superconducting quantum interference device magnetometer; basic electrical rhythm; spectral analysis; volume conductor; small intestine; electrical control activity; electrical recordings SMOOTH MUSCLE in the gastrointestinal tract displays two types of electrical activities. A high-frequency spiking activity known as the electrical response activity is associated with muscle contraction, whereas an oscillating slow wave, known as the electrical control activity or basic electrical rhythm (BER), is present continuously (8, 9, 24). The BER has typically been detected with invasive serosal electrodes that measure the local potential difference of a section of bowel. Recently, Chen et al. (6) reported detecting the human small bowel BER using cutaneous electrodes. To record small bowel BER, a band-pass filter was used to filter out all other biologically active tissues (stomach, heart, and colon). The results indicate that the small bowel potentials result in low-amplitude cutaneous potentials that give a low signal-to-noise ratio. Moreover, the studies by Chen et al. (6) involved subjects in whom the jejunum had been attached to the abdominal wall, providing for increased electrical contact between the bowel and the cutaneous electrode. Presumably, even lower signal-to-noise ratios than Chen et al. (6) reported would be obtained if the direct electrical contact between the bowel and the cutaneous electrode, provided by attaching the bowel segment to the abdominal wall, were broken.Because magnetic fields are not attenuated by lowconductivity layers such as those found in the abdominal wall, another option is to measure the magnetic field produced by small bowel electrical activity (27). Biomagnetic fields are typically several orders of magnitude smaller than the magnetic field of the Earth, so a sensitive detection device is necessary. Superconducting quantum interference device (SQUID) magnetometers are able to detect the weak magnetic fields of biological origin. Previous studies have shown that SQUIDS can detect small intestine BER in vitro (25) and in vivo (11). Magnetoencephalography researchers routinely measure the magnetic field of the brain (27), which is at least an order of magnitude smaller than the magnetic field of the small intestine (21).We present an experiment to examine the effect of a nonconducting layer placed between the small bowel and the abdominal wall on the transabdominal magnetic field, the serosal potential, and the cutaneous potential. We hypothesized that the magnetic fields of small bowel electrical activity recorded with the SQUID would correlate with the potentials measured with invasive electrodes and would not be attenuated by the placement of a nonconducting layer between the bowel and the abdominal wall. Transabdominal magnetic measurements are not expected to suffer from the same problems (attenuation and smoothing by electrically insulating layers) as cutaneous electrical recordings. A previous study utilizing a mathematical model for the magnetic fields and electric poten...

An analysis of the relative capabilities of methods for magnetic and electric detection of gastrointestinal electrical activity is presented. The model employed is the first volume conductor model for magnetic fields from GEA to appear in the literature. A mathematical model is introduced for the electric potential and magnetic field from intestinal electrical activity in terms of the spatial filters that relate the bioelectric sources with the external magnetic fields and potentials. The forward spatial filters are low-pass functions of spatial frequency, so more superficial external fields and potentials contain less spatial information than fields and potentials near the source. Inverse spatial filters, which are reciprocals of the forward filters, are high-pass functions and must be regularised by windowing. Because of the conductivity discontinuities introduced by low-conductivity fat layers in the abdomen, the electric potentials recorded outside these layers required more regularisation than the magnetic fields, and thus, the spatial resolution of the magnetic fields from intestinal electrical activity is higher than the spatial resolution of the external potentials. In this study, two smooth muscle sources separated by 5cm were adequately resolved magnetically, but not resolved electrically. Thus, sources are more accurately localized and imaged using magnetic measurements than using measurements of electric potential.

Electrical activity in the gastrointestinal system produces magnetic fields that may be measured with superconducting quantum interference device magnetometers. Although typical magnetometers have detection coils that measure a single component of the magnetic field, gastric and intestinal magnetic fields are vector quantities. We recorded gastric and intestinal magnetic fields from nine abdominal sections in nine normal human volunteers using a vector magnetometer that measures all three Cartesian components of the magnetic field vector. A vector projection technique was utilized to separate the magnetic field vectors corresponding to gastric and intestinal activity. The gastric magnetic field vector was oriented in a cephalad direction, consistent with previously observed data, and displayed oscillatory characteristics of gastric electrical activity (f = 3.03 +/- 0.18 cycles/min). Although the small bowel magnetic field vector showed no consistent orientation, the characteristic frequency gradient of the small bowel electrical activity was observed. Gastric and intestinal magnetic field vectors were oriented in different directions and were thus distinguished by the vector projection technique. The observed difference in direction of gastric and intestinal magnetic field vectors indicates that vector recordings dramatically increase the ability to separate physiological signal components from nonphysiological components and to distinguish between different physiological components.

SQUIDs can noninvasively detect bowel ischemia early in a free-lying segment of small bowel in this animal model with a high degree of sensitivity and specificity.