Spiking neuron computation with the time machine.

The Time Machine (TM) is a spike-based computation architecture that represents synaptic weights in time. This choice of weight representation allows the use of virtual synapses, providing an excellent tradeoff in terms of flexibility, arbitrary weight connections and hardware usage compared to dedicated synapse architectures. The TM supports an arbitrary number of synapses and is limited only by the number of simultaneously active synapses to each neuron. SpikeSim, a behavioral hardware simulator for the architecture, is described along with example algorithms for edge detection and objection recognition. The TM can implement traditional spike-based processing as well as recently developed time mode operations where step functions serve as the input and output of each neuron block. A custom hybrid digital/analog implementation and a fully digital realization of the TM are discussed. An analog chip with 32 neurons, 1024 synapses and an address event representation (AER) block has been fabricated in 0.5 µm technology. A fully digital field-programmable gate array (FPGA)-based implementation of the architecture has 6,144 neurons and 100,352 simultaneously active synapses. Both implementations utilize a digital controller for routing spikes that can process up to 34 million synapses per second.

Limb-length discrepancy as a cause of plantar fasciitis.

BACKGROUND: Anisomelia, or limb-length discrepancy, has disruptive effects on gait, posture, and ambulation. Limb-length discrepancy has been shown to be a factor in stress fractures in the femur and tibia, and the longer limb, a contributing factor in the development of low-back pain, a cause of scoliosis. We sought to determine whether limb-length discrepancy contributes to the frequency and severity of plantar fasciitis. METHODS: We enrolled 26 patients who met the inclusion criteria. Direct and indirect methods were used to measure limb-length discrepancy. We took measurements from the anterior superior iliac spine to the medial malleolus and from the umbilicus to the medial malleolus and performed the block test. Body mass index (the weight in kilograms divided by the square of the height in meters) was also recorded for all of the patients. RESULTS: There is enough evidence to support the fact that the pain location and the longer limb are associated (Fisher test P < .0001). There was not enough evidence in this study to illustrate that body mass index was related to pain location (Fisher test P = .7411). CONCLUSIONS: There has been little research on etiology and treatment correlation. These results indicate a strong correlation between a longer limb and unilateral plantar fasciitis pain.

A sawtooth waveform inspired pitch estimator for speech and music.

A sawtooth waveform inspired pitch estimator (SWIPE) has been developed for speech and music. SWIPE estimates the pitch as the fundamental frequency of the sawtooth waveform whose spectrum best matches the spectrum of the input signal. The comparison of the spectra is done by computing a normalized inner product between the spectrum of the signal and a modified cosine. The size of the analysis window is chosen appropriately to make the width of the main lobes of the spectrum match the width of the positive lobes of the cosine. SWIPE('), a variation of SWIPE, utilizes only the first and prime harmonics of the signal, which significantly reduces subharmonic errors commonly found in other pitch estimation algorithms. The authors' tests indicate that SWIPE and SWIPE(') performed better on two spoken speech and one disordered voice database and one musical instrument database consisting of single notes performed at a variety of pitches.

Automatic speech recognition using a predictive echo state network classifier.

We have combined an echo state network (ESN) with a competitive state machine framework to create a classification engine called the predictive ESN classifier. We derive the expressions for training the predictive ESN classifier and show that the model was significantly more noise robust compared to a hidden Markov model in noisy speech classification experiments by 8+/-1 dB signal-to-noise ratio. The simple training algorithm and noise robustness of the predictive ESN classifier make it an attractive classification engine for automatic speech recognition.

Prediction of intrauterine pressure from electrohysterography using optimal linear filtering.

We propose a method of predicting intrauterine pressure (IUP) from external electrohysterograms (EHG) using a causal FIR Wiener filter. IUP and 8-channel EHG data were collected simultaneously from 14 laboring patients at term, and prediction models were trained and tested using 10-min windows for each patient and channel. RMS prediction error varied between 5-14 mmHg across all patients. We performed a 4-way analysis of variance on the RMS error, which varied across patients, channels, time (test window) and model (train window). The patient-channel interaction was the most significant factor while channel alone was not significant, indicating that different channels produced significantly different RMS errors depending on the patient. The channel-time factor was significant due to single-channel bursty noise, while time was a significant factor due to multichannel bursty noise. The time-model interaction was not significant, supporting the assumption that the random process generating the IUP and EHG signals was stationary. The results demonstrate the capabilities of optimal linear filter in predicting IUP from external EHG and offer insight into the factors that affect prediction error of IUP from multichannel EHG recordings.

Acoustic detection and classification of Microchiroptera using machine learning: lessons learned from automatic speech recognition.

Current automatic acoustic detection and classification of microchiroptera utilize global features of individual calls (i.e., duration, bandwidth, frequency extrema), an approach that stems from expert knowledge of call sonograms. This approach parallels the acoustic phonetic paradigm of human automatic speech recognition (ASR), which relied on expert knowledge to account for variations in canonical linguistic units. ASR research eventually shifted from acoustic phonetics to machine learning, primarily because of the superior ability of machine learning to account for signal variation. To compare machine learning with conventional methods of detection and classification, nearly 3000 search-phase calls were hand labeled from recordings of five species: Pipistrellus bodenheimeri, Molossus molossus, Lasiurus borealis, L. cinereus semotus, and Tadarida brasiliensis. The hand labels were used to train two machine learning models: a Gaussian mixture model (GMM) for detection and classification and a hidden Markov model (HMM) for classification. The GMM detector produced 4% error compared to 32% error for a baseline broadband energy detector, while the GMM and HMM classifiers produced errors of 0.6 +/- 0.2% compared to 16.9 +/- 1.1% error for a baseline discriminant function analysis classifier. The experiments showed that machine learning algorithms produced errors an order of magnitude smaller than those for conventional methods.

Exploiting independent filter bandwidth of human factor cepstral coefficients in automatic speech recognition.

Mel frequency cepstral coefficients (MFCC) are the most widely used speech features in automatic speech recognition systems, primarily because the coefficients fit well with the assumptions used in hidden Markov models and because of the superior noise robustness of MFCC over alternative feature sets such as linear prediction-based coefficients. The authors have recently introduced human factor cepstral coefficients (HFCC), a modification of MFCC that uses the known relationship between center frequency and critical bandwidth from human psychoacoustics to decouple filter bandwidth from filter spacing. In this work, the authors introduce a variation of HFCC called HFCC-E in which filter bandwidth is linearly scaled in order to investigate the effects of wider filter bandwidth on noise robustness. Experimental results show an increase in signal-to-noise ratio of 7 dB over traditional MFCC algorithms when filter bandwidth increases in HFCC-E. An important attribute of both HFCC and HFCC-E is that the algorithms only differ from MFCC in the filter bank coefficients: increased noise robustness using wider filters is achieved with no additional computational cost.

Ascertaining the importance of neurons to develop better brain-machine interfaces.

In the design of brain-machine interface (BMI) algorithms, the activity of hundreds of chronically recorded neurons is used to reconstruct a variety of kinematic variables. A significant problem introduced with the use of neural ensemble inputs for model building is the explosion in the number of free parameters. Large models not only affect model generalization but also put a computational burden on computing an optimal solution especially when the goal is to implement the BMI in low-power, portable hardware. In this paper, three methods are presented to quantitatively rate the importance of neurons in neural to motor mapping, using single neuron correlation analysis, sensitivity analysis through a vector linear model, and a model-independent cellular directional tuning analysis for comparisons purpose. Although, the rankings are not identical, up to sixty percent of the top 10 ranking cells were in common. This set can then be used to determine a reduced-order model whose performance is similar to that of the ensemble. It is further shown that by pruning the initial ensemble neural input with the ranked importance of cells, a reduced sets of cells (between 40 and 80, depending upon the methods) can be found that exceed the BMI performance levels of the full ensemble.

A bio-amplifier with pulse output.

A low-power fully integrated bioamplifier is presented that can amplify signals in the range from mHz to kHz while rejecting large DC offsets generated at the electrode-tissue interface. The novel aspect of this amplifier is that its analog output is represented by a series of pulses which provide a low-power, noise-resistant means for coding and transmission. The original analog signal can be reconstructed from the resulting pulse train with 13 bit precision at a remote site where power consumption is not so crucial. The fabricated analog amplifier exhibits a gain of 39.5 dB from 0.3 Hz to 5.4 kHz. The power consumption of the whole system is less than 300 microW/channel from a 5-V supply. The fully integrated system was designed in the AMI 0.6 microm CMOS process and it consumes 0.088 mm(2) channel of chip area.