CA2602218C - Method and system for characterization of knee joint morphology - Google Patents
Method and system for characterization of knee joint morphology Download PDFInfo
- Publication number
- CA2602218C CA2602218C CA2602218A CA2602218A CA2602218C CA 2602218 C CA2602218 C CA 2602218C CA 2602218 A CA2602218 A CA 2602218A CA 2602218 A CA2602218 A CA 2602218A CA 2602218 C CA2602218 C CA 2602218C
- Authority
- CA
- Canada
- Prior art keywords
- loci
- time point
- joint
- parameters
- statistical template
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 210000000629 knee joint Anatomy 0.000 title claims abstract description 32
- 238000012512 characterization method Methods 0.000 title claims description 8
- 210000003127 knee Anatomy 0.000 claims description 34
- 238000005259 measurement Methods 0.000 claims description 32
- 230000008859 change Effects 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000007619 statistical method Methods 0.000 claims description 5
- 230000000149 penetrating effect Effects 0.000 claims description 4
- 230000005855 radiation Effects 0.000 claims description 4
- 230000001225 therapeutic effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims 1
- 238000004458 analytical method Methods 0.000 description 10
- 201000008482 osteoarthritis Diseases 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 210000000845 cartilage Anatomy 0.000 description 5
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 238000007620 mathematical function Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000000513 principal component analysis Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 206010061818 Disease progression Diseases 0.000 description 2
- 208000003947 Knee Osteoarthritis Diseases 0.000 description 2
- 230000005750 disease progression Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 241000228740 Procrustes Species 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011979 disease modifying therapy Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 230000005251 gamma ray Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 208000037821 progressive disease Diseases 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 210000002303 tibia Anatomy 0.000 description 1
- 210000000689 upper leg Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B23/00—Models for scientific, medical, or mathematical purposes, e.g. full-sized devices for demonstration purposes
- G09B23/28—Models for scientific, medical, or mathematical purposes, e.g. full-sized devices for demonstration purposes for medicine
- G09B23/30—Anatomical models
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/50—Clinical applications
- A61B6/505—Clinical applications involving diagnosis of bone
-
- G06T3/18—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/77—Determining position or orientation of objects or cameras using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/24—Indexing scheme for image data processing or generation, in general involving graphical user interfaces [GUIs]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Abstract
A method and system for characterizing a knee joint in terms of its skeletal morphology. A plurality of loci associated with a model of skeletal structure of a knee joint are fitted and used to parameterize positions of the plurality of loci in a given subject and, thereby, to derive parameters of a deformable statistical template. The skeletal morphology is then characterized on the basis of the derived parameters of the deformable statistical template.
Description
Method and System for Characterization of Knee Joint Morphology Technical Field The invention relates generally to the characterization of the morphology of knee joints, each knee joint at one or more times, for purposes of research or for diagnosis of pathologies in a particular individual.
Background of the Invention Studies of the anatomy of the knee require quantitative characterization of structural parameters. As one particular example, anatomical studies of knees in identified populations require a measure of knee osteoarthritis (OA), a slow progressive disease characterized by loss of cartilage in the joint and leads to loss of joint movement and increased pain. Of the two primary compartments (lateral and medial) of the knee, OA is seen mainly in the medial compartment, due to the higher weight-bearing load borne here. Longitudinal evaluation of the disease in an individual relies on clinical and radiographic features, chiefly pain, disability and structural changes.
Disease-modifying therapies are currently under development and these will rely upon the accurate and precise assessment of the progression of the disease.
In current practice, the primary endpoint used in population studies, clinical trials, and epidemiological studies of OA of the knee, is the surrogate measure of radiographic minimum Joint Space Width (mJSW), measured between either of the weight-bearing surfaces of the femoral condyles and the tibial plateau from a radiograph taken in a semi-flexed position, as shown in the radiograph of Fig. I. The progressive loss of cartilage is measured by narrowing of the mJSW. Since OA is seen mainly in the medial compartment, the mJSW is usually measured in this compartment only.
The measurement of mJSW is usually made by a trained physician using a graduated hand-held lens while reviewing the radiograph on a light-box. Using this method, it is difficult to avoid significant inter- and intra-observer variation due to the subjectivity of the human observer. Moreover, since the mJSW is usually only measured in the medial compartment, it is possible that OA in the lateral compartment will elude detection, and indeed, some patients have primary lateral compartment disease.
Finally, in that the mJSW is a single measure of disease progression, effects in the whole joint may be masked by reliance on a single indicator.
In order to address the major problem of human subjectivity, computer analysis of digitized knee radiographs for the measurement of mJSW has been employed by several authors. One method of computer analysis for deriving mJSW from X-ray images was described by Duryea et al., 27 Med. Phys. , pp. 580-91 (March, 2000).
Moreover, surrogate outcome measures for characterizing the knee joint space other than the mJSW, have been studied and compared with mJSW, for example, by Duryea et al., 11 Osteoarthritis & Cartilage, pp. 102-110 (2003). The foregoing methods are examples of feature-based analytical techniques.
It is desirable, however, to supplement radiographic mJSW measurement with a technique that contributes to greater sensitivity to OA in quantitative measures than is available using non model-based approaches, particularly with a view to clinical evaluation of potential therapies that may be performed more quickly with fewer patients by virtue of the enhanced sensitivity.
Summary of the Invention In accordance with preferred embodiments of the present invention, a method is provided for characterizing a knee joint in terms of a model parameterization.
The method has steps of:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with the specified features to allow parameterization in terms of a deformable statistical template;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template; and d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
It is to be understood that where the term "subset" is used in the description of __ the present invention or in any appended claim, in connection with elements comprising a set, the term "subset" is to be understood as encompassing either a proper, or a full, subset of the entirety of the set of elements. Furthermore, the term "locus"
is used to mean one of the plurality of points within some "distance" of a point (in the vector space) that define a feature. "Distance" refers, generally, to a norm defined over the vector space.
In accordance with other embodiments of the invention, the two-dimensional image may be a radiograph obtained by transmission of penetrating radiation through the knee joint, such as by transmission of x-rays through the knee joint. The method may further include predicting a clinical outcome of a therapeutic modality based on characterization __ of the skeletal morphology in terms of parameters of the deformable statistical template.
In accordance with further embodiments of the invention, the step of creating a deformable statistical template may include defining the set of parameters on the basis of statistical analysis of a set of two-dimensional images of knees, while the step of fitting, at runtime, a plurality of loci associated with specified model features may include __ receiving operator input. Steps (c) and (d) of the method recited above may be repeated at successive points in time for describing evolution of the skeletal morphology over time.
In accordance with yet further embodiments of the invention, an additional step may include performing a clinical intervention between successive iterations of step (c).
The step of parameterizing positions may include successive approximation of positions __ of the plurality of loci. The step of characterizing the skeletal morphology of the knee may include estimating the joint separation width at a specified position in the medial or lateral compartment, or estimating the minimum joint separation in at least one of the medial and lateral compartment.
A further aspect of the invention provides a method for characterizing a knee joint __ in terms of its skeletal morphology. This method has steps of:
Background of the Invention Studies of the anatomy of the knee require quantitative characterization of structural parameters. As one particular example, anatomical studies of knees in identified populations require a measure of knee osteoarthritis (OA), a slow progressive disease characterized by loss of cartilage in the joint and leads to loss of joint movement and increased pain. Of the two primary compartments (lateral and medial) of the knee, OA is seen mainly in the medial compartment, due to the higher weight-bearing load borne here. Longitudinal evaluation of the disease in an individual relies on clinical and radiographic features, chiefly pain, disability and structural changes.
Disease-modifying therapies are currently under development and these will rely upon the accurate and precise assessment of the progression of the disease.
In current practice, the primary endpoint used in population studies, clinical trials, and epidemiological studies of OA of the knee, is the surrogate measure of radiographic minimum Joint Space Width (mJSW), measured between either of the weight-bearing surfaces of the femoral condyles and the tibial plateau from a radiograph taken in a semi-flexed position, as shown in the radiograph of Fig. I. The progressive loss of cartilage is measured by narrowing of the mJSW. Since OA is seen mainly in the medial compartment, the mJSW is usually measured in this compartment only.
The measurement of mJSW is usually made by a trained physician using a graduated hand-held lens while reviewing the radiograph on a light-box. Using this method, it is difficult to avoid significant inter- and intra-observer variation due to the subjectivity of the human observer. Moreover, since the mJSW is usually only measured in the medial compartment, it is possible that OA in the lateral compartment will elude detection, and indeed, some patients have primary lateral compartment disease.
Finally, in that the mJSW is a single measure of disease progression, effects in the whole joint may be masked by reliance on a single indicator.
In order to address the major problem of human subjectivity, computer analysis of digitized knee radiographs for the measurement of mJSW has been employed by several authors. One method of computer analysis for deriving mJSW from X-ray images was described by Duryea et al., 27 Med. Phys. , pp. 580-91 (March, 2000).
Moreover, surrogate outcome measures for characterizing the knee joint space other than the mJSW, have been studied and compared with mJSW, for example, by Duryea et al., 11 Osteoarthritis & Cartilage, pp. 102-110 (2003). The foregoing methods are examples of feature-based analytical techniques.
It is desirable, however, to supplement radiographic mJSW measurement with a technique that contributes to greater sensitivity to OA in quantitative measures than is available using non model-based approaches, particularly with a view to clinical evaluation of potential therapies that may be performed more quickly with fewer patients by virtue of the enhanced sensitivity.
Summary of the Invention In accordance with preferred embodiments of the present invention, a method is provided for characterizing a knee joint in terms of a model parameterization.
The method has steps of:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with the specified features to allow parameterization in terms of a deformable statistical template;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template; and d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
It is to be understood that where the term "subset" is used in the description of __ the present invention or in any appended claim, in connection with elements comprising a set, the term "subset" is to be understood as encompassing either a proper, or a full, subset of the entirety of the set of elements. Furthermore, the term "locus"
is used to mean one of the plurality of points within some "distance" of a point (in the vector space) that define a feature. "Distance" refers, generally, to a norm defined over the vector space.
In accordance with other embodiments of the invention, the two-dimensional image may be a radiograph obtained by transmission of penetrating radiation through the knee joint, such as by transmission of x-rays through the knee joint. The method may further include predicting a clinical outcome of a therapeutic modality based on characterization __ of the skeletal morphology in terms of parameters of the deformable statistical template.
In accordance with further embodiments of the invention, the step of creating a deformable statistical template may include defining the set of parameters on the basis of statistical analysis of a set of two-dimensional images of knees, while the step of fitting, at runtime, a plurality of loci associated with specified model features may include __ receiving operator input. Steps (c) and (d) of the method recited above may be repeated at successive points in time for describing evolution of the skeletal morphology over time.
In accordance with yet further embodiments of the invention, an additional step may include performing a clinical intervention between successive iterations of step (c).
The step of parameterizing positions may include successive approximation of positions __ of the plurality of loci. The step of characterizing the skeletal morphology of the knee may include estimating the joint separation width at a specified position in the medial or lateral compartment, or estimating the minimum joint separation in at least one of the medial and lateral compartment.
A further aspect of the invention provides a method for characterizing a knee joint __ in terms of its skeletal morphology. This method has steps of:
a. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization of a deformable statistical template;
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template; and c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
In accordance with the invention, a computer program product may be provided for use on a computer system for characterizing the skeletal morphology of a knee joint of a subject. The computer program product has program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint into a computer memory, program code for parameterizing positions of the plurality of loci in a given subject to derive parameters of a deformable statistical template, and program code for characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
A further aspect of the present invention provides a method for characterizing a knee joint in terms of a model parameterization. The method comprises the steps of:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization in terms of a deformable statistical template, wherein said fitting includes receiving operator input;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template;
d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
e. carrying out steps b to d(ii) at a first time point and a second time point; and f. determining a change in said joint space width measurement at said position between said first time point and said second time point.
A further aspect of the present invention provides a method for characterizing a knee joint in terms of its skeletal morphology. The method comprises the steps of:
a. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization of a deformable statistical template;
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template;
c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
A further aspect of the present invention provides computer program product for use on a computer system for characterizing the skeletal morphology of a knee joint of a subject. The program product comprises:
a. program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint into a computer memory;
b. program code for parameterizing positions of the plurality of loci in a given subject to derive parameters of a deformable statistical template;
c. program code for characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and 4a e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
Brief Description of the Drawings The foregoing features of the invention will be more readily understood by reference to the following detailed description, taken with reference to the accompanying drawings, in which:
FIG. 1 depicts the prior art standard measurement of the mJSW in the medial compartment as surrogate measure of knee osteoarthritis;
FIG. 2 depicts a typical set of anatomical landmarks defined for application of the invention;
FIG. 3 is a flow chart depicting an application of preferred embodiments of the invention to image data of a knee; and FIG. 4 depicts software analysis of specified measures of knee joint morphology, in accordance with preferred embodiments of the present invention.
4b Detailed Description of Specific Embodiments In accordance with preferred embodiments of the current invention, a computer analysis of a digitized knee radiograph is carried out either automatically or semi-automatically using a deformable statistical template that has been produced, for example, by the statistical analysis of a number of hand-annotated example radiographs of the knee.
The invention described herein and claimed in any appended claims is applied to data obtained by imaging of a knee joint, of a person or animal, by the use of penetrating electromagnetic radiation such as x-rays, for example. Typically, two-dimensional radiographs representing the transmission of penetrating radiation through the joint are employed. While it is to be understood that the invention is not limited in scope to a particular modality of imaging nor to a particular modality for storing and manipulating the image, or images, obtained, there are advantages that arise from particular modalities, such as the high spatial resolution advantageously provided by high energy (x-ray or gamma ray) radiation.
Analysis of the image of the knee joint (referred to, herein, without limitation, as a 'radiograph') proceeds, as described below, on the basis of a 'model' which is applied to the input data. As used herein and in any appended claims, the term 'model,' generally, refers to any mathematical description that provides for parameterization of the position and/or motion of a subject or its component parts. The application of the methods described herein to any model of knee image data is within the scope of the present invention as claimed. When a model is referred to herein as "statistical," it is to be understood as based on an analysis of variation of parameters among members of a population of subjects.
More particularly, the invention will be described with reference to a class of models wherein the model represents the average relative positions of a specified set of 2D point positions on the knee, along with a mathematical description of the way these relative positions vary in normal circumstances among individuals or in a particular individual with the passage of time or due to an intervening circumstance, such, for example, as the progression of a disease. Practice of the present invention is posited upon the existence of a mathematical model of 'plausible' morphologies, wherein morphology encompasses shapes and shape variation, and may also encompass other aspects of appearance such as the texture of a modeled object. A method, described below, is employed for applying the model to data obtained from an image of an actual knee. The method is not, however, specific to any particular placement of the point set, and is illustrated in Fig. 2, purely by way of example, and without limitation, as a set of points (or 'landmarks') 20 placed automatically, or semi-automatically, on the tibial spines, peripheral boundaries of the joint, margins of the femoral condyles and tibial plateau. A
ball-bearing 25, used as a calibration target of known diameter, to allow measurements to be expressed in standard units of length, may also be found in the image shown in Fig. 2.
During run-time application of embodiments of the present invention, the specified points are preliminarily identified (a process referred to, herein, as segmentation), in an image of a knee joint, by a program element trained to identify these positions. In semi-automatic analysis, the user of the application is asked to define some subset (proper or full) of the landmarks on the knee radiograph (larger dots 26 in Figure 2) that were identified, either automatically or semi-automatically, in creation of the deformable statistical template. These positions are defined in such a manner as to effectively 'describe' the radiographic appearance of the knee ¨ whether by relation to extrema' features or otherwise.
Even though the precise morphology of the knee joint varies among subjects and changes with time, these landmarks remain identifiable, for the most part. The template is statistical in that it models the distribution (with the mean and 'normal' variation, or other moments serving, without limitation, as representative characterizations) of the radiographic appearance (as discussed above) of the knee across an ensemble of subjects.
The template, thus, allows for parameterization of the morphology in terms of a finite number of values, with the present invention independent of any particular scheme of parameterization.
A mathematical model of the plausible positions of points may be built, for application in the present invention, as now described. A set of training 2D
data blocks, are taken from an ensemble of radiographic images of knees. These data may be augmented by manual adjustment prior to the process of model building.
For the purpose of building a model, the relative positions of the 2D points are consequential rather than their 'absolute' space-referenced positions. Thus, in building the model, the first step is typically to align each frame of 2D data to a common reference frame, as may be achieved by using one of various standard alignment techniques, such as by 'Procrustes Analysis', which is described by Horn, Closed Form Solution of Absolute Orientation Using Unit Quaternions, J. Optical Society, vol. A 4, pp.
(April, 1987).
The model provides for a compact mathematical description of the variation in relative 2D point positions among frames of the training data. Once the data are aligned, this can be done by one or more of several types of statistical modeling techniques, including, for example, 'Principal Component Analysis' as described by Johnson and Wichem, in Applied Multivariate Statistical Analysis, pp. 458-513 (5th Edition, 2002).
In one set of embodiments, the model may consist of an 'average' shape for the data along with a set of mathematical functions which describe how the shapes can change. By feeding a vector of control numbers or 'model parameters' into the mathematical functions any plausible set of 2D point coordinates can be generated. While the model parameters may span a space of model shape excursions, such is not necessarily the case for practice of the invention as claimed. Moreover, the model may be linear, in the sense in which motions correspond to linear combinations of points moving along straight lines (rather than, for example, arcs or curves). However, the invention is not limited in its applicability to such models.
As will now be discussed with reference to the flow diagram of Fig. 3, embodiments of the present invention use the model, once trained as heretofore described, to segment the image, i.e., to locate the 2D point coordinates that characterise the skeletal morphology of the knee. In accordance with preferred embodiments of the invention, a number of sets of putative input point locations are generated, such as randomly or across a set of predefined possible locations, for example. Each of these sets is tested in order to identify a suitable set of initial input point locations based upon their positioning within 'distances' of neighboring points that the model recognizes as 'reasonable.' The term 'distance', as used herein and in any appended claims refers to a norm with respect to the parameterized variables of the model, and may correspond to a Euclidean norm, but need not, within the scope of the invention. Once initial input 2D
point locations have been identified, the best set of alignment parameters is found, to match these locations with the locations of corresponding points in the model, i.e., to transform all coordinates to the reference frame of the model, in such a way as to maximize the probability that the model parameters describe the actual image.
Iterations, as described below, are then employed for localized fitting. An algorithm for performing such a segmentation step is described in T.F. Cootes and C.J. Taylor, Statistical Models of Appearance for Medical Image Analysis and Computer Vision, in Proc. SPIE
Medical Imaging, (2001).
An initial set of points can be described as a vector X, X = {Xi, x2...xn, YI3 Y2.= =Yn 3 where (xi, yi) are the 2D coordinates of the point with index i.
The points when aligned to the reference frame of the model using, e.g., Horn (supra) are described as a vector X where X' is the result of applying the computed alignment transformation, T, to X, = T(X) , where T is the matrix of computed transformation parameters.
The model is some function, F, which generates a vector of parameters, b, given a set of input point coordinates b = F(X') In one set of embodiments, where the model consists of an 'average' shape for the 2D data along with a set of mathematical functions which describe how the shapes can change, b is calculated using:
b = A(X'- X'm) where X. is the vector of 2D point coordinates for the 'average' shape and A
is a matrix learned during the training phase using, for example, Principal Components Analysis, as described in 'Principal Component Analysis' as described by Johnson and Wichern, in Applied Multivariate Statistical Analysis, pp. 458-513 (5th Edition, 2002).
To estimate a set of point coordinates given a set of model parameters the 'inverse', (which, in most cases, can only be an approximate inverse) of F, F
is used X,' = F" (b) Where X, is the estimated value of the 2D coordinates for a given set of parameters, b. If the model is built using Principal Components Analysis then this is written as:
X, = X. +A"(b) Where A is the pseudoinverse of A which in the case of Principal Component Analysis is identical to the transpose of A.
Various model-fitting algorithms may be used to accomplish the foregoing step.
In one embodiment, a simple, unweighted least squares estimate of the model values is computed:
b A(V-X. ) These values are used predict the values of the entire 2D point vector in the model frame of reference according to:
X,' = F (b) T , the inverse of the transformation matrix T, is used to estimate the entire point vector in the original frame of reference:
X, = T" (Xe") .
This subsequent set of points is now realigned with the model frame of reference and the process that has been described is repeated. A best set of alignment parameters is found and then the fitting algorithm is applied to derive a best set of model parameters.
The best set of model parameters is then used to generate another set of points.
This iterative process is repeated until there is convergence (within a specified criterion) or else until a specified maximum number of iterations have been executed.
When the iterations have finished, the output of the final step is a solution for the full set of 2D points as fit by the model parameters.
A final segmentation (i.e., identification, in the radiograph, of the specified initialization points) may now be used to extract a number of measurements from the radiograph, such as those are outlined below, which are presented as examples only and not as a comprehensive set:
1. The mJSW in either the lateral or medial compartment.
2. The JSW at any position in the medial or lateral compartment, the position of the JSW measurement may be parameterized along a line from the tibial spine to a specified position at an edge of the joint, thus this measurement may be compared at various time points in a longitudinal study.
3. A defined measure, having the dimensions of an area, characterizing a specified region subtended by either compartment.
Since a deformable statistical template has been used to detect and annotate the joint, a parameterization of the joint shape may be determined from optimized fit of this template. Therefore, the 'shape', in an abstract sense, of the joint may be compared to that of a 'universe' of 'normal' joints, or else the changes in shape parameters at time points in a longitudinal study may be used as a novel measurement of disease progression. Changes may be mapped in specified measures over the course of a period of time, whether in the presence of a medical intervention, or otherwise. The case of a medical intervention, in the most general sense, will be referred to herein, and in any .
appended claims, as a "therapeutic modality," and will include the administration of medicinal agents, but will not be limited thereto.
A screenshot of a software application demonstrating several of these measurements being performed is shown in Figure 4. In particular, the following measures are examples of measures that may be employed, in accordance with the invention, for characterization of knee morphology:
Cartilage Area Measurement For both baseline and follow-up images, and for either compartment, an area is measured, defined by the tibial plateau, the femoral condyle and the joint space widths (JSWs) at either end of the compartment.
Joint Space Width (JSW) Profile Measurement For both baseline and follow-up images, and for either compartment, a profile of Joint Space Width (JSW) may be measured along the entire length of the tibial plateau.
The JSW at a given point is the minimum distance between the tibial boundary and the femoral boundary. The JSW profile is measured (in mm) as a function of the distance along the medial axis of the joint from the tibial spine.
Minimum JSW Measurement For both baseline and follow-up images, a minimum JSW may be measured in both the medial and the lateral compartment. The minimum JSW is the minimum value of the JSW profile measurement in a given range along the tibial plateau. By default, the minimum is found between the inner extent of the cartilage area measurement, and the outer extent of the JSW profile measurement.
Equivalent JSW Measurement For follow-up images, and for either compartment, the JSW may be measured at the estimated position (along the tibial plateau) at which the mJSW was calculated on the baseline image. This measurement requires that both baseline and follow-up images for a patient are attached to the study.
Cross-over Flag The cross-over flag is a flag which indicates whether the boundary of the tibia crosses over the boundary of the femur, for either the medial compartment or the lateral compartment. In these cases where cross-over is found, it may be desirable to ignore the results.
Shape Measure For a given pair of baseline and follow-up images, the system may calculate a statistic corresponding to how much the overall shape of the knee has changed between the baseline knee and the follow-up knee. The higher the value of this statistic, the more change there is between baseline and follow-up. The units of the Shape Measure are mm, and the figure corresponds to the mean amount by which each point on the segmented knee boundary has moved between the baseline and follow-up knee, measured in the same frame of reference. This measurement requires that both baseline and follow-up images for a patient be attached to the study.
This should allow the rapid identification of 'interesting' images that may warrant extended manual investigation.
The disclosed methods for characterizing the morphology of a knee joint may be implemented as a computer program product for use with a computer system. Such implementations may include a series of computer instructions fixed either on a tangible medium, such as a computer readable medium (e.g., a diskette, CD-ROM, ROM, or fixed disk) or transmittable to a computer system, via a modem or other interface device, such as a communications adapter connected to a network over a medium. The medium may be either a tangible medium (e.g., optical or analog communications lines) or a medium implemented with wireless techniques (e.g., microwave, infrared or other transmission techniques). The series of computer instructions embodies all or part of the functionality previously described herein with respect to the system. Those skilled in the art should appreciate that such computer instructions can be written in a number of programming languages for use with many computer architectures or operating systems.
Furthermore, such instructions may be stored in any memory device, such as semiconductor, magnetic, optical or other memory devices, and may be transmitted using any communications technology, such as optical, infrared, microwave, or other transmission technologies. It is expected that such a computer program product may be distributed as a removable medium with accompanying printed or electronic documentation (e.g., shrink wrapped software), preloaded with a computer system (e.g., on system ROM or fixed disk), or distributed from a server or electronic bulletin board over the network (e.g., the Internet or World Wide Web). Of course, some embodiments of the invention may be I implemented as a combination of both software (e.g., a computer program product) and hardware. Still other embodiments of the invention are implemented as entirely hardware, or entirely software (e.g., a computer program product).
The described embodiments of the invention are intended to be merely exemplary and numerous variations and modifications will be apparent to those skilled in the art. All such variations and modifications are intended to be within the scope of the present invention as defined in the appended claims.
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template; and c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
In accordance with the invention, a computer program product may be provided for use on a computer system for characterizing the skeletal morphology of a knee joint of a subject. The computer program product has program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint into a computer memory, program code for parameterizing positions of the plurality of loci in a given subject to derive parameters of a deformable statistical template, and program code for characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template.
A further aspect of the present invention provides a method for characterizing a knee joint in terms of a model parameterization. The method comprises the steps of:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization in terms of a deformable statistical template, wherein said fitting includes receiving operator input;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template;
d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
e. carrying out steps b to d(ii) at a first time point and a second time point; and f. determining a change in said joint space width measurement at said position between said first time point and said second time point.
A further aspect of the present invention provides a method for characterizing a knee joint in terms of its skeletal morphology. The method comprises the steps of:
a. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization of a deformable statistical template;
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template;
c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
A further aspect of the present invention provides computer program product for use on a computer system for characterizing the skeletal morphology of a knee joint of a subject. The program product comprises:
a. program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint into a computer memory;
b. program code for parameterizing positions of the plurality of loci in a given subject to derive parameters of a deformable statistical template;
c. program code for characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and 4a e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
Brief Description of the Drawings The foregoing features of the invention will be more readily understood by reference to the following detailed description, taken with reference to the accompanying drawings, in which:
FIG. 1 depicts the prior art standard measurement of the mJSW in the medial compartment as surrogate measure of knee osteoarthritis;
FIG. 2 depicts a typical set of anatomical landmarks defined for application of the invention;
FIG. 3 is a flow chart depicting an application of preferred embodiments of the invention to image data of a knee; and FIG. 4 depicts software analysis of specified measures of knee joint morphology, in accordance with preferred embodiments of the present invention.
4b Detailed Description of Specific Embodiments In accordance with preferred embodiments of the current invention, a computer analysis of a digitized knee radiograph is carried out either automatically or semi-automatically using a deformable statistical template that has been produced, for example, by the statistical analysis of a number of hand-annotated example radiographs of the knee.
The invention described herein and claimed in any appended claims is applied to data obtained by imaging of a knee joint, of a person or animal, by the use of penetrating electromagnetic radiation such as x-rays, for example. Typically, two-dimensional radiographs representing the transmission of penetrating radiation through the joint are employed. While it is to be understood that the invention is not limited in scope to a particular modality of imaging nor to a particular modality for storing and manipulating the image, or images, obtained, there are advantages that arise from particular modalities, such as the high spatial resolution advantageously provided by high energy (x-ray or gamma ray) radiation.
Analysis of the image of the knee joint (referred to, herein, without limitation, as a 'radiograph') proceeds, as described below, on the basis of a 'model' which is applied to the input data. As used herein and in any appended claims, the term 'model,' generally, refers to any mathematical description that provides for parameterization of the position and/or motion of a subject or its component parts. The application of the methods described herein to any model of knee image data is within the scope of the present invention as claimed. When a model is referred to herein as "statistical," it is to be understood as based on an analysis of variation of parameters among members of a population of subjects.
More particularly, the invention will be described with reference to a class of models wherein the model represents the average relative positions of a specified set of 2D point positions on the knee, along with a mathematical description of the way these relative positions vary in normal circumstances among individuals or in a particular individual with the passage of time or due to an intervening circumstance, such, for example, as the progression of a disease. Practice of the present invention is posited upon the existence of a mathematical model of 'plausible' morphologies, wherein morphology encompasses shapes and shape variation, and may also encompass other aspects of appearance such as the texture of a modeled object. A method, described below, is employed for applying the model to data obtained from an image of an actual knee. The method is not, however, specific to any particular placement of the point set, and is illustrated in Fig. 2, purely by way of example, and without limitation, as a set of points (or 'landmarks') 20 placed automatically, or semi-automatically, on the tibial spines, peripheral boundaries of the joint, margins of the femoral condyles and tibial plateau. A
ball-bearing 25, used as a calibration target of known diameter, to allow measurements to be expressed in standard units of length, may also be found in the image shown in Fig. 2.
During run-time application of embodiments of the present invention, the specified points are preliminarily identified (a process referred to, herein, as segmentation), in an image of a knee joint, by a program element trained to identify these positions. In semi-automatic analysis, the user of the application is asked to define some subset (proper or full) of the landmarks on the knee radiograph (larger dots 26 in Figure 2) that were identified, either automatically or semi-automatically, in creation of the deformable statistical template. These positions are defined in such a manner as to effectively 'describe' the radiographic appearance of the knee ¨ whether by relation to extrema' features or otherwise.
Even though the precise morphology of the knee joint varies among subjects and changes with time, these landmarks remain identifiable, for the most part. The template is statistical in that it models the distribution (with the mean and 'normal' variation, or other moments serving, without limitation, as representative characterizations) of the radiographic appearance (as discussed above) of the knee across an ensemble of subjects.
The template, thus, allows for parameterization of the morphology in terms of a finite number of values, with the present invention independent of any particular scheme of parameterization.
A mathematical model of the plausible positions of points may be built, for application in the present invention, as now described. A set of training 2D
data blocks, are taken from an ensemble of radiographic images of knees. These data may be augmented by manual adjustment prior to the process of model building.
For the purpose of building a model, the relative positions of the 2D points are consequential rather than their 'absolute' space-referenced positions. Thus, in building the model, the first step is typically to align each frame of 2D data to a common reference frame, as may be achieved by using one of various standard alignment techniques, such as by 'Procrustes Analysis', which is described by Horn, Closed Form Solution of Absolute Orientation Using Unit Quaternions, J. Optical Society, vol. A 4, pp.
(April, 1987).
The model provides for a compact mathematical description of the variation in relative 2D point positions among frames of the training data. Once the data are aligned, this can be done by one or more of several types of statistical modeling techniques, including, for example, 'Principal Component Analysis' as described by Johnson and Wichem, in Applied Multivariate Statistical Analysis, pp. 458-513 (5th Edition, 2002).
In one set of embodiments, the model may consist of an 'average' shape for the data along with a set of mathematical functions which describe how the shapes can change. By feeding a vector of control numbers or 'model parameters' into the mathematical functions any plausible set of 2D point coordinates can be generated. While the model parameters may span a space of model shape excursions, such is not necessarily the case for practice of the invention as claimed. Moreover, the model may be linear, in the sense in which motions correspond to linear combinations of points moving along straight lines (rather than, for example, arcs or curves). However, the invention is not limited in its applicability to such models.
As will now be discussed with reference to the flow diagram of Fig. 3, embodiments of the present invention use the model, once trained as heretofore described, to segment the image, i.e., to locate the 2D point coordinates that characterise the skeletal morphology of the knee. In accordance with preferred embodiments of the invention, a number of sets of putative input point locations are generated, such as randomly or across a set of predefined possible locations, for example. Each of these sets is tested in order to identify a suitable set of initial input point locations based upon their positioning within 'distances' of neighboring points that the model recognizes as 'reasonable.' The term 'distance', as used herein and in any appended claims refers to a norm with respect to the parameterized variables of the model, and may correspond to a Euclidean norm, but need not, within the scope of the invention. Once initial input 2D
point locations have been identified, the best set of alignment parameters is found, to match these locations with the locations of corresponding points in the model, i.e., to transform all coordinates to the reference frame of the model, in such a way as to maximize the probability that the model parameters describe the actual image.
Iterations, as described below, are then employed for localized fitting. An algorithm for performing such a segmentation step is described in T.F. Cootes and C.J. Taylor, Statistical Models of Appearance for Medical Image Analysis and Computer Vision, in Proc. SPIE
Medical Imaging, (2001).
An initial set of points can be described as a vector X, X = {Xi, x2...xn, YI3 Y2.= =Yn 3 where (xi, yi) are the 2D coordinates of the point with index i.
The points when aligned to the reference frame of the model using, e.g., Horn (supra) are described as a vector X where X' is the result of applying the computed alignment transformation, T, to X, = T(X) , where T is the matrix of computed transformation parameters.
The model is some function, F, which generates a vector of parameters, b, given a set of input point coordinates b = F(X') In one set of embodiments, where the model consists of an 'average' shape for the 2D data along with a set of mathematical functions which describe how the shapes can change, b is calculated using:
b = A(X'- X'm) where X. is the vector of 2D point coordinates for the 'average' shape and A
is a matrix learned during the training phase using, for example, Principal Components Analysis, as described in 'Principal Component Analysis' as described by Johnson and Wichern, in Applied Multivariate Statistical Analysis, pp. 458-513 (5th Edition, 2002).
To estimate a set of point coordinates given a set of model parameters the 'inverse', (which, in most cases, can only be an approximate inverse) of F, F
is used X,' = F" (b) Where X, is the estimated value of the 2D coordinates for a given set of parameters, b. If the model is built using Principal Components Analysis then this is written as:
X, = X. +A"(b) Where A is the pseudoinverse of A which in the case of Principal Component Analysis is identical to the transpose of A.
Various model-fitting algorithms may be used to accomplish the foregoing step.
In one embodiment, a simple, unweighted least squares estimate of the model values is computed:
b A(V-X. ) These values are used predict the values of the entire 2D point vector in the model frame of reference according to:
X,' = F (b) T , the inverse of the transformation matrix T, is used to estimate the entire point vector in the original frame of reference:
X, = T" (Xe") .
This subsequent set of points is now realigned with the model frame of reference and the process that has been described is repeated. A best set of alignment parameters is found and then the fitting algorithm is applied to derive a best set of model parameters.
The best set of model parameters is then used to generate another set of points.
This iterative process is repeated until there is convergence (within a specified criterion) or else until a specified maximum number of iterations have been executed.
When the iterations have finished, the output of the final step is a solution for the full set of 2D points as fit by the model parameters.
A final segmentation (i.e., identification, in the radiograph, of the specified initialization points) may now be used to extract a number of measurements from the radiograph, such as those are outlined below, which are presented as examples only and not as a comprehensive set:
1. The mJSW in either the lateral or medial compartment.
2. The JSW at any position in the medial or lateral compartment, the position of the JSW measurement may be parameterized along a line from the tibial spine to a specified position at an edge of the joint, thus this measurement may be compared at various time points in a longitudinal study.
3. A defined measure, having the dimensions of an area, characterizing a specified region subtended by either compartment.
Since a deformable statistical template has been used to detect and annotate the joint, a parameterization of the joint shape may be determined from optimized fit of this template. Therefore, the 'shape', in an abstract sense, of the joint may be compared to that of a 'universe' of 'normal' joints, or else the changes in shape parameters at time points in a longitudinal study may be used as a novel measurement of disease progression. Changes may be mapped in specified measures over the course of a period of time, whether in the presence of a medical intervention, or otherwise. The case of a medical intervention, in the most general sense, will be referred to herein, and in any .
appended claims, as a "therapeutic modality," and will include the administration of medicinal agents, but will not be limited thereto.
A screenshot of a software application demonstrating several of these measurements being performed is shown in Figure 4. In particular, the following measures are examples of measures that may be employed, in accordance with the invention, for characterization of knee morphology:
Cartilage Area Measurement For both baseline and follow-up images, and for either compartment, an area is measured, defined by the tibial plateau, the femoral condyle and the joint space widths (JSWs) at either end of the compartment.
Joint Space Width (JSW) Profile Measurement For both baseline and follow-up images, and for either compartment, a profile of Joint Space Width (JSW) may be measured along the entire length of the tibial plateau.
The JSW at a given point is the minimum distance between the tibial boundary and the femoral boundary. The JSW profile is measured (in mm) as a function of the distance along the medial axis of the joint from the tibial spine.
Minimum JSW Measurement For both baseline and follow-up images, a minimum JSW may be measured in both the medial and the lateral compartment. The minimum JSW is the minimum value of the JSW profile measurement in a given range along the tibial plateau. By default, the minimum is found between the inner extent of the cartilage area measurement, and the outer extent of the JSW profile measurement.
Equivalent JSW Measurement For follow-up images, and for either compartment, the JSW may be measured at the estimated position (along the tibial plateau) at which the mJSW was calculated on the baseline image. This measurement requires that both baseline and follow-up images for a patient are attached to the study.
Cross-over Flag The cross-over flag is a flag which indicates whether the boundary of the tibia crosses over the boundary of the femur, for either the medial compartment or the lateral compartment. In these cases where cross-over is found, it may be desirable to ignore the results.
Shape Measure For a given pair of baseline and follow-up images, the system may calculate a statistic corresponding to how much the overall shape of the knee has changed between the baseline knee and the follow-up knee. The higher the value of this statistic, the more change there is between baseline and follow-up. The units of the Shape Measure are mm, and the figure corresponds to the mean amount by which each point on the segmented knee boundary has moved between the baseline and follow-up knee, measured in the same frame of reference. This measurement requires that both baseline and follow-up images for a patient be attached to the study.
This should allow the rapid identification of 'interesting' images that may warrant extended manual investigation.
The disclosed methods for characterizing the morphology of a knee joint may be implemented as a computer program product for use with a computer system. Such implementations may include a series of computer instructions fixed either on a tangible medium, such as a computer readable medium (e.g., a diskette, CD-ROM, ROM, or fixed disk) or transmittable to a computer system, via a modem or other interface device, such as a communications adapter connected to a network over a medium. The medium may be either a tangible medium (e.g., optical or analog communications lines) or a medium implemented with wireless techniques (e.g., microwave, infrared or other transmission techniques). The series of computer instructions embodies all or part of the functionality previously described herein with respect to the system. Those skilled in the art should appreciate that such computer instructions can be written in a number of programming languages for use with many computer architectures or operating systems.
Furthermore, such instructions may be stored in any memory device, such as semiconductor, magnetic, optical or other memory devices, and may be transmitted using any communications technology, such as optical, infrared, microwave, or other transmission technologies. It is expected that such a computer program product may be distributed as a removable medium with accompanying printed or electronic documentation (e.g., shrink wrapped software), preloaded with a computer system (e.g., on system ROM or fixed disk), or distributed from a server or electronic bulletin board over the network (e.g., the Internet or World Wide Web). Of course, some embodiments of the invention may be I implemented as a combination of both software (e.g., a computer program product) and hardware. Still other embodiments of the invention are implemented as entirely hardware, or entirely software (e.g., a computer program product).
The described embodiments of the invention are intended to be merely exemplary and numerous variations and modifications will be apparent to those skilled in the art. All such variations and modifications are intended to be within the scope of the present invention as defined in the appended claims.
Claims (9)
1. A
method for characterizing a knee joint in terms of a model parameterization, the method comprising:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization in terms of a deformable statistical template, wherein said fitting includes receiving operator input;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template;
d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i.
generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
e. carrying out steps b to d(ii) at a first time point and a second time point; and f. determining a change in said joint space width measurement at said position between said first time point and said second time point.
method for characterizing a knee joint in terms of a model parameterization, the method comprising:
a. creating a deformable statistical template characterized by a set of parameters that, as a set, span an abstract vector space representing the set of spatial positions of specified features of the knee joint, each vector uniquely describing its variation from a population mean;
b. fitting at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization in terms of a deformable statistical template, wherein said fitting includes receiving operator input;
c. parameterizing positions of the plurality of loci in a given subject to derive values for parameters of the deformable statistical template;
d. characterizing the skeletal morphology of the knee of the given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i.
generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
e. carrying out steps b to d(ii) at a first time point and a second time point; and f. determining a change in said joint space width measurement at said position between said first time point and said second time point.
2. A
method in accordance with claim 1, wherein the two-dimensional image is a radiograph obtained by transmission of penetrating radiation through the knee joint.
method in accordance with claim 1, wherein the two-dimensional image is a radiograph obtained by transmission of penetrating radiation through the knee joint.
3. A
method in accordance with claim 1, wherein the two-dimensional image is a radiograph obtained by transmission of x-rays through the knee joint.
method in accordance with claim 1, wherein the two-dimensional image is a radiograph obtained by transmission of x-rays through the knee joint.
4. A
method in accordance with claim 1, further including predicting a clinical outcome of a therapeutic modality based on characterization of the skeletal morphology in terms of parameters of the deformable statistical template.
method in accordance with claim 1, further including predicting a clinical outcome of a therapeutic modality based on characterization of the skeletal morphology in terms of parameters of the deformable statistical template.
5. A
method in accordance with claim 1, wherein the step of creating a deformable statistical template includes defining the set of parameters on the basis of statistical analysis of a set of two-dimensional images of knees.
method in accordance with claim 1, wherein the step of creating a deformable statistical template includes defining the set of parameters on the basis of statistical analysis of a set of two-dimensional images of knees.
6. A
method in accordance with claim 1, further comprising a step of performing a clinical intervention between the extraction of a joint space width measurement at the first time point and the second time point.
method in accordance with claim 1, further comprising a step of performing a clinical intervention between the extraction of a joint space width measurement at the first time point and the second time point.
7. A
method in accordance with claim 1, wherein the step of fitting includes successive approximation of positions of the plurality of loci.
method in accordance with claim 1, wherein the step of fitting includes successive approximation of positions of the plurality of loci.
8. A
method for characterizing a knee joint in terms of its skeletal morphology, the method comprising:
a. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization of a deformable statistical template;
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template;
c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i.
generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
method for characterizing a knee joint in terms of its skeletal morphology, the method comprising:
a. fitting, at runtime, in a two-dimensional image of the skeleture of a knee, a plurality of loci associated with specified features to allow parameterization of a deformable statistical template;
b. parameterizing positions of the plurality of loci in a given subject to derive parameters of the deformable statistical template;
c. characterizing the skeletal morphology on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i.
generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and ii. extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
d. carrying out steps a to c(ii) at a first time point and a second time point; and e. determining a change in said joint space width measurement at said position between said first time point and said second time point.
9. A
computer readable memory having recorded thereon statements and instructions for execution by a computer, the statements and instructions comprising:
a.
program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint of a given subject;
b. program code for parameterizing positions of the plurality of loci in said given subject to derive parameters of a deformable statistical template; and c. program code for characterizing a skeletal morphology of said knee joint of said given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
wherein steps a to c(ii) are carried out at a first time point and a second time point; and wherein a change in said joint space width measurement at said position is determined between said first time point and said second time point.
computer readable memory having recorded thereon statements and instructions for execution by a computer, the statements and instructions comprising:
a.
program code for storing loci associated with specified skeletal features in a two-dimensional image of a knee joint of a given subject;
b. program code for parameterizing positions of the plurality of loci in said given subject to derive parameters of a deformable statistical template; and c. program code for characterizing a skeletal morphology of said knee joint of said given subject on the basis of either a subset of the plurality of loci, and/or the derived parameters of the deformable statistical template by:
i. generating a set of points based on the subset of the plurality of loci and/or on the derived parameters of the deformable statistical template; and extracting a joint space width measurement at a position parameterized along a line from the tibial spine to a specified position at an edge of the joint based on the set of points;
wherein steps a to c(ii) are carried out at a first time point and a second time point; and wherein a change in said joint space width measurement at said position is determined between said first time point and said second time point.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US66491205P | 2005-03-24 | 2005-03-24 | |
US60/664,912 | 2005-03-24 | ||
PCT/US2006/009565 WO2006104707A2 (en) | 2005-03-24 | 2006-03-16 | Method and system for characterization of knee joint morphology |
Publications (2)
Publication Number | Publication Date |
---|---|
CA2602218A1 CA2602218A1 (en) | 2006-10-05 |
CA2602218C true CA2602218C (en) | 2015-02-10 |
Family
ID=36791636
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA2602218A Expired - Fee Related CA2602218C (en) | 2005-03-24 | 2006-03-16 | Method and system for characterization of knee joint morphology |
Country Status (7)
Country | Link |
---|---|
US (2) | US7760923B2 (en) |
EP (1) | EP1861821A2 (en) |
JP (1) | JP5054669B2 (en) |
CN (1) | CN101189638B (en) |
AU (1) | AU2006229785A1 (en) |
CA (1) | CA2602218C (en) |
WO (1) | WO2006104707A2 (en) |
Families Citing this family (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1861821A2 (en) | 2005-03-24 | 2007-12-05 | Image Metrics Limited | Method and system for characterization of knee joint morphology |
CA2644574C (en) | 2006-03-17 | 2016-11-08 | Zimmer, Inc. | Methods of predetermining the contour of a resected bone surface and assessing the fit of a prosthesis on the bone |
JP4934786B2 (en) * | 2006-10-13 | 2012-05-16 | 国立大学法人 東京大学 | Knee joint diagnosis support method, apparatus and program |
EP2126841A2 (en) * | 2007-01-16 | 2009-12-02 | Optasia Medical, Limited | Image processing systems and methods |
US9179983B2 (en) | 2007-08-14 | 2015-11-10 | Zimmer, Inc. | Method of determining a contour of an anatomical structure and selecting an orthopaedic implant to replicate the anatomical structure |
US8126249B2 (en) * | 2008-05-30 | 2012-02-28 | Optasia Medical Limited | Methods of and system for detection and tracking of osteoporosis |
WO2010018406A1 (en) | 2008-08-12 | 2010-02-18 | Wyeth Pharmaceuticals Inc. | Morphometry of the human hip joint and prediction of osteoarthritis |
EP2385792A1 (en) | 2008-08-12 | 2011-11-16 | Wyeth Pharmaceuticals Inc. | Morphometry of the human knee joint and prediction for ostearthritis |
JP5467747B2 (en) * | 2008-09-16 | 2014-04-09 | 永一 坂東 | Cause physical quantity estimation method and apparatus |
US8294767B2 (en) * | 2009-01-30 | 2012-10-23 | Microsoft Corporation | Body scan |
US8675939B2 (en) | 2010-07-13 | 2014-03-18 | Stryker Leibinger Gmbh & Co. Kg | Registration of anatomical data sets |
EP2603136B1 (en) * | 2010-08-13 | 2023-07-12 | Smith & Nephew, Inc. | Detection of anatomical landmarks |
US9236024B2 (en) | 2011-12-06 | 2016-01-12 | Glasses.Com Inc. | Systems and methods for obtaining a pupillary distance measurement using a mobile computing device |
US9311746B2 (en) | 2012-05-23 | 2016-04-12 | Glasses.Com Inc. | Systems and methods for generating a 3-D model of a virtual try-on product |
US9286715B2 (en) | 2012-05-23 | 2016-03-15 | Glasses.Com Inc. | Systems and methods for adjusting a virtual try-on |
US9483853B2 (en) | 2012-05-23 | 2016-11-01 | Glasses.Com Inc. | Systems and methods to display rendered images |
US20140270449A1 (en) * | 2013-03-15 | 2014-09-18 | John Andrew HIPP | Interactive method to assess joint space narrowing |
CN103186901A (en) * | 2013-03-29 | 2013-07-03 | 中国人民解放军第三军医大学 | Full-automatic image segmentation method |
TWI700473B (en) | 2014-06-04 | 2020-08-01 | 美商康寧公司 | Method and system for measuring thickness of glass article |
US10376182B2 (en) | 2015-10-30 | 2019-08-13 | Orthosensor Inc. | Spine measurement system including rod measurement |
US10595941B2 (en) | 2015-10-30 | 2020-03-24 | Orthosensor Inc. | Spine measurement system and method therefor |
GB2544531A (en) * | 2015-11-20 | 2017-05-24 | Imorphics Ltd | Image processing method |
JP6933521B2 (en) * | 2016-08-05 | 2021-09-08 | キヤノンメディカルシステムズ株式会社 | Medical image processing equipment and X-ray CT equipment |
CN109377478B (en) * | 2018-09-26 | 2021-09-14 | 宁波工程学院 | Automatic grading method for osteoarthritis |
Family Cites Families (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5360446A (en) * | 1992-12-18 | 1994-11-01 | Zimmer, Inc. | Interactive prosthesis design system for implantable prosthesis |
US5682886A (en) * | 1995-12-26 | 1997-11-04 | Musculographics Inc | Computer-assisted surgical system |
US5880976A (en) * | 1997-02-21 | 1999-03-09 | Carnegie Mellon University | Apparatus and method for facilitating the implantation of artificial components in joints |
DE69922317D1 (en) * | 1998-09-29 | 2005-01-05 | Koninkl Philips Electronics Nv | Image processing method for ultrasonic medical images of the bone structure, and a computer-aided surgery device |
FR2798760B1 (en) * | 1999-09-17 | 2002-03-29 | Univ Joseph Fourier | RECONSTRUCTION OF THREE-DIMENSIONAL SURFACES USING STATISTICAL MODELS |
US7027650B2 (en) | 1999-12-10 | 2006-04-11 | Christian Williame | Dynamic computing imagery, especially for visceral osteopathy and for articular kinetics |
US6711432B1 (en) * | 2000-10-23 | 2004-03-23 | Carnegie Mellon University | Computer-aided orthopedic surgery |
US6701174B1 (en) * | 2000-04-07 | 2004-03-02 | Carnegie Mellon University | Computer-aided bone distraction |
US6963825B1 (en) * | 2000-10-16 | 2005-11-08 | Kabushiki Kaisha Armonicos | Outline extracting apparatus and method and computer readable memory medium |
EP1293925A1 (en) | 2001-09-18 | 2003-03-19 | Agfa-Gevaert | Radiographic scoring method |
JP2003144454A (en) * | 2001-11-16 | 2003-05-20 | Yoshio Koga | Joint operation support information computing method, joint operation support information computing program, and joint operation support information computing system |
EP1923809A3 (en) * | 2002-03-27 | 2010-12-29 | Agfa HealthCare NV | Method of performing geometric measurements on digital radiological images |
US20050168460A1 (en) | 2002-04-04 | 2005-08-04 | Anshuman Razdan | Three-dimensional digital library system |
JP2004041694A (en) * | 2002-05-13 | 2004-02-12 | Fuji Photo Film Co Ltd | Image generation device and program, image selecting device, image outputting device and image providing service system |
GB2393625B (en) * | 2002-09-26 | 2004-08-18 | Internet Tech Ltd | Orthopaedic surgery planning |
US20040101186A1 (en) * | 2002-11-27 | 2004-05-27 | Xin Tong | Initializing model-based interpretations of digital radiographs |
GB2402470B (en) | 2003-04-30 | 2005-11-30 | Image Metrics Plc | A method of and apparatus for classifying images |
US8290564B2 (en) | 2003-09-19 | 2012-10-16 | Imatx, Inc. | Method for bone structure prognosis and simulated bone remodeling |
EP1598778B1 (en) * | 2004-05-18 | 2008-08-13 | Agfa HealthCare NV | Method for automatically mapping of geometric objects in digital medical images |
EP1861821A2 (en) | 2005-03-24 | 2007-12-05 | Image Metrics Limited | Method and system for characterization of knee joint morphology |
CN1710612A (en) * | 2005-07-08 | 2005-12-21 | 北京航空航天大学 | Skin-top affecting weight distribution method based on standby joint point collection |
US7848592B2 (en) | 2006-07-31 | 2010-12-07 | Carestream Health, Inc. | Image fusion for radiation therapy |
EP2126841A2 (en) | 2007-01-16 | 2009-12-02 | Optasia Medical, Limited | Image processing systems and methods |
US8126249B2 (en) | 2008-05-30 | 2012-02-28 | Optasia Medical Limited | Methods of and system for detection and tracking of osteoporosis |
-
2006
- 2006-03-16 EP EP06738603A patent/EP1861821A2/en not_active Withdrawn
- 2006-03-16 WO PCT/US2006/009565 patent/WO2006104707A2/en active Application Filing
- 2006-03-16 US US11/376,868 patent/US7760923B2/en not_active Expired - Fee Related
- 2006-03-16 AU AU2006229785A patent/AU2006229785A1/en not_active Abandoned
- 2006-03-16 CA CA2602218A patent/CA2602218C/en not_active Expired - Fee Related
- 2006-03-16 JP JP2008503042A patent/JP5054669B2/en not_active Expired - Fee Related
- 2006-03-16 CN CN2006800094375A patent/CN101189638B/en not_active Expired - Fee Related
-
2010
- 2010-04-16 US US12/761,693 patent/US7929745B2/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
US20100272332A1 (en) | 2010-10-28 |
US7929745B2 (en) | 2011-04-19 |
AU2006229785A1 (en) | 2006-10-05 |
WO2006104707A2 (en) | 2006-10-05 |
CN101189638B (en) | 2011-11-30 |
CA2602218A1 (en) | 2006-10-05 |
AU2006229785A2 (en) | 2006-10-05 |
JP5054669B2 (en) | 2012-10-24 |
EP1861821A2 (en) | 2007-12-05 |
JP2008534055A (en) | 2008-08-28 |
US20060216681A1 (en) | 2006-09-28 |
US7760923B2 (en) | 2010-07-20 |
CN101189638A (en) | 2008-05-28 |
WO2006104707A3 (en) | 2006-12-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2602218C (en) | Method and system for characterization of knee joint morphology | |
Sarkalkan et al. | Statistical shape and appearance models of bones | |
Kozic et al. | Optimisation of orthopaedic implant design using statistical shape space analysis based on level sets | |
Karade et al. | 3D femur model reconstruction from biplane X-ray images: a novel method based on Laplacian surface deformation | |
Zheng | Statistical shape model‐based reconstruction of a scaled, patient‐specific surface model of the pelvis from a single standard AP x‐ray radiograph | |
Maken et al. | 2D-to-3D: a review for computational 3D image reconstruction from X-ray images | |
Zhang et al. | 3-D reconstruction of the spine from biplanar radiographs based on contour matching using the hough transform | |
Patias | Medical imaging challenges photogrammetry | |
Ghafurian et al. | A computationally efficient 3D/2D registration method based on image gradient direction probability density function | |
CN109350059B (en) | Combined steering engine and landmark engine for elbow auto-alignment | |
Zhou et al. | Transfer learning from an artificial radiograph-landmark dataset for registration of the anatomic skull model to dual fluoroscopic X-ray images | |
Zheng et al. | Lumbar spine visualisation based on kinematic analysis from videofluoroscopic imaging | |
Strickland et al. | Development of subject-specific geometric spine model through use of automated active contour segmentation and kinematic constraint-limited registration | |
Schmid et al. | Musculoskeletal simulation model generation from MRI data sets and motion capture data | |
Liu et al. | Rigid model‐based 3D segmentation of the bones of joints in MR and CT images for motion analysis | |
KR20160057024A (en) | Markerless 3D Object Tracking Apparatus and Method therefor | |
Ün et al. | An analytical method to create patient-specific deformed bone models using X-ray images and a healthy bone model | |
Kadoury et al. | Self-calibration of biplanar radiographic images through geometric spine shape descriptors | |
Kumar et al. | Improving visibility of stereo-radiographic spine reconstruction with geometric inferences | |
Saadat et al. | An efficient image registration method for 3D post-operative analysis of total knee arthroplasty | |
EP3843629A1 (en) | Detecting spinal shape from optical scan | |
Agrawal et al. | Neuro-fuzzy approach for reconstruction of 3-D spine model using 2-D spine images and human anatomy | |
Ha et al. | 2D-3D Reconstruction of a Femur by Single X-Ray Image Based on Deep Transfer Learning Network | |
Zheng et al. | Reconstruction of 3D vertebral models from a single 2D lateral fluoroscopic image | |
Benjelloun et al. | A New semi-automatic approach for X-Ray cervical images segmentation using active shape model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request | ||
MKLA | Lapsed |
Effective date: 20190318 |