Visualization for epidemiological modelling: challenges, solutions, reflections and recommendations.

We report on an ongoing collaboration between epidemiological modellers and visualization researchers by documenting and reflecting upon knowledge constructs-a series of ideas, approaches and methods taken from existing visualization research and practice-deployed and developed to support modelling of the COVID-19 pandemic. Structured independent commentary on these efforts is synthesized through iterative reflection to develop: evidence of the effectiveness and value of visualization in this context; open problems upon which the research communities may focus; guidance for future activity of this type and recommendations to safeguard the achievements and promote, advance, secure and prepare for future collaborations of this kind. In describing and comparing a series of related projects that were undertaken in unprecedented conditions, our hope is that this unique report, and its rich interactive supplementary materials, will guide the scientific community in embracing visualization in its observation, analysis and modelling of data as well as in disseminating findings. Equally we hope to encourage the visualization community to engage with impactful science in addressing its emerging data challenges. If we are successful, this showcase of activity may stimulate mutually beneficial engagement between communities with complementary expertise to address problems of significance in epidemiology and beyond. See https://ramp-vis.github.io/RAMPVIS-PhilTransA-Supplement/. This article is part of the theme issue 'Technical challenges of modelling real-life epidemics and examples of overcoming these'.

Ultrafast Structural Dynamics along the ß-γ Phase Transition Path in MnAs.

We investigate the orthorhombic distortion and the structural dynamics of epitaxial MnAs layers on GaAs(001) using static and time-resolved x-ray diffraction. Laser-induced intensity oscillations of Bragg reflections allow us to identify the optical phonon associated with orthorhombic distortion and to follow its softening along the path towards an undistorted phase of hexagonal symmetry. The frequency of this mode falls in the THz range, in agreement with recent calculations. Incomplete softening suggests that the ß-γ transformation deviates from a purely second-order displacive transition.

Differential roles for parietal and frontal cortices in fixed versus evolving temporal expectations: Dissociating prior from posterior temporal probabilities with fMRI.

The ability to predict when an event will occur allows us to respond optimally to that event. Temporal predictability can be either fixed (prior probability) or evolving (posterior probability), in which case it is dynamically updated as a function of the elapse of time itself ("hazard function"). We used fMRI to identify the brain regions involved in either form of temporal prediction, within a single experimental paradigm. Participants performed a cued reaction time (RT) task, in which the target appeared after one of four intervals ("foreperiods") that was either predictable (temporal condition) or variable (neutral condition). As expected, RTs were faster in temporal versus neutral conditions, indicating the behavioural benefit of fixed temporal predictability. RTs also got faster as a function of foreperiod in the neutral, but not temporal, condition, reflecting the evolving temporal predictability of the hazard function. We confirmed that left inferior parietal cortex was preferentially activated by the fixed temporal predictability of temporal (versus neutral) cues. Then, by directly comparing how activity varied as a function of foreperiod in the neutral versus time conditions, we identified the neural substrates of the changes in temporal probability defined by the hazard function, while simultaneously controlling for changes related simply to the elapse of time itself. Whole-brain fMRI analyses (independently confirmed by anatomically guided ROI analyses) showed that activity in left inferior parietal cortex tracked the evolving temporal probabilities of the hazard function. ROI analysis further revealed a similar role for right inferior frontal cortex. Our data highlight a key role for left parietal cortex in instantiating the behavioural benefits of temporal predictability, whether predictions are fixed or dynamically evolving.

SMA Selectively Codes the Active Accumulation of Temporal, Not Spatial, Magnitude.

Estimating duration depends on the sequential integration (accumulation) of temporal information in working memory. Using fMRI, we directly compared the accumulation of information in temporal versus spatial domains. Participants estimated either the duration or distance of the dynamic trajectory of a moving dot or, in a control condition, a static line stimulus. Comparing the duration versus distance of static lines activated an extensive cortico-striatal network. By contrast, comparing the duration versus distance of dynamic trajectories, both of which required sequential integration of information, activated SMA alone. Indeed, activity in SMA, as well as right inferior occipital cortex, increased parametrically as a function of stimulus duration and also correlated with individual differences in the propensity to overestimate stimulus duration. By contrast, activity in primary visual cortex increased parametrically as a function of stimulus distance. Crucially, a direct comparison of the parametric responses to duration versus distance revealed that activity in SMA increased incrementally as a function of stimulus duration but not as a function of stimulus distance. Collectively, our results indicate that SMA responds to the active accumulation of information selectively in the temporal domain.

Effects of hyperbaric nitrogen-induced narcosis on response-selection processes.

Certain underwater circumstances carry risk of inert gas narcosis. Impairment of sensorimotor information processing due to narcosis, induced by normobaric nitrous oxide or high partial nitrogen pressure, has been broadly evidenced, by a lengthening of the reaction time (RT). However, the locus of this effect remains a matter of debate. We examined whether inert gas narcosis affects the response-selection stage of sensorimotor information processing. We compared an air normobaric condition with a hyperbaric condition in which 10 subjects were subjected to 6 absolute atmospheres of 8.33% O2 Nitrox. In both conditions, subjects performed a between-hand choice-RT task in which we explicitly manipulated the stimulus-response association rule. The effect of this manipulation (which is supposed to affect response-selection processes) was modified by inert gas narcosis. It is concluded, therefore, that response selection processes are among the loci involved in the effect of inert gas narcosis on information processing.

Why is there an error negativity on correct trials? A reappraisal.

In healthy subjects, the Error Negativity (Ne) was initially reported on errors and on partial errors, only. Later on, application of the Laplacian transformation to EEG data unmasked a Ne-like wave (Nc) that shares a main generator with the Ne, suggesting that the Nc is just a small Ne. However, the reason why a small Ne would persist on correct responses remains unclear. Now, sometimes, subthreshold EMG activations in the muscles corresponding to correct responses (not strong enough to reach the response threshold) can precede full-blown correct responses. These "partially correct" activities seem to correspond to (force) execution errors, as they evoke a sizeable Ne. Within the frames of the Reward Value and Prediction Model or of the Predicted Response-Outcome model we propose that the action monitoring system evokes a Ne/Nc on correct responses because, even when a correct choice has been made, the accuracy of response (force) execution cannot be fully predicted. If this interpretation is correct, it can be assumed that, once these execution errors have been corrected, the correctness of the (full-blown) correcting response is highly predictable. Consequently, they should evoke a smaller Nc/Ne than "pure" correct responses. We show, that for the response thresholds set in the present experiment, the correcting response of the trials containing a partially correct activation evoke no identifiable Nc at all. Therefore it seems that there usually is an Error Negativity on correct trials because the correctness of response (force) execution cannot be fully predicted.

Top-down modulation of brain activity underlying intentional action and its relationship with awareness of intention: an ERP/Laplacian analysis.

Intentional actions are executed with the peculiar experience of "I decide to do that." It has been proposed that intentional actions involve a specific brain network involving the supplementary motor areas (SMAs). Here, we manipulated the internal representation participants attended to (intention vs. movement) in order to (1) examine the activity of SMAs and of the primary motor cortex (M1) during intentional action preparation and execution, and (2) investigate the temporal relationship between activity in these structures and intention awareness. Participants performed self-paced key presses. After each key press, participants were asked to report either the time they had the first intention to press the key (W-condition) or the time they actually started the movement (M-condition). We then estimated surface Laplacians from brain electrical potentials recorded while participants were performing the task. Activity in SMAs was greater in the W-condition than in the M-condition more than 1 s before electromyographic (EMG) activation, suggesting that this region is indeed associated to the formation of conscious intention. Conversely, activity in primary motor cortex (M1) contralateral to the responding hand was larger in the M-condition than in the W-condition, revealing that this region is also modulated by top-down processes. In addition, waveforms time-locked to the W-judgement revealed that M1 as well as EMG activation preceded the time at which participants become aware of their intention by about 0.3 s. This observation argues against the possibility that the temporal delay between motor-related activation and intention awareness results from smearing artifacts.

Simulation of X-ray projections on GPU: Benchmarking gVirtualXray with clinically realistic phantoms.

BACKGROUND AND OBJECTIVES: This study provides a quantitative comparison of images created using gVirtualXray (gVXR) to both Monte Carlo (MC) and real images of clinically realistic phantoms. gVirtualXray is an open-source framework that relies on the Beer-Lambert law to simulate X-ray images in realtime on a graphics processor unit (GPU) using triangular meshes. METHODS: Images are generated with gVirtualXray and compared with a corresponding ground truth image of an anthropomorphic phantom: (i) an X-ray projection generated using a Monte Carlo simulation code, (ii) real digitally reconstructed radiographs (DRRs), (iii) computed tomography (CT) slices, and (iv) a real radiograph acquired with a clinical X-ray imaging system. When real images are involved, the simulations are used in an image registration framework so that the two images are aligned. RESULTS: The mean absolute percentage error (MAPE) between the images simulated with gVirtualXray and MC is 3.12%, the zero-mean normalised cross-correlation (ZNCC) is 99.96% and the structural similarity index (SSIM) is 0.99. The run-time is 10 days for MC and 23 ms with gVirtualXray. Images simulated using surface models segmented from a CT scan of the Lungman chest phantom were similar to (i) DRRs computed from the CT volume and (ii) an actual digital radiograph. CT slices reconstructed from images simulated with gVirtualXray were comparable to the corresponding slices of the original CT volume. CONCLUSIONS: When scattering can be ignored, accurate images that would take days using MC can be generated in milliseconds with gVirtualXray. This speed of execution enables the use of repetitive simulations with varying parameters, e.g. to generate training data for a deep-learning algorithm, and to minimise the objective function of an optimisation problem in image registration. The use of surface models enables the combination of X-ray simulation with real-time soft-tissue deformation and character animation, which can be deployed in virtual reality applications.

Dynamics of executive control and motor deficits in parkinsonian rats.

While there is general agreement that in Parkinson's disease (PD), striatal dopamine (DA) depletion causes motor deficits, the origin of the associated cognitive impairments remains a matter of debate. The present study aimed to decipher the influence of a partial 6-hydroxydopamine (6-OHDA) lesion of striatal DA nerve terminals in rats performing a reaction time task previously used to assess cognitive deficits in PD patients. The effects of two behavioral manipulations-foreperiod duration and stimulus-response congruence-known to affect motor processes and executive control, respectively, were studied over 8 weeks postsurgery in control and lesion animals. Two weeks after surgery, the lesion abolished the effect of foreperiod, confirming the direct involvement of striatal DA in motor processes, but failed to alter the effect of congruence. During the following weeks, the effect of foreperiod was reinstated, indicating a recovery of lesion-induced motor symptoms. This recovery was accompanied by a progressive increase of the congruence effect, signaling an executive control deficit in lesion animals. This result provides the first evidence that 6-OHDA lesioned rats exhibit the same cognitive impairment as PD patients in this task. The deficit, however, built up progressively after the lesion and may result from adaptations mitigating lesion-induced motor deficits.

A method to compute respiration parameters for patient-based simulators.

We propose a method to automatically tune a patient-based virtual environment training simulator for abdominal needle insertion. The key attributes to be customized in our framework are the elasticity of soft-tissues and the respiratory model parameters. The estimation is based on two 3D Computed Tomography (CT) scans of the same patient at two different time steps. Results are presented on four patients and show that our new method leads to better results than our previous studies with manually tuned parameters.

Interactive teaching environment for diagnostic radiography with real-time X-ray simulation and patient positioning.

PURPOSE: Traditional undergraduate radiographer training mixes academic lectures and clinical practice. Our goal is to bridge the current disconnection between theory and practice in a safe environment, avoiding the risk of radiation for both practitioners and patients. To this end, this research proposes a new software to teach diagnostic radiography using real-time interactive X-ray simulation and patient positioning. METHODS: The proposed medical simulator is composed of three main modules. A fast and accurate character animation technique is in charge of simulating the patient positioning phase and adapts their internal anatomy accordingly. gVirtualXRay is an open-source X-ray simulation library and generates the corresponding radiographs in real time. Finally, the courseware allows going through all the diagnostic radiology steps from the patient positioning and the machine configuration to the final image enhancing. RESULTS: A face and content validation study has been conducted; 18 radiology professionals were recruited to evaluate our software using a questionnaire. The results show that our tool is realistic in many ways (72% of the participants agreed that the simulations are visually realistic), useful (67%) and suitable (78%) for teaching X-ray radiography. CONCLUSIONS: The proposed tool allows simulating the most relevant steps of the projectional radiography procedure. The virtual patient posing system and X-ray simulation module execute at interactive rates. These features enable the lectures to show their students the results of good and bad practices in a classroom environment, avoiding radiation risk.

Orienting attention in time activates left intraparietal sulcus for both perceptual and motor task goals.

Attention can be directed not only toward a location in space but also to a moment in time ("temporal orienting"). Temporally informative cues allow subjects to predict when an imminent event will occur, thereby speeding responses to that event. In contrast to spatial orienting, temporal orienting preferentially activates left inferior parietal cortex. Yet, left parietal cortex is also implicated in selective motor attention, suggesting its activation during temporal orienting could merely reflect incidental engagement of preparatory motor processes. Using fMRI, we therefore examined whether temporal orienting would still activate left parietal cortex when the cued target required a difficult perceptual discrimination rather than a speeded motor response. Behaviorally, temporal orienting improved accuracy of target identification as well as speed of target detection, demonstrating the general utility of temporal cues. Crucially, temporal orienting selectively activated left inferior parietal cortex for both motor and perceptual versions of the task. Moreover, conjunction analysis formally revealed a region deep in left intraparietal sulcus (IPS) as common to both tasks, thereby identifying it as a core neural substrate for temporal orienting. Despite the context-independent nature of left IPS activation, complementary psychophysiological interaction analysis revealed how the functional connectivity of left IPS changed as a function of task context. Specifically, left IPS activity covaried with premotor activity during motor temporal orienting but with visual extrastriate activity during perceptual temporal orienting, thereby revealing a cooperative network that comprises both temporal orienting and task-specific processing nodes.

On the Comparison Between the Nc/CRN and the Ne/ERN.

After the Error Negativity (Ne or ERN) has been described on full-blown errors and on partial error, a smaller Error Negativity-like wave (CRN or Nc) has also been evidenced on correct trials, first in patients with schizophrenia and, later on, in healthy subjects. The functional significance of the Nc as compared to the Ne is of critical importance since most models accounting for the genesis of the Ne on errors and partial errors cannot account for the existence of the Nc if this Nc simply corresponds to a small Ne. On the contrary, if the Nc and the Ne are two completely distinct components, then the existence of a Nc poses no constraint to the existing models. To this end, we examine in the present review the similarities and the differences existing between the Ne and the Nc regarding their functional properties and their anatomical origin.

Mechanisms and dynamics of cortical motor inhibition in the stop-signal paradigm: a TMS study.

The ability to stop ongoing motor responses in a split-second is a vital element of human cognitive control and flexibility that relies in large part on prefrontal cortex. We used the stop-signal paradigm to elucidate the engagement of primary motor cortex (M1) in inhibiting an ongoing voluntary motor response. The stop-signal paradigm taps the ability to flexibly countermand ongoing voluntary behavior upon presentation of a stop signal. We applied single-pulse TMS to M1 at several intervals following the stop signal to track the time course of excitability of the motor system related to generating and stopping a manual response. Electromyography recorded from the flexor pollicis brevis allowed quantification of the excitability of the corticospinal tract and the involvement of intracortical GABA(B)ergic circuits within M1, indexed respectively by the amplitude of the motor-evoked potential and the duration of the late part of the cortical silent period (SP). The results extend our knowledge of the neural basis of inhibitory control in three ways. First, the results revealed a dynamic interplay between response activation and stopping processes at M1 level during stop-signal inhibition of an ongoing response. Second, increased excitability of inhibitory interneurons that drives SP prolongation was evident as early as 134 msec following the instruction to stop. Third, this pattern was followed by a stop-related reduction of corticospinal excitability implemented around 180 after the stop signal. These findings point to the recruitment of GABA(B)ergic intracortical inhibitory circuits within M1 in stop-signal inhibition and support the notion of stopping as an active act of control.

Rostral Cingulate Zone and correct response monitoring: ICA and source localization evidences for the unicity of correct- and error-negativities.

Falkenstein et al. (1991) first described a negative wave occurring just after an erroneous response in choice Reaction time tasks ("Error Negativity"-Ne or "Error Related Negativity"-ERN). Thanks to Laplacian transform of the data, Vidal et al. (2000, 2003a) described a wave on correct trials with similar topography and latency, although of smaller amplitude compared to the errors. A critical question is whether the Ne observed on errors and the negativity reported on correct trials reflect the same (modulated) activity, or whether they reflect completely different mechanisms. These two alternative possibilities were tested thanks to Independent Component Analysis (ICA) and source localization. ICA results showed that the waves recorded on errors and correct trials can be accounted for by the same independent component, corresponding to a dipolar source located within the Rostral Cingulate Zone. Source localization on the raw data also confirmed a common generator for correct and error trials. These data suggest that the waves on errors and correct trials reflect the same brain activity, whose amplitude varies as a function of the correctness of the response. The implications of this result for cognitive control are discussed.

Direct evidence for cortical suppression of somatosensory afferents during visuomotor adaptation.

Upon exposure to novel visuomotor relationships, the information carried by visual and proprioceptive signals becomes discrepant, often disrupting motor execution. It has been shown that degradation of the proprioceptive sense (arising either from disease or experimental manipulation) enhances performance when drawing with mirror-reversed vision. Given that the central nervous system can exert a dynamic control over the transmission of afferent signals, reducing proprioceptive inflow to cortical areas could be part of the normal adaptive mechanisms deployed in healthy humans upon exposure to novel visuomotor environments. Here we address this issue by probing the transmission of somatosensory afferents throughout the course of adaptation to a visuomotor conflict, by recording median nerve somatosensory evoked potentials. We show that early exposure to tracing with mirror-reversed vision is accompanied by substantial proprioceptive suppression occurring in the primary somatosensory cortex (S1). This proprioceptive gating is gradually alleviated as performance increases with adaptation, returning to baseline levels. Peripheral and spinal evoked potentials were not modulated throughout, suggesting that the gating acted to reduce cortico-cortico excitability directly within S1. These modulations provide neurophysiological evidence for flexibility in sensory integration during visuomotor adaptation, which may functionally serve to reduce the sensory conflict until the visuo-proprioceptive mapping is updated.

Sequential adjustments before and after partial errors.

In choice reaction time tasks, subjects speed up before making an error, but slow down afterward to prevent the occurrence of a new error. In some trials, the correct response is preceded by an incorrect electromyographic (EMG) activation too small to reach the response threshold. In this article, we show that these incorrect EMG activations give rise to the same sequential effects as overt errors: Before a trial containing an incorrect EMG activation, subjects speed up, whereas after that trial, they slow down. These activations reflect errors that have been detected, inhibited, and corrected in time. As such, they index the involvement of online executive control.

Cost effective ultrasound imaging training mentor for use in developing countries.

This paper reports on a low cost system for training ultrasound imaging techniques. The need for such training is particularly acute in developing countries where typically ultrasound scanners remain idle due to the lack of experienced sonographers. The system described below is aimed at a PC platform but uses interface components from the Nintendo Wii games console. The training software is being designed to support a variety of patient case studies, and also supports remote tutoring over the internet.

Developing an immersive ultrasound guided needle puncture simulator.

We present an integrated system for training ultrasound guided needle puncture. Our aim is to provide a cost effective and validated training tool that uses actual patient data to enable interventional radiology trainees to learn how to carry out image-guided needle puncture. The input data required is a computed tomography scan of the patient that is used to create the patient specific models. Force measurements have been made on real tissue and the resulting data is incorporated into the simulator. Respiration and soft tissue deformations are also carried out to further improve the fidelity of the simulator.