Patent No. 4951674 Biomagnetic analytical system using fiber-optic magnetic sensors

 

Patent No. 4951674  Biomagnetic analytical system using fiber-optic magnetic sensors (Zanakis, et al., Aug 28, 1990)

Abstract

A biomagnetic analytical system for sensing and indicating minute magnetic fields emanating from the brain or from any other tissue region of interest in a subject under study. The system includes a magnetic pick-up device constituted by an array of fiber-optic magnetic sensors mounted at positions distributed throughout the inner confines of a magnetic shield configured to conform generally to the head of the subject or whatever other body region is of interest. Each sensor yields a light beam whose phase or other parameter is modulated in accordance with the magnetic field emanating from the related site in the region. The modulated beam from each sensor is compared in an interferometer with a reference light beam to yield an output signal that is a function of the magnetic field being emitted at the related site. The output signals from the interferometer are processed to provide a display or recording exhibiting the pattern or map of magnetic fields resulting from emanations at the multitude of sites encompassed by the region.

Notes:

BACKGROUND OF INVENTION

1. Field of Invention

This invention relates generally to biomagnetic analytic systems for sensing and indicating minute magnetic fields emanating from the brain and other tissue regions of the human body, and more particularly to a system using fiber-optic magnetic sensor pick-up devices for this purpose.

2. Status of Prior Art

Biomagnetic fields arise from three principal sources, the first being electric currents produced by the movement of ions. The second source is remanent magnetic movement of contaminants, and the third is paramagnetic or diamagnetic constituents of the body.

The first source is of primary significance in human brain activity in which the currents creating the magnetic fields result from signals generated by neurons as they communicate with each other and with sensory organs of the body. The intensity of extracranial magnetic field produced by such currents is extremely minute, having a strength no more than about a billionth of the magnetic field at the earth's surface. It is usually measured in terms of tesla (T) or gauss (G), one T being equal to 10.sup.4 G.

The magnetic field arising from spontaneous brain activity (alpha waves) is about one picotesla (IpT=10.sup.-12 T), whereas the magnetic field at the earth's surface is about 6.times.10.sup.-5 T. The magnetic field emanating from the brain has a strength much below that emitted by the heart. Hence monitoring of brain magnetic activity presents formidable difficulties.

A major concern of the present invention is magnetoencephalography (also commonly referred to as MEG). This is the recording of magnetic fields emanating from the brain resulting from neuronal electric currents, as distinguished from an electroencephalogram (EEG) in which electric potentials originating in the brain are recorded. With an EEG measurement, it is difficult to extract the three-dimensional distribution of electrically active brain sites from potentials developed at the scalp. While this difficulty can be overcome by inserting electrodes through apertures bored in the skull, this invasive technique is not feasible in the study of normal brain functions or to diagnose functional brain disorders or brain dysfunctions. Thus ionic currents associated with the production of electrically measurable epileptic seizures generate detectable extracranial magnetic fields, and these can be detected externally without invading the skull.

Non-invasive MEG procedures are currently used in epilepsy research to detect the magnetic field distribution over the surface of the head of a patient with a view to localizing the seizure foci and spread patterns. This analysis serves as a guide to surgical intervention for the control of intractable seizures. (See: "Magnetoencephalography and Epilepsy Research"--Rose et al.; Science--16 Oct. 1987--Volume 238, pp. 329-335.)

MEG procedures have been considered as a means to determine the origin of Parkinson's tremor, to differentiate at the earliest possible stage Alzheimer's disease from other dementias, and to localize the responsible cortical lesions in visual defects of neurological origin. MEG procedures are also of value in classifying active drugs in respect to their effects on specific brain structures, and to in this way predict their pharmaceutical efficacy. And with MEG, one can gain a better understanding of the recovery process in head trauma and strokes by observing the restoration of neurological functions at the affected site.

But while MEG holds great promise in the above-noted clinical and pharmaceutical applications, practical considerations, mainly centered on limitations inherent in magnetic sensors presently available for this purpose, have to a large degree inhibited these applications.

The characteristics of biomagnetic activity that are measurable are the strength of the field, the frequency domain and the nature of the field pattern outside of the body. In magnetoencephalography, measurement of all three of these components are important. Ideally, simultaneous measurement of three orthogonal components of the magnetic field provides a complete description of the field as a function of space and time. Coincident measurement of the magnetic field along the surface of the skull can provide a magnetic field map of the cortical and subcortical magnetic activity. With spontaneous activity, the brain emits magnetic fields of about 10.sup.-8 to 10.sup.-9 Gauss, compared with approximately 10.sup.-6 Gauss emitted by the heart. Thus, monitoring of the brain's magnetic activity places heavy demand upon the required hardware.

In brain activity, the current dipole or source is generated by the current flow associated within a neuron or group of neurons. Volume current is analogous to the extracellular component of the current source. In MEG, the net magnetic field measured depends on the magnetic field generated by the current dipole itself. The contribution from volume conduction is small in which approximations to spherical symmetry are made. However, there are tangential magnetic components originating from secondary sources representing perturbations of the pattern by the volume current at boundaries between regions of different conductivity. Contributions from these secondary sources to the tangential component of the field become relatively more pronounced with distance from the current dipole. But there is no interference from these secondary sources when measurement is confined to the magnetic fields perpendicular to the skull.

In biomagnetic analysis, three types of magnetic sensors are known to have adequate sensitivy and discrimination against ambient noise for this purpose. (See: "Magnetoencephalography"--Sato et al.--Journal of Clinical Neurophysiology--Vol. 2, No. 2--1985.) The first is the induction coil. But because of Nyquist noise associated with the resistance of the windings and its loss of sensitivity at frequencies below a few Herz, the induction coil is rarely used in MEG studies.

The second is the Fluxgate magnetometer; and while this has been used in geophysical studies, it has certain drawbacks when used in MEG applications. It is for this reason that the third type, the SQUID system, is presently used almost exclusively in MEG applications.

A SQUID (Superconducting QUantum Interference Device) comprises a superconducting loop incorporating a "weak link" highly sensitive to the magnetic field encompassed within the area of the loop. While the loop itself can act as a magnetic field sensor, use is made of a detection coil tightly coupled to the superconducting loop, the coil acting as a flux transformer. Both the coil and the loop are immersed in a bath of liquid helium contained within a dewar.

With the advent of so-called high-temperature superconductors operating at liquid nitrogen temperatures, a SQUID magnetometer has been developed using such superconductors. (See: "The Impact of High Temperature Superconductivity on SQUID Magnetometers"--Clarke et al.--Science--Vol. 242--14 Oct. 1988.)

In the booklet published by Biomagnetic Technologies, Inc., of San Diego, Calif., entitled "Introduction to Magnetoencephalography--A New Window on The Brain," there is disclosed a SQUID-type sensor for MEG studies. This SQUID is especially suited to measure magnetic fields in the frequency range from DC to 20 kHz, the magnetic field being converted into a signal that is amplified, filtered and displayed for subsequent analysis.

Because the brain's field falls off sharply with distance from the head, the dewar for the cryogenic liquid, which is inherently bulky, is provided with a tail section of reduced diameter to house the pick-up coil and to minimize the distance of the coil from the head of the patient being studied, thereby maximizing the detected field.

As pointed out in the above-identified booklet, in order to produce a contour map of the brain, the magnetic field must be measured simultaneously at a number of points outside the head. While it is possible with SQUIDS to sample the magnetic field emanating from the brain at one to seven points separated laterally from each other by several centimeters, a complete mapping of the field pattern at a given instant requires forty or more pick-up points. It is proposed, therefore, in the booklet to move SQUID sensors from one point to another to accumulate the required field data. But a measurement taken at a point X will not reveal magnetic brain activity taking place concurrently at a point Y if one has to physically shift the sensor from point X to point Y.

The booklet notes that the ultimate goal of MEG measurement is to simultaneously observe all areas of the brain to produce real-time activity maps responding instantaneously to changes as they occur. However, the booklet concedes that this goal has not yet been realized with SQUID sensors.

The present invention attains this goal by means of fiber-optic magnetometers (FOM). In a FOM sensor, a magnetostrictive alloy is interfaced with an optical fiber to produce a magnetometer whose principle of operation is based on the transference of strain from the magnetostrictive material to the core of the optical fiber via mechanical bonding. This results in modulation of the phase or other parameters of the light propagated in the fiber which is subsequently detected by a fiber-optic interferometer. Integrated fiber-optic magnetometers in which all components are fabricated on or around the optical fibers are now known.

FOM sensors of the type currently available are far less expensive to manufacture and maintain than SQUID sensors; they are considerably more compact, and they operate at room temperature. Their sensitivity to weak magnetic fields, which can be greater than that of a SQUID, renders them suitable for MEG and other applications.

The following publications disclose various forms of FOM sensors:

1. "Single-Mode Fiber-Optic Magnetometer with DC Bias Field Stabilization"--Kersey et al.--Journal of Lightwave Technology--Vol. LT-3, No. 4--August 1985.

2. "Fiber-Optic Polarimetric DC Magnetometer Utilizing a Composite Metallic Glass Resonator"--Mermelstein--Journal of Lightwave Technology, Vol. LT-4, No. 9--September 1986.

3. "Optical Fiber Sensors Using The Method of Polarization-Rotated Reflection"--Enokihara et al.--Journal of Lightwave Technology--Vol. LT-5--No. 11--November 1987.

4. "An Analysis of A Fiber-Optic Magnetometer with Magnetic Feedback"--Koo et al.--Journal of Lightwave Technology--Vol. LT-5--No. 12--December 1987.

The disclosures of these publications are incorporated herein by reference.

SUMMARY OF INVENTION

In view of the foregoing, the main object of this invention is to provide a biomagnetic analytical system which includes a pick-up device employing an array of fiber-optic magnetic (FOM) sensors for measuring and indicating minute magnetic fields emanating from a multitude of sites in the brain or in other tissue regions of interest in a subject being diagnosed.

A significant advantage of a fiber-optic magnetic sensor (FOM) over a SQUID is that the former is a solid-state device that is considerably smaller than the latter and requires no cryogenics, thereby making it possible to distribute a multitude of the sensors (i.e., in excess of forty) around the skull of the patient or about any other tissue region of interest to effect more accurate localization of magnetic activity, as well as a more precise determination of the physiological condition of the region being studied.

More particularly, an object of this invention is to provide a system of the above type for MEG analysis in which the FOM sensors are so distributed in a three-dimensional array as to pick up magnetic fields emanating from a multitude of brain sites simultaneously and to spatially localize the field signals.

Also an object of the invention is to provide a shielded magnetic pick-up device in which the FOM sensors in the array are magnetically shielded from each other to prevent magnetic interaction therebetween, as well as from magnetic fields extraneous to the region of interest, thereby obviating the need for a shielded room to conduct studies on biomagnetic activity.

Yet another object of the invention is to provide a biomagnetic system in which the outputs of the FOM sensors in the array are multiplexed, whereby a common interferometer can be used for the multitude of sensors in the array thereof.

Briefly stated, these objects are attained in a biomagnetic analytical system for sensing and indicating minute magnetic fields emanating from the brain or from any other tissue region of interest in a subject under study. The system includes a magnetic pick-up device constituted by an array of fiber-optic magnetic sensors mounted at positions distributed throughout the inner confines of a magnetic shield configured to conform generally to the head of the subject or whatever other body region is of interest.

Each sensor yields a light beam whose phase or other parameter is modulated in accordance with the magnetic field emanating from the related site in the region. The modulated beam from each sensor is compared in an interferometer with a reference light beam to yield an output signal that is a function of the magnetic field being emitted at the related site. The output signals from the interferometer are processed to provide a display or recording exhibiting the pattern or map of magnetic fields resulting from emanations at the multitude of sites encompassed by the region.