AU746410B2 - Magnetic determination of position and orientation - Google Patents

Magnetic determination of position and orientation Download PDF

Info

Publication number
AU746410B2
AU746410B2 AU64421/99A AU6442199A AU746410B2 AU 746410 B2 AU746410 B2 AU 746410B2 AU 64421/99 A AU64421/99 A AU 64421/99A AU 6442199 A AU6442199 A AU 6442199A AU 746410 B2 AU746410 B2 AU 746410B2
Authority
AU
Australia
Prior art keywords
orientation
probe
image
body part
reference
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired
Application number
AU64421/99A
Other versions
AU6442199A (en
Inventor
David Ellis Acker
Wayne Grandner
Ian Mcnulty
Robert Pacheco
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Biosense Inc
Original Assignee
Biosense Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US132479 priority Critical
Application filed by Biosense Inc filed Critical Biosense Inc
Priority to AU64421/99A priority patent/AU746410B2/en
Publication of AU6442199A publication Critical patent/AU6442199A/en
Application granted granted Critical
Publication of AU746410B2 publication Critical patent/AU746410B2/en
Anticipated expiration legal-status Critical
Application status is Expired legal-status Critical

Links

Description

AUSTRALIA

Patents Act COMPLETE SPECIFICATION

(ORIGINAL)

Class Int. Class Application Number: Lodged: Complete Specification Lodged: Accepted: Published: Priority Related Art: Name of Applicant: Biosense, Inc.

Actual Inventor(s): DAVID ELLIS ACKER, IAN MCNULTY, ROBERT PACHECO, WAYNE GRANDNER Address for Service: PHILLIPS ORMONDE FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne 3000 AUSTRALIA Invention Title: MAGNETIC DETERMINATION OF POSITION AND ORIENTATION Our Ref: 603711 POF Code: 1421/277213 The following statement is a full description of this invention, including the best method of performing it known to applicant(s): -1- 6006q -2-

DESCRIPTION

MAGNETIC DETERMINATION OF POSITION AND ORIENTATION This is a divisional application divided out of Australian patent application 709793 the entire contents of which are incorporated herein by reference.

TECHNICAL FIELD The present invention relates generally to methods for determining the position and orientation of an object by means of magnetic fields, and includes in particular methods for displaying the location of a probe within a living subject and depicting a body part of that subject by means of magnetic fields. The methods are applicable for use with medical patients.

BACKGROUND OF THE INVENTION Various methods have been utilized for determining the position and orientation of an object in space. For example, it is often necessary to determine the position and orientation of a medical instrument within the body of a patient as, for example, to monitor the position and orientation of a catheter, endoscope or other probe. It is desirable to present data showing the object superposed on a picture of the patient such as on an image showing internal organs. One simple method of accomplishing this is to take a direct Xray image, such as a fluoroscopic image, showing both the patient and the inserted probe. This method has the disadvantage that the entire imaging procedure to acquire an image of the patient must be repeated whenever the probe is moved. X-ray imaging such as fluoroscopy can be performed during some surgical procedures, but inherently expose the patient to unwanted ionizing radiation. Some types of imaging procedures, such as MRI imaging 25 cannot be repeated during surgery or other treatment procedures.

As illustrated for example in GB patent application 2,094,590 and in U.S.

Patent 5,186,174, the probe may be mechanically constrained or linked to an S•arm or frame so that the position of the probe is constrained with respect to a fixed frame of reference and known with respect to that frame of reference. An image of the probe may be superposed on an image of C:AMy DoCmmentsW(ATE\SPECITP709793.doc

II:"

-3the patient, using the positional data derived from the fixed frame of reference. However, these systems make the probe inflexible and hence impose severe disadvantages in the medical procedure itself. For example, one such system requires that a probe be advanced in a straight line to the area of interest.

Systems of this nature are more suitable for positioning a probe on the outside of the body than inside the body.

Bryer et al. U.S. Patent 4,697,595 and Bryer et al., Ultrasonically Marked Catheter a Method for Positive Echographic Catheter Position Identification, Medical and Biological Engineering and Computer, vol.

22, No. 3 (1984), pp. 268-271, both disclose intracardiac catheters equipped with ultrasonic detectors. The position of the catheter is deduced from time of flight measurements from an ultrasonic transducer on the exterior of the patient to the catheter, and the deduced position is superimposed on 20 an ultrasonically generated image.

Van Steenwyk et al. U.S. Patent 4,173,228; Pfeiler et al. U. S. Patent 5,042,486 and Dumoulin et al. U.S. Patent 5,211,165 all disclose arrangements wherein electromagnetic signals are propagated between one antenna on the tip of a medical catheter inserted in the body and several antennas outside of the body.

The position and orientation of the catheter tip are assertedly determined from the signals transmitted between these antennas. That is, the relative position and orientation is deduced from the properties of the signal propagation path between these antennas as, for example, from the degree of signal attenuation in transit from one antenna to the others. The Van Steenwyck patent notes the possibility of using a magnetic field and a Hall effect transducer sensor, but offers no details as to how this might be accomplished in a practical device.

Dumoulin suggests that the radio frequency derived V 1 "I position of a catheter tip can be superposed on an image acquired by an imaging system.

Numerous systems have been proposed for locating articles outside of the human body by use of magnetic fields. Thus Blood, U.S. Patents 4,945,305; 4,849,692 and 4,613,866 all disclose systems for determining the position and orientation of objects in three-dimensional' space using magnetic coils on the object to be located and stationary coils in a fixed frame of reference. Other systems of this type include Voisin U.S. Patents 5,172,056 and 5,168,222; Constant U.S. Patent 4,396,885; Cantaloube U.S. Patent 5,109,194; Weed et al. U.S. Patent 4,317,078; Hansen U.S. Patent 4,642,786 and Morgenstern U.S. Patent 5,047,715. These systems typically employ a magnetic field transmitter incorporating several coils wound on orthogonal axes about an iron core, and a similar structure used as a receiver. The coils of the transmitter are actuated in sequence and/or at 20 different frequencies, and the signals detected by the coils of the receiver are analyzed to determine the position and orientation of the transmitter relative to the receiver. Among the uses for such systems are three-dimensional data entry devices for computers and systems for detecting the position and orientation of a helmet.

Additionally, as disclosed in Remel, An Inexpensive Eye Movement Monitor Using the Scleral Search Coil Technique, IEEE Transactions On Biomedical 30 Engineering, vol. BME-34, 14 April 1984, pp 388-390, researchers attempting to follow the rotation of the eyeball have mounted small loop-like sensing coils on the surface of the eyeball, as by suturing or by incorporating the coil in a contact lens. The subject having such a coil is placed between pairs of orthogonally oriented Helmholtz coils which are energized with high-frequency alternating currents having two different frequencies. The voltage induced in the coil on the eye will include components at both frequencies, and the relative magnitudes of these components will depend upon the orientation of the eye.

Despite all of these efforts in the art, there still has been a need heretofore, for improved methods for determining the position and orientation of an object in space, and, particularly, for improved methods for determining the position and orientation of a probe within the body of a living subject.

SUMMARY OF THE INVENTION One aspect of the present invention provides a method of displaying the location of a probe in a living subject. That method includes the steps of: providing at least one fiducial marker in fixed position relative to the subject; acquiring the image of the subject including a depiction of said at least one fiducial marker.

determining the locations of said at least one fiducial marker and of said probe in a common frame of reference by measuring magnetic fields transmitted to or from said probe and said fiducial marker, whereby the position of said probe relative to said at least one fiducial marker is known; and S(d) superposing a representation of said probe on said image of said 20 subject at a position relative to said depiction of said at least one fiducial marker corresponding to the position of said probe relative to said at least one fiducial marker in said common frame of reference.

This method may further include the steps of determining the orientation of the monitoring probe and of the fiducial marker in a common frame of 25 reference in the magnetic fields and orienting the superposed representation of the monitoring probe on the image of the subject, accordingly, so that the orientation of the monitoring probe representation relative to the depiction of the reference probe in the superposed images corresponds to the true orientation of the monitoring probe relative to the reference probe.

Another aspect of the present invention provides a method of depicting a body part. That method includes the steps of: C:\My DacumerdsXKATEMPECftP7O9793.doc -6providing a sensor in fixed position relative to the body part and determining the orientation of said sensor in a fixed frame of reference; acquiring an image of the body part in a starting orientation; monitoring the orientation of said sensor in said frame of reference by monitoring magnetic fields transmitted to or from said reference sensor to thereby determine a moved orientation of said body part after movement of the body part from said starting orientation; transforming said image of said body part in said starting orientation into an image of said body part in said moved orientation and displaying the transformed image.

Preferably, the monitoring, transforming and displaying steps are repeated as the body part is moved through a range of moved orientations so that a transformed image corresponding to each such moved orientation is displayed substantially in real time when the body part is in such moved orientation.

The method may be performed with a plurality of body parts simultaneously. Thus, separate sensors may be affixed to each of a plurality of body parts, and the steps of acquiring an image of the body part may include the step of acquiring the images of all of the body parts in respective first or starting positions. The steps of monitoring the orientation of the sensor and determining the orientation of the body part may include the step of monitoring the orientation of all of the sensors and determining separate moved orientations for each body part. The step of transforming the image may include the step of transforming the image of each body part into an image of 25 that body part in its respective moved orientation. The displaying step may be conducted so as to display all of the transformed images together, so that the orientations of the transformed images relative to one another correspond to 1the orientations of the body parts relative to one another. For example, where the body parts are the bones connected to one another at a skeletal joint, the physician can observe the transformed images of the bones, and thus observe the relative orientations of the parts constituting the joint during a medical procedure without taking additional x-rays as the bones are moved.

C:\My Documents'KATE\SPECI\P709793.doc -7- Preferably, the method includes the step of determining the position of each sensor in a fixed frame of reference while the body parts are in their respective starting orientations and while the body parts are in the respective move orientations, so that the steps of transforming and displaying the images may include the step of adjusting the positions of the displayed transformed images relative to one another to compensate for relative movement of the body parts. Thus, the displayed images will correctly represent the positions of the body parts relative to one another.

The foregoing and other features and advantages of the present invention will be more readily apparent from the detailed description of the preferred embodiments set forth below, taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a diagrammatic, partially prospective view of one form of apparatus for carrying out a method in accordance with an embodiment of the invention.

Fig. 2 is a diagrammatic view depicting a portion of the apparatus illustrated in Fig. 1.

Each of Figs. 3 5 is a graph illustrating certain magnetic fields occurring in the apparatus of Fig. 1 2.

Fig. 6 is a vector diagram relating to operation of the apparatus of Figs.

1 -2.

Figs. 7 and 8 are fragmentary diagrammatic elevational views depicting portions of another form of apparatus for carrying out a method in accordance 25 with an embodiment of the invention.

Fig. 9 is a fragmentary perspective view depicting a portion of a further form of apparatus for carrying out a method in accordance with an embodiment S•of the invention.

Fig. 10 is a fragmentary elevational view depicting another portion of the apparatus of Fig. 9.

C:\My Documerts\KATE\SPECI\P709793.doc -8- BEST MODE OF CARRYING OUT INVENTION Apparatus for carrying out one embodiment of the invention includes a frame structure C:lly Documents~ATE\SPECKP709793.doc -9substantially surrounding a sensing or patient receiving volume 32. Frame structure 30 supports three pairs of Helmholtz coils 34, 36 and 38 so that the coils of each pair are disposed coaxially with one another on three mutually orthogonal pair axes X, Y and Z. The intersecting pair axes thus form a conventional Cartesian coordinate system having an origin 39 at the center of the sensing volume and having positive and negative directions from the origin along each of the X, Y and Z axes. The coils of each pair are disposed on opposite sides of the sensing volume 32. All of the coils are circular, of equal diameter and have equal numbers of turns. The two coils of each pair are wound co-directionally with one another. Thus, both coils 34 arranged along the X axes have windings which wrap around the X axes in a right-handed or counterclockwise direction as seen from the positive end of the X axis. The spacings between the coils are exaggerated for clarity of illustration in Fig. i. Preferably, the distance .between the coils of the pair as measured along the axes of the pair is between about 0.5 and about 1.4 times the diameter of each coil. Rules for design of Helmholtz coils to provide uniform fields and to 25 provide fields having substantially linear gradients are set forth in the article "Modifying Helmholtz coils to Magnetically Drive *a Small Dipole", Lucas, Am.J. Phys., 54(7), pp.666-667 (1986), the disclosure of which is hereby incorporated by reference herein. The coils, as well as frame are constructed and arranged to permit insertion of a oo body part of a living patient into the sensing volume 32. Desirably, the frame has openings 40 coaxial with :'the various coils so that the patient's body may be 35 inserted into the sensing volume 32 through one such opening 40 and so that the physician may gain access to the patient while the patient is within the sensing volume through the other openings. The frame may be constructed of essentially any rigid material.

Dielectric materials are preferred because they do not create eddy currents when magnetic fields are changed.

The frame may include quick-release or breakable connections between the various coils and frame elements. This allows the physician to move the coils and frame away from the patient immediately in the event of an emergency requiring totally unimpeded access to the patient.

All of the coils are connected to a coil driver 42, which in turn is connected via an output interface 44 to receive output from a computer 46.

Computer 46 may be a digital computer of the type commonly referred to as a workstation or dedicated computer. Output interface 44 is arranged to accept the commands from the computer and actuate coil driver 42 in response to such commands. Coil driver 42 includes a conventional DC power supply 47 and conventional switching devices schematically indicated 20 at 48 for connecting the coils of each pair to the .o power supply 47. either in a forward configuration or in a reverse configuration as commanded by computer 46 through interface 44. In the forward configuration, the coils are connected in a series-aiding arrangement such that current passing through both coils of the pair flows in the same direction about the pair axis.

For example, when the coils 34 of the X-axis coil set are in the forward or series aiding configuration, they are connected in series so that current passing 30 through both coils will flow counterclockwise around the X axes as seen from the positive-X end thereof.

When the two coils of a pair are in the reverse configuration, they are connected in series opposition :0.6 so that a current will pass through both coils in series, but will flow in opposite directions around the axis in the two coils of the pair. Current will flow clockwise around the X axis in one coil 34 but -11counterclockwise around the X axis in the other coil 34.

The apparatus further includes a patient monitoring probe 50. Monitoring probe 50 includes an elongated body 52 having a proximal end 54 and a distal end 56 adapted for insertion into the body of the patient. In the particular embodiment illustrated, the probe body 52 is an elongated, flexible shaft adopted to fit inside a conventional endoscope 53, which in turn may be introduced into the body. Endoscope 53 has one or more bores or passageways for conducting surgical instruments, fluids and the like so as to introduce the same into the of the patient's body, and probe body 52 may fit within one such bore. Such bores or passageways may also be used for passage of endoscopic surgical instruments or other treatment devices or substances.

Alternatively or additionally, probe body 52 or the endoscope 53 may include known electrical, optical or electro-optical devices for providing a local image of the tissues surrounding distal end 56 such as video cameras and fiber-optic bundles. The endoscope may be connected to conventional medical treatment apparatus 58 for introducing and/or withdrawing 25 fluids, for observing the tissues surrounding the tip by means of the electrical or electro-optical devices and for performing other conventional medical treatment applications commonly performed through probes inserted into the body.

A sensor 60 is mounted to the probe body 52 adjacent the distal end 56. As best seen in Fig. 2, sensor 60 includes a plurality of planar sensing elements 62, 64 and 66. Element 62 is arranged normal to a common sensor axis Thus, the vector 3: 5 X' illustrated in Fig. 2 extending on the X' axiss of the sensor is normal to planar sensing element 62.

Likewise, element 64 is normal to the Y' direction of the sensor and element 66 is normal to the Z' -12direction. The directions Y' and Z' are orthogonal to one another, and hence the planes of elements 62, 64 and 66 are perpendicular to one another as well. Each sensing element incorporates material providing a Hall effect so that the sensing element provides an output voltage proportional to the component of the magnetic field in the vicinity of the sensor normal to the face of the particular sensing element. Thus, sensing element 62 incorporates a thin film of galvanomagnetic material such as indium arsenide, gallium arsenide or others, on its surface, a pair of excitation leads 68 connected to two opposite edges of the film and a pair of output voltage leads 70 connected to the remaining edges. In operation, when a current is passed between edges 68 and a magnetic field is present in the vicinity of the sensor, a voltage appears between leads proportional to the component of the field perpendicular to the sensor surface. Stated another way, where B is the local magnetic field vector and X' is the vector in the X' along the axis of the sensor perpendicular to elements 62, the output voltage V 062 from element 62 is:

V

06 2 K VI 6 2 25 Where the V 062 is the output voltage, K is a proportionality factor and VI 62 is the output voltage from element 62 at zero applied magnetic field.

Element 64 has similar leads (not shown) and similarly produces an output voltage: S 30 V 06 4

VI

6 4 Where V 064 is the output voltage from element 64; Y' is the unit vector in the Y' direction; and V 164 is the output voltage from element 64 at zero applied magnetic field. Likewise, element 66 produces 35 output voltage V 0 66

V

06 6 K

V

1 6 6 Where Z' is the unit vector in the Z' direction and VI 6 6 is the output voltage from element -13- 66 at zero applied magnetic field. Because the sensor is disposed in the distal tip 56 of probe body 52, the position and orientation of sensor 60 are not fixed with respect to the X, Y and Z reference directions of frame structure 30 and coils 34 38. Thus, the local axes Y' and Z' of sensor 60 can be at any orientation with respect to the axes X, Y and Z of the coils and frame structure.

The leads of sensor 60, including leads 68 and 70 and other leads associated with other sensing elements (not shown) are connected to a drive and amplification unit 72 (Fig. 1) via cable 75 extending through the body 52 of the probe. Drive and amplification unit 72 is arranged to apply the appropriate excitation voltages to the elements of sensor 60 and to amplify the output voltages

V

066

V

064 0 V 062 of the various sensing elements. The drive and amplification unit 72 is connected to an analog to digital converter 74, which in turn is connected to an input of computer 46. Although only a single connection is shown, it should be appreciated that converter 74 is a multi-channel device, with multiple connections and/or with multiplexing arrangements for delivering digital representations of all of the 25 output signals from all of the sensing elements in sensor 60 to computer 46.

The apparatus further includes a fiducial marker 76 (Fig. 1) which incorporates a sensor 80 and a marker body 78. Marker body 78 is arranged for mounting to a body part of the patient in a substantially fixed position relative to such body part. Thus, the marker body 78 may incorporate clamps, bands, rings or straps for fastening it onto the exterior of a patient's head, limb or torso.

35 Alternatively or additionally, body 78 may include apertures or notches to accommodate sutures, pins, surgical staples or other fastening devices. The body of the fiducial marker may also be fastened to the -14patient's body with conventional surgical adhesive tape, bandages or thi like. At least one part of the fiducial marker, such as the marker body 78, the sensor 80 or both is formed from a material which is detectable in a patient imaging technique such as Xray, magnetic resonance imaging computerized axial tomography or other commonly utilized imaging modalities. Sensor 80 of the fiducial marker is substantially the same as the sensor 60 of the monitoring probe 50. Sensor 80 is connected to a drive and amplification unit 82, which in turn is connected through an analog to digital converter 84 to computer 46. Unit 82 and converter 84 are similar to drive and amplification unit 72 and analog to digital converter 74.

The apparatus further includes an image storage unit 86 for storing an image of a patient, or of a body part of a patient, in digital form..

Typically, image storage unit 86 includes conventional computer memory devices. An image input device 88 is also provided. Input device 88 may include conventional digital data input devices such as disk drives, data communication links or the like for accepting images in digital form from any suitable 25 imaging equipment such as x-ray equipment, MRI, CAT or video equipment, and for transferring the input images into image storage device 86. A display monitor 90 is linked to computer 46.. Display monitor 90 is equipped with manual input devices such as knobs 92, 94 and 96, and linked to computer 46 so that the settings of those knobc can be communicated to the computer. The data input devices may also include a mouse, trackball, joystick or keyboard.

In a method according to one embodiment of 35 the invention, a body part of the patient, such as the head, is imaged using conventional imaging apparatus such as MRI, CAT, X-ray or the like, while the fiducial marker 76 is mounted on the body part. Thus, *oo the image data so acquired includes a depiction of the fiducial marker. In this regard, it is not essential that the entire fiducial marker be mounted on the patient's body part at this stage of the process.

Thus, if the magnetic sensor 80 is detachable from marker body 78, and if marker body 78 can be visualized by the imaging modality employed, then the sensor 80 may be disconnected from the marker body during this stage of the process. Also, the sensor of the fiducial marker may be disconnected from the remainder of the apparatus during this stage of the process. In this stage of the process, the patient need not be, and typically is not, positioned in frame or in the imaging volume 32. Typically, the patient is imaged in an entirely separate room or facility. After this imaging step, the fiducial marker remains in place on the patient throughout the remainder of the process. The image data depicting the patient's body part, including the depiction of fiducial marker 76, is transferred through input device 88 to image stc-age unit 86. The image data may include an image in a single plane or, more desirably, images in plural planes or a fully threedimensional image, such as a map of radioopacity or S 25 magnetic resonance activity over a full threedimensional volume incorporating a portion of the body part. The image data may be stored as output data wherein individual data elements represent densities or colors to be displayed on a pictorial representation of the image. Alternatively, the image may be stored in the form of input data such as time domain or frequency domain representations of magnetic resonance signals, or unprocessed tomographic data, trom which a pictorial representation can be reconstructed.

In the next stage of the method, the body part of the patient is positioned within sensing volume 32. Probe 50 is advanced into the patient's -16body using conventional techniques of the medical arts. Computer 46 actuates coil driver 48 to operate the coils in a predetermined, repetitive sequence.

The sequence includes a null state, in which all of the coils are turned off and the only magnetic fields in sensing volume 32 are those introduced by extraneous sources, such as the earth's magnetic field, stray magnetic fields from nearby items and the like. Also, the sequence includes a forward or homogeneous-field stage for each set of coils. When each set of coils is being driven in its forward stage, in the aforementioned series-aiding configuration with codirectional current flow in both coils, such set provides a substantially homogeneous unidirectional magnetic field oriented along its axis.

For example, the X-direction coils 34 in the forward state provides a field consisting essentially of flux directed parallel to the X-axis, of substantially equal magnitude throughout a central region of sensing volume 32, adjacent the midpoint of the sensing volume, the origin or intersection 39 of the X, Y and Z axes. Typically, the field is homogeneous to within about 1% over a region extending from the origin and encompassing about 62% of the distance 25 between the coils and extending laterally outwardly from the axes for a distance equal to about 25% of the coil diameter. The three coil sets are actuated separately, at different times during the sequence, so that only one coil set is active in the forward mode at any given time.

The sequence also includes gradient or reverse configuration mode actuation of each coil set.

In the reverse or gradient mode, with the currents flowing counterdirectionally in the two coils of each set, the field generated by each coil set incorporates a component directed along the axis between the coils and having a substantially linear, uniform gradient along that axis. For example, when the X-direction -17coils 34 are actuated in the gradient mode, the field has the configuration illustrated in Figs. 3, 4 and In Fig. 3, R, is the component of the reverse or gradient field in the X-direction, parallel to the X-axis. As seen in Fig. 3, this component has a low negative value at one end of the sensing volume.

The X-directed component Rx increases monotonically and linearly with increasing values of X, passing through 0 at approximately X 0, at the origin 39. The value or magnitude of the X-directed component at any given location along the X-axes is the same for any Y and Z. Rx, thus is a function only of X, and is constant with respect to Y and Z.

The reverse or gradient field generated by X-axis coils 34 also includes a radial component perpendicular to the X axis. The radial component is symmetrical about the X-axis and is directed away from the X axis. The radial component resolves into a Ydirected component and a Z-directed component. As seen in Fig. 4, the Y-directed component Ry of (the reverse or gradient field generated by the X-coils 34 varies monotonically with Y, reaching 0 at Y 0, at the X Z plane. The Z-direction component Rz of the X-coil reverse or gradient field has a 25 similar configuration (Fig. Thus, R z is proportional to Z. The proportionality constant or slope of Ry versus Y is about one third of the proportionality constant for Rx versus X. Likewise, the proportionality constant or slope for R z versus Z is also about one third of the proportionality constant for Rxversus X.

The foregoing discussion with respect to Figs. 3, 4 and 5 refers to the various components of a single reverse or gradient field, generated by X-axis 35 coils 34. The reverse or gradient fields generated by the Y-axis coils 36 and by the Z-axes coils 38 have the same configurations, but along different axes.

Thus, the reversed or gradient field generated by Y- -18axis coils 36 has a component R, which increases monotonically with increasing values of Y (with increasing displacement along the Y-axis in the +Y direction) but which it does not substantially vary with X or Z at any given value of Y. This component also has a value of about 0 at the Y 0 point, i.e., at the origin 39 and along the X-Z plane. The same Yaxis gradient field has a redial component perpendicular to the Y axis. The redial component resolves to a component in the X-direction which is proportional to X and a component in the 2-direction proportional to Z. The proportionality constants of the X and Z components with X and Z distance, respectively, about one third the proportionality constant of the Y component with Y distance.

Likewise, the reverse or gradient field provided by Z-axis coils 38 has a 2-direction component varying monotonically with Z and reaching about 0 at the origin (Z this component being constant with respect to X and Y within the central region of the sensing volume 32. The reverse or gradient field generated by coils 38 includes an Xdirection component proportional to X and a Y-directed component proportional to Y both of these 25 proportionality constants being less than the proportionality constant of the-Z-direction component.

With the reverse or gradient fields as well, the individual coil sets are actuated separately, at different times from one another, so that only one coil set is actuated at any given time, and so that none of the coil sets is actuated for gradient-field mode while another set is in homogeneous-field mode.

The various actuations are repeated cyclically. The particular sequence within each cycle is unimportant.

Thus, each cycle may include actuation of each coil set to produce its respective uniform field, followed by actuation of each coil set to produce its reversed or gradient field followed by deactivation of all coil -19sets. Alternatively, the sequence may include actuation of the X-direction coil said to produce its uniform field, then to produce its reversed or gradient field, followed by similar actuation of the Y and Z coil sets and deactivation of all coil sets, or any combination of these patterns. However, computer 46 maintains a record of the coil actuation for each moment in the sequence, so that incoming data from sensors 60 and 80 can be associated with particular coil set activations. The full sequence is repeated rapidly so that, ieach sequence takes about 10 to about 100 milliseconds.

As the fields are switched on and off in the sequence, computer 46 controls acquisition of data from sensors 60 and 80 so that signals representing each field are acquired at a preselected acquisition time during each switching cycle. Thus, as each coil pair is switched on in a particular mode, the magnitude of the field generated by that coil pair rises exponentially to a constant value according to the equation Mi-Mc(l-e(-t/tau) where: t is time; Mi is the instantaneous field magnitude at time •t; Mc is the constant or steady-state magnitude; and tau is a time constant proportional to the inductance of the coil set and inversely 30 proportional to the series resistance.

Because the acquisition time is the same for all switching cycles, the instantaneous field magnitude Mi at the acquisition time is always a constant proportion of the steady-state field magnitude Mc. For example, the delay between the start of a switching cycle and the acquisition time may be about equal to one times the time constant tau, about percent of the delay time which would be required to reach a substantially steady-state condition. In this arrangement, the instantaneous field magnitude at the acquisition time in each cycle is about 63 percent of the steady-state value. Because each switching cycle can be terminated, and a new switching cycle begun, immediately after the acquisition time, this arrangement allows many more measurements per unit time than the alternative arrangement in which each coil pair is allowed to come to steady state before data acquisition. In a variant of this scheme, the acquisition time can be controlled by monitoring the instantaneous current through the coil sets. The system may be arranged to acquire data from sensors and 80 on each cycle when the instantaneous current in the particular coil set being activated rises to a preselected trigger value, preferably less than the equilibrium current. In this arrangement as well, the sensor data samples will be taken at a consistent field magnitude from cycle to cycle. While the coils are repeatedly activated, signals are obtained from sensors 60 and 80, amplified, digitized and fed to computer 46. The signals are used to ascertain the orientation and position of the sensor, and the attached probe, within the sensing volume 32, in the 25 X, Y and Z coordinate system. At each stage of each cycle, the computer acquires data representing the component of the field in the Y' and 2' directions of sensor 60, so that there is a separate data entry for the field in each such local direction during each coil actuation. Each data entry is corrected by subtracting the corresponding data entry for the coils-off of null-field condition in the same cycle.

For example, all of the readings for elements 62 giving the field in the X' direction are corrected by 35 subtracting the value for the field component in the X' direction acquired during the null-field condition.

This removes any effect of fields other than those applied by the coils. References hereinbelow to -21- "null-corrected" values of sensor readings should be understood as referring to the values corrected in this manner. The null-corrected values for the sensor readings acquired during generation of homogeneous fields are referred to as "Hm,n" where m is an letter indicating which sensing element acquired the value and n is a letter indicating which coil pair generated the homogeneous field during acquisition of the values. The notation is. indicated in table I, below: TABLE I SENSING ELEMENT READINGS FOR HOMOGENEOUS

FIELDS

X Y Z SENSING Coil Coil Coil ELEMENT Pair Pair Pair X'-62 Hx',x Hxiy Hxi, Y'-64 Hy ,x Hy,y Hy'z Z'-66 HZ',X HZ'iy HZ'i The values acquired during all of the homogeneous-fields thus form a 3 X 3 matrix. These values are normalized by dividing each such value by the magnitude of the flux generated by-the particular pair, by the magnitude of the total flux vector 25 applied by the coil pair. That magnitude in turn is the square root of the sum of the squares of the individual sensor readings: Hx sqrt[ (Hs x) 2

(H

z 2 Hy sqrt( (Hx iy)' 2 2 Hz sqrt[ (Hx, z)'2 (Hy, z)^ 2 (Hz, z)^ 2 Where Hx is the magnitude of the magnetic flux applied by the X-axis coil set 34 during uniform field activation and Hy2 and Hz3 are similar magnitudes for the flux applied by the Y-axis set 36 S* 35 and by the Z-axis set 38 respectively. The term "sqrt" means "square root of" the expression in brackets; "2 means the square of the preceding expression. These magnitudes are then used to compute -22normalized values. That is, each sensor reading taken during a given homogeneous-field actuation is divided by the magnitude of the flux generated during that actuation: H'x, Hx, x/HX Hx, =y Hx y/Hy H'xz Hx ,z/Hz H,y,,x H y',x/Hx H'y',y Hy,y/Hy Hy,,z/HZ H'z,x HZ,,x/Hx H'z,y Hz',y/Hy H'zz HZ,,/Hz The orientation of the sensor the angles of the sensor Y' and Z' axes with respect to the X, Y and Z axes of the frame 30 and coil sets can be determined directly from the normalized values. Less than all of the values are needed for calculation of the orientation from all of the homogeneous fields.

The system selects a set of values having the highest magnitude, and hence the greatest signal to noise ratio, for use in the calculation.

The orientation of the sensor with respect to the X,Y,Z coordinate system of the coils can be expressed in many ways. Preferably, however, the orientation is expressed as three angles referred to a pitch, roll and yaw. All of these are referenced to an assumed "home" or zero position in which the local directions of the sensor are aligned with the X,Y,Z reference directions of the coil system. Pitch is the angle 93 (Fig. 2) between the Z' local direction of the sensor and the Z reference direction. It can be 30 calculated as: Pitch-arccos(H', z) Yaw is the angle 94 between the projection of the sensor Z' axis onto the X-Y plane and the Y reference direction. It is calculable as: Yaw-arctan(H'zx/H'z ,y) In this notation, yaw is undefined for a zero-pitch condition. Roll is defined as the angle 95 between the 0* -23local direction X' of the sensor and the system reference Z direction. It can be calculated as: Roll=arctan(H'y, z/H'x, z Any other internally consistent set of angles can be used to specify orientation. Moreover, orientation of the sensor can be fully specified by specifying any two unit vectors of the sensor local coordinate system in terms of the X,Y,Z reference coordinate system.

Merely by way of example: where X' is the unit vector in the X' direction; and i,j,k are the unit vectors in the X,Y,Z reference directions.

The system also acquires data representing the sensor readings acquired during application of the gradient fields. As with the data acquired during generation of the homogeneous fields, above, the gradient field data is corrected by subtracting the null-field values for each sensor. The corrected values are then taken as data representing sensor readings, and hence, field components in particular X' Y' and Z' directions during application of the gradient fields by the different coil sets. Thus, as indicated in Table II, below, Rx, x represents the X' sensor reading, and hence the field component in the X' local direction of sensor 60 while the X-axis coils are operating in the reversed or gradient field mode, Rx,,y represents the similar component in the X' local direction during gradient field operation of the Y- 30 axis coil pair and so on: *0 Soo oooo -24- TABLE II SENSING ELEMENT READINGS FOR GRADIENT FIELDS x Y Z SENSING Coil Coil Coil ELEMENT Pair Pair Pair X'-62 Rx',x Rx',y Rx',z Y'-64 Ry',x Ry',y Ry',z Z'-66 Rz',x Rz',z Rz',z The magnitude of the total flux vectors at sensor during actuation of each coil set in a gradient mode are calculated by computer 46: IBx sqrt[ (Rx, x)^2 (Ry, 2 JByl sqrt[ (Rx, y)' 2 (Ry'y)A 2 IBz sqrt[ (Rx,,z) 2 (Ry,z)' 2 Where IBx| is the magnitude of the flux at the sensor during operation of the X-axis coil set 34 in a gradient or reversed mode, and where IByl is the magnitude of the flux vector at sensor 60 during gradient field operation of the Y-direction coil set 36 and where IBz| is the magnitude of the flux vector at the sensor during operation of the 2-axis coil set 38.

Although the magnitude of the total flux vectors for each gradient field can be readily calculated as described above from the gradient field sensor readings Rx, x *.Rzz the direction of the gradient fields flux vector in the XYZ reference coordinate system defined by the coils is not ascertainable from these values alone. Thus, the Rx,,x Rz values represent components in the X', Y' and 2' local directions of the sensor. The normalized homogeneous-field values H' 1

H'

33 are used in conjunction with the gradient field component values for each coil set to derive a project 4 on of the 35 total gradient field vector for each coil set onto the axis of that coil set, and thus derive the component of the gradient field vector on the axis of the coil set which was activated to produce that gradient field.

First, the system computes the magnitude of a difference vector between the total gradient field vector and the normalized unit magnitude homogeneousfield vector generated by the same coil pair. For example, as shown in Fig. 6, the system computes the magnitude of the difference vector d X representing the vector difference between the total flux B. during operation of the X-axis coil set in gradient mode and the normalized, unit length vector, representing a unit vector in the direction of the flux during operation of the X-axis coil set in the homogeneousfield mode. That latter direction of course is along the X-axis of the coil sets. The components of the difference vector in the Z' local coordinate system of the sensor are calculable directly from the gradient field sensor values RX,,x' Ry',x Rzx acquired during operation of the X coil set in gradient mode and from the normalized homogeneousfield sensor values H'y,,x H',x acquired during operation of the same coil set in homogeneousfield mode. Thus: Idxj sqrt[(Rx,, 2 2 25 (Rz, X) 1 2 oo. Where jdXI is the magnitude of the .difference vector. Inasmuch as the magnitudes of the unit vector Hx, of the gradient field total flux vector Bx and of the difference vector dX are all 30 known, the angles between these vectors can be calculated by application of the law of cosines. In particular, the angle Tx between the total gradient field vector B. at the sensor during operation of the X-direction coil set in gradient mode and the unit 35 vector H' can be calculated as: Tx arccos[ (1 IBxl'2 -IdXI^2)/(2*lBxl)] Because the homogeneous field generated by the X-axis coil set has flux directed along the X- I- ra IPL~Y~ -26axis, angle Tx represents the angle between the gradient field total flux vector B, and the X-axis.

The component of the gradient field total flux vector

B

x in the direction of the X-axis reference direction is simply: Rx (IBxI cos (Tx) This represents the component of flux in the Xdirection at sensor 60 during operation of the X-axis coil set 34 in the gradient or reverse field mode. As noted above, the component of flux during this operation of the X-direction coil set is substantially proportional to position in the X-direction (Fig. 3).

Thus: X Position (mm) (Mx) (Rx) Where Mx is a conversion factor The conversion factor represents the slope of the function relating the X component of flux R x to distance in the X direction. Mx may also be scaled to take account of the sensitivity of the various elements in the sensor and the gains of the various amplification channels, assuming all of these are equal for all of the sensors. Alternatively, additional sensitivity scaling factors may be applied to convert the voltages

V

0621

V

064

V

066 derived from each sensing element 25 into actual numerical values of the magnetic flux components, so that separate scaling factors are applied to the readings from the different sensing elements.

The difference vectors and angles are 30 computed in the exactly analogous way to derive the Yaxis component Ry of the flux vector By detected at sensor 60 during operation of the Y-direction coil set in the gradient mode and to derive the Z-axis component Rz in the z direction during gradient operation of the Z-axis coils. The magnitudes of the difference vectors are calculated as: *•go• -27- Idyl sqrt H'x, y)2 and IdzI sqrt (Ry,,z (RzIz-H'zz)^ 2 Where IdYl and jdZj are the magnitudes of the difference vectors associated with the Y-axis and Zaxis coils, in the same way as IdXl is calculated above. Angles Ty and Tz are calculated in the same manner as Tx, using the law of cosines: Ty arccos IBy*2 IdY|I2)/(2*|Byl)] and Tz arccos ((l+IBz|2 IdZ|^2)/(2lBz|] Y-direction and Z-direction positions are calculated from these components and angles in exactly the same way as X-direction positions. Thus: Ry (IByl) cos Ty Rz (IBzI) cos Tz Y Position (My)(Ry) Z Position (Mz)(Rz) Thus, each time the coils cycle through the null, homogeneous and gradient field states, computer 46 recalculates the position and orientation of sensor 60. Inasmuch as sensor 60 is attached to the distal 25 tip 56 of probe body 52, these calculations yield the position and orientation of the probe body distal tip.

In foregoing discussion, the conversion factor Mx relating X-direction position to the magnitude of the X-direction component Rx is assumed 30 to be exactly constant and exactly uniform for all values of Y and Z. Likewise, the corresponding conversion factors relating Y-direction position to magnitude of the Y-direction flux component and relating Z-direction position to magnitude of the Z- 35 direction flux component are assumed constant. The accuracy of the position determination can be enhanced by calibrating the instrument to derive a map of actual conversion factors for various positions in the '4'4 AL 110) -28sensing volume. In the calibration step, the actual magnitude of the field component in each axial direction is measured at various known values of X,Y and Z during reverse or gradient-field operation of the coil set associated with that axis. For example, the Y-direction component is measured at known locations during gradient-field operation of the Yaxis coil set. Each gradient field is substantially symmetrical about the axis of the associated coil set.

Therefore, the calibrating measurements can be taken at points on a single plane including the axis of the coil set, and the same values can be assumed to apply to the points corresponding to rotation of the actually-measured points about the axis of the coil set. The measured magnitudes are used to derive actual conversion factors applicable at each of the known positions. The same data is used to derive average conversion factors for each direction, applicable to the sensing volume as a whole, as by finding the single factor for a given direction which best fits all of the data acquired in calibration.

In operation, the X, Y and Z coordinates of the sensor are determined as a first approximation using the average conversion factors and the measured values of gradient-field components Rx, RYy and R These first-approximation coordinates are then used to determine actual conversion factors by interpolation between the actual conversion factors for the closest points where measurements were taken during the calibration step of the process. The resulting conversion factors are then applied with the measured values of the gradient-field components Rx, Ryy and R z to determine second approximated coordinates, which are in turn used to determine new values for the actual conversion factors by interpolation in the calibration data. These steps are repeated until the approximation converges to constant values of the S q% N A i -29various coordinates, which represent the best approximation of position.

In exactly the same way, the computer acquires flux component values from the sensing elements of sensor 80 on fiducial marker 76. On each cycle of the coils, the computer drives the position and orientation of the fiducial marker as well.

Computer 46 takes the data defining the image of the patient's body part from storage device 86 and displays a pictorial representation of the image I on monitor 90. Because the image of the patient's body part acquired during the previous imaging step includes a depiction of fiducial marker 76, the displayed image I will include a pictorial depiction 102 of the fiducial marker 76. The computer also actuates monitor 90 to display a pictorial representation 104 of fiducial marker 76 at a location on the monitor screen. Representation 104 is displayed at a position corresponding to its position in the XYZ frame of reference, to the position of the fiducial marker as derived from the magnetic field measurements. The orientation of representation 104 is set to correspond to the orientation of the fiducial marker determined from the magnetic field 25 measurements. If the representations 104 of the fiducial marker is in registration withdepiction 102 of the same marker on the monitor screen, this indicates that the image of the patient's body part is displayed at a location and orientation on the monitor screen corresponding to the location and orientation of the body part within the XYZ reference coordinate system.

At the start of the displaying step, the image I of the patient's body part is typically misregistered. Thus, the depiction 102 of the fiducial marker incorporated in the image is out of registration with the representation 104 of the fiducial marker derived from the magnetic positioning and orientation data. The user manually instructs the computer to reorient the image by applying appropriate inputs to manual input devices 92, 94 and 96 to shift and turn the image until the depiction 102 overlies the representation 104 and so that these two pictorial elements are properly aligned with one another. Where the stored image includes images or "slices" on a plurality of planes through the body part, the manual adjustments to the image may include selection of an image in the correct plane to show the depiction 102 of the fiducial sensor in a true size, with the imaging plane cutting through the sensor. By means of these manual control inputs,. the user effectively inputs relationship data into the system informing the system of the relationship between the initial frame of reference of the image and the X-Y-Z frame of reference of the system, and the system transforms the image into an image in the X-Y-Z frame of reference.

Once the manual controls have been actuated to bring depiction 102 and representation 104 of the fiducial marker with one another, a further "lock" signal is sent to the computer through a conventional data input device such as a keyboard 106 indicating that such registration has been accomplished. After 25 receipt of the "lock" signal the computer continually maintains registration of the fiducial marker depiction 102 in the image with the magnetic dataderived repr'sentation 104 of the fiducial marker.

Thus, if the position and orientation data for S" 30 fiducial marker 76 obtained through the magnetic field measurements indicate a change in the position, orientation or both of the fiducial marker 76, the computer will shift the magnetic field derived representation 104 of the fiducial marker on monitor 90 accordingly, and will also transform the image I, including the depiction 102 of the fiducial marker, in the same way. Thus, there is no need for the patient to remain in a fixed position during the procedure.

i~ Cl~; r(irillArl~ -31- For example, the particular image I shown in Fig. 1 is an image taken on a cutting plane parallel to the Y-Z axes. If the position and orientation data for the fiducial marker 76 derived from the magnetic field measurements indicates that the patient's body part has moved by in the +Y direction and has rotated clockwise around the +X axes since the image was initially brought into registration, then computer 46 will transform the image and actuate monitor 90 to show the image displaced and rotated in the corresponding directions on the monitor screen.

Methods of transforming stored images to show displacements and rotations. in two and three dimensions are well known in the art, and need not be described in detail here. However, such methods are commonly employed to show the image transformed in response to arbitrarily input displacements and positions. Here, the transformation techniques are applied in rjesponse to true displacements and rotations of the body part measured by the magnetic field monitoring system.

At this point, the image I of the patient's body part is displayed on the screen of monitor 90 in a position corresponding to its true position and orientation in the XYZ reference coordinates of the *frame and coils. The computer actuates display 90 to show a representation 108 of at least the distal end 56 of the probe 50. The position and orientation of representation 108 correspond to the position and orientation of the distal tip 56 derived from magnetic field measurements by sensor 60. As both the representation 108 and the image I are displayed on monitor 90 in positions and orientations corresponding to the true position and orientation of the probe and 35 of the patient's body part in the X, Y, Z reference direction coordinate system, the combined representation 108 and image I on the monitor screen accurately depict the position of the probe distal tip 32- 56 relative to the patient's body part. Also, as the position and orientation of probe distal tip 56 changes, the computer will alter the position and orientation of the probe representation 108 on the monitor screen. The representation of the monitoring probe tip 56 and of fiducial marker 76 may be pictorial representations or else may be schematic representations such as lines, arrows or the like.

Thus, the system will continually show the true relationship between the probe tip and the patient's body part. In effect, the system provides a result similar to that which would be achieved by continually imaging the patient during the medical procedure, as by continually taking X-ray, CAT, or MRI images to show the probe tip. However, the magnetic field monitoring steps according to this embodiment of the invention do not interfere with the ongoing medical procedure. No ionizing radiation is employed.

Moreover, the magnetic fields used are so small that they do not exert any significant, perceptible forces on magnetic materials in the sensing region. The system is also substantially insensitive to the presence of magnetic materials and/or stray electromagnetic fields in the sensing ;volume. The coils typically provide good access to the sensing volume for the position to reach the patient through the openings 40, so that the physician can conduct the required procedure. Further, the coils and the supporting structure can be removed immediately for even better access of an emergency crises during the procedure.

The ability of the present system to provide accurate positional information, and to superpose a representation of the probe on an image of the patient's body is useful in many medical procedures, but is especially valuable in procedures where an endoscopic instrument is to be used in the brain.

Although endoscopic instruments typically are equipped Ll o -33with devices such as miniature video cameras, fiberoptics or the like which provide an image of the tissues immediately surrounding the distal tip of the endoscope, such images do not permit the physician to establish the location of the tip in the brain, simply because many regions of the brain present the same endoscopic appearance. Using methods as discussed above, the physician can monitor the location and orientation of the instrument by viewing the representation of the distal tip of the probe against an image of the brain. Thus, the endoscopic instrument equipped with the probe can be used to perform surgical operations or other treatments in the brain, as by positioning the endoscope using the probe, removing the probe from the bore of the endoscope body 53 and then passing conventional endoscopic surgical instruments through the bore of the endoscope. Alternatively, the endoscope may have two bores, or one large bore, so that instruments may be inserted while the probe remains in place. Also, the sensor 60 may be mounted at the distal tip of the endoscope itself, so that the endoscope body 53 serves as the probe body. The methods and apparatus discussed above can also be used in all-other areas of 25 the body as, for example, in the spinal column; heart; otolaryngological tract; urogenital tract and organs; abdomen; and lungs. Additional embodiments of the invention provide a frameless stereotactic system. In a conventional stereotactic system, a rigid probe is guided by a frame mounted outside of the patient's body and fixed to a body part, so that the tip of the probe can be positioned at a precise location within the body. Thus the frame may be fixed to the skull and the probe advanced along a preselected line set by the 35 frame. Using systems as described above, no frame is necessary; the position of the probe relative to the patient is determined from the measured magnetic field components at the probe and at the fiducial marker on -34the patient. Where the probe is rigid, the magnetic sensor on the probe may be disposed at the proximal end of the probe, which remains outside of the patient's body during the procedure. In this case, the computer system is arranged to calculate the position of the probe tip from the position and orientation of the sensor.

In a further method according to the invention, a plurality of body parts such as the portions of a patient's limb 174 and 175 (Fig. 7) joined at a common joint 177 may be provided with probes 176A and 176B respectively. Each such probe is similar to the fiducial marker .76 discussed above with reference to Fig. i. Probe 176A is fastened by a band 180 to limb 174 so that the probe is held in substantially fixed position relative to the bone 182 within limb portion 174. To assure even greater accuracy, the probe 176 may be fastened directly to the bone, as by surgically inserting the probe into the limb, or by connecting the probe to pins (not shown) extending through the soft tissues of the limb to the bone. Likewise, probe 176B is fastened to limb portion 175 so that the probe is held in fixed position relative to the bone 183. Here again, an image of the body parts is acquired, as by X-ray,

CAT,

MRI or other imaging modality. The image includes images of probes 176A and 176B, as well as images of bones 182 and 183, and may also include images of other body parts such as the surrounding soft tissues.

30 In the same way as discussed above with reference to Figs. 1 6, the image is input and stored in an image storage unit associated with computer 46. The limb is then placed in the sensing volume 32 of the apparatus. Sensors incorporated in probes 176A and 176B are connected to the drive and amplification units 72 and 82 of the apparatus. The positions and orientations of these probes are acquired by the apparatus based upon magnetic field measurements in the same way as discussed above with reference to Figs. 1 6. The positions and orientations of the two sensors are monitored simultaneously. As shown in Fig. 8, the images I182 and 183 showing the two bones 182 and 183 are displayed on the monitor screen 90 by the computer.

The displayed images include a depiction 186A of probe 176A and a further depiction of 186B of probe 176B.

In the same manner as described above, the computer also displays representations 188A and 188B of probes 176A and 176B at positions corresponding to the positions of the probes determined by the magnetic field measurements. Utilizing the manual input controls associated with the computer, the user actuates the computer to change the position of Image 1182 until the depiction 186A of probe 176A in the image overlies the magnetic-field based representation 188A and is aligned therewith, whereupon the user inputs a lock signal. The user separately manually adjusts the position of image I 183 until the depiction 186B of probe 176B is superposed on the representation 188B derived from the magnetic field data in the correct alignment therewith. At this point, the user inputs a further lock signal.

25 Once both images have been aligned and placed in registration with the magnetic field frame of reference in this matter, both body parts are in initial, reference positions. Any subsequent movement of either probe 176A or 176B detected by magnetic S* 30 field measurements is taken as showing movement of the associated body part 174 or 175. Thus, when probe 176A moves to the moved position indicated in broken lines in Fig. 7 at 176A prime, the computer transforms the image 1182 of the associated bone 182 to the 35 transformed image I'182 (Fig. The rotation and/or translation applied to image 1182 to form transformed image I'182 correspond to the rotation and translation -36of the probe in going from its starting position 176A to its moved position 176A'.

The image of each body part is treated independently, in accordance with position and orientation data derived from the associated probe.

The system shows the images of the body parts, in particular, of bones 182 and 183, in true relative position and orientation as the bones move relative to one another. In effect, the system provides the equivalent of a continuous imaging system, such as a fluoroscopic imaging system, without the disadvantages thereof. The same system can also show the orientation and position of. one or more medical instruments, such as one or more probes equipped with sensors as discussed above with reference to probe of Fig. 1. As the images of both body parts are shown in true position and orientation relative to one another and relative to the XYZ reference direction coordinate system of the frame and coils, and as the representation of the instrument is likewise shown in :true position and orientation, the instrument is presented in proper position and orientation relative to both body parts.

In the illustration of Fig. 8, the images 1182 and 1183 are shown as including the depictions 186A and 186B of the probes. However, after the images have been brought into registration with the XYZ coordinate system by aligning the depictions of the probes with the magnetic field-based representations 188A and 188B, there is no need for further visible display of the probe depictions 186 or of the probe representations 188. At this stage, the display may depict only the area of interest, such as at joint 177.

In each of the systems discussed above, the image of the patient is brought into registration with the XYZ coordinate system by manual alignment, of the magnetic field derived representation of the probe -37with or marker with a depiction of the probe or marker in the image itself. However, this step can be automated, as by using automatic pattern recognition equipment to detect a depiction of the fiducial marker or probe in the patient image. With such automation, it is unnecessary to display a visible depiction or representation of a fiducial marker or probe. In the system discussed above with reference to Fig. 1 only one fiducial marker is provided on the patient's body part. Registration of the image data with the coordinate system thus depends upon the positional and orientation data for this single marker. For greater accuracy and versatility, a plurality of fiducial markers, and preferably three fiducial markers are employed at spaced apart, non-colinear locations on the patient's body part. Where a plurality of independently movable body parts are to be located, as in the embodiment of Figs. 7-8, plural probes may be mounted to each.

Alternatively or additionally, the patient's body part may be registered with the XYZ reference coordinate system through means other than a fiducial marker. In the simplest arrangement, the patient's body part is fixed to frame 30 at a known location and orientation, and the image I of the body part is displayed on monitor 90 at a location corresponding to this known location. Alternatively, the position and orientation of the patient's body part, such as the head shown in Fig. 1 is determined by mechanical or 30 other non-magnetic measuring devices and that data is used to determine the proper position and orientation of the image I on screen A sensor used in a further embodiment of the invention includes three field-sensing elements in the 35 form of semiconductor chips 250.(Fig. 9) Each chip includes one or more elongated bars of magnetoresistive material 252. Each such chip is sensitive to magnetic field components in the -38direction of the bars. The sensor further includes a foldable, sheetlike support 254 including a dielectric layer and having conductors 256 on the dielectric layer. Support 254 desirably is formed from a polyimide material of the type commonly used for flexible, tapelike electronic circuitry. Support 254 is initially in a generally cruciform shape as seen in Fig.9. The support includes a rectangular or square central panel 262, a tip panel 264 projecting from one edge of central panel 262, and a pair of side panels 266 and 268 projecting from opposite edges of the central panel. The support further includes an elongated strip 266 projecting from the edge of central panel 262 opposite from tip panel 264.

One chip 250 is mounted on central panel 262; another is mounted on tip panel 264 and the third chip 250 is mounted on side panel 266. The magnetoresistive bars 252 of the chip mounted on side panel 266 are parallel to the bars 252 of the chip mounted on tip panel 264, but both are orthogonal to the bars of the chip mounted on center panel 262. A temperature sensitive chip 258, which may incorporate one or more thermistors, thermoresistive elements or other temperature-sensitive devices, is mounted to the 25 other side panel 268 of the support. An amplification chip 260 is mounted to the elongated strip 266 of the support adjacent its juncture with the central panel 262. The components can be mounted on the support using conventional semiconductor mounting and connecting techniques adapted to operate on planar circuit panels.. After the components are mounted on the support, the support is folded over a core 270.

Core 270 is substantially in the form of a rectangular solid or cube, and preferably is formed from a dielectric material. The central panel 262 is positioned on one face 272 of the support. The tip panel 264, side panels 266 and 268 and strip 266 are folded over the edges of the core so as to position -39these on faces of the core adjoining face 272. The panels and strips are secured in position on the core.

In this condition, the finished sensor has magnetoresistive sensing elements 250 on three orthogonal faces of the cube, and the bars of the three elements extend in three orthogonal directions.

The entire sensor, including the magnetoresistive elements, the temperature sensor 258 and amplifier 260 desirably is less than about 5mm, and more preferably less than about Imm in width and thickness.

The sensor according to this embodiment may be mounted in an elongated probe body 280 so that the sensing elements are disposed at or near the distal tip of the probe body, and so that the elongated strip 266 of- the sensor, and the conductors 256 thereon, extend towards the proximal end of the probe body for connection to external conductors (not shown). The compact construction of the sensor permits use of a probe body having a very small diameter, desirably about 0.8 mm or less. A probe according to this embodiment can be used in methods as discussed above.

In this sensor, each sensing element is sensitive to field components in one direction parallel to the face of the sensor, rather than normal to the face as in 25 the sensors discussed above. Also, 'the built-in *amplifier at the sensor enhances the signal-to-noise ratio. Otherwise, the probe is used in the same manner as discussed above.

In a method according to a further aspect of the invention, the probe of Figs. 9 and 10 is employed to take physiologic temperature measurements within the patient's body. As discussed above, the probe is inserted into the patient's body, and the distal tip of the probe is moved over a range of positions within the body. As the probe is moved, the magnetic field monitoring system keeps track of the location of the distal tip within the XYZ reference direction coordinate system of the frame, so that each temperature measurement is associated with a position in that coordinate system. Thus, a map of temperature versus position is stored in the memory of computer 46. Preferably, such a map is displayed or stored in association with an image of the relevant body part of the patient. As discussed above, such an image is brought into registration with the reference direction coordinate system of the magnetic field apparatus through use of a fiducial marker or through other means. Thus, the temperatures measured by the probe can be mapped onto the features of the body part as, for example, by shading or coloring various portions of the image to correspond with different temperatures or by showing contour lines overlaid on the body image as illustrated in Fig. 11.

Probes and methods as described with reference to Figs. 9 11 can be utilized for example by inserting the probes into the body through natural body orifices to probe cavities such as the intestinal tract, urinary tract, and respiratory tract and detect regions of locally elevated temperature. Such regions are often associated with pathological conditions such as malignant tumors. Further, once such a region has been located and precisely mapped, the physician can use the same or a different probe to examine the area visually and/or to perform biopsy, surgical, excision or other treatment.

The temperature sensitive element 258 may be ee.. replaced or supplemented by one or more elements S 30O sensitive to any other physiologic variables, such as pH, ionic concentration, or others. Also the element or elements sensitive to temperature or to other physiologic variables can be formed separately from the magnetic sensor.

35 As an alternative to display of the mapped physiologic data superposed on an image of the body part, the mapped data may be displayed alone in pictorial form or otherwise. For example, a pictorial -41representation showing a region of elevated temperature can show the size and shape of a tumor even without superposition. The maps of the physiologic data desirably are derived and displayed in real time, during the procedure, so that the displayed maps (with or without superposed body part images) can guide the physician in further mapping.

Thus, the physician may first move the probe over a limited number of widely-spaced locations in the region of interest to derive a coarse map. This coarse map can be used to locate regions requiring fine examination by mapping of closely-spaced points.

In further variants of the apparatus and methods discussed above, the Z-axes coils, and the corresponding uniform and gradient fields are omitted.

Thus, the system uses only two orthogonally directed uniform fields and gradient fields from only two coil sets. However, the system can still derive the position and orientation of the probes in three dimensions. Because the flux vectors for the uniform X, Y and Z fields are orthogonal to one S. another, the flux vector for any one uniform field can be calculated by calculating the cross-product of the other two. Stated another way, the components which a 25 hypothetical third uniform field flux' vector would have in the coordinate system of the sensor can be calculated from the actual components of the other two uniform-field flux vectors in the same coordinate system. Thus, the normalized components H m of a hypothetical Z-direction uniform field flux vector can be calculated from the actual normalized components measured for the X-direction and Y-direction fields: H'x',z H'y',x 35 H'y',z H'z',y H'x',x) H'z',z H'x',y H'y',x) -42-- Thus, all of the components H'mn can be acquired by operating only the X and Y coil sets in uniform-field mode.

In the methods discussed above using three actual coil sets, the system calculates the magnitude of the projection of the reverse-field or gradient flux vector for each coil set onto axis of that coil set. However, the measured values for the reversefield flux vector from one coil set can be eliminated.

To find position for the axis corresponding to the missing coil set, the system calculates a projection of one of the flux vectors from one of the two remaining coil sets onto the third axis, as well as onto the axis of the coil set itself. This can be accomplished computationally by simply copying the values for one coil pair (one column in Table II) into the values for the third coil pair. Thus, where the Zdirection coil is omitted, the system sets: Rx ,z Rx',y RyZ Ry',y Rzi, 2 Rzi,y and otherwise performs the computation in the same manner as described above. However, because the Ydirection field has its principal component along the 25 Y-axis, and only a radial component along the Z-axis, the magnitude Rz of the Z-axis component at any given Z-direction distance from the origin will be about 0.36 times the magnitude R. of the Y-direction .**:component at an equal Y-direction distance. Thus, the Z-direction gradient dRz/dZ is about -0.36 the Ydirection gradient dRy/dY, and the conversion factor Mz used to convert Z-axis flux component to Z-axis distance is adjusted accordingly. Where a calibration step as discussed above is employed, the 2-axis 35 conversion factor will be measured during Y-axis coil set activation.

i -43- The system according to this variant needs only two orthogonal coil sets, and thus provides even better access to the patient for the physician.

In the apparatus and methods discussed above, the various coil sets are actuated alternately, so that only one coil set is operating at any given time. However, the predetermined time sequence of magnetic fields used in other embodiments of the invention may include two or more fields applied simultaneously. The simultaneously-applied fields desirably are varied at different frequencies. In this case, the signals from the sensors representing the magnetic fields will incorporate separate components varying at these different frequencies. Each such component represents the sensor reading for one field.

These components can be segregated by conventional filtering techniques and then evaluated separately to provided separate readings for the separate fields.

For example, the magnitude of the AC waveform at each frequency can be evaluated separately. In each of the variants discussed above, the magnetic fields ar; applied by coaxially positioned Helmholtz coils.

However, other magnetic field generating devices, such as permanent magnets or other coil arrangements can be 25 employed. Also, it is particularly desirable to use uniform fields and fields having a component varying linearly with distance (uniform gradient) as discussed above. However, other combinations of magnetic fields can be employed. The fields need not necessarily include uniform fields, but can instead include a plurality of different fields with different, known gradients, provided that it is possible to mathematically deconvolute position and orientation from the sensor readings acquired during the various 35 fields Also, fields with non-uniform gradients, and with non-linear variation of one or more components with distance can be employed. Such fields are less preferred. Moreover, the magnitude of each field -44component which is measured in the present invention desirably is quasilinear s with distance in a direction within the sensing volume. As used in this disclosure, a component magnitude is "quasilinear" with distance in a particular direction if the magnitude can be described by a polynomial function of position in such direction of the form: a o qO al1q a 2 q 2 a 3 q 3 aqn in which q is position and in which the constant (zero-power) term, the first-power term or the sum of these terms predominates over the other terms. Stated another way, when evaluated for any distance q within the sensing volume the sum (a 0 q 0 alql) should be at least twice the sum (a2q2 a3q 3 aqn).

Preferably, (aoq 0 alql) is at least 5 times, and more preferably at least 10 times the sum (a 2 q 2 a 3 q 3 aqn). In the most preferred arrangements discussed above, a 0 qO predominates in the homogeneous fields, the other terms being as close to zero as is practicable, whereas in the gradient fields, alql predominates, and all of the other terms are as close to zero as is practicable. If a particular field includes components which are not measured in determining position or orientation, those extraneous 25 components need not be quasilinear.

As these and other variations and combinations of the features discussed above can be utilized without departing from the present invention as defined by the claims, the foregoing description of 30 the preferred embodiments should be taken by way of illustration rather than by way of limitation of the present invitation.

•INDUSTRIAL APPLICABILITY Medical, surgical, veterinary and similar 35 procedures.

Claims (13)

1. A method for displaying the location of a probe in a living subject including the steps of: providing at least one fiducial marker in fixed position relative to the subject; acquiring the image of the subject including a depiction of said at least one fiducial marker. determining the locations of said at least one fiducial marker and of said probe in a common frame of reference by measuring magnetic fields transmitted to or from said probe and said fiducial marker, whereby the position of said probe relative to said at least one fiducial marker is known; and superposing a representation of said probe on said image of said subject at a position relative to said depiction of said at least one fiducial marker corresponding to the position of said probe relative to said at least one fiducial marker in said common frame of reference.
2. A method as claimed in claim 1, further including the steps of determining the orientation of said probe and of said at least one fiducial marker in said common frame of reference, said step of superposing said representation of said monitoring probe on said image of said subject being 20 performed so that the orientation of said monitoring probe representation relative to the depiction of said at least one fiducial marker corresponds to the orientation of said monitoring probe relative to said at least one fiducial marker in said common frame of reference.
3. A method as claimed in claim 1, wherein said superposing step includes 25 the step of displaying said representation of said monitoring probe in a display frame of reference so that the position of said monitoring probe representation corresponds to the position of said monitoring probe in said common frame of reference and displaying said image of said subject in said display frame of reference so that the location of said representation of each fiducial marker in said display frame of reference corresponds to the location of such fiducial marker in said common frame of reference.
4. A method for depicting a body part including the steps of: C:Wy Documents\KATESPECI\P709793.doc -46- providing a sensor in fixed position relative to the body part and determining the orientation of said sensor in a fixed frame of reference; acquiring an image of the body part in a starting orientation; monitoring the orientation of said sensor in said frame of reference by monitoring magnetic fields transmitted to or from said reference sensor to thereby determine a moved orientation of said body part after movement of the body part from said starting orientation; transforming said image of said body part in said starting orientation into an image of said body part in said moved orientation and displaying the transformed image.
A method as claimed in claim 4, wherein said monitoring, transforming and depicting steps are repeated as said body part is moved through a range of moved orientations so that a transformed image corresponding to said each said moved orientation is displayed substantially in real time while the body part is in such moved orientation.
6. A method as claimed in claim 4, wherein said monitoring, transforming and displaying steps are performed during a medical treatment procedure and said displaying step is performed so that the displayed image is visible to the physician performing such procedure, wherein such physician may view a 20 transformed image corresponding to the current actual orientation of the body part at any time during the procedure.
7. A method as claimed in claim 5 or claim 6, wherein said step of providing a sensor includes the step of providing a separate sensor fixed to each of a *....plurality of body parts, said steps of acquiring an image and determining the orientation of the sensor including the steps of acquiring images of all of said body parts and determining the orientations of all of said sensors while said Sbody parts are in respective starting positions, said step of monitoring the orientation of said sensor and determining moved orientation of the body part including the step of monitoring the orientation of all of said sensors and determining a separate moved orientation for each of said body parts, said transforming step including the step of transforming the image of each said body part into an image of that body part in its moved orientation, said C:VMy DocumentSKATESPECRP7O979doc -47- displaying step including the step of displaying all of said transformed images together so that the orientations of said transformed images relative to one another correspond to the orientations of said body parts relative to one another.
8. A method as claimed in claim 7, further including the step of determining the position of said sensors in said fixed frame of reference while said body parts are in said starting orientations and determining the positions of said sensors in said frame of reference when said body parts are in said moved orientations, said steps of transforming and displaying said images including the step of adjusting the positions of said displaying transformed images of said body parts relative to one another to compensate for movement of said body parts relative to one another so that said displayed images correctly represent the positions of said body parts relative to one another.
9. A method as claimed in claim 7, wherein said body parts are bones connected to one another at a common joint.
A method as claimed in claim 6, further including the steps of providing an instrument probe on a medical instrument and monitoring the orientation of said instrument during said procedure by monitoring magnetic fields transmitted to or from said instrument probe and displaying a representation of said 20 instrument in conjunction with said image of said body part so that the orientation of said instrument representation relative to said displayed image of said body part corresponds to the orientation of said instrument relative to said body part.
11. A method as claimed in claim 10, further including the step of 25 determining the positions of said sensors and said instrument probe in said fixed frame of reference while said body part and said instrument are in said starting orientations, determining the positions of said sensors and said instrument probe in said frame of reference when said body part and said instrument are in said moved orientations, said steps of displaying said image and said representation of said instrument including the step of adjusting the position of said representation of said instrument relative to said displayed transformed image of said body part so that said displayed image and C:\My Documents\KATEkSPECIlP709793.doc -48- representation correctly represent the position of said instrument relative to said body part.
12. A method for displaying the location of a probe in a living subject, substantially as hereinbefore described with reference to what is shown in the accompanying drawings.
13. A method for depicting a body part, substantially as hereinbefore described with reference to what is shown in the accompanying drawings. DATED :9 December, 1999 PHILLIPS ORMONDE FITZPATRICK Attorneys For: BIOSENSE, INC. D4A&d^i S .e S S. S. C 5*5* S S SS .me S @5 c C C S C:Wy Documents\KATE\SPECI\P709793.doc
AU64421/99A 1993-10-06 1999-12-09 Magnetic determination of position and orientation Expired AU746410B2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US132479 1993-10-06
AU64421/99A AU746410B2 (en) 1993-10-06 1999-12-09 Magnetic determination of position and orientation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
AU64421/99A AU746410B2 (en) 1993-10-06 1999-12-09 Magnetic determination of position and orientation

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
AU63578/98A Division AU709793B2 (en) 1993-10-06 1998-04-23 Magnetic determination of position and orientation

Publications (2)

Publication Number Publication Date
AU6442199A AU6442199A (en) 2000-02-24
AU746410B2 true AU746410B2 (en) 2002-05-02

Family

ID=3749115

Family Applications (1)

Application Number Title Priority Date Filing Date
AU64421/99A Expired AU746410B2 (en) 1993-10-06 1999-12-09 Magnetic determination of position and orientation

Country Status (1)

Country Link
AU (1) AU746410B2 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4945914A (en) * 1987-11-10 1990-08-07 Allen George S Method and apparatus for providing related images over time of a portion of the anatomy using at least four fiducial implants
US5318025A (en) * 1992-04-01 1994-06-07 General Electric Company Tracking system to monitor the position and orientation of a device using multiplexed magnetic resonance detection

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4945914A (en) * 1987-11-10 1990-08-07 Allen George S Method and apparatus for providing related images over time of a portion of the anatomy using at least four fiducial implants
US5318025A (en) * 1992-04-01 1994-06-07 General Electric Company Tracking system to monitor the position and orientation of a device using multiplexed magnetic resonance detection

Also Published As

Publication number Publication date
AU6442199A (en) 2000-02-24

Similar Documents

Publication Publication Date Title
EP1103229B1 (en) System and method for use with imaging devices to facilitate planning of interventional procedures
US6014580A (en) Device and method for specifying magnetic field for surgical applications
CA2414914C (en) Wireless position sensor
CA2325540C (en) Three-axis coil sensor
DE19629890B4 (en) Magnetic resonance imaging and tracking system
US6669635B2 (en) Navigation information overlay onto ultrasound imagery
US7302288B1 (en) Tool position indicator
US6359959B1 (en) System for determining target positions in the body observed in CT image data
US9179827B2 (en) Systems and methods for determining the position and orientation of medical devices inserted into a patient
EP0427358B1 (en) Mechanical arm for and interactive image-guided surgical system
US5309913A (en) Frameless stereotaxy system
EP0922438B1 (en) Image guided interventional procedures
US5285787A (en) Apparatus for calculating coordinate data of desired point in subject to be examined
US7869865B2 (en) Current-based position sensing
JP6559748B2 (en) System and method for identifying a target
CA2505464C (en) Catheter tracking with phase information
US5913820A (en) Position location system
Patriciu et al. Automatic brachytherapy seed placement under MRI guidance
US7834621B2 (en) Electromagnetic tracking employing scalar-magnetometer
US6920347B2 (en) Trajectory storage apparatus and method for surgical navigation systems
US6226547B1 (en) Catheter tracking system
US6774624B2 (en) Magnetic tracking system
EP1174082B1 (en) Calibration system for magnetic fields
US6493574B1 (en) Calibration phantom and recognition algorithm for automatic coordinate transformation in diagnostic imaging
US20050119556A1 (en) Catheter navigation within an MR imaging device

Legal Events

Date Code Title Description
FGA Letters patent sealed or granted (standard patent)