A cerebellar granule cell-climbing fiber computation to learn to track long time intervals.

In classical cerebellar learning, Purkinje cells (PkCs) associate climbing fiber (CF) error signals with predictive granule cells (GrCs) that were active just prior (∼150 ms). The cerebellum also contributes to behaviors characterized by longer timescales. To investigate how GrC-CF-PkC circuits might learn seconds-long predictions, we imaged simultaneous GrC-CF activity over days of forelimb operant conditioning for delayed water reward. As mice learned reward timing, numerous GrCs developed anticipatory activity ramping at different rates until reward delivery, followed by widespread time-locked CF spiking. Relearning longer delays further lengthened GrC activations. We computed CF-dependent GrC→PkC plasticity rules, demonstrating that reward-evoked CF spikes sufficed to grade many GrC synapses by anticipatory timing. We predicted and confirmed that PkCs could thereby continuously ramp across seconds-long intervals from movement to reward. Learning thus leads to new GrC temporal bases linking predictors to remote CF reward signals-a strategy well suited for learning to track the long intervals common in cognitive domains.

Multi-level spatial-temporal and attentional information deep fusion network for retinal vessel segmentation.

Objective.Automatic segmentation of fundus vessels has the potential to enhance the judgment ability of intelligent disease diagnosis systems. Even though various methods have been proposed, it is still a demanding task to accurately segment the fundus vessels. The purpose of our study is to develop a robust and effective method to segment the vessels in human color retinal fundus images.Approach.We present a novel multi-level spatial-temporal and attentional information deep fusion network for the segmentation of retinal vessels, called MSAFNet, which enhances segmentation performance and robustness. Our method utilizes the multi-level spatial-temporal encoding module to obtain spatial-temporal information and the Self-Attention module to capture feature correlations in different levels of our network. Based on the encoder and decoder structure, we combine these features to get the final segmentation results.Main results.Through abundant experiments on four public datasets, our method achieves preferable performance compared with other SOTA retinal vessel segmentation methods. Our Accuracy and Area Under Curve achieve the highest scores of 96.96%, 96.57%, 96.48% and 98.78%, 98.54%, 98.27% on DRIVE, CHASE_DB1, and HRF datasets. Our Specificity achieves the highest score of 98.58% and 99.08% on DRIVE and STARE datasets.Significance.The experimental results demonstrate that our method has strong learning and representation capabilities and can accurately detect retinal blood vessels, thereby serving as a potential tool for assisting in diagnosis.

Discovering the neuronal dynamics in major depressive disorder using Hidden Markov Model.

Introduction: Major Depressive Disorder (MDD) is a leading cause of worldwide disability, and standard clinical treatments have limitations due to the absence of neurological evidence. Electroencephalography (EEG) monitoring is an effective method for recording neural activities and can provide electroneurophysiological evidence of MDD. Methods: In this work, we proposed a probabilistic graphical model for neural dynamics decoding on MDD patients and healthy controls (HC), utilizing the Hidden Markov Model with Multivariate Autoregressive observation (HMM-MAR). We testified the model on the MODMA dataset, which contains resting-state and task-state EEG data from 53 participants, including 24 individuals with MDD and 29 HC. Results: The experimental results suggest that the state time courses generated by the proposed model could regress the Patient Health Questionnaire-9 (PHQ-9) score of the participants and reveal differences between the MDD and HC groups. Meanwhile, the Markov property was observed in the neuronal dynamics of participants presented with sad face stimuli. Coherence analysis and power spectrum estimation demonstrate consistent results with the previous studies on MDD. Discussion: In conclusion, the proposed HMM-MAR model has revealed its potential capability to capture the neuronal dynamics from EEG signals and interpret brain disease pathogenesis from the perspective of state transition. Compared with the previous machine-learning or deep-learning-based studies, which regarded the decoding model as a black box, this work has its superiority in the spatiotemporal pattern interpretability by utilizing the Hidden Markov Model.

Odor encoding by signals in the olfactory bulb.

To understand the operation of the olfactory system, it is essential to know how information is encoded in the olfactory bulb. We applied Shannon information theoretic methods to address this, with signals from up to 57 glomeruli simultaneously optically imaged from presynaptic inputs in glomeruli in the mouse dorsal (dOB) and lateral (lOB) olfactory bulb, in response to six exemplar pure chemical odors. We discovered that, first, the tuning of these signals from glomeruli to a set of odors is remarkably broad, with a mean sparseness of 0.83 and a mean signal correlation of 0.64. Second, both of these factors contribute to the low information that is available from the responses of even populations of many tens of glomeruli, which was only 1.35 bits across 33 glomeruli on average, compared with the 2.58 bits required to perfectly encode these six odors. Third, although there is considerable interest in the possibility of temporal encoding of stimulus including odor identity, the amount of information in the temporal aspects of the presynaptic glomerular responses was low (mean 0.11 bits) and, importantly, was redundant with respect to the information available from the rates. Fourth, the information from simultaneously recorded glomeruli asymptotes very gradually and nonlinearly, showing that glomeruli do not have independent responses. Fifth, the information from a population became available quite rapidly, within 100 ms of sniff onset, and the peak of the glomerular response was at 200 ms. Sixth, the information from the lOB was not additive with that of the dOB.NEW & NOTEWORTHY We report broad tuning and low odor information available across the lateral and dorsal bulb populations of glomeruli. Even though response latencies can be significantly predictive of stimulus identity, such contained very little information and none that was not redundant with information based on rate coding alone. Last, in line with the emerging notion of the important role of earliest stages of responses ("primacy"), we report a very rapid rise in information after each inhalation.

Arterial spin labeling using spatio-temporal encoding readout for robust perfusion imaging in inhomogenous magnetic fields.

PURPOSE: To evaluate the feasibility of spatio-temporal encoding (SPEN) readout for pseudo-continuous ASL (pCASL) in brain, and its robustness to susceptibility artifacts as introduced by aneurysm clips. METHODS: A 2D self-refocused T2 *-compensated hybrid SPEN scheme, with super-resolution reconstruction was implemented on a 1.5T Philips system. Q (=BWchirp *Tchirp ) was varied and, the aneurysm clip-induced artifact was evaluated in phantom (label-images) as well as in vivo (perfusion-weighted signal (PWS)-maps and temporal SNR (tSNR)). In vivo results were compared to gradient-echo EPI (GE-EPI) and spin-echo EPI (SE-EPI). The dependence of tSNR on TR was evaluated separately for SPEN and SE-EPI. SPEN with Q Ëœ 75 encodes with the same off-resonance robustness as EPI. RESULTS: The clip-induced artifact with SPEN decreased with increase in Q, and was smaller compared to SE-EPI and GE-EPI in vivo. tSNR decreased with Q and the tSNR of GE-EPI and SE-EPI corresponded to SPEN with a Q-value of approximately ˜85 and ˜108, respectively. In addition, SPEN perfusion images showed a higher tSNR (p < 0.05) for TR = 4000 ms compared to TR = 2100 ms, while SE-EPI did not. tSNR remained relatively stable when the time between SPEN-excitation and start of the next labeling-module was more than ˜1000 ms. CONCLUSION: Feasibility of combining SPEN with pCASL imaging was demonstrated, enabling cerebral perfusion measurements with a higher robustness to field inhomogeneity (Q > 75) compared to SE-EPI and GE-EPI. However, the SPEN chirp-pulse saturates incoming blood, thereby reducing pCASL labeling efficiency of the next acquisition for short TRs. Future developments are needed to enable 3D scanning.

See, Hear, or Feel - to Speak: A Versatile Multiple-Choice Functional Near-Infrared Spectroscopy-Brain-Computer Interface Feasible With Visual, Auditory, or Tactile Instructions.

Severely motor-disabled patients, such as those suffering from the so-called "locked-in" syndrome, cannot communicate naturally. They may benefit from brain-computer interfaces (BCIs) exploiting brain signals for communication and therewith circumventing the muscular system. One BCI technique that has gained attention recently is functional near-infrared spectroscopy (fNIRS). Typically, fNIRS-based BCIs allow for brain-based communication via voluntarily modulation of brain activity through mental task performance guided by visual or auditory instructions. While the development of fNIRS-BCIs has made great progress, the reliability of fNIRS-BCIs across time and environments has rarely been assessed. In the present fNIRS-BCI study, we tested six healthy participants across three consecutive days using a straightforward four-choice fNIRS-BCI communication paradigm that allows answer encoding based on instructions using various sensory modalities. To encode an answer, participants performed a motor imagery task (mental drawing) in one out of four time periods. Answer encoding was guided by either the visual, auditory, or tactile sensory modality. Two participants were tested outside the laboratory in a cafeteria. Answers were decoded from the time course of the most-informative fNIRS channel-by-chromophore combination. Across the three testing days, we obtained mean single- and multi-trial (joint analysis of four consecutive trials) accuracies of 62.5 and 85.19%, respectively. Obtained multi-trial accuracies were 86.11% for visual, 80.56% for auditory, and 88.89% for tactile sensory encoding. The two participants that used the fNIRS-BCI in a cafeteria obtained the best single- (72.22 and 77.78%) and multi-trial accuracies (100 and 94.44%). Communication was reliable over the three recording sessions with multi-trial accuracies of 86.11% on day 1, 86.11% on day 2, and 83.33% on day 3. To gauge the trade-off between number of optodes and decoding accuracy, averaging across two and three promising fNIRS channels was compared to the one-channel approach. Multi-trial accuracy increased from 85.19% (one-channel approach) to 91.67% (two-/three-channel approach). In sum, the presented fNIRS-BCI yielded robust decoding results using three alternative sensory encoding modalities. Further, fNIRS-BCI communication was stable over the course of three consecutive days, even in a natural (social) environment. Therewith, the developed fNIRS-BCI demonstrated high flexibility, reliability and robustness, crucial requirements for future clinical applicability.

Efficient learning with augmented spikes: A case study with image classification.

Efficient learning of spikes plays a valuable role in training spiking neural networks (SNNs) to have desired responses to input stimuli. However, current learning rules are limited to a binary form of spikes. The seemingly ubiquitous phenomenon of burst in nervous systems suggests a new way to carry more information with spike bursts in addition to times. Based on this, we introduce an advanced form, the augmented spikes, where spike coefficients are used to carry additional information. How could neurons learn and benefit from augmented spikes remains unclear. In this paper, we propose two new efficient learning rules to process spatiotemporal patterns composed of augmented spikes. Moreover, we examine the learning abilities of our methods with a synthetic recognition task of augmented spike patterns and two practical ones for image classification. Experimental results demonstrate that our rules are capable of extracting information carried by both the timing and coefficient of spikes. Our proposed approaches achieve remarkable performance and good robustness under various noise conditions, as compared to benchmarks. The improved performance indicates the merits of augmented spikes and our learning rules, which could be beneficial and generalized to a broad range of spike-based platforms.

The Same Hippocampal CA1 Population Simultaneously Codes Temporal Information over Multiple Timescales.

It has long been hypothesized that a primary function of the hippocampus is to discover and exploit temporal relationships between events. Previously, it has been reported that sequences of "time cells" in the hippocampus extend for tens of seconds. Other studies have shown that neuronal firing in the hippocampus fluctuates over hours and days. Both of these mechanisms could enable temporal encoding of events over very different timescales. However, thus far, these two classes of phenomena have never been observed simultaneously, which is necessary to ascribe broad-range temporal coding to the hippocampus. Using in vivo calcium imaging in unrestrained mice, we observed sequences of hippocampal neurons that bridged a 10 s delay. Similar sequences were observed over multiple days, but the set of neurons participating in those sequences changed gradually. Thus, the same population of neurons that encodes temporal information over seconds can also be used to distinguish periods of time over much longer timescales. These results unify two previously separate paradigms of temporal processing in the hippocampus that support episodic memory.

Duration adaptation modulates EEG correlates of subsequent temporal encoding.

Repetitive exposure to relatively long or short sensory events has been shown to shorten or lengthen the perceived duration of a subsequent event. However, the neural basis of this phenomenon, called duration adaptation, remains unclear. In this study, we used electroencephalography (EEG) to investigate whether duration adaptation could modulate the subsequent temporal encoding represented by the contingent negative variation (CNV). Participants were asked to reproduce the duration of a test stimulus after adapting duration (Experiment 1) or after anchor duration (Experiment 2). We found that both adapting duration and anchor duration affected the reproduction duration of a subsequently presented test stimulus. The simultaneously recorded event-related potentials (ERPs) revealed that test stimuli evoked clearly identifiable N1, P2 and CNV components in the fronto-central scalp. Further analyses showed that the CNV amplitude was modulated by duration adaptation: adaptation to shorter duration (200ms) increased whereas adaptation to longer duration (800ms) decreased the CNV amplitude. These findings suggest that the neural correlates of temporal encoding represented by the CNV amplitude reflect the duration aftereffect. Additionally, the duration adaptation effect observed on the P2 component also suggests an early expectancy effect on subsequent encoding processes. Finally, no effect of anchor duration was observed on the CNV amplitude, which implies that different mechanisms underlie the duration aftereffect and the anchor effect.

Analysis of Nociceptive Information Encoded in the Temporal Discharge Patterns of Cutaneous C-Fibers.

The generation of pain signals from primary afferent neurons is explained by a labeled-line code. However, this notion cannot apply in a simple way to cutaneous C-fibers, which carry signals from a variety of receptors that respond to various stimuli including agonist chemicals. To represent the discharge patterns of C-fibers according to different agonist chemicals, we have developed a quantitative approach using three consecutive spikes. By using this method, the generation of pain in response to chemical stimuli is shown to be dependent on the temporal aspect of the spike trains. Furthermore, under pathological conditions, gamma-aminobutyric acid resulted in pain behavior without change of spike number but with an altered discharge pattern. Our results suggest that information about the agonist chemicals may be encoded in specific temporal patterns of signals in C-fibers, and nociceptive sensation may be influenced by the extent of temporal summation originating from the temporal patterns.

Dual temporal encoding mechanisms in human auditory cortex: Evidence from MEG and EEG.

Current hypotheses about language processing advocate an integral relationship between encoding of temporal information and linguistic processing in the brain. All such explanations must accommodate the evident ability of the perceptual system to process both slow and fast time scales in speech. However most cortical neurons are limited in their capability to precisely synchronise to temporal modulations at rates faster than about 50Hz. Hence, a central question in auditory neurophysiology concerns how the full range of perceptually relevant modulation rates might be encoded in the cerebral cortex. Here we show with concurrent noninvasive magnetoencephalography (MEG) and electroencephalography (EEG) measurements that the human auditory cortex transitions between a phase-locked (PL) mode of responding to modulation rates below about 50Hz, and a non-phase-locked (NPL) mode at higher rates. Precisely such dual response modes are predictable from the behaviours of single neurons in auditory cortices of non-human primates. Our data point to a common mechanistic explanation for the single neuron and MEG/EEG results and support the hypothesis that two distinct types of neuronal encoding mechanisms are employed by the auditory cortex to represent a wide range of temporal modulation rates. This dual encoding model allows slow and fast modulations in speech to be processed in parallel and is therefore consistent with theoretical frameworks in which slow temporal modulations (such as rhythm or syllabic structure) are akin to the contours or edges of visual objects, whereas faster modulations (such as periodicity pitch or phonemic structure) are more like visual texture.

The duality of temporal encoding - the intrinsic and extrinsic representation of time.

While time is well acknowledged for having a fundamental part in our perception, questions on how it is represented are still matters of great debate. One of the main issues in question is whether time is represented intrinsically at the neural level, or is it represented within dedicated brain regions. We used an fMRI block design to test if we can impose covert encoding of temporal features of faces and natural scenes stimuli within category selective neural populations by exposing subjects to four types of temporal variance, ranging from 0% up to 50% variance. We found a gradual increase in neural activation associated with the gradual increase in temporal variance within category selective areas. A second level analysis showed the same pattern of activations within known brain regions associated with time representation, such as the Cerebellum, the Caudate, and the Thalamus. We concluded that temporal features are integral to perception and are simultaneously represented within category selective regions and globally within dedicated regions. Our second conclusion, drown from our covert procedure, is that time encoding, at its basic level, is an automated process that does not require attention allocated toward the temporal features nor does it require dedicated resources.

Input coding for neuro-electronic hybrid systems.

Liquid State Machines have been proposed as a framework to explore the computational properties of neuro-electronic hybrid systems (Maass et al., 2002). Here the neuronal culture implements a recurrent network and is followed by an array of linear discriminants implemented using perceptrons in electronics/software. Thus in this framework, it is desired that the outputs of the neuronal network, corresponding to different inputs, be linearly separable. Previous studies have demonstrated this by either using only a small set of input stimulus patterns to the culture (Hafizovic et al., 2007), large number of input electrodes (Dockendorf et al., 2009) or by using complex schemes to post-process the outputs of the neuronal culture prior to linear discriminance (Ortman et al., 2011). In this study we explore ways to temporally encode inputs into stimulus patterns using a small set of electrodes such that the neuronal culture's output can be directly decoded by simple linear discriminants based on perceptrons. We demonstrate that network can detect the timing and order of firing of inputs on multiple electrodes. Based on this, we demonstrate that the neuronal culture can be used as a kernel to transform inputs which are not linearly separable in a low dimensional space, into outputs in a high dimension where they are linearly separable. Thus simple linear discriminants can now be directly connected to outputs of the neuronal culture and allow for implementation of any function for such a hybrid system.

Effects of congenital blindness on the subcortical representation of speech cues.

Human modalities play a vital role in the way the brain produces mental representations of the world around us. Although congenital blindness limits the understanding of the environment in some aspects, blind individuals may have other superior capabilities from long-term experience and neural plasticity. This study investigated the effects of congenital blindness on temporal and spectral neural encoding of speech at the subcortical level. The study included 26 congenitally blind individuals and 24 normal-sighted individuals with normal hearing. Auditory brainstem response (ABR) was recorded with both click and speech synthetic 40-ms /da/ stimuli. No significant difference was observed between the two groups in wave latencies or amplitudes of click ABR. Latencies of speech ABR D (p=0.012) and O (p=0.014) waves were significantly shorter in blind individuals than in normal-sighted individuals. Amplitudes of the A (p<0.001) and E (p=0.001) speech ABR (sABR) waves were also significantly higher in blind subjects. Blind individuals had significantly better results for duration (p<0.001) amplitude (p=0.015) and slope of the V-A complex (p=0.004), signal-to-noise ratio (p<0.001), and amplitude of the stimulus fundamental frequency (F0) (p=0.009), first formant (F1) (p<0.001) and higher-frequency region (HF) (p<0.001) ranges. Results indicate that congenitally blind subjects have improved hearing function in response to the /da/ syllable in both source and filter classes of sABR. It is possible that these subjects have enhanced neural representation of vocal cord vibrations and improved neural synchronization in temporal encoding of the onset and offset parts of speech stimuli at the brainstem level. This may result from the compensatory mechanism of neural reorganization in blind subjects influenced from top-down corticofugal connections with the auditory cortex.

Super-resolved parallel MRI by spatiotemporal encoding.

Recent studies described an "ultrafast" scanning method based on spatiotemporal (SPEN) principles. SPEN demonstrates numerous potential advantages over EPI-based alternatives, at no additional expense in experimental complexity. An important aspect that SPEN still needs to achieve for providing a competitive ultrafast MRI acquisition alternative, entails exploiting parallel imaging algorithms without compromising its proven capabilities. The present work introduces a combination of multi-band frequency-swept pulses simultaneously encoding multiple, partial fields-of-view, together with a new algorithm merging a Super-Resolved SPEN image reconstruction and SENSE multiple-receiving methods. This approach enables one to reduce both the excitation and acquisition times of sub-second SPEN acquisitions by the customary acceleration factor R, without compromises in either the method's spatial resolution, SAR deposition, or capability to operate in multi-slice mode. The performance of these new single-shot imaging sequences and their ancillary algorithms were explored and corroborated on phantoms and human volunteers at 3 T. The gains of the parallelized approach were particularly evident when dealing with heterogeneous systems subject to major T2/T2* effects, as is the case upon single-scan imaging near tissue/air interfaces.

Direct design of 2D RF pulses using matrix inversion.

Multi-dimensional pulses are frequently used in MRI for applications such as targeted excitation, fat-water separation or metabolic imaging with hyperpolarised (13)C compounds. For the design, the problem is typically separated into the different dimensions. In this work, a method to directly design two-dimensional pulses using the small-tip angle approximation is introduced based on a direct matrix representation. The numerical problem is solved in a single step directly in two dimensions by matrix inversion. Exemplary spectral-spatial excitation and spatio-temporal encoding (SPEN) pulses are designed and validated. The main benefits of the direct design approach include a reduction of artefacts in case of spectral-spatial pulses, a simple and straightforward computer implementation and high flexibility in the pulse design.

Pheromone transduction in moths.

Calling female moths attract their mates late at night with intermittent release of a species-specific sex-pheromone blend. Mean frequency of pheromone filaments encodes distance to the calling female. In their zig-zagging upwind search male moths encounter turbulent pheromone blend filaments at highly variable concentrations and frequencies. The male moth antennae are delicately designed to detect and distinguish even traces of these sex pheromones amongst the abundance of other odors. Its olfactory receptor neurons sense even single pheromone molecules and track intermittent pheromone filaments of highly variable frequencies up to about 30 Hz over a wide concentration range. In the hawkmoth Manduca sexta brief, weak pheromone stimuli as encountered during flight are detected via a metabotropic PLCß-dependent signal transduction cascade which leads to transient changes in intracellular Ca(2+) concentrations. Strong or long pheromone stimuli, which are possibly perceived in direct contact with the female, activate receptor-guanylyl cyclases causing long-term adaptation. In addition, depending on endogenous rhythms of the moth's physiological state, hormones such as the stress hormone octopamine modulate second messenger levels in sensory neurons. High octopamine levels during the activity phase maximize temporal resolution cAMP-dependently as a prerequisite to mate location. Thus, I suggest that sliding adjustment of odor response threshold and kinetics is based upon relative concentration ratios of intracellular Ca(2+) and cyclic nucleotide levels which gate different ion channels synergistically. In addition, I propose a new hypothesis for the cyclic nucleotide-dependent ion channel formed by insect olfactory receptor/coreceptor complexes. Instead of being employed for an ionotropic mechanism of odor detection it is proposed to control subthreshold membrane potential oscillation of sensory neurons, as a basis for temporal encoding of odors.