 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to an improved method and apparatus for determining,
in real time, the position of the tip of an invasive probe inside a
three-dimensional object and displaying its position relative to a
geometrical model of that object visually displayed on a computer screen.
More specifically, this invention relates to an improved method and
apparatus of interactively determining the position of a probe tip inside
the head of a patient during intracranial surgery relative to a
three-dimensional internal diagnostic image of that patient.
2. Description of the Prior Art
Computed tomography (CT), magnetic resonance imaging (MRI), and other
methods provide important detailed internal diagnostic images of human
medical patients. However, during surgery there often is no obvious,
clear-cut relationship between points of interest in the diagnostic images
and the corresponding points on the actual patient. While anomalous tissue
may be obviously distinct from normal healthy tissue in the images, the
difference may not be as visible in the patient on the operating table.
Furthermore, in intracranial surgery, the region of interest may not
always be accessible to direct view. Thus, there exists a need for
apparatus to help a surgeon relate locations in the diagnostic images to
the corresponding locations in the actual anatomy and vice versa.
The related prior art can be divided into art which is similar to the
present invention as a whole and art which is related to individual
components of this invention.
Prior art similar to the present invention as a whole includes methods of
correlating three-dimensional internal medical images of a patient with
the corresponding actual physical locations on the patient in the
operating room during surgery. U.S. Pat. No. 4,791,934 does describe a
semi-automated system which does that, but it requires additional
radiographic imaging in the operating room at the time of surgery as the
means to correlate the coordinate systems of the diagnostic image and the
live patient. Furthermore, the system uses a computer-driven robot arm to
position a surgical tool. In particular, it does not display the location
of an input probe positioned interactively by the surgeon.
There have been other attempts to solve the three-dimensional localization
problem specifically for stereotactic surgery. One class of solutions has
been a variety of mechanical frames, holders, or protractors for surgery
(usually intracranial surgery). For examples see U.S. Pat. Nos. 4,931,056;
4,875,478; 4,841,967; 4,809,694; 4,805,615; 4,723,544; 4,706,665;
4,651,732; and 4,638,798. Generally, these patents are intended to
reproduce angles derived from the analysis of internal images, and most
require rigidly screwing a stereotactic frame to the skull. In any case,
these methods are all inconvenient, time-consuming, and prone to human
error.
A more interactive method uses undesirable fluoroscopy in the operating
room to help guide surgical tools (U.S. Pat. No. 4,750,487).
More relevant prior art discloses a system built specifically for
stereotactic surgery and is discussed in the following reference:
David W. Roberts, M.D., et al; "A Frameless Stereotaxic Integration of
Computerized Tomographic Imaging and the Operating Microscope", J.
Neurosurgery 65, October 1986.
It reports how a sonic three-dimensional digitizer was used to track the
position and orientation of the field of view of a surgical microscope.
Superimposed on the view in the microscope was the corresponding internal
planar slice of a previously obtained computed tomographic (CT) image. The
major disadvantages reported about this system were the inaccuracy and
instability of the sonic mensuration apparatus.
Although the present invention does not comprise the imaging apparatus used
to generate the internal three-dimensional image or model of the human
patient or other object, the invention does input the data from such an
apparatus. Such an imaging device might be a computed tomography (CT) or
magnetic resonance (MRI) imager. The invention inputs the data in an
electronic digital format from such an imager over a conventional
communication network or through magnetic tape or disk media.
The following description concentrates on the prior art related
specifically to the localizing device, which measures the position of the
manual probe and which is a major component of this invention. Previous
methods and devices have been utilized to sense the position of a probe or
object in three-dimensional space, and employ one of various mensuration
methods.
Numerous three-dimensional mensuration methods project a thin beam or a
plane of light onto an object and optically sense where the light
intersects the object. Examples of simple distance rangefinding devices
using this general approach are described in U.S. Pat. Nos. 4,660,970;
4,701,049; 4,705,395; 4,709,156; 4,733,969; 4,743,770; 4,753,528;
4,761,072; 4,764,016; 4,782,239; and 4,825,091. Examples of inventions
using a plane of light to sense an object's shape include U.S. Pat. Nos.
4,821,200, 4,701,047, 4,705,401, 4,737,032, 4,745,290, 4,794,262,
4,821,200, 4,743,771, and 4,822,163. In the latter, the accuracy of the
surface sample points is usually limited by the typically low resolution
of the two-dimensional sensors usually employed (currently about 1 part in
512 for a solid state video camera). Furthermore, these devices do not
support the capability to detect the location and orientation of a
manually held probe for identifying specific points. Additionally, because
of line-of-sight limitations, these devices are generally useless for
locating a point within recesses, which is necessary for intracranial
surgery.
The internal imaging devices themselves (such as computed tomography,
magnetic resonance, or ultrasonic imaging) are unsuited for tracking the
spatial location of the manually held probe even though they are
unencumbered by line-of-sight restrictions.
A few other methods and apparatus relate to the present invention. They
track the position of one or more specific moveable points in
three-dimensional space. The moveable points are generally represented by
small radiating emitters which move relative to fixed position sensors.
Some methods interchange the roles of the emitters and sensors. The
typical forms of radiation are light (U.S. Pat. No. 4,836,778), sound
(U.S. Pat. No. 3,821,469), and magnetic fields (U.S. Pat. No. 3,983,474).
Other methods include clumsy mechanical arms or cables (U.S. Pat. No.
4,779,212). Some electro-optical approaches use a pair of video cameras
plus a computer to calculate the position of homologous points in a pair
of stereographic video images (for example, U.S. Pat. Nos. 4,836,778 and
4,829,373). The points of interest may be passive reflectors or flashing
light emitters. The latter simplify finding, distinguishing, and
calculating the points.
Probes with a pointing tip and sonic localizing emitters on them have been
publicly marketed for several years. The present invention also utilizes a
stylus, but it employs tiny light emitters, not sound emitters, and the
method of sensing their positions is different.
Additional prior art related to this patent is found in these references:
Fuchs, H.; Duran, J.; Johnson, B.; "Acquisition and 10 Modeling of Human
Body Form Data", Proc. SPIE, vol. 166, 1978, pp. 94-102.
Mesqui, F.; Kaeser, F.; Fischer, P.; "Real-time, Non-invasive Recording and
3-D Display of the Functional Movements of an Arbitrary Mandible Point",
SPIE Biostereometrics, Vol. 602, 1985, pp. 77-84.
Yamashita, Y.; Suzuki, N.; Oshima, M. "Three-Dimensional Stereometric
Measurement System Using Optical Scanners, Cylindrical Lenses, and Line
Sensors", Proc. SPIE, vol. 361, 1983, pp. 67-73.
The paper by Fuchs, et al., (1978) best describes the method used by the
present invention to track the surgical probe in three-dimensional space.
It is based on using three or more one-dimensional sensors, each
comprising a cylindrical lens and a linear array of photodetectors such as
a charge-coupled semiconductor device (CCD) or a differential-voltage
position sensitive detector (PSD).
The sensors determine intersecting planes which all contain a single
radiating light emitter. Calculation of the point of intersection of the
planes gives the location of the emitter. The calculation is based on the
locations, orientations, and other details concerning the one-dimensional
sensors and is a straightforward application of analytic geometry. This
electro-optical method, however, has not been previously used for the
purpose of the present invention.
Thus, there still remains a need for a complete apparatus which provides
fast, accurate, safe, convenient mensuration of the three-dimensional
position of a manual probe and which visually relates that position to the
corresponding position on the image of a previously-generated
three-dimensional model of an object.
SUMMARY OF THE INVENTION
A first objective of the present invention is to provide accurate,
three-dimensional mensuration of the location and orientation of an
instrument on or inside an object, which could be (but is not limited to)
a surgical patient in an operating room.
A second objective of this invention is to provide an electro-optical
mensuration system which is inexpensive, easy to use, reliable, and
portable and which employs a manually positioned probe or other pointing
instrument.
A third objective of this invention is to provide a simple, non-invasive
means of establishing a correspondence between a predetermined coordinate
system of the object and a coordinate system of a three-dimensional,
geometrical compu | | |
