern the modulation tuning of neuronal populations throughout the auditory cortex. Specifically, our fMRI results suggest that neuronal populations in posterior/dorsal auditory regions preferably encode coarse spectral information with high temporal precision. Vice-versa, neuronal populations in anterior/ventral auditory regions preferably encode fine-grained spectral information with low temporal precision. We propose that such a multi-resolution analysis may be crucially relevant for flexible and behaviorally-relevant sound processing and may constitute one of the computational underpinnings of functional specialization in auditory cortex.