NoC simulation steered by NEST: McAERsim and a Noxim patch.

Introduction: Great knowledge was gained about the computational substrate of the brain, but the way in which components and entities interact to perform information processing still remains a secret. Complex and large-scale network models have been developed to unveil processes at the ensemble level taking place over a large range of timescales. They challenge any kind of simulation platform, so that efficient implementations need to be developed that gain from focusing on a set of relevant models. With increasing network sizes imposed by these models, low latency inter-node communication becomes a critical aspect. This situation is even accentuated, if slow processes like learning should be covered, that require faster than real-time simulation. Methods: Therefore, this article presents two simulation frameworks, in which network-on-chip simulators are interfaced with the neuroscientific development environment NEST. This combination yields network traffic that is directly defined by the relevant neural network models and used to steer the network-on-chip simulations. As one of the outcomes, instructive statistics on network latencies are obtained. Since time stamps of different granularity are used by the simulators, a conversion is required that can be exploited to emulate an intended acceleration factor. Results: By application of the frameworks to scaled versions of the cortical microcircuit model-selected because of its unique properties as well as challenging demands-performance curves, latency, and traffic distributions could be determined. Discussion: The distinct characteristic of the second framework is its tree-based source-address driven multicast support, which, in connection with the torus topology, always led to the best results. Although currently biased by some inherent assumptions of the network-on-chip simulators, the results suit well to those of previous work dealing with node internals and suggesting accelerated simulations to be in reach.

Verification of a neuromorphic computing network simulator using experimental traffic data.

Simulations are a powerful tool to explore the design space of hardware systems, offering the flexibility to analyze different designs by simply changing parameters within the simulator setup. A precondition for the effectiveness of this methodology is that the simulation results accurately represent the real system. In a previous study, we introduced a simulator specifically designed to estimate the network load and latency to be observed on the connections in neuromorphic computing (NC) systems. The simulator was shown to be especially valuable in the case of large scale heterogeneous neural networks (NNs). In this work, we compare the network load measured on a SpiNNaker board running a NN in different configurations reported in the literature to the results obtained with our simulator running the same configurations. The simulated network loads show minor differences from the values reported in the ascribed publication but fall within the margin of error, considering the generation of the test case NN based on statistics that introduced variations. Having shown that the network simulator provides representative results for this type of -biological plausible-heterogeneous NNs, it also paves the way to further use of the simulator for more complex network analyses.

Extracting Cardiac Information From Medical Radar Using Locally Projective Adaptive Signal Separation.

Electrocardiography is the gold standard for electrical heartbeat activity, but offers no direct measurement of mechanical activity. Mechanical cardiac activity can be assessed non-invasively using, e.g., ballistocardiography and recently, medical radar has emerged as a contactless alternative modality. However, all modalities for measuring the mechanical cardiac activity are affected by respiratory movements, requiring a signal separation step before higher-level analysis can be performed. This paper adapts a non-linear filter for separating the respiratory and cardiac signal components of radar recordings. In addition, we present an adaptive algorithm for estimating the parameters for the non-linear filter. The novelty of our method lies in the combination of the non-linear signal separation method with a novel, adaptive parameter estimation method specifically designed for the non-linear signal separation method, eliminating the need for manual intervention and resulting in a fully adaptive algorithm. Using the two benchmark applications of (i) cardiac template extraction from radar and (ii) peak timing analysis, we demonstrate that the non-linear filter combined with adaptive parameter estimation delivers superior results compared to linear filtering. The results show that using locally projective adaptive signal separation (LoPASS), we are able to reduce the mean standard deviation of the cardiac template by at least a factor of 2 across all subjects. In addition, using LoPASS, 9 out of 10 subjects show significant (at a confidence level of 2.5%) correlation between the R-T-interval and the R-radar-interval, while using linear filters this ratio drops to 6 out of 10. Our analysis suggests that the improvement is due to better preservation of the cardiac signal morphology by the non-linear signal separation method. Hence, we expect that the non-linear signal separation method introduced in this paper will mostly benefit analysis methods investigating the cardiac radar signal morphology on a beat-to-beat basis.

Variational Bayesian inversion for hierarchical unsupervised generative embedding (HUGE).

A recently introduced hierarchical generative model unified the inference of effective connectivity in individual subjects and the unsupervised identification of subgroups defined by connectivity patterns. This hierarchical unsupervised generative embedding (HUGE) approach combined a hierarchical formulation of dynamic causal modelling (DCM) for fMRI with Gaussian mixture models and relied on Markov chain Monte Carlo (MCMC) sampling for inference. While well suited for the inversion of complex hierarchical models, MCMC-based sampling suffers from a computational burden that is prohibitive for many applications. To address this problem, this paper derives an efficient variational Bayesian (VB) inversion scheme for HUGE that simultaneously provides approximations to the posterior distribution over model parameters and to the log model evidence. The face validity of the VB scheme was tested using two synthetic fMRI datasets with known ground truth. Additionally, an empirical fMRI dataset of stroke patients and healthy controls was used to evaluate the practical utility of the method in application to real-world problems. Our analyses demonstrate good performance of our VB scheme, with a marked speed-up of model inversion by two orders of magnitude compared to MCMC, while maintaining a similar level of accuracy. Notably, additional acceleration would be possible if parallel computing techniques were applied. Generally, our VB implementation of HUGE is fast enough to support multi-start procedures for whole-group analyses, a useful strategy to ameliorate problems with local extrema. HUGE thus represents a potentially useful practical solution for an important problem in clinical neuromodeling and computational psychiatry, i.e., the unsupervised detection of subgroups in heterogeneous populations that are defined by effective connectivity.

Multiple Vital-Sign-Based Infection Screening Outperforms Thermography Independent of the Classification Algorithm.

GOAL: Thermography-based infection screening at international airports plays an important role in the prevention of pandemics. However, studies show that thermography suffers from low sensitivity and specificity. To achieve higher screening accuracy, we developed a screening system based on the acquisition of multiple vital-signs. This multimodal approach increases accuracy, but introduces the need for sophisticated classification methods. This paper presents a comprehensive analysis of the multimodal approach to infection screening from a machine learning perspective. METHODS: We conduct an empirical study applying six classification algorithms to measurements from the multimodal screening system and comparing their performance among each other, as well as to the performance of thermography. In addition, we provide an information theoretic view on the use of multiple vital-signs for infection screening. The classification methods are tested using the same clinical data, which has been analyzed in our previous study using linear discriminant analysis. A total of 92 subjects were recruited for influenza screening using the system, consisting of 57 inpatients diagnosed to have seasonal influenza and 35 healthy controls. RESULTS: Our study revealed that the multimodal screening system reduces the misclassification rate by more than 50% compared to thermography. At the same time, none of the multimodal classifiers needed more than 6 ms for classification, which is negligible for practical purposes. CONCLUSION: Among the tested classifiers k-nearest neighbors, support vector machine and quadratic discriminant analysis achieved the highest cross-validated sensitivity score of 93%. SIGNIFICANCE: Multimodal infection screening might be able to address the shortcomings of thermography.

Individualized biomonitoring in heart failure--Biomon-HF "Keep an eye on heart failure--especially at night".

In the project "Individualized Biomonitoring in Heart Failure (Biomon-HF)," innovative sensors and algorithms for measuring vital signs, i.e., during the nocturnal sleep period, have been developed and successfully tested in five clinical feasibility studies involving 115 patients. The Biomon-HF sensor concepts are an important step toward future patient-customized telemonitoring and sensor-guided therapy management in chronic heart failure, including early detection of upcoming HF exacerbation and comorbidities at home. The resulting preventable disease complications and emergencies and reduction of consequences of disease are very important advantages for the patients, causing relief for medical staff and, thus, offer an enormous potential for improvements and cost savings in healthcare systems.

Model-based verification of a non-linear separation scheme for ballistocardiography.

The current rise in popularity of ballisto-cardiography-related research has led to the development of new sensor concepts and recording methods. Measuring the ballistocardiogram using bed mounted pressure sensors opens up new possibilities for home monitoring applications. The signals measured with these sensors contain a mixture of cardiac and respiratory components, which can be used for detection of comorbidities of heart failure like apnea or arrhythmia. However, the separation of the cardiac and respiratory components has proven to be difficult, since there is significant overlap in the spectra of both components. In this paper, an algorithm for the separation task is presented, which can overcome the problem of overlapping spectra. Additionally, a model has been developed for the generation of artificial ballistocardiograms, which are used to analyze the separation performance. Furthermore, the algorithm is tested on preliminary data from a clinical study.

Distributed intelligent sensor network for the rehabilitation of Parkinson's patients.

The coordination between locomotion and respiration of Parkinson's disease (PD) patients is reduced or even absent. The degree of this disturbance is assumed to be associated with the disease severity [S. Schiermeier, D. Schäfer, T. Schäfer, W. Greulich, and M. E. Schläfke, "Breathing and locomotion in patients with Parkinson's disease," Eur. J. Physiol., vol. 443, No. 1, pp. 67-71, Jul. 2001]. To enable a long-term and online analysis of the locomotion-respiration coordination for scientific purpose, we have developed a distributed wireless communicating network. We aim to integrate biofeedback protocols with the real-time analysis of the locomotion-respiration coordination in the system to aid rehabilitation of PD patients. The network of sensor nodes is composed of intelligent network operating devices (iNODEs). The miniaturized iNODE contains a continuous data acquisition system based on microcontroller, local data storage, capability of on-sensor digital signal processing in real time, and wireless communication based on IEEE 802.15.4. Force sensing resistors and respiratory inductive plethysmography are applied for motion and respiration sensing, respectively. A number of experiments have been undertaken in clinic and laboratory to test the system. It shall facilitate identification of therapeutic effects on PD, allowing to measure the patients' health status, and to aid in the rehabilitation of PD patients.

Integration of amplitude and phase statistics for complete artifact removal in independent components of neuromagnetic recordings.

In magnetoencephalography (MEG) and electroencephalography (EEG), independent component analysis is widely applied to separate brain signals from artifact components. A number of different methods have been proposed for the automatic or semiautomatic identification of artifact components. Most of the proposed methods are based on amplitude statistics of the decomposed MEG/EEG signal. We present a fully automated approach based on amplitude and phase statistics of decomposed MEG signals for the isolation of biological artifacts such as ocular, muscle, and cardiac artifacts (CAs). The performance of different artifact identification measures was investigated. In particular, we show that phase statistics is a robust and highly sensitive measure to identify strong and weak components that can be attributed to cardiac activity, whereas a combination of different measures is needed for the identification of artifacts caused by ocular and muscle activity. With the introduction of a rejection performance parameter, we are able to quantify the rejection quality for eye blinks and CAs. We demonstrate in a set of MEG data the good performance of the fully automated procedure for the removal of cardiac, ocular, and muscle artifacts. The new approach allows routine application to clinical measurements with small effect on the brain signal.

Online cardiac arrhythmia classification by means of circle maps analysis implemented on an intelligent miniaturized sensor.

The intermittent occurrence of cardiac arrhythmias like e.g. atrial fibrillation hampers their diagnosis and hence the treatment. Since persons suffering from atrial fibrillation are known to have a remarkable increased risk of stroke the diagnosis of atrial fibrillation is a matter of great importance. Easy and comfortable to use long term ECG recording systems capable of online arrhythmia classification might help to solve this problem. We developed an intelligent, miniaturized, and wireless networking sensor which allows lossless local data recordings up to 4 GB. With its outer dimensions of 20mm per rim and less than 15g of weight including the Lithium-Ion battery our modular designed sensor node is thoroughly capable of up to eight channel recordings with 8 kHz sample rate each and provides sufficient computational power for online digital signal processing. For online arrhythmia classification we will record one ECG channel and 3-axis accelerometer data with 512 Hz each, the later being used for activity classification based artifact identification. We adapted our recently developed circle maps analysis of short term heart rate variation to run on this miniaturized intelligent sensor powered by the Texas Instruments MSP430 microcontroller derivate F1611. With this configuration we started to evaluate the cardiac arrhythmia classification in long term ECG recordings.

iSANLA: intelligent sensor and actuator network for life science applications.

In the fields of neurological rehabilitation and neurophysiological research there is a strong need for miniaturized, multi channel, battery driven, wireless networking DAQ systems enabling real-time digital signal processing and feedback experiments. For the scientific investigation on the passive auditory based 3D-orientation of Barn Owls and the scientific research on vegetative locomotor coordination of Parkinson's disease patients during rehabilitation we developed our 'intelligent Sensor and Actuator Network for Life science Application' (iSANLA) system. Implemented on the ultra low power microcontroller MSP430 sample rates up to 96 kHz have been realised for single channel DAQ. The system includes lossless local data storage up to 4 GB. With its outer dimensions of 20mm per rim and less than 15 g of weight including the Lithium-Ion battery our modular designed sensor node is thoroughly capable of up to eight channel recordings with 8 kHz sample rate each and provides sufficient computational power for digital signal processing ready to start our first mobile experiments. For wireless mobility a compact communication protocol based on the IEEE 802.15.4 wireless standard with net data rates up to 141 kbit/s has been implemented. To merge the lossless acquired data of the distributed iNODEs a time synchronization protocol has been developed preserving causality. Hence the necessary time synchronous start of the data acquisition inside a network of multiple sensors with a precision better than the highest sample rate has been realized.

Preserved parasympathetic cardiac innervation after atrioventricular node modification: evidence from circle maps of respiratory sinus arrhythmia.

INTRODUCTION: Respiratory sinus arrhythmia (RSA) and heart rate variability (HRV) are parameters of autonomic cardiac innervation. They decrease with age and after atrioventricular nodal modification (AVNM) suggesting vagal denervation in both situations. We hypothesized, however, that AVNM causes only a transient, functional decline in vagal activity, whereas aging causes permanent vagal denervation. A new method of analyzing RSA phase dynamics based on circle maps (CM) can potentially differentiate between both forms of reduced vagal activity. METHODS: In 18 younger and 14 older healthy control subjects 24-hour Holter ECGs were recorded for HRV analysis. Repeated measurements of RSA were acquired during paced breathing (PB). In 16 consecutive patients undergoing AVNM the same measurements were applied before, 1 day and 3 months after the procedure. CM were calculated from consecutive RR intervals and the similarity between different CM quantified by the Kullback information gain (KIG). RESULTS: HRV analysis revealed lower HF bands, LF bands and RSA amplitudes in older vs. younger control subjects. KIG revealed less similarity between younger and older control subjects than within the respective age groups. After AVNM a decrease in HF bands was noted in HRV analysis. Three months after AVNM, HF bands returned to pre-ablation values. CM obtained before and 1 day after AVNM displayed comparable similarity to CM acquired 1 day before and 3 months after ablation. CONCLUSIONS: In contrast to conventional HRV parameters, CM of RSA are not altered by ablation in the posteroseptal space but by aging. Thus, this new method appears to differentiate between transient autonomic modification and chronic denervation.