CA2437822A1

CA2437822A1 - Method and apparatus for the non-invasive imaging of anatomic tissue structures

Info

Publication number: CA2437822A1
Application number: CA002437822A
Authority: CA
Inventors: David Roberts
Original assignee: Individual
Current assignee: Individual
Priority date: 2001-02-12
Filing date: 2001-02-12
Publication date: 2002-08-22
Also published as: US20040076317A1; US20100222665A1; AU2007203153B2; JP2004528873A; IL157349A0; CN1505490A; EP1361820A1; US8195270B2; WO2002064034A1; IL157349A; EP1361820B1; AU2007203153A1; US7672708B2

Abstract

A method for the non-invasive imaging of an anatomic tissue structure in isolation from surrounding tissues, including: receiving from an input device magnetic imaging data from a patient of the anatomic tissue structure and surrounding tissues; segmenting the imaging data to isolate the anatomic tissue structure imaging data from the imaging data for the surrounding tissues; separating the anatomic tissue structure imaging data into data populations corresponding to tissue microstructures; constructing an image from the imaging data for at least one of the tissue microstructures; and storing or displaying the image. An apparatus embodying the disclosed method is also described, as well as a method for the quantitative measurement of a nerve tissue suspected of demyelination.

Description

METHOD AND APPARATUS FOR THE NON-INVASIVE IMAGING OF
ANATOMIC TISSUE STRUCTURES
Background Of The Invention The present invention relates to the non-invasive mag-S netic imaging of anatomic tissue structures in isolation from surrounding tissues, and, in particular, to the non-invasive magnetic imaging of nerve tissues. The present invention more specifically relates to the non-invasive quantitative measurement of damage to nerve tissue, such as the damage that occurs upon compression.
Clinically it is possible to estimate which neural pathway is involved in trauma by determining the end result loss of motor function and either abberant sensory function or complete loss of it. ~It is very cTif~ficult -to deterrriine where the lesion is and to what extent it has caused deteri-oration of the nerves involved.
Peripheral nerves are compound structures_ Each nerve trunk consists of a very large number of nerve "fibers"
(neurites) that are protoplasmic extensions of neuronal cell bodies located in the dorsal root ganglia (sensory neurons) and ventral horn of the spinal cord (motor neurons).. All nerve fibers (except fine pain fibers) are surrounded by a fatty sheath of myelin which serves to enhance nerve impulse SUBSTITUTE SHEET (RULE 26) conduction (neural flow). The nerve fibers are grouped into fasciculi, bundles of fibers bound together by a fine connective tissue membrane (epineureum). The bundles of fasciculi comprising the nerve trunk are bound by a thicker membrane, the perineureum. The nerve trunk may be surrounded by other layers of protective connective tissue or fat.
Neuropathy in peripheral nerves may occur through the agency of traumatic distortion, including compression, stretch and breakage, and/or associated ischemia. The trauma may be acute, especially when the surrounding tissues are injured, om it may be chronic, and due to pathologic mechan-ical relations between the nerve and surrounding tissues:
regional compression is considered to be the principal etiol-ogy in entrapment syndromes. There is rapid recovery from transient compression, but prolonged compression extends the recovery period because the nerve fibers are damaged. Remye-lination and re-establishment of conduction must take place.
Chronic and repetitive compression prevents recovery and results in severe neurologic impairment.
zn chronic and repetitive compressions, which are more common than acute compressions, the damage is more extensive.
Nerve trunks become thinned at the site of the compression, but may swell proximally. Characteristically conduction is slowed and conduction blocks occur as long segments of myelin are damaged. Depending on the severity of the chronic com-pression, the nerve trunk proximally will contain a full com-..._,..- p~lement of myelinated fibers within its fasciculi; ~ but at they level of compression and distally the number of myelinated fibers may be reduced significantly. The relative density of the myelinated fibers between proximal and distal portions of the nerve is a measure of loss of neurophysiologic function in its distal distribution.
Although nerve damage can occur anywhere within the body, some nerves are predisposed to chronic damage because of their structural relationships with other tissues_ Clas-sic sites for chronic nerve compression include the brachial plexus - especially where the cords of the plexus pass be-tween the rib cage and clavicle, the elbow - where the ulna nerve winds around the olecranon process of the ulna bone, the carpal tunnel - where the median nerve lies between deep flexor tendons and the flexor retinaculum, the sacro-iliac region - where the lumbo-sacral trunk and sacral plexus lie on the bony surfaces of the pelvis, the gluteal region -where the sciatic nerve passes between the powerful gluteal musculature, the neck of the fibula - where the common pero-veal nerve winds about the external surface of the bone, and the tarsal canal - where the tibial nerve winds beneath the tarsal bones to enter the sole of the foot.
Some nerves are vulnerable to acute trauma as a result of their intimate relationship to bones subject to fracture.
These include that portion of the brachial plexus lying beneath the clavicle, the radial nerve - which winds about the shaft of the humerus, and the common peroneal nerve, which winds about the neck of the fibula_ Carpal Tunnel Syndrome (CTS) is defined as a pathologic condition in which the tunnel size is reduced, causing press-ure on the median nerve with resultant pain and slight sens-ory impairment in the digits supplied by the nerve, and some-times accompanied by slight wasting of the thenar muscles.
The carpal groove is a deep concavity on the palmar surface of the carpus formed by lateral and medial projections of the carpal bones. The medial projection is formed by the pisi--~~~~""'- ~'forni boize~ and the hamulus - a hook-like projection of the hamate bone. The lateral projection is formed by the tuber cles of the scaphoid and trapezium bones. The carpal groove is converted into an osteofibrous carpal tunnel by a strong fibrous retinaculum attached to the lateral and medial mar-gins of the carpal groove. The carpal tunnel transmits the tendons of the deep and superficial digital flexor muscles which in this region fcrm a firm bundle_ The median nerve lies within the tunnel between the flexor retinaculum and the flexor tendon bundle. It emerges from behind the flexor digitorum superficialis distally in the forearm about 5 cm proximal to the flexor retinaculum of the carpus. It lies superficial to the deep digital flexor tendons and between the tendons of the superficial flexor muscle and the flexor carpi radialis. The nerve then passes deep to the flexor retinaculum of the wrist where normally it lies superficial to the stout bundle of flexor tendons.
Overt morphology associated with CTS includes carpal dislocations, carpal arthritis and tenosynovitis of the long flexor tendons. These conditions clearly restrict the space within the tunnel. There is often no apparent morphologic cause for the syndrome, but as complete division of the flex-or retinaculum results in relief from symptoms in most cases it is assumed that mechanical compression that is not obvious from the gross morphology of the tunnel or its contents causes ischemia or directly compresses the median nerve.
Clinical symptoms includes weakening of the thenar mus-cles,,parasthesias of the lateral palm, thumb and first two fingers. Other symptoms may include stiffness of the wrist and some pain_ Clinical examination consists of identifying those symptoms known to be associated with nerve damage. The anatomic distribution of the branches of peripheral nerves is constant and well known. The distribution of abnormal sensation or inability to utilize one or more muscles is a "~"'w ~"s'urew'iWdioator of whic°h ne~v~~ 'is involved:
Electromyography is used as a supplement to clinical ex-amination. The technique uses needle electrodes inserted into suspect muscle. The intensity of the potentials measured is an indirect measure of the :~.evel of muscle activity. Neural flow stimulation also involves the use of needle electrodes.
The technique compares input and output values of electric potentials.

None of these diagnostic techniques definitively identi-fies the focus of the neuropathy. In most cases, if the ex-amination and electromyographic results are typical of those associated with a particular syndrome, the focus of the neu-5 ropathy can be estimated with some certainty based on exper-ience; but false positive diagnoses are relatively common and may lead to unnecessary surgery or incomplete relief of the symptoms. In addition, electromyography is a relatively complex technique requiring some expertise to obtain a satis-factory diagnosis. 'Furthermore, the use of needle electrodes is invasive, and may prove painful. This can lead to false readings or patient rejection of the study.
A further problem is experienced with clinical examina-tions. Numerous patients attempt to simulate symptomatology in order to qualify for insurance compensation. Identifying these patients using an essentially subjective examination is difficult and time consuming for the diagnostician.
Despite conflicting reports in the literature, there is no convincing evidence that nerve morphology viewed in unas-listed two-dimensional magnetic images is a clinically pract-ical means of obtaining information about histological degen-eration and impaired neural flow. At present, CTS patients are not referred for magnetic imaging scans as a diagnostic procedure.
Summary Of The Invention It has now been discovered that post-scan data process-ing may be employed to resolve the problem of obtaining phys-iological information with significant clinical diagnostic value from magnetic imaging data. By first segmenting the imaging data to isolate anatomic tissue structures from sur-rounding tissues, high resolution three-dimensional images are obtained in which tissue microstructures are visible and provide clinically significant information relating to phys-iology. When imaging nerve tissues, it is possible to obtain neurophysiological information with significant clinical diagnostic value relating to the efficacy of nerve flow.
Therefore, according to one aspect of the present invention, a method is provided for the non-invasive imaging of an anatomic tissue structure in isolation from surrounding tissues, including the steps of:
receiving from an input device, magnetic imaging data from a patient,of an anatomic tissue structure and the surrounding tissues;
segmenting the imaging data to isolate the anatomic tissue structure imaging data from the imaging data for the surrounding tissues;
separating the anatomic tissue structure imaging data into data populations corresponding to tissue microstruc-tures;
constructing an image from imaging data for at least one of the tissue microstructures; and storing or displaying the image.
The imaging data my be acquired by either magnetic resonance or magnetic transfer imaging. Applications for the method of the present invention are not limited to the imag-..~......_._...._.~.ing of nerve tissue ..,but... apply_~ b.:.._~ss~it~ially arry.. tissue. -_~w._. ___..
structure for which it is difficult to obtain a high resolu-tion image that is not obscured by surrounding tissues.
Thus, the method of the present invention is particularly well suited for the imaging of vascular tissue and tumor tissue. The present invention makes possible high resolution images of tumor tissue boundaries that assist the surgeon in achieving the goal of complete tumor removal.
Therefore, according to another aspect of the present invention, an apparatus is provided for preforming the method of the invention. An apparatus in accordance with the present invention includes:
an input device for receiving magnetic imaging data from a patient;
a processing unit having:
a first program for segmenting the magnetic imaging data to isolate imaging data for an anatomic tissue structure from imaging data for surrounding tissues;
a second program for separating the anatomic tissue structure imaging data into data populations correspond ing to tissue microstructures;
a third program for constructing an image from ima-ging data for at least one of the microstructures; and a processor for sequentially executing the first, second and third programs;
wherein the processor executes:
the first program using the imaging data from the input device;
the second program using the anatomic tissue 2Ø. . ,....... ~. _. .'_. _.._.
s~ruct'iire...im~glmg,.._aata._.1.~..ol~twd~"'~byW.e~e~u'Cim"of' the'°.'-- .
first program; and the third program using the data populations obtained by execution of the second program; and an output device for displaying the image or a storage device for storing the image.
The present invention includes apparatuses for calculat-ing the presence or loss of myelin in a nerve trunk or tract_ Such apparatuses include devices that measure the myelin without first constructing an image of the nerve trunk or tract. An apparatus in accordance with this aspect of the present invention includes:
an input device for receiving imaging data from a pa-tient of a nerve trunk or tract and the surrounding tissues;
a processing unit having:
a first program for segmenting the magnetic imaging data to isolate imaging data for the nerve trunk or tract from the imaging data for the surrounding tissues;
a second program for separating the nerve trunk or tract imaging data into data populations corresponding to fasciculi and other nerve trunk or tract microstruct-ures;
a third program for calculating the presence or loss of myelin in the nerve trunk or tract from the fasciculi imaging data; and a processor for sequentially executing the first, second and third programs;
wherein the processor executes:
the first program using iinagirig'°'c~ata °~'rom-"~he-~"~°'"'~""
input device;
the second program using the nerve trunk or tract imaging data isolated by execution of the first program; and the third program using the fasciculi data population obtained by execution of the second program; and ' 9 an output device for displaying the myelin calculation or a storage device for storing the myelin calculation.
The method and apparatus of the present invention are particularly useful for segmenting nerve tissue imaging data, enabling three-dimensional visualization of individual nerve fasciculi, and quantification of the percentage of fascicular tissue present in cross-sections of the nerve trunk. The numerical values obtained are represented in bivariate plots of percentage and cross-sectional location and directly indicate to the physician those parts of the nerve trunk where myelin content is reduced and neural flow is impaired, and the extent of the degeneration.
Therefore, according to another aspect of the present invention, a method is provided for the imaging of a nerve trunk or tract suspected of demyelination, including:
receiving from an input device, magnetic imaging data from a patient of a nerve trunk or tract suspected of demye-lination and the surronding tissues;
segmenting the imaging data to separate the imaging data for the nerve trunk or tract from the i!;~aging data for the surrounding tissues;
separating the imaging data for~the nerve trunk or tract into data populations corresponding to the fasciculi and ~~ ~ ~ other microstructures of the nerve trunk ~ or- tra~ct'~ ~ " '~~'"-constructing an image from imaging data for at least the fasciculi of the nerve trunJc or tract; and storing or displaying the image.
The present invention incorporates the discovery that the imaging data for the fasciculi can be used to calculate the presence or loss of ~~yelin in a nerve trunk or tract_ Therefore, preferred methods according to this aspect of the present~invention further include the steps of calculating from the imaging data for the fasciculi, the presence or loss of myelin in the nerve trunk or tract suspected of demyelin-ation, and storing or displaying the myelin calculation.
5 According to another aspect of the present invention, the presence or loss of myelin in a nerve trunk or tract may be calculated from fasciculi imaging data independently of the construction of the nerve trunk image. The present in-vention therefore also includes a method for the quantitative 10 measurement of a nerve trunk or tract suspected of demyelina-tion, including the steps of:
receiving from an input device, magnetic imaging data from a patient of a nerve trunk or tract suspected of demye-lination and the surrounding tissues;
l5 segmenting the imaging data to separate the imaging data for the nerve trunk or tract from the imaging data for the surrounding tissues;
separating the imaging data for the nerve trunk or tract into data populations corresponding to the fasciculi and other microstructures of the nerve trunk or tract;
calculating from the fasciculi imaging data, the presence or loss of myelin in the nerve trunk or tract; and storing or displaying the myelin calculation. ~ w~w "' The method of the present invention for quantitatively measuring nerve tissue demyelination is particularly useful for the non-invasive diagnosis of CTS. Therefore, according to yet another aspect of the present invention, a method is provided for the imaging of a median nerve suspected of demy-elination within one of a patient's carpal tunnels, includ-ing the steps of:

receiving from an input device, magnetic imaging data from a patient including at least a portion of the median nerve passing through a carpet tunnel;
segmenting the imaging data to isolate the median nerve imaging data from any imaging data for surrounding tissues;
separating the median nerve imaging data into data popu-lations corresponding to the fasciculi and other microstruct-ures of the median nerve;
constructing an image from imaging data for at least the fasciculi of the median nerve; and storing or displaying the image.
This method can also be employed to calculate the pres-ence or loss of myelin in the median nerve within a carpal tunnel from the fasciculi imaging data. Methods in accord-ante with this aspect of the present invention further in-clude the steps of calculating from the fasciculi imaging the presence or loss of median nerve myelin within the carpal tunnel, and storing or displaying the myelin calculation.
The presence or loss of myelin in the median nerve with-in a carpal tunnel can also be calculated from the fasciculi imaging data independently of the construction of a median nerve image. The present invention therefore also includes a ......_._...~.._....._.method. -for the quantitative' measurement of a' median nerve suspected of demyelination within one of a patient's carpal tunnels, including the steps of:
receiving from an input device, magnetic imaging data from a patient including at least a portion of the median nerve passing through a carpet tunnel;
segmenting the imaging data to isolate the median nerve imaging data from any imaging data for surrounding tissues;

l2 separating the median nerve imaging data into data populations corresponding to the fasciculi and other micro-structures of the median nerve;
calculating from the imaging data for the median nerve fasciculi, the presence or loss of myelin in the median nerve within the carpal tunnel; and storing or displaying the myelin calculation.
A more complete appreciation of the invention and many more other intended advantages can be readily obtained by reference to the following detailed description of the preferred embodiments and claims, taken in conjunction with the accompanying drawings, which disclose the.principles of the invention, and the best modes which are presently contemplated for carrying them out.
Brief Description Of The Drawings FIG_ 1 is a block diagram of the components of an exem-plary apparatus in accordance with the present invention;
FIG. 2 is a flow chart illustrating an exemplary set of steps in accordance with the method of the present invention;
FIG. 3 is a prior art axial cross-section of a wrist, including the median nerve, obtained by magnetic resonance .. . .,...-.._..~,~._.__.imaging; ° ... . . _ . . .. .... . . .. _~ ,..
. _ . _ .. .. ... . ..
FIG. 4 is a magnetic resonance image of a healthy median nerve within a carpal tunnel of a patient, constructed by an .
apparatus of the present invention employing the method of the present invention; and FIG. 5 is a magnetic resonance image of a median nerve within a carpal tunnel of a patient suffering from CTS, con-structed by the same apparatus and method employed to construct the image of FIG. 4.

Detailed Description Of The Preferred Embodiments Referring to the drawings, in which like reference num-erals identify similar or identical elements, as shown in Figure 1, the present disclosure describes a method and app-s aratus implementing an imaging device 10, which includes a processing unit 12 having processor l4, memory 16 and stored programs 18, including post-scan data processing programs 20.
The processing device also includes an input device 26 and an output device 28.
The present invention is based on the discovery that by first segmenting imaging data to isolate anatomic tissue structures from surronding tissues before histographically resegmenting the imaging data, it is possible to construct images of the tissue microstructures having exceptional clarity. The new methods and apparatus disclosed herein are particularly adapted for use in the diagnosis of neuropathy, especially the neuropathy resulting from compression and the associated demyelination, such as occurs with CTS.
' In an exemplary embodiment, the processing unit 12 is an 81000 workstation from SGI having about 128 MB associated RAM
memory and a 4 GB hard or fixed drive memory 16. The proces-sor operates using a UNIX operating system to run application software 18 implemeiZting the disclosed imaging device 10 and method.
I , Y~ 25~ ~ . ~ ' It should ~nbe understood that the above described proces-sor is merely exemplary. Any computer or processing system having sufficient processing ability and memory may be used.
Imaging data acquired by magnetic resonance or magnetic transfer imaging is provided by imaging source 24 through an input device 2& to the processor 14. The data source may be any source for providing magnetic resonance or magnetic transfer imaging data to a computer, such as a floppy disk containing an imaging data file, or the processor of a mag-netic resonance or magnetic transfer imaging device (not shown), directly linked to a input port of processor 14, or by means of a modem interface. The input device 26 may be a floppy disk drive, modem, or other data reading or data input device, and also includes an interactive, semi-automatic or automatic contouring device for segmenting of anatomic struc-tures. For purposes of the present invention "anatomic structures" are defined as the functional substructures within a gross organ, such as the functional microstructures within a nerve trunk, i.e., fasciculi, and the like.
Input device 26 may also include a keyboard or similar input device for supplying commands to processor 14. The command-supplying input device may also be provided as a second input device (not shown).
The processor 14 receives commands and imaging data from the imaging data source through the input device(s)_ The im-aging data my be stored in memory 16 for further processing or used by the processor 14 to construct an image of a tissue target. The constructed image is sent to an output device 28 such as a visual image display device, for example, a video display terminal. A video terminal for displaying the origi-nal magnetic resonance or magnetic transfer image may also be included. The constructed image may also be displayed numer-ically or graphically through an appropriate output device.
The imaging device 10 performs the applications programs described herein in conjunction with Figure 2, which are im-plemented from compiZed~ source code's in they V"C" programming language. Referring to Figure 2, the~imaging device 10 in-cludes conventional imaging software, such as software for the display of original magnetic images. The imaging device also includes post-scan data processing programs, such as "clipping" software for discarding irrelevant data, and soft-ware for operating interactive,. semi-automatic or automatic contouring devices for segmentation of anatomic structures_ The imaging device also includes post-scan data process-ing software for the creation of binary images of segmented structures and binary slices and software for the creation of three-dimensional binary masks and software for the segmenta-tion of gray-level data representing the segmented structure via the binary mask. Post-scan data processing software for 5 the histographic analysis of the gray-level data is also included, which performs single or iterative segmentation of the gray-level anatomic tissue structure into the intrinsic tissue microstructures, as well as software for the defini-tion of material opacity, emission and surface reflection.
10 The imaging device post-scan data processing software also includes software for the creation of three-dimensional images of individual intrinsic tissue microstructures, or compound images of complete tissue structures, and software for image smoothing (aliasing). Post-scan data processing 15 software is also included for the calculation of the central trunk (tract) axis (axes), cross-sections and tissue percent-ages and the graphical display of values.
Figure 2 depicts an applications program for the imaging of a nerve trunk. In step 28 magnetic imaging data is acqu-fired in 1 mm slices along the nerve trunk axial plane. In step 30 the data is transferred to a processor for post scan processing. In step 32 gray-level images of the magnetic imaging slices are constructed by the image display software.
In step 34 the interactive clipping software identifies _25 the nerve trunk in the gray.-,Level slices , and in.,.step 36 cl.ip.:_,..
_...
the first nerve trunk slice to eliminate the surrounding tis-sue imaging data. The rest of the slice stack is then auto-matically clipped, with modification of the clipping boun~xary as required_ In step 38 the semi-interactive segmentation software identifies the nerve trunk bou~,dary to the computex in the initial one or two slices, after which the software automatically segments the remaining slices and then checks and modifies the identified boundaries.
In step 40 the binary masking soft~~~are creates two-dimensional binary masks for the segmented nerve boundaries.

In step 42 the two-dimensional binary masks are used by the gray-level data segmentation software to segment the gray- .
level data representing the nerve trunk from the original magnetic resonance or magnetic transfer image slices.
In step 44 the histographic analysis software resegments the nerve trunk gray-level data for the segmented nerve trunk into nerve trunk microstructure data populations, including the fascicular bundles, the epineurium and inter-fascicular fascia. Interpolation of the resegmented data populations is performed in step 45. Software for the definition of mater-ial opacity, emission and surface reflection adjusts these features for individual microstructures to enhance their vis-ualization within compound tissue images.
Still and movie images are then constructed by the three-dimensional imaging software in step 48. Both individ-ual nerve tissue microstructure images and compound images of the nerve trunk composed of the individual nerve tissue microstructures are constructed.
In step 50 images are displayed, either as printed ima-ges or by viewing software as video displays. Optionally, the tissue calculation software determir~es in step 52, from the fasciculi imaging data, the amount of fascicular material in selected cross-sections, the values for which are present-ed graphically in step 54 by the display software.
Nerve trunks are readily visualized in high resolution T1 or T2 weighted magnetic resonance (MR) images. In cross-sections of a nerve trunk the fascicles appear as bright "spots" interspersed in a darker gray matrix which represeni_~
interstitial tissues, surrounded by a dark border of variable thickness that represents the perineureum and several layers of thin supporting fascia.
The relatively high amplitude signal from the fasciculi relates to the presence of myelin. Myelin is composed of lipids and protein, as in the plasma membranes of other l7 cells, but the percentage of lipid is higher; and in addition to the usual suite of lipids found in plasma membranes, mye-lin contains a characteristic glycolipid. Myelin also con-tains up to 30% bound water.
When demyelination occurs, the loss of lipid molecules and the free interstitial water (which will affect ionic ex-change rates between the free and bound water) reduces the amplitude of the MR signal within the bounding perineureum.
Specifically, a percentage of the bright "spots" will become no longer visible. Pathologic tissues will appear darker in MR images than do normal tissues, and the presence of fasciculi will be less apparent.
The presence of fasciculi can be partially detected by direct visual inspection of cross-sectional MR images. But determination of the number or percentage of fasciculi pres-ent in a cross-section, comparisons between different sect-ions is not satisfactory via visual inspection alone. It is often not possible to resolve darker fasciculi and small changes in signal amplitude may go unobserved.
The method of the present invention is thus used to gen-erate three-dimensional images of the fasciculi, which can be visualized in a static format, or which may be rotated about the (approximate) long axis of the trunk. The latter enhan-ces the visualization process. Because the three-dimensional image can be limited to the fasciculi, the local density of these structures within the trunk, and their brightness (sig-.
nal amplitude) are now relatively easy to visualize directly.
With the present invention, it is :iow possible to quant-ify nerve trunk demyelination values from the generated ima-ges, which were heretofore measured eiti:e~~ invasively or sub-jectively. Nerve tissue demyelination is determined by first calculating the volume of fascicular tissue as a percentage of the volume of the gross trunk (tract) structure. Next, the area of fascicular tissue as a percentage of the area of the gross trunk (tract) structure as seen in selected cross-l8 sections of the trunk (tract) is calculated. The contiguous cross sectional areas of the trunk (tract) of specified thickness (i.e., 1 mm) perpendicular to the long axis is then calculated.
The percentage of volume of fascicular tissue as a per-centage of total volume of the trunk (tract) in contiguous cross-sections is calculated next. A graphical display of the percentage volumes obtained vs. location of the sections on the central axis of the trunk is then constructed. The variation of myelin content as indicated by these percentage values is a measure of the neurophysiologic (f low efficiency) status of the trunk (nerve) and thus a diagnostically significant index.
This technique enables a comparison of the percentage of fascicular tissue present, for example, in the forearm (where entrapment and compression is unlikely), with the value for fascicular tissue within the carpal tunnel, and in regions of the trunk distal to the carpal tunnel. It is also possible to compare fascicular percentages in homologous regions from different individuals (compare between patients).
The method of the present invention thus allows for the.
fascieular percentages to be calculated for cross-sectional areas of a nerve trunk. For the median nerve, fascicular percentages for 1 mm thick contiguous sections normal to the w.w..25,.~long axis. of the trunk may be calculated, and then represent-ed as a bivariate plot of percentage of fascicular material present vs. location within the trunk_ This graphical rep-resentation provides the CTS diagnostician with a simple and direct means of determining in which precise regions of the median nerve trunk a loss of fascicular material has occurr-ed, and the amount of fascicular tissue lost.
A drawback in using MR images for inter-patient compari-sons is that the scale of gray-level values representing dif-ferent types of tissue is scanner dependant and consequently varies between scanners. Magnetic Transfer Imaging (MTI) is l9 a new technique that has two advantages over conventional MR
imaging when applied to analysis of nerve trunks. First, the in-plane resolution is improved. Second, the data values obtained with this technique are not scanner dependant and consequently direct inter-patient comparisons are possible.
The possibility of using MTI for the analysis of nerve tissue has been discussed in the literature. The loss of bound protons in the degenerating myelin sheath has the effect, via magnetic transfer, of depressing the MR signal from free protons in the immediate vicinity of the fasciculi.
The Magnetic Transfer function can be calculated, and the resultant values used to create MT images. Thus, MTI is particularly well-suited for use with the methods and apparatus of the present invention_ The method and apparatus of the present invention are also well-suited for the diagnosis of nerve pathology. In diagnostic methods according to the,present invention, a clinical examination is typically first performed to deter-mine the likely location of nerve pathology. A high-resolu-tion magnetic image of the region is then obtained.
The MR data comprises thin (1 mm) contiguous slices in a plane which is more or Iess normal to the axis of the nerve.
High resolution images can be obtained lay using scan times up to ten minutes.
Post-scan data processing is then applied to the MR
data. MR hard-copy (acetate) is not required. The three-dimensional MR data is loaded into the ~:~ree-dimensional imaging software. Steps in the subsequ-=_zt data processing are as follows:
The original MR slices are displayed. The nerve is identified and image contrast adjusted ~o maximize its visualizatiori by the operator. This co=:trast adjustment does not affect the signal differential marlc_mg the boundary of the nerve, and is not a factor in subse~aent segmentation.

The size of the data set is reduced by data clipping to discard all data except that in the immediate vicinity of the nerve. Once a region of interest has been defined, clipping 5 is automatic; thus, the size of the region of interest relates to the position of the nerve on successive images.
Interpolation of the imaging data is then performed.
The original voxels are prismatic; i.e., in the Z direction (thickness of image-slice) they are usually 1 mm. The dimen-10 sions of X and Y (in the plane of the slice) depend on the imaging protocol. Typically X and Y are 0.5-0.7 mm. Cubile voxels are required for 3-dimensional rendering; thus, it is necessary to interpolate the data set in the Z direction in order to obtain slices corresponding in thickness to the X,Y
15 dimensions. The interpolation process is not simple data av-eraging; it utilizes a sophisticated algorithm which identi-fies the "shape" of regions of consistent signal in contig-uous slices and calculates intermediate "shapes" (assuming linearity in shape change between slices). Shape based in-20 terpolation has been shown to produce highly accurate infor-mation in the interpolated slices. Interpolation is usually performed after clipping and segmentation on the binary image produced. This reduces the time required to perform inter-active contouring during the process of segmentation.
Automatic segmentation of MR data is difficult because of the non-linear signal. At present several different tech-niques are used to contour the boundary of the region of interest on the original MR images. They are interactive, live-wire and live-path contouring.
With interactive segmentation the operator traces the boundary of the nerve trunk section, recording coordinates on the boundary in so doing. In live-wire segmentation, the boundary of the nerve as seen in cross-section 'is identified by the operator and indicated to the computer. The computer is thus able to identify subsequently contiguous voxels with the same boundary signal characteristics. Once the boundary has been identified, the operator can indicate a starting point on it, then move the cursor to a next point some distance away from the first.
The computer is able to track the boundary between the two points using the boundary information provided, and draws a colored line to mark it. Because the boundary signal var-ies, and the presence of adjacent non-boundary pixels with similar signal characteristics, the boundary selected auto-matically may vary from that which the operator judges to be correct. In this case the boundary is redefined at the point of divergence (i.e., the boundary-contour is corrected) and the technique of indicating points at intervals along the boundary is continued until the boundary is closed. Boundary identification during contouring is used to refine the suite of signal characteristics used lay the computer to determine the boundary automatically. After one or two boundaries have been contoured it is usually only necessary to indicate two or three points to obtain a complete, accurate boundary_ This technique can be extended to automatic identifica-tion of the boundary in subsequent slices, which speeds up the procedure. Usually the operator examines the automatic-ally generated boundary contours and can correct any errors.
The continually refined boundary characteristics are stored by the computer and can be utilized to segment other data sets obtained via similar imaging protocols.
I .. I ' Live-path contouring ~is ~simllar to live-wire contouring, but the computer only seeks boundary characteristics coincid-ing with those defined for it in a "path" of limited width on either side of the cursor_ It is possibl;_ to run the cursor rapidly around the region of interest, vri:hout following the boundary exactly, to describe a contour w=thin the limits of the defined path, for a very rapid method of contouring.
Live-wire and live-path contouring techniques are norm-ally carried out on all the original MR slices.. However, these techniques may be developed so that contours are only defined interactively on the first two or three slices, after which the computer will search for the corresponding boundary in the remaining slices and contour them automatically.
The segmentation techniques output the surface boundary of the region of interest by a binary separation of the three dimensional data based on those voxels within the boundary and those without it. Three-dimensional imaging of the bin-ary data therefore yields a r"solid" object. However, the boundary surface can be used as a "mask" to extract MR data within the region of interest from the original three-dimen-sional MR data matrix. The extracted MR data set is referred to as a gray-level image.
The gray-level image thus contains information about the tissue microstructures contained within the nerve trunk. The gray-level values for nerve fasciculi, matrix, and perineur-eum comprise distinct and overlapping populations. In the region of signal overlap the percentage o~ signal belonging to each material can be ranges of signal -ralues assigned accordingly. This allows the different tissue types to be resegmented automatically based on this h~stographic data.
The histographic resegmentation may be pe_formed iteratively in order to resolve any signal overlap to obtain the best visualization of the nerve tissue microstructures (i.e., visualization of the very distinctive fasciculi).
Opacity, emission and surface strength parameters opti-mal for visualization of the~nerve microstructures.areJthen determined by creation of three-dimensional images and exam-ination of the anatomy. Once values have been determined for these parameters, they may be used for a,~o~s automated imaging in subsequent data sets. Improvement of ,:,isualization by ad-justing these parameters is a distinct j~rt~cess from segmenta-tion and does not select the tissue conce~neci from its sur-rounding matrix.
The output is visual and numerical. Visual results con-silt of three-dimensional renderings of tze trunk, including a compound image of the gross structure, which includes the perineurium, fascicles and interfascicular tissue. The opac-ity of the perineurium is set so that this tissue is trans-lucent. This enables the content of the nerve trunk to be viewed in the context of the perineurium. Similarly, the op-acity of the interfascicular tissue is set to be greater than that of the perineurium, but less than that of the fascicles.
The result is to display the microstructures of the nerve in their correct anatomic relationships.
To enhance the three-dimensionality of the image, and to enable the anatomic relationships to be examined from all an-gles, the image is rotated about its long axis and recorded as multiple "frames" which are stored in a "cine" file. This is performed using conventional MR imaging software. Images are also created containing each of the material sets separa-tely, and tine files created for each. These images allow direct examination of each microstructure separately.
A compound image consisting of the perineureum and fas-ciculae is also created and rendered in a cine file. This image allows the texture and thickness of the perineureum to be examined in the context of the fasciculi, and regions of membrane thickening or thinning compared with regions of MR
signal fluctuation in the fasciculi.
The numerical output includes the cross-sectional area value for the nerve trunk normal to.its long axis, calculated..,~,._ continuously throughout its length and displayed in a bivari-ate plot of cross-sectional area vs. section location. A
similar bivariate plot is calculated to show the density of fascicular material within the cross-section vs. location.
The parameters measured by the r::et-zoc.i and apparatus of the present invention for nerve tissa~ include the general morphology of the trunk (tract), the cross-sectional morphol-ogy, the size and distribution of the nerve fasciculi and the density of the fasciculi within cross-sections of the nerve trunk. These parameters are associated with physiologic changes accompanying nerve damage and dysfunction.
Gross morphology and cross-sectional morphology are phy-sical reflections of the pressure environment_ It is well known that in regions of pressure a peripheral nerve reorgan-izes by becoming flattened in a plane normal to the compress-ive force. The amount of flattening almost certainly relates to altered function, but the sensitivity of the relationship may not be adequate for the purpose of early diagnosis.
In regions of compression the size of the fascicles de-creases, but they increase in number, forming smaller bund-les. The cross-section of the neurites can be measured.
However, cross-sectional fascicular density is a measure of MR signal from the "fascicular material" present in the section. The MR signal may be expected to vary with the his-tochemistry of the nerve fibers (neurites) and their interst-itial environment_ This includes changes in the structure and chemistry of the myelin sheaths (composed of complex lipids and a high percentage of bound water) and in the ionic environment of the neurite and interstitial tissues. Thus, variation in the MR signal is an indicator of functional changes in the neurites, including those due to trauma.
Signal variation can be detected in two ways. One way is by direct examination of the final image_ Regions of nerve.4"
trauma are likely to be represented by localized disappear-ance of the fasciculi. The extent of fascicular disappear-ance is an indication of the severity of the nerve damage.
The MR signal variation can also be quantified by measuring the density of the fascicular material in cross-sections of the nerve trunk and plotting this data against the locations of the cross-sections.
Neural flow studies can be made on tl~e median nerve in the region of the carpal tunnel because of the very superfic-ial placement of the trunk in this region_ However, neuro-pathies occur in other regions, both within and outside the central nervous system where neural flow studies are impract-ical because of the depth of the trunk_ The methods and app-aratus of the present invention provide an innovative means 5 of diagnosing and assessing neuropathy in such deep periph-eral nerves, and in portions of the central nervous system.
The neuropathy need not be associated with compression or other trauma, but may also result from other causes, such as multiple sclerosis and related diseases. Until now it was 10 not possible to assess quantitatively the progression of a disease like multiple sclerosis, or the effectiveness of treatment. Instead, practitioners were forced to rely upon the subjective assessment of the patient. This was particu-larly inaccurate in most circumstances, in which patients 15 want to believe that a treatment is working and relate their well-being from this perspective.
The methods and apparatus of the present invention may also be used to assess other tissues that are inaccessible because of their depth within the body, including vascular 20 tissues and tumor tissues. The present invention makes it possible to visualize vascular blockage, as well as tumor tissue boundaries_ The ability to visualize tumor tissue boundaries makes the present invention a particularly useful tool for the cancer surgeon.
25 Proof of the concept of method of the invention for the analysis of neuropathy was obtained via a feasibility study of the median nerve in patients diagnosed with CTS and non-symptomatic subjects. The feasibility study is presented to illustrate certain aspects of the invention, and should not be interpreted as limiting the scope of the invention as it is defined by the claims_ Twenty wrists, including four cadave-= wrists, were scanned and analyzed using the technique described above_ Subjects scanned in vivo included clinically diagnosed patients with severe CTS, less severe CTS, and control (asymptomatic) subjects.
Results showed:
a. Clear feasibility of "dissecting" arid quantifying nerve trunk structures is possible using this analytic sys-tem. A prior art axial cross-section of a wrist obtained by magnetic resonance imaging is shown in Fig. 3. The cross-sectional image includes the median nerve and associated tendons and carpal bones. This Figure illustrates how such images provided little information of clinical diagnostic value relating to neurophysiology.
A magnetic resonance image of a healthy median nerve trunk constructed by the method and apparatus of the present invention is shown in Fig. 4. The white streaks are bundles of myelinated nerve fibers, i.e., fasciculi. This is in contrast to the magnetic resonance image of a median nerve trunk of a patient afflicted with CTS, shown in Fig. 5, and also constructed by the method and apparatus of the present invention. The nerve fiber demyelination is dramatically evident.
b. MR signal amplitude was reduced where the nerve trunk passes beneath the flexor retinaculum_ Reduction of signal amplitude is modest in subjects with no clinical symptoms of ,.".,s "... _ CTS, but. becomes increasingly significant with CTS symptoms,.
and is pronounced in subjects diagnosed w=th severe CTS.
The four cadaver wrists were also subjected to MR
scanning. Evidence of demyelination in these wrists was partially confirmed by extirpation of the median nerve from the carpal tunnel and histologic secticns.:ng and staining.
Conclusions:
The results strongly supported tl~e h~..~pothesis that nerve degeneration can be detected and localize<:~ in CTS. The ex-tent of degeneration determined by this procedure in CTS pat-ients appears to be associated with the severity of symptoms.
Thus, using the methods and apparatus of the present invention, it is possible to determine if demyelination has occurred, the percentage of myelin loss in the affected region compared with other "functionally normal" regions, and to precisely localize the neuropathy. This provides a non-invasive, objective and quantified diagnostic procedure.
As will be readily appreciated, numerous variations and combinations of the features set forth above can be utilized without departing from the pre-sent invention as set forth in the claims. Such variations are not regarded as a departure from the spirit and scope of the invention, and all such variations are intended to be included within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method for the non-invasive imaging of an anatomic tissue structure in isolation from surrounding tissues, comprising:
receiving from an input device, magnetic imaging data from a patient of said anatomic tissue structure and said surrounding tissues, characterized in that said imaging data includes imaging data for bound water;
segmenting said imaging data to isolate said anatomic tissue structure imaging data from said imaging data for said surrounding tissues;
separating said anatomic tissue structure imaging data into data populations corresponding to tissue microstruc-tures;
constructing an image from imaging data for at least one of said tissue microstructures; and storing or displaying said image:

2. The method of claim 1, wherein said magnetic imaging-data is acquired by magnetic resonance imaging.

3. The method of Claim 1, wherein said magnetic imaging data is acquired by magnetic transfer imaging.

4. The method of claim 1, wherein said anatomic tissue structure imaging data is separated into said microstructure data populations by histographic resegmenting.

5. The method of claim 4, wherein said histographic resegmenting is performed iteratively to separate any over-lapping microstructure data populations.

6. The method of claim 1, wherein said image is displayed as a visual image.

7. The method of claim 1, wherein said anatomic tissue structure is segmented by interactive contouring.

8. The method of claim 1, wherein said anatomic tissue structure is segmented by live-wire contouring.

9. The method of claim 1, wherein said anatomic tissue structure is segmented by live-path contouring.

10. The method of claim 1, wherein said constructing step comprises constructing a compound image of said anatomic tissue structure from imaging data for a plurality of said tissue microstructures.

11. The method of claim 10, further comprising the step of adjusting at least one of the material opacity, emission and surface reflection for at least one of said tissue micro-structures to enhance the visualization of said compound image.

12. The method of claim 1, wherein said target tissue is a nerve trunk or tract.

13. The method of claim 12, wherein said microstructures comprise the perineureum, fasciculi and inter-fascicular tissue.

14. The method of claim 13, further including the step of calculating from fasciculi imaging data the presence or loss of myelin in said nerve trunk or tract.

15. The method of claim 14, further including the step of comparing said myelin calculation to a myelin calculation obtained for another portion of said nerve trunk or tract.

16. The method of claim 12, wherein said nerve trunk or tract is the median nerve.

17. The method of claim 1, wherein said anatomic tissue is tumor tissue.

18. The method of claim 1, wherein said anatomic tissue is vascular tissue.

19. An apparatus comprising:
an input device for receiving magnetic imaging data from a patient;
a processing unit comprising:
a first program for segmenting magnetic imaging data to isolate, imaging data for a anatomic tissue structure from imaging data for surrounding tissue, characterized in that said imaging data includes imaging data for bound water;
A second program for separating said anatomic tissue structure imaging data into data populations corresponding to tissue microstructures;
a third program for constructing an image from ima-ging data for at least one of said microstructures; and a processor for sequentially executing said first, second and third programs;
wherein said processor executes:
said first program using said imaging data from said input device;
said second program using said anatomic tissue structure imaging data isolated by execution of said first program; and said third program using at least one of said data populations obtained by execution of said second program; and an output device for displaying said image or a storage device for storing said image.

20. The apparatus of claim 19, wherein said second program separates said anatomic tissue structure imaging data into said microstructure data populations by histographic resegmenting.

21. The apparatus of claim 20, wherein said second pro-gram is executed iteratively by said processor to separate any overlapping microstructure data populations.

22. The apparatus of claim 19, wherein said first program segments said imaging data by interactive contouring.

23. The apparatus of claim 19, wherein said first program segments said imaging data by live-wire contouring.

24. The apparatus of claim 19, wherein said first program segments said imaging data by live-path contouring.

25. The apparatus of claim 19, wherein said output device is a visual display device.

26. The apparatus of claim 20, further comprising an output device for displaying the histograph obtained by said histographic resegmenting.

27. The apparatus of claim 19, wherein said input device comprises a device for receiving imaging data of a nerve trunk or tract and surronding tissues; said second program separates said nerve trunk or tract imaging data into data populations corresponding to fasciculi and other nerve trunk or tract microstructures; said processing unit further comprises a fourth program for calculating from fasciculi imaging data the presence or loss of myelin in a nerve trunk or tract, wherein said processor executes said fourth program after the execution of said second program using imaging data for nerve trunk fasciculi obtained by execution of said second program; and said output device comprises a device for displaying said myelin calculation or a storage device for storing said myelin calculation.

28. The apparatus of claim 27, wherein. said processing unit further comprises a fifth program for comparing myelin calculations from different nerve trunks or tracts or differ-ent portions of the same nerve trunk or tract, wherein said processor executes said fifth program after the execution of said fourth program using the myelin calculations obtained by execution of said fourth program.

29. The apparatus of claim 26, wherein said output device is a visual display device.

30. A method for imaging demyelination of the median nerve within one of a patient's carpal tunnels, comprising:
receiving from an input device, magnetic imaging data from a patient comprising at least a portion of the median nerve passing through a carpal tunnel, characterized in that said imaging data includes imaging data for bound water;
segmenting said imaging data to isolate the median nerve imaging data from imaging data for surrounding tissues;
separating said median nerve imaging data into data pop-ulations corresponding to the fasciculi and other microstruc-tures of said median nerve;
constructing an image from at least said imaging data for said fasciculi of said median nerve; and storing or displaying said image.

31. The method of claim 30, wherein said method further includes the step of calculating from said imaging data for said median nerve fasciculi the presence or loss of myelin in said portion of said median nerve passing through said carpal tunnel, and sorting or displaying said myelin calculation.

32. The method of claim 30, wherein said median nerve imaging data is separated into microstructure data popula-tions by histographic resegmenting.

33. The method of claim 32, wherein said histographic resegmenting is performed iteratively to separate any over-lapping data populations.

34. The method of claim 30, wherein said median nerve imaging data is segmented by interactive contouring.

35. The method of claim 30, wherein said median nerve imaging data is segmented by live-wire contouring.

36. The method of claim 30, wherein said median nerve imaging data is segmented by live-path contouring.

37. The method of claim 30, wherein said median nerve microstructures also include the perineurium and inter-fascicular tissue.

38. The method of claim 30, wherein said imaging data is acquired by magnetic resonance imaging.

39. The method of claim 30, wherein said imaging data is acquired by magnetic transfer imaging.

40. The method of claim 30, wherein said constructing step comprises constructing a compound image of said median nerve comprising imaging data from a plurality of said nerve tissue microstructures.

41. The method of claim 40, wherein at least one of the material opacity, emission and surface reflection for at least one of said microstructures of said compound image are adjusted to enhance the visualization of said median nerve.

42. The method of claim 41, wherein said compound image also includes the perineurium and inter-fascicular tissue, and the opacity of said inter-fascicular tissue is set to be greater than the opacity of said perineurium, but less than that of said fasciculi.

43. The method of claim 42, wherein said opacity of said, perineurium is set to appear translucent.

44. The method of claim 30, further comprising the step of comparing said myelin calculation to a myelin calculation obtained for another portion of said median nerve.

45. The method of claim 30, further comprising the step of comparing said myelin calculation to a myelin calculation obtained for the median nerve within said patient's other carpal tunnel.

46. A method for imaging a nerve trunk or tract suspected of demyelination comprising:
receiving from an input device, magnetic imaging data from a patient of a nerve trunk or tract suspected of demyelination and the surrounding tissue, characterized in that said imaging data includes imaging data for bound water;
segmenting said imaging data to separate sand imaging data for said nerve trunk or tract from said imaging data for said surrounding tissues;
separating said imaging data for said nerve trunk or tract into data populations corresponding to the fasciculi and other microstructures of said nerve trunk or tract;
constructing an image from at least said imaging data for said fasciculi of said nerve trunk or tract; and storing or displaying said image.

47. The method of claim 46, further including the step of calculating from said fasciculi imaging data the presence or loss of myelin in said nerve trunk or tract suspected of demyelination, and storing or displaying said myelin calculation.

48. The method of claim 46, wherein said imaging data for said nerve trunk or tract is separated into microstruc-ture data populations by histographic resegmenting.

49. The method of claim 48, wherein said histographic resegmenting is performed iteratively to separate any overlapping microstructure data populations.

50. The method of claim 46, wherein said imaging data for said nerve trunk or tract is segmented by interactive contouring.

51. The method of claim 46, wherein said imaging data for said nerve trunk or tract is segmented by live-wire contouring.

52. The method of claim 46, wherein said imaging data for said nerve trunk or tract is segmented by live-path contouring.

53. The method of claim 46, wherein said microstructures of said nerve trunk or tract also include the perineureum and inter-fascicular tissue.

54. The method of claim 46, wherein said imaging data is acquired by magnetic resonance imaging.

55. The method of claim 46, wherein said imaging data is acquired by magnetic transfer imaging.

56. The method of claim 46, wherein said constructing step comprises constructing a compound image of said nerve trunk or tract comprising imaging data from a plurality of nerve tissue microstructures.

57. The method of Claim 56, wherein at least one of the material opacity, emission and surface reflection for at least one of said microstructures of said compound image is adjusted to enhance the visualization of said nerve trunk or tract.

58. The method of claim 57, wherein said compound image also includes the perineurium and inter-fascicular tissue, and the opacity of said inter-fascicular tissue is set to be greater that the opacity of said perineurium, but less than that of said fasciuli.

59. The method of claim 58, wherein said opacity of said perineurium is set to appear translucent.

60. The method of claim 47, further comprising the step of comparing said myelin calculation to a myelin calculation obtained for another portion of said nerve trunk or tract.

61. The method of claim 46, wherein said nerve trunk or tract is the median nerve.

62. An apparatus comprising:
an input device for receiving magnetic imaging data from a patient for a nerve trunk or tract and the surrounding tissues, characterized in that said imaging data includes imaging data for bound water;
a processing unit comprising:
a first program for segmenting said magnetic imaging data for said nerve trunk or tract from said magnetic imaging data for said surrounding tissues;
a second program for separating said nerve trunk or nerve trunk imaging data into data populations corres-ponding to fasciculi and other nerve trunk or tract microstructures;

a third program for calculating the presence or loss of myelin in said nerve trunk or tract from said fasciculi imaging data; and a processor for sequentially executing said first, second and third programs;
wherein said processor executes:
said first program using imaging data from said input device;
said second program using said nerve trunk or track imaging data isolated by execution of said first program; and said third program using said fasciculi data population obtained by execution of said second program; and an output device for displaying said myelin calculation or a storage device for storing said myelin calculation.

63. The apparatus of claim 62, wherein said second program separates said nerve trunk or tract imaging data into said microstructure data populations by histographic resegmenting.

64. The apparatus of claim 63, wherein said second program is executed iteratively by said processor to separate any overlapping microstructure data populations.

65. The apparatus of claim 62, wherein said first program segments said imaging data by interactive contouring.

66. The apparatus of claim 62, wherein said first program segments said data by live-wire contouring.

67. The apparatus of claim 62, wherein said first program segments said data by live-path contouring.

68. The apparatus of claim 62, wherein said output device is a visual display device.

69. The apparatus of claim 63, further comprising an output device for displaying the histograph obtained by said histographic resegmenting.

70. The apparatus of claim 62, wherein said processing unit comprises a fourth program for comparing myelin calcula-tions obtained from different nerve trunks or tracts or diff-erent portions of the same nerve trunk or tract, wherein said processor executes said fourth program after the execution of said third program using myelin calculations obtained by execution of said third program.

71. A method for the quantitative measurement of a nerve trunk or tract suspected of demyelination, comprising:
receiving from an input device, magnetic imaging data from a patient of a nerve trunk or tract susptected of demyel-ination and the surrounding tissues, characterized n that said imaging data includes imaging data for bound water;
segmenting said imaging data to separate said imaging data for said nerve trunk or tract from said imaging data for said surrounding tissues;
separating said imaging data for said nerve trunk or tract into data populations corresponding to the fasciculi and other microstructures of said nerve trunk or tract;
calculating from said fasciculi imaging data the presence or loss of myelin in said nerve trunk or tract; and storing or displaying said myelin calculation.

72. The method of claim 71, wherein said imaging data for said nerve trunk or tract is separated into microstruc-ture data populations by histographic resegmenting.

73. The method of claim 72, wherein said histographic resegmenting is performed iteratively to separate any overlapping microstructure data populations.

74. The method of claim 71, wherein said imaging data for said nerve trunk or tract is segmented by interactive contouring.

75. The method of claim 71, wherein said imaging data for said nerve trunk or tract is obtained by live-wire contouring.

76. The method of claim 71, wherein said imaging data for said nerve trunk or tract is obtained by live-path contouring.

77. The method of claim 71, wherein said imaging data is acquired by magnetic resonance imaging.

78. The method of claim 71, wherein said imaging data is acquired by magnetic transfer imaging.

79. The method of claim 71, further comprising the step of comparing said myelin calculation to a myelin calculation for another portion of said nerve trunk or tract.

80. The method of claim 71, wherein said nerve trunk or tract is the median nerve.

81. A method for the quantitative measurement of median nerve demyelination within one of a patient's carpal tunnels, comprising:

receiving from an input device, magnetic imaging data from a patient including at least a portion of the median nerve passing through a carpal tunnel:
segmenting said imaging data to isolate said median nerve imaging data from any imaging data for surrounding tissues;
separating said median nerve imaging data into data populations corresponding to the fasciculi and other micro-structures of said median nerve;
calculating from said imaging data for said median nerve fasciculi, the presence or loss of myelin in said median nerve within said carpal tunnel; and storing or displaying said myelin calculation.

82. The method of claim 81, wherein said median nerve imaging data is separated into microstructure data populations by histographic resegmenting.

83. The method of claim 82, wherein said histographic resegmenting is performed iteratively to separate any over-lapping data populations.

84. The method of claim 81, wherein said median nerve imaging data is segmented by interactive contouring.

85. The method of claim 81, wherein said median nerve imaging data is segmented by live-wire contouring.

86. The method of claim 81, wherein said median nerve imaging data is segmented by live-path contouring.

87. The method of claim 81, wherein said imaging data is acquired by magnetic resonance imaging.

88. The method of claim 81, wherein said imaging data is acquired by magnetic transfer imaging.

89. The method of claim 81, further comprising the step of comparing said myelin calculation to a myelin calculation obtained for another portion of said median nerve.

90. The method of claim 81, further comprising the step of comparing said myelin calculation to a myelin calculation obtained for the median nerve within said patient's other carpal tunnel.