Criteria for Optimal Averaging of Cardiac Signals

The averaging process is modeled as a linear system whose low-pass filter characteristics are determined by the degree in temporal misalignment of signals. Assuming the errors in temporal alignment of successive cardiac cycles are random, then the model transfer function is equivalent to the probability density function. The response of the model to a step input is equivalent to the probability distribution function, which can be readily quantified. To validate the model, a high resolution ECG amplifier and QRS recognition system was constructed that synchronizes a step input with a point on the QRS. Design criteria for optimal amplification, filtering, and triggering of the ECG are determined. Test of the model reveals a close correspondence between observed and predicted step responses. From the average step response, the recording fidelity of any average can be determined-rapidly while the alignment is adjusted for optimal precision. Using ECG signals from patients, our model system demonstrates that alignment errors can both add and subtract signal components. Methods for estimating the extent of signal distortion induced by averaging as well as criteria for minimizing it are presented.     

Biomimetic finger control by filtering of distributed forelimbpressures

A linear filter was developed for decoding finger commands from volitional pressures distributed within the residual forelimb. Filter parameters were based on dynamic pressures recorded from the residual limb within its socket, during specific finger commands. A matrix of signal features was derived from eight-dimensional pressure vectors, and its pseudoinverse comprised the filter parameters. Results with amputees showed that the filter could discriminate specific finger flexion commands, suggesting that pressure vector decoding can provide them with biomimetic finger control     

Rhythm Analysis of Arterial Blood Pressure

Rhythms identified in the power spectra of blood pressure and ECG recordings were used as probes of the intact cardiovascular control systems. A prominent vasomotor rhythm was detected in human subjects and experimental dogs, with a period ranging between 15 and 30 s. This rhythm did not depend on specific rhythms of heart rate but was dependent on the sympathetic nervous system, and was identified as a third-order rhythm of blood pressure. The parasympathetic nervous system appears to mediate a separate rhythm having a slightly shorter period. Another rhythm studied was a subharmonic of heart rate that appeared during episodes of tachycardia. Electrophysiological mapping of the ventricular surface in dogs revealed that tachycardia induced an alternating pattern of electrical conduction in ischemic areas of the ventricle, coincident with the appearance of subharmonics in pressure and ECG. Our results illustrate the potential utility of spectral analysis of cardiovascular signals in assessing cardiovascular regulation.     

Mechanoelectrical feedback in cardiac myocytes fromstretch-activated ion channels

Stretch-activated ion channels (SACS) in cardiac myocytes from neonatal rats were studied in cell-attached patches. Stretch of membrane patches by suction in the recording pipette caused the triggering of action potentials that were recorded as action currents (ACs). The significance of a temporal correlation between SAC open probability and ACs was tested using the Kolmogorov-Smirnov and Poisson distributions. It was shown that the 50-ms epoch immediately preceding the action current has unique kinetics and represented a peak in SAC open probability (p<0.001). Thus it appears that current from a small number of SAC's injects sufficient charge (0.2 pC during 50 ms) to trigger action potentials in myocytes. These data strengthen the hypothesis that passive mechanical stretch of myocardium can be arrhythmogenic     

A logical state model of reentrant ventricular activation

The ventricular surface of the heart was modeled as two-dimensional, 4096 element, network of cells connected logically to each other. An ischemic area was represented by a central core of prolonged refractoriness, distributed into eccentrically-layered elliptical contours such that refractoriness declined along varying gradients to the surrounding normal area. Propagation of cardiac action potentials was stimulated by five sequential states ranging from activation to inactivation. Reentrant activation was induced by premature stimulation of the network and resembled a "figure 8" type reentry seen experimentally. Activation patterns of reentry appeared as two propagation wavefronts which traveled around the ends of a continuous line of functional conduction block, merged into a single wavefront, then conducted slowly along a retrograde path to reactivate a region proximal to the block. Reentry could be prevented by modifying the distribution of recovery of excitability through stimulation at two strategically located sites during basic rhythm. Prevention occurred when the second site was situated in an area of prolonged refractoriness, just distal to the line of block. These simulations indicate that reentrant activation is characterized by the formation of long lines of conduction block which occur along a border of steeply graded refractoriness, and retrograde slow conduction which occurs along a more shallow refractory gradient. The occurrence of reentry is dependent on: 1) the coupling interval of the premature stimulus, 2) the location of the stimulus relative to the maximum refractory gradient, and 3) the activation sequence of the basic paced beats. Thus, this paper presents an efficient logical state model of cardiac activation which simulates experimentally observed activation patterns of reentry and its prevention.     

A biomimetic controller for a multifinger prosthesis.

A novel controller for a multifinger hand prosthesis was developed and tested to measure its accuracy and performance in transducing volitional signals for individual "phantom" fingers. Pneumatic sensors were fabricated from open-cell polymeric foam, and were interposed between the prosthetic socket and superficial extrinsic tendons associated with individual finger flexion. Test subjects were prompted to move individual fingers or combinations thereof to execute either taps or grasps. Sensor outputs were processed by a computer that controlled motions of individual fingers on a mechanical prosthesis. Trials on three upper-limb amputees showed that after brief training sessions, the TAP controller was effective at producing voluntary flexions of individual fingers and grasping motions. Signal energies were between 5 and 25 dB relative to noise from all sources, including adjacent sensors, indicating high degrees of both sensitivity and specificity for tendon-associated transduction. Finger flexions at up to three repetitions per second, and rhythmic tapping of sequential fingers were readily transduced. One amputee subject was able to play a short piano piece with three fingers, at approximately one-quarter normal tempo. TAP sensors responded linearly to graded forces from individual fingers, indicating proportional force control. Our results demonstrate the feasibility of restoring some degree of finger dexterity by noninvasive sensing of extrinsic tendons.