 |
Description  |
|
|
FIELD OF THE INVENTION
The present invention relates to the field of imaging devices which produce images based on non-invasive methods such as computed tomography, magnetic resonance imaging and ultrasonography. More specifically, the present invention relates to a
system for manipulatively displaying images in the operating theater and indicating the position on those displayed images of a magnetic field probe manipulable by a surgeon, especially to assist in locating problem areas during the performance of
surgery.
BACKGROUND OF THE INVENTION
The term stereotaxis describes a technique, most often applied to the nervous system, in which the contents of the patient's skull (or body) are considered in a precise three dimensional space defined by a measuring instrument, such as a
stereotactic frame, which is fixed to the patient's skull or body. Stereotactic frames are mechanical devices typically based upon a Cartesian or polar coordinate system. These systems typically include a means for securing the stereotactic frame
device to the patient, at least one measuring scale for determining and confirming target coordinates and probe trajectories, and a probe holder/carrier.
The probe holder/carrier directs a surgical probe or some other instrument to a desired three dimensional location within the work space which is defined with respect to the geometry of the stereotactic frame. In practice, the stereotactic frame
is also used to position a probe or other instrument inside the body into an anatomic or pathologic structure. The frame coordinates of the target structure are determined from stereotactic imaging studies including Computed Tomography (CT), Magnetic
Resonance Imaging (MRI), ultrasonography, etc., and radiographically based procedures such as positive contrast ventriculography, or stereotactic atlases.
The use of stereotactic methods in the management of human brain tumors was first proposed in the early 1900's by the British physiologist, Robert Henry Clarke. Clarke patented a device for human stereotactic neurosurgery in 1912. However, the
first human stereotactic procedure was not performed until 1947 when Spiegel and Wycis of Philadelphia attempted a ventriculography based dorsal median thalamotomy for psychiatric disease. Stereotactic instruments, methods and indications rapidly
evolved thereafter. The three dimensional locations of intracranial targets were determined by means of stereotactic radiographically based methods, most commonly positive contrast ventriculography and stereotactic atlases based on the identification of
radiographically established intracranial landmarks. The most common clinical use of stereotactic instrumentation in the late 1950's and 1960's was the placement of subcortical lesions to treat movement disorders, primarily the tremor of Parkinson's
disease. However, following the introduction of L-Dopa in 1968 indications for stereotaxis decreased and the number of stereotactic procedures declined precipitously.
Nevertheless, old concepts in stereotactic frame design which were influenced by radiographically based point-in-space procedures carried over into the next phase of evolutionary stereotaxis. The advent of CT scanning in the early 1970's, and
later magnetic resonance imaging, MRI scanning, rekindled interest in stereotactic surgery for two reasons: First, Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) provided a precise three dimensional data base which could be incorporated
into the three dimensional coordinate system of a stereotactic frame. Secondly, in contrast to radiographically based examinations, one could actually see the complete intracranial tumors on these new computer based imaging modalities.
A progressive and expanding interest in CT and MRI based stereotactic procedures for intracranial tumors and other indications has been noted in neurosurgery since late 1979. Old stereotactic frames were modified to provide CT and MRI
compatibility. New imaging compatible stereotactic frames were developed and introduced into the medical market. Nonetheless, the old concept of the stereotactic frame as a rigid device fixed to the patient's skull which was used to mechanically direct
probes and other instruments to defined intracranial target points, persisted.
However, in the contemporary era of CT-and MRI-based stereotaxis, the coordinates for the intracranial target are derived from stereotactic CT and MRI examinations instead of a data base consisting only of projection radiographs. But during the
stereotactic CT and MRI examinations the patient's head must still be fixed in the rigid, confining, stereotactic frame. CT and MRI opaque external fiducial reference marker systems are frequently applied to the frame to facilitate and simplify the
calculation of stereotactic coordinates from the imaging modalities.
The mechanical constraints of a rigid stereotactic frame and the complexity of the computer-based imaging data bases act to limit most contemporary stereotactic procedures to point-in-space targets for biopsy, placing interstitial catheters,
localizing a bone flap over a tumor and the like. However, the availability of high capacity and low cost computer work stations have provided a method for reducing the complexity of the image reconstructions and target point cross registration between
imaging modalities.
In addition, computer interactive stereotactic methods allow tumors defined by CT and MRI to be considered as volumes in space and provide to surgeons graphical displays which indicate the CT and MRI defined boundaries of the lesion within a
defined stereotactic surgical field. This volumetric display technique is superior because it is necessary to remove the entire volume of the tumor, especially at its outer boundary, to ensure that further growth does not take place after the operation.
Such technique for volumetric stereotaxis was first proposed by Kelly et al in 1982. In these procedures, a tumor volume is reconstructed from stereotactic CT or MRI data and reformatted along a surgical viewing trajectory defined by a
stereotactic frame. During surgery, an operating room computer system displays cross sections of the reformatted tumor volume with respect to surgical instruments directed into the surgical field using the stereotactic frame as a reference source.
Intra-operatively, the surgeon monitors the computer generated image of the surgical field which was derived from CT or MRI scans, as well as the surgical field itself.
In 1986 a system was developed for superimposition of the computer image upon the surgical field by means of a heads-up display unit attached to the operating microscope. This allowed the surgeon to simultaneously view updated reformatted and
scaled images of the CT or MRI defined surgical field visually superimposed upon the actual surgical field. Since 1984 over 2300 computer-assisted stereotactic procedures and more than 800 computer-assisted volumetric stereotactic tumors resections have
been performed at the Mayo Clinic. It has been found that these procedures allow a more complete removal of a tumor in a minimally invasive way.
However, stereotactic systems to date have had a series of associated problems. Stereotactic frames are cumbersome in general. They are especially cumbersome for procedures requiring more than a few target points and in volumetric stereotactic
procedures where the demands of the procedure dictate the need for a larger working area, yet where such demand comes into conflict with the physical structure of the frame. A conventional stereotactic frame is typically a cage structure extending about
the patient's head and therefore inherently restricts the freedom of movement of the surgeon. Changing a target point or trajectory to reach that new target point involves a separate mechanical adjustment of the stereotactic frame. This is not usually
a problem in point-in-space stereotaxis, as in biopsy procedures for example.
But many mechanical adjustments become very cumbersome when a surgeon is confronted with an infinite number of points which define the boundary of a volumetric lesion. In addition, a stereotactic reference frame must be applied to a patient's
head in order to acquire a pre-surgical data base.
Some surgeons find the stereotactic frame application procedure difficult and time consuming. Patients also find this uncomfortable. In addition, the necessity to repeat CT and MRI examinations for the pre-surgical data base increases the cost
to the patient. Finally many surgeons are intimidated by mechanically complex devices in general and stereotactic frames in particular. Furthermore, although the mathematics in stereotaxis are understandable, many surgeons are uncomfortable with these
also.
Since 1987 there has been an interest in so-called frameless stereotactic procedures. In the procedures described so far, a multi-jointed digitizing arm is indexed to the patient's head. Typically, precision potentiometers or optical encoders
on each of the joints of the digitizing arm provide feedback from which real world coordinates of a three dimensional point are determined by a host computer system.
In some systems, reference marks are placed on the patient's scalp. Imaging studies are performed at surgery. The surgeon uses the digitizing arm to touch these registration points. The coordinates of these known points correspond to reference
marks on the imaging studies. The computer can then calculate a transformation matrix to allow transformation of the real world coordinate system to the coordinate system of the imaging study. In practice, a cursor, which corresponds to the position of
the tip of the pointer in the surgical field, is displayed on CT or MRI slices or three dimensionally rendered images based on CT or MRI.
The problem with these multi-jointed digitizing devices is that they also are cumbersome and restrict the surgeon's freedom of movement. In addition, the 5 or 6 encoders at each of the 5 or 6 joints in the multi-jointed digitizing device can
occasionally combine to form a significant non-offsetting error resulting in unpredictable results in surgery.
Although advances in technology include faster hardware and software, improved error detection and improved mathematical treatment in the techniques of producing the images produced and better resolution with respect to individual images selected
from an object to be scanned, few such improvements have been directed toward the surgical practitioner to facilitate his working through the procedure. Such improvements are needed to balance the improvements in hardware and software to improve the
overall effectiveness of the surgeon's skill.
Several advances in the software and hardware areas have been made, however none enable the effective use and manipulation of two dimensional imaging system as keyed to a three dimensional locational system to be used with a patient during
surgery. The following patents outline some of these improvements.
U.S. Pat. No. 4,849,692 issued on Jul. 18, 1989 to Ernest B. Blood and entitled "Device for Quantitatively Measuring the Relative Position and Orientation of Two Bodies in the Presence of Metals Utilizing Direct Current Magnetic Fields" uses
two or more transmitting antenna of known position and orientation. Each transmitting antenna is driven, one at a time by a pulsed direct current signal. The receiving antenna measure the transmitted signals, one axis at a time, and then measures the
earth's magnetic signal, one axis at a time.
U.S. Pat. No. 4,945,305, issued on Jul. 31, 1990 to Ernest B. Blood and also entitled "Device for Quantitatively Measuring the Relative Position and Orientation of Two Bodies in the Presence of Metals Utilizing Direct Current Magnetic Fields"
describes improvements to the '692 patent and improved locational and orientational data. Both the U.S. Pat. Nos. 4,849,692 and 4,945,305 patents relate to the use of magnetic fields to determine location in a three dimensional area despite the
occasional and changing presence of metallic bodies which usually serve to distort the very field which is being relied upon for measurement. Both the U.S. Pat. Nos. 4,849,692 and 4,945,305 relate to locating a point in three dimensional space and
are unrelated to either imaging or surgery.
U.S. Pat. No. 4,951,653 to Fry et al entitled "Ultrasonic Brain Lesioning System" discloses the use of ultrasound, CT (computerized axial tomography), or MRI (magnetic resonance imaging) in probing for site localization in conjunction with a
skull fixation system. A precision ball provides linear and rotary positioning data by way of a cup fitting over a plurality of spheres and a linear encoder which interfaces with the cups. A bulky apparatus is utilized with a series of joystick type
"precision balls" to direct an ultrasound signal through a cooling media to produce volume lesions in the brain at the site of identified brain tumors. The device is not utilized with open surgery, and the production of lesions is based upon the machine
targeting of pre-existing tumors and the production of lesions without cutting the body.
U.S. Pat. No. 4,959,610 to Suzuki et al, entitled "Magnetic Resonance Apparatus" discloses details pertinent to NMR electromagnetic field atomic theory. U.S. Pat. No. 5,050,608 to Watanabe et al, entitled "System for indicating a Position to
be Operated in a Patient's Body," discloses the use of an articulated probe as a control source for displaying a series of tomographical images on a cathode ray tube. This device assumes the orientation of the patient with respect to the device. If the
patient moves, especially during surgery, the mechanical pointer may be pointing to a portion of the patient's anatomy such that the computer controls produce a tomographical image which is keyed to another portion of the patient's anatomy.
Further, since the Watanabe device includes four joints, a significant error is introduced in the resolution since the spatial Location of the pointer is dependent on the positions of the various angular displacement sensors at the joints of the
articulated arm of the pointer. Further, the pointer is made collapsible so that it can be pushed toward the scalp to indicate an affected part in order for a surgeon to be enabled to make an incision. A solid tip can be pushed into the brain to
indicate the depth of the tip into the brain while using the notches on the pointer to measure depth into the brain. This device is stated as being useful to assist in the initiation of surgery rather than assist during surgery, particularly since there
is no provision to account for movement of the patient during the surgery.
U.S. Pat. No. 5,094,241 to Allen, entitled "Apparatus for Imaging the Anatomy," involves the placement of implants below the skin level and on the bones in order to key a patient to an imaging system. This method is utilized to account for
problems in re-imaging for circumstances where significant amounts of time pass between a first and subsequent examinations. In such cases, a shift in viewing angle which might make a volume of interest appear larger or smaller compared to a subsequent
examination, is corrected for by using the implants. The implants utilized must be of the type which will show up on a tomographic imaging system in order to register it to the images produced in a given examination.
U.S. Pat. No. 5,107,839 to Houdek et al and entitled "Computer Controlled Stereotaxic Radiotherapy System and Method," discloses a low frequency electromagnetic position detection means. The patient's head is repetitively position re located
using a halo attached to the patient's head with skin piercing screws and which operates within a large multi-structured stereotaxic cage. The halo and cage would significantly interfere with surgery, and would be affected by the presence of metal
surgical tools within the stereotaxic cage.
U.S. Pat. No. 4,791,934 to Brunnett, entitled "Computer Tomography Assisted Stereotactic Surgery System and Method" discloses a system using a multi-jointed referencing system. In the Brunette patent, a CT scan occurs at one location and is
digitally stored in a computer. At a second location the patient is positioned in a digital radiographic imaging device utilized to produce a shadowgrahic image which is also stored. The shadowgraphic image is then registered with the scan image which
may be accomplished visually on a video monitor on an operating table and "registered" in 3 dimensions with the CT data using shadowgraphic images shown on a display. The Brunette patent teaches that once the patient is "registered" the surgeon can then
plan the best path of entry with a biopsy needle. The shadowgraphic image is formed with an X-ray device, thus presenting concerns about excess radioactive exposure of the patient.
Most of these systems are bulky, cumbersome, and generally more related to matching the coordinates of one system with those of another system. Each one presents an advancement in the art, but all fail to teach the construction of a system which
joins the coordinate resolution of a physical three dimensional system with the two dimensional nature of non-invasive imaging.
SUMMARY OF THE INVENTION
The present invention enables stereotactic surgery with a minimal structure present in order to facilitate surgery yet with enhanced accuracy in stereotactic imaging assistance to the surgeon. This device and method is known as "frameless"
stereotaxis system, and uses coordinate referencing between the CT or MRI scans and a stereotactic referencing system which uses an electromagnetically driven system to determine the location of a stylus.
Magnetic field distortion is reduced through the use of a low frequency magnetic field. One such system utilizing low frequency radiation is the "Flock of Birds.TM." system currently available from Ascension Technology Corporation of Burlington,
Vt. The frame of reference through which this system operates is coordinate referenced to the coordinates of a composite image formed from a CT, MRI, or other system. Slice images from the CT, MRI, or other system scan are arranged and may be generated
in two dimensional fashon in response to the position of a pointer as they are referenced to the orientation of a patient during an operation. As has been established in the literature, the precision and exactitude with which surgery can be performed is
directly related to the surgeon's ability to pinpoint the tissue mass to be removed or cut. In the present device and method, registration of the scan images illustrating the lesion, with the physical position of the patient is performed by first
synchronizing to two coordinate systems. Such synchronization can be performed in several ways, including the calibration of the stylus to several landmarks on the patient's anatomy.
The device and method of the invention also utilizes a transformation matrix which corrects for both angular displacement of the patient with respect to the "bird" system coordinates, and angular displacement of the visual image slice with
respect to the angular displacement of the patient. The invention facilitates the use of peripheral devices such as the use of a "heads up" display, or other line of sight referencing structure.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a perspective view of a standard Mayfield or Gardner headholder showing the positions of attachment for the electromagnetic source and the use of the electromagnetic receiver and stylus;
FIG. 2 is a perspective schematic view of the headholder of FIG. 1 with a patient, having a volumetric lesion, in operating position and the stylus employed;
FIG. 3 is an expanded view of the electromagnetic transmitter and the electromagnetic receiver, without the stylus, and illustrating the coordinate system in which they operate;
FIG. 4 illustrates the use of a calibration grid into which the electromagnetic receiver shown in Figures 1-3 is placed when calibrating the electromagnetic signal to account for non linearities in the electromagnetic field from a transmitter;
FIG. 5 is a graphical representation of the cartesian coordinate system utilized in conjunction with the present invention, and the angular displacement values .beta., .alpha., and .gamma., associated with rotation about the X, Y, and Z axes,
respectively;
FIG. 6 illustrates the computation of a midpoint MP, utilizing the reference points R1, R2, and R3;
FIG. 7 illustrates the cubic reference points utilized in the computation of warp correction; and
FIG. 8 is a schematic representation of one possible system configuration in which the method and system of the present invention can be practiced.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
The system of the present invention includes a system and method which correlates (1) a three dimensional magnetic field digitizer, (2) the anatomical features of a patient and (3) diagnostic images of the patient such as Computed Tomography (CT)
and Magnetic Resonance Imaging (MRI). The explanation of the system can be best explained by initially referring to FIG. 1.
FIG. 1 is a perspective view of a standard Mayfield or Gardner headholder, hereinafter headholder 11, but with some of the head support structure removed for clarity. The perspective taken in FIG. 1 would best enable the surgeon to be seated at
the lower right portion of FIG. 1. A "Y" shaped structure 13 includes a slot 15 at its upper portion to interfit with the head support structure which was removed for clarity.
Attached to this "Y" shaped structure 13 is a transmitter structural support 17 which is shown extending away from the "Y" shaped structure 13 to the left and curvingly upward. The transmitter structural support 17 terminates into a square
shaped end 19. The square end 19 has a dimensions matching the base dimensions of an electromagnetic transmitter 21. Transmitter 21 is a frusto-pyramidal shaped box, and is supported by the transmitter structural support 17 in a manner to enable it to
be supported as near as possible to the position occupied by the patient's head, as will be shown in FIG. 2.
Permanent mounting of the transmitter is desirable for a number of reasons. First, it will always bear the same relationship, during surgery, to the operating table, the headholder support 15 as well as the further headholder structural supports
not shown in FIG. 1. Further, since most of the structure shown in FIG. 1 is ferromagnetic, tending to distort any magnetic field present, the constant position of the transmitter 21 will enable these distortions to be accounted for at one time, with
out the necessity to re-calibrate each time the transmitter 21 is shifted.
Also shown in FIG. 1 is the electromagnetic receiver 23 and attached stylus 25. As will be shown, the point of reference in this system is the center of the electromagnetic receiver 23. However, since the center of the electromagnetic receiver
23 cannot be positioned immediately adjacent the structure to be measured, computational adjustments are performed to enable the referencing to be accomplished with respect to the tip of the stylus 25. Such adjustments are easily accomplished once the
length of the stylus 25 is known. Also shown leading away from the electromagnetic receiver 23 is a lead 27, through which are passed signals which were received from the electromagnetic receiver 23. The lead 27 is connected to a computer (not shown in
FIG. 2) in order that the position of the electromagnetic receiver 23 may be utilized to reference the relevant scan section, as will be shown.
In FIG. 1, the electromagnetic transmitter 21 and receiver 23 are part of a commercially available system known as the "Bird" Position and Orientation Measurement System, which is manufactured by Ascension Technology Corporation of Burlington,
Vt. The "Y" shaped structure is part of a COMPASS.TM. Stereotactic Positioner which is commercially available from Stereotactic Medical Systems, Inc of Rochester, Minn. The use of the electromagnetic transmitter attached directly to a stereotactic
positioner allows the coordinates being acquired by the "Bird" to be translated to the coordinate system of the stereotactic positioner because the orientation of the two coordinate systems remain in constant alignment. The value of this approach is
that the stereotactic positioner, a proven tool for localizing surgical targets, acts to independently confirm the position reported by the "Bird" to verify that the positioning system is performing correctly. In doing so, the patient's head 31 is
placed within the magnetic field produced by the electromagnetic transmitter 21, allowing coordinate determination, both with respect to the "Bird" and the COMPASS.TM. of specific anatomical references and that are also seen on CT or MR images. In
doing so, calculations can be made to relate the location of the stylus to correlate the patient's anatomy to the diagnostic images.
Referring to FIG. 2, an enlarged view, taken from a position similar to that shown in FIG. 1, illustrates the position of the electromagnetic transmitter 21, headholder support 15, electromagnetic receiver 23 and stylus 25 with respect to a
patient's head 31 having a volumetric lesion 33 shown at the approximate center of the patient's head 31. Also shown are additional head support structures 37 used to grasp and fix the position of the patient's head 31. Note the openness of the area
about the patient's head 31 and the associated freedom with which surgery may be performed.
The importance of this "frameless" device in the surgical environment enables the surgeon to use this device as a guide to relate the surgical field to the diagnostic images helping him/her to better determine the location and margins of the
legion. This is especially important when the lesion is adjacent to areas that are sensitive to surgical invasion.
The "Bird" tracking device, as it is presently commercially available, is a six degree of freedom measuring device which can be configured to simultaneously track the position and orientation of up to thirteen electromagnetic receivers 23 with a
single electromagnetic transmitter 21.
Referring to FIG. 3, an enlarged view of the transmitter 21 and receiver 23 is illustrated. Within the transmitter 21, and shown partially in phantom is the center of the X, Y, and Z coordinate system with which the receiver 23 measures its
spatial displacement. The positive Y direction is shown extending to the left and toward the observer, the positive X direction is shown extending to the right and toward the observer, while the positive Z direction is shown as being directed
downwardly.
The orientation of the receiver 23 matches the orientation of the transmitter 21 in FIG. 3, and is shown in a parallel position. The spike design, and the frusto-pyramidal shape of the transmitter 21 assist in the ready identification of the
orientation of the transmitter 21 with regard to its coordinate system shown in FIG. 3.
Each electromagnetic receiver 23 is capable of making from 10 to 144 measurements per second of its position and orientation when a electromagnetic receiver 23 is located within a finite, reasonably close distance from the electromagnetic
transmitter 21. A pulsed direct current, or DC magnetic field enables the electromagnetic receivers 23 to determine position and orientation. From the measured magnetic field characteristics, each electromagnetic receiver 23 independently computes its
position and orientation and makes this information available to a computer (not shown).
Each electromagnetic receiver 23 contains two independent serial interfaces, includin | | |
