James J. Ackmann

Complex Bioelectric Impedance Measurement System for the Frequency Range from 5 Hz to 1 MHz

Analytic techniques that have been successfully employed in materials science, and to a lesser extent in the study of biologic systems, have potential for improving the application of bioelectric impedance provided that both real and imaginary impedance components can be measured with sufficient accuracy over a given frequency range. Since biologic tissue, particularly animal tissue, is typically highly conductive, phase angles are small, making accurate measurements difficult. A practical four-terminal system employing commercial lock-in amplifiers is described and error sources and corrective tech-niques are discussed.

Early somatosensory evoked potentials

The early somatosensory evoked potential secondary to median nerve stimulation in the human had an onset latency of 9--12 msec when recorded from scalp electrodes at vertex-to-mastoid, vertex-to-inion or at the base of the skull. Similar latencies were observed from responses recorded over the cervical dorsal columns during neurologic surgery. A latency difference of 1.5 msec was observed between the early response and the responses recorded from the junction of medial lemniscus and nucleus ventralis posterior lateralis of the thalamus during human stereotaxic surgery. Cervical cord transections and transection at the midpontine levels of the monkey showed that the evoked potential was due to generators between these levels. Depth recording of the monkey indicate that the early evoked potential originates in the region of dorsal column nuclei, while the later components are secondary to generators in cerebral cortex.

Quantitative Evaluation of Long-Term Parkinson Tremor

Because of wide temporal fluctuations in the severity of Parkinson tremor, long-term observation is required to objectively assess the effects of therapy. This study was conducted to design a system for long-term recording of tremor in the digits or wrist, to examine the stochastic properties of tremor, and to develop a means for temporally compressing the data for rapid clinical interpretation. Forty patients were studied. A telemetry system was used for data acquistion to allow the patients freedom of movement during the recording session. Recording length ranged from ¿to 6 h with a mean of 2 h. The data were found to be self-stationary in the wide sense over 1-min epochs. The statistical properties varied with time in individual subjects and there was little uniformity among different subjects. Cross-correlation and cross-spectral density analyses indicated that tremor at corresponding sites in contralateral limbs is relatively independent. Graphs of percent on time, mean frequency, and mean amplitude for contiguous 3-5-min epochs were found most suitable for clinical purposes.

Spinal Cord Implant Studies

A loaded probe technique was used to measure the current density distribution resulting from application of electrical current to the spinal cords of live anesthetized stumptail macaque monkeys and fresh human cadaver spinal cords via the electrode arrays of four commercially available spinal implant systems used for management of intractable pain.

In Vivo Measurement of Real-Time Aortic Segmental Volume Using the Conductance Catheter

The goal of this investigation was to determine if the conductance catheter technique for chamber volume measurement could be applied in vivo to determine real-time phasic aortic segmental volume. A four-electrode conductance catheter was used to measure time-varying resistance of the descending thoracic aorta in open-chest, anesthetized dogs. Resistance was converted to segmental volume and the slope correction factor (α) and parallel conductance volume (VP) were determined. The results showed excellent linear correlation between conductance and sonomicrometric segmental volume. The correction factors α and VP were found to be empirically related to average vessel diameter. The relatively high values for the slope correction factor (α=4.59±0.17 SEM) were found to be primarily related to low-resistivity shunt paths probably originating in the periadventitial aortic wall and to a lesser extent to changes in flow-induced increases in blood resistivity, hematocrit, catheter position, and other adjacent tissue resistivity. The results demonstrate that correction factors empirically derived from measurements of mean aortic diameter could be used to determine absolute real-time phasic segmental volume, cross-sectional area, or diameter. The conductance technique may possess the same potential for determining aortic mechanical properties which has already been demonstrated for determining ventricular mechanical properties.

Effects of physical parameters on the cylindrical model for volume measurement by conductance

Despite its undisputed utility for determining changes in ventricular pressure-volume relationships, the conductance catheter technique has not been proven reliable for measuring absolute volume. This limitation is due to violations of the assumptions inherent in the cylindrical model on which the method is based (i.e., homogeneous electric field and no leakage current). The purpose of this investigation was to relate cylindrical model correction factors to the physical environment of the catheter and to the cylindrical equation. Physical measurements of saline-filled, nonconductive cylinders using a four-electrode conductance catheter were compared with a three-dimensional finite element model of the physical apparatus. These measurements were incorporated into a parallel conductance model to relate physical parameters to corrections in the cylindrical equation for volume measurement. Excellent agreement between measured and modeled data was found. Results demonstrated a nonlinear relationship between the field nonhomogeneity correction factor (α) and cylinder diameter. The relationship between α and diameter was consistent with a theoretical extrapolation of cylinder diameter toward infinity. An inverse relationship between α and the parallel conductance volume (VP) was also clarified. The parallel conductance model was able to demonstrate opposite effects of the physical presence of the catheter body and electrodes, which tended to cancel out any net effect on measured conductance. Results of this investigation and the developed finite element model clarify the nature of the correction terms in the cylindrical model and may lead to greater application of the conductance technique.

In Vitro and finite-element model investigation of the conductance technique for measurement of aortic segmental volume

This investigation examined the feasibility of applying the conductance catheter technique for measurement of absolute aortic segmental volume. Aortic segment volume was estimated simultaneouslyin vitro by using the conductance catheter technique and sonomicrometer crystals. Experiments were performed in five isolated canine aortas. Vessel diameter and pressure were altered, as were the conductive properties of the surrounding medium. In addition, a three-dimensional finite-element model of the vessel and apparatus was developed to examine the electric field and parallel conductance volume under different experimental conditions. The results indicated that in the absence of parallel conductance volume, the conductance catheter technique predicted absolute changes in segmental volumes and segmental pressure-volume relationships that agreed closely with those determined by sonomicrometry. The introduction of parallel conductance volume added a significant offset error to measurements of volume made with the conductance catheter that were nonlinearly related to the conductive properties of the surrounding medium. The finite-element model was able to predict measured resistance and parallel conductance volume, which correlated strongly with those measuredin vitro. The results imply that absolute segmental volume and distensibility may be determined only if the parallel conductance volume is known. If the offset volume is not known precisely, the conductance catheter technique may still be applied to measure absolute changes in aortic segmental volume and compliance.

Peak-detection algorithm for EEG analysis

A peak-detection method is described for computer analysis of the the electroencephalogramme (EEG). The technique consists of measuring the amplitude and time interval between successive maxima (peaks) and minima (troughs) in the signal. A critical feature of the peak-detection algorithm is the inclusion of an amplitude threshold criterion which eliminates the registration of low-voltage activity riding on EEG waves. The peak-detection procedure permits the formulation of a variety of intra-band and inter-band EEG statistics which can be useful in on-line computer applications. The peak-detection algorithm has been successfully applied to a number of normal and clinical EEG recordings. Although no computer procedure for EEG analysis has yet been universally adopted, the peak-detection algorithm reported in this paper presents a standardised approach which can be used between EEG clinics.

Cerebellar Implant Studies

Electrical currents were applied to the cerebellum of anesthetized monkeys using techniques similar to those employed in human cerebellar electrode implant systems. Alterations in the bloodbrain barrier were not observed. Histological damage, when present, appeared secondary to mechanical injury rather than to application of current. Measurement of current density in the monkey and human cerebellum indicated that with the electrode configuration employed, most of the current is retained in the vicinity of the electrodes. In human patients with spasticity and dyskinesia, application of current through implanted cerebellar electrode units relieves spasticity and also has been associated with reduction of somatosensory evoked potential amplitude. The amount of current required for clinical improvement and for alteration of evoked potential amplitude has remained stable in all patients, even those followed for more than a year. These observations suggest that the methods described do not produce appreciable cerebellar damage.

Specific Impedance of Canine Blood

The specific impedance of canine erythrocytes suspended in plasma was measured in the frequency range from 5 kHz to 1 MHz in samples from three animals in the hematocrit range from zero to packed cells at a temperature of 39°C; measurements were made with a conductivity cell using four electrodes and a current density of 21 μA/cm2. With the use of impedance spectroscopy, data were fitted to an equivalent circuit model; model parameters in turn were fitted as functions of hematocrit. The resultant model can be used to predict specific impedance (real and reactive components) as a function of hematocrit and frequency over a frequency range from 5 kHz to 1 MHz and a hematocrit range from 0 to 80. Over a normal range of hematocrits and at frequencies less than 100 kHz., the current is almost exclusively confined to the plasma, and the specific impedance is nearly equal to the real component; however, at higher frequencies, the complex nature of specific impedance becomes important.