Efficient image reconstruction algorithm for the circle and arc cone beam computer tomography

Abstract

Methods, systems and processes for providing efficient, accurate and exact image reconstruction using portable and easy to use C-arm scanning devices and rotating gantries, and the like, that combines both a circle and a curve scan. The invention can provide exact convolution-based filtered back projection (FBP) image reconstruction by combining two curved scans of the object. The curved scan can be less than or greater than a full circle about an object being scanned. The invention can be done by a first curve within a first plane followed by a second curve within a second plane that is transversal to the first plane.

Description

FIELD OF INVENTION

This invention relates to computer tomography, and in particular to processes, methods and systems for reconstructing three dimensional images from the data obtained by a circle and arc scan of an object, such as when the C-arm rotates around an object within a first plane, and then the C-arm rotates around the object within another plane that is transversal to the first plane.

BACKGROUND AND PRIOR ART

Over the last thirty years, computer tomography (CT) has gone from image reconstruction based on scanning in a slice-by-slice process to spiral scanning to also include non-spiral scanning techniques such as those performed with C-arm devices, with all techniques and devices experiencing problems with image reconstruction.

From the 1970s to 1980s the slice-by-slice scanning was used. In this mode the incremental motions of the patient on the table through the gantry and the gantry rotations were performed one after another. Since the patient was stationary during the gantry rotations, the trajectory of the x-ray source around the patient was circular. Pre-selected slices through the patient were reconstructed using the data obtained by such circular scans.

From the mid 1980s to present day, spiral type scanning has become the preferred process for data collection in CT. Under spiral scanning a table with the patient continuously moves at a constant speed through the gantry that is continuously rotating about the table. At first, spiral scanning has used one-dimensional detectors, which receive data in one dimension (a single row of detectors). Later, two-dimensional detectors, where multiple rows (two or more rows) of detectors sit next to one another, have been introduced. In CT there have been significant problems for image reconstruction especially for two-dimensional detectors. Data provided by the two-dimensional detectors will be referred to as cone-beam (CB) data or CB projections.

In addition to spiral scans there are non-spiral scans, in which the trajectory of the x-ray source is different from spiral. In medical imaging, non-spiral scans are frequently performed using a C-arm device, which is usually smaller and more portable than spiral type scanning systems. For example, C-arm scanning devices have been useful for being moved in and out of operating rooms, and the like.

FIG. 1 shows a typical prior art arrangement of a patient on a table that moves through a C-arm device, that is capable of rotating around the patient, having an x-ray tube source and a detector array, where cone beam projection data sets are received by the x-ray detector, and an image reconstruction process takes place in a computer with a display for the reconstructed image.

There are known problems with using C-arm devices to reconstruct data. See in particular for example, pages 755-759 of Kudo, Hiroyuki et al., Fast and Stable Cone-Beam Filtered Backprojection Method for Non-planar Orbits, IOP Publishing LTD, 1998, pages 747-760. The Kudo paper describes image reconstruction using C-arm devices for various shift-variant filtered back projection (FBP) structures, which are less efficient than convolution-based FBP algorithms.

For three-dimensional (also known as volumetric) image reconstruction from the data provided by spiral and non-spiral scans with two-dimensional detectors, there are two known groups of algorithms: Exact algorithms and Approximate algorithms, that each have known problems. Under ideal circumstances, exact algorithms can provide a replication of an exact image. Thus, one should expect that exact algorithms would produce images of good quality even under non-ideal (that is, realistic) circumstances.

However, exact algorithms can be known to take many hours to provide an image reconstruction, and can take up great amounts of computer power when being used. These algorithms can require keeping considerable amounts of cone beam projections in memory.

Approximate algorithms possess a filtered back projection (FBP) structure, so they can produce an image very efficiently and using less computing power than Exact algorithms. However, even under the ideal circumstances these algorithms produce an approximate image that may be similar to but still different from the exact image. In particular, Approximate algorithms can create artifacts, which are false features in an image. Under certain circumstances and conditions these artifacts could be quite severe.

To date, there are no known algorithms that can combine the beneficial attributes of Exact and Approximate algorithms into a single algorithm that is capable of replicating an exact image under the ideal circumstances, uses small amounts of computer power, and reconstructs the exact images in an efficient manner (i.e., using the FBP structure) in the cases of complete circle and arc and incomplete circle and arc scanning.

If the C-arm rotates 360 degrees around the patient, this produces a complete circle. If the C-arm rotates less than 360 degrees around the patient, this produces an incomplete circle. In what follows, the word circle covers both complete and incomplete cases. Here and everywhere below by the phrase that the algorithm of the invention reconstructs an exact image we will mean that the algorithm is capable of reconstructing an exact image. Since in real life any data contains noise and other imperfections, no algorithm is capable of reconstructing an exact image.

Image reconstruction has been proposed in many U.S. patents. See for example, U.S. Pat. Nos. 5,663,995 and 5,706,325 and 5,784,481 and 6,014,419 to Hu; U.S. Pat. Nos. 5,881,123 and 5,926,521 and 6,130,930 and 6,233,303 and 6,292,525 to Tam; U.S. Pat. No. 5,960,055 to Samaresekera et al.; U.S. Pat. No. 5,995,580 to Schaller; U.S. Pat. No. 6,009,142 to Sauer; U.S. Pat. No. 6,072,851 to Sivers; U.S. Pat. No. 6,173,032 to Besson; U.S. Pat. No. 6,198,789 to Dafni; U.S. Pat. Nos. 6,215,841 and 6,266,388 to Hsieh. Other U.S. patents have also been proposed for image reconstruction as well. See U.S. Pat. No. 6,504,892 to Ning; U.S. Pat. No. 6,148,056 to Lin; U.S. Pat. No. 5,784,481 to Hu; U.S. Pat. No. 5,706,325 to Hu; U.S. Pat. No. 5,170,439 to Zeng et al.

However, none of the patents overcome all of the deficiencies to image reconstruction referenced above. The inventor is not aware of any known processes, methods and systems that combines the beneficial attributes of Exact and Approximate algorithms into a single algorithm that is capable of replicating an exact image under the ideal circumstances, uses small amounts of computer power, and reconstructs the exact images in an efficient manner (i.e., using the FBP structure) in the cases of complete circle and arc and incomplete circle and arc scanning.

SUMMARY OF THE INVENTION

A primary objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects that have been scanned with two-dimensional detectors.

A secondary objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that is able to reconstruct an exact image and not an approximate image.

A third objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that creates an exact image in an efficient manner using a filtered back projection (FBP) structure.

A fourth objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that creates an exact image with minimal computer power.

A fifth objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that creates an exact image utilizing a convolution-based FBP structure.

A sixth objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that is CB projection driven allowing for the algorithm to work simultaneously with the CB data acquisition.

A seventh objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that does not require storing numerous CB projections in computer memory.

An eighth objective of the invention is to provide improved processes, methods and systems for reconstructing images of objects scanned with a circle and arc x-ray source trajectory that allows for almost real time imaging to occur where images are displayed as soon as a slice measurement is completed.

A preferred embodiment of the invention uses a seven overall step process for reconstructing the image of an object under a circle and arc scan. In a first step a current CB projection is measured. Next, a family of lines is identified on a detector according to a novel algorithm. Next, a computation of derivatives between neighboring projections occurs and is followed by a convolution of the derivatives with a filter along lines from the selected family of lines. Next, using the filtered data, the image is updated by performing back projection. Finally, the preceding steps are repeated for each CB projection until an entire object has been scanned. This embodiment works with keeping several (approximately 2 to approximately 4) CB projections in memory at a time and uses one family of lines.

The invention is not limited to an object that undergoes a scan consisting of a single circle and a single arc. The invention can be applied to trajectories consisting of several circles and arcs by applying it to various circle and arc pairs, and then combining the results.

The circle and arc scanning can include a partial planar curve scan before or after an arc scan. The planar curved scan can be less than a full circle and even greater than a full circle. Additional and subsequent circle and arc scans can be done consecutively after a first circle and arc scan.

Unlike the prior art, the subject invention does not require the patient having to be moved during the measurements. Patients already hooked up to intravenous feeding lines and/or other machines are not easily moveable. Thus, the subject invention can be used as a portable device where the patient does not need to be moved while measurements are taking place.

Further objects and advantages of this invention will be apparent from the following detailed description of the presently preferred embodiments, which are illustrated schematically in the accompanying drawings.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 shows a typical prior art view arrangement of a patient on a table that moves through a C-arm device, that is capable of rotating around the patient, having an x-ray tube source and a detector array, where cone beam projection data sets are received by the x-ray detector, and an image reconstruction process takes place in a computer with a display for the reconstructed image.

FIG. 2 shows an overview of the basic process steps of the invention.

FIG. 3 shows mathematical notations of the circle and arc scan.

FIG. 4 illustrates a stereographic projection from the current source position on to the detector plane used in the algorithm for the invention.

FIG. 5 illustrates finding of a filtering line for a reconstruction point x when the x-ray source is on the circle.

FIG. 6 illustrates finding of a filtering line for a reconstruction point x when the x-ray source is on the arc.

FIG. 7 illustrates a family of lines used in the algorithm of the invention corresponding to the case when the x-ray source is on the circle.

FIG. 8 illustrates a family of lines used in the algorithm of the invention corresponding to the case when the x-ray source is on the arc.

FIG. 9 is a three substep flow chart for identifying the set of lines, which corresponds to step 30 of FIG. 2.

FIG. 10 is a seven substep flow chart for preparation for filtering, which corresponds to step 40 of FIG. 2.

FIG. 11 is a seven substep flow chart for filtering, which corresponds to step 50 of FIG. 2. FIG. 12 is an eight substep flow chart for backprojection, which corresponds to step 60 of FIG. 2.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Before explaining the disclosed embodiments of the present invention in detail it is to be understood that the invention is not limited in its application to the details of the particular arrangements shown since the invention is capable of other embodiments. Also, the terminology used herein is for the purpose of description and not of limitation.

This invention is a Continuation-In-Part of PCT (Patent Cooperation Treaty) Serial No. PCT/US04/12536 filed Apr. 23, 2004, which is a Continuation-In-Part of U.S. patent application Ser. No. 10/728,136, filed Dec. 4, 2003 which claims the benefit of priority to U.S. Provisional Application Ser. No. 60/430,802 filed, Dec. 4, 2002, and is a Continuation-In-Part of U.S. patent application Ser. No. 10/389,534 filed Mar. 14, 2003, now U.S. Pat. No. 6,804,321, which is a Continuation-In-Part of Ser. No. 10/389,090 filed Mar. 14, 2003, now U.S. Pat. No. 6,771,733, which is a Continuation-In-Part of Ser. No. 10/143,160 filed May 10, 2002 now U.S. Pat. No. 6,574,299, which claims the benefit of priority to U.S. Provisional Application 60/312,827 filed Aug. 16, 2001, all of which are incorporated by reference

As previously described, FIG. 1 shows a typical prior art view arrangement of a patient on a table that moves through a C-arm device such as the AXIOM Artis MP, manufactured by Siemens, that is capable of rotating around the patient, having an x-ray tube source and a detector array, where CB projections are received by the x-ray detector, and an image reconstruction process takes place in a computer 4 with a display 6 for displaying the reconstructed image. For the subject invention, the detector array can be a two-dimensional detector array. For example, the array can include two, three or more rows of plural detectors in each row. If three rows are used with each row having ten detectors, then one CB projection set would be thirty individual x-ray detections.

Alternatively, a conventional gantry, such as ones manufactured by Siemens, Toshiba, General Electric, and the like, can be used, as shown by the dotted concentric lines, for the X-ray sources and detector array. The gantry can rotate partially, and up to a full circle, or greater than a full circle.

FIG. 2 shows an overview of the basic process steps of the invention that occur during the image reconstruction process occurring in the computer 4 using a first embodiment. An overview of the invention process will now be described.

A preferred embodiment works with keeping several (approximately 2 to approximately 4) CB(cone beam) projections in computer memory at a time and uses one family of lines.

In the first step 10, a current CB (cone beam) projection set is taken. The next steps 20 and 30 identify sets of lines on a virtual x-ray detector array according to the novel algorithm of the invention, which will be explained later in greater detail. In the given description of the algorithm the detector array can be considered to be flat, so the selected line can be a straight tilted line across the array.

The next step 40 is the preparation for the filtering step, which includes computations of the necessary derivative of the CB projection data for the selected lines.

The next step 50 is the convolution of the computed derivative (the processed CB data) with a filter along lines from the selected family of lines. This step can also be described as shift-invariant filtering of the derivative of the CB projection data. In the next step 60, the image of the object being computed is updated by performing back projection. In the final step 70 it is indicated that the above steps 10-60 are repeated, unless image reconstruction is completed or there are no more CB projections to process.

The invention will now be described in more detail by first describing the main inversion formula followed by the novel algorithm of the invention.

Referring to FIG. 3, we consider a source trajectory consisting of a circle C₁ and an arc C₂ attached to C₁. Let y₀ be the point where they intersect. FIG. 3 shows mathematical notations of the scan. Such a trajectory can be implemented with a C-arm by first rotating the C-arm around the patient in one plane, and then rotating the C-arm within a perpendicular plane.

Suppose the trajectory is described by the following parametric equations 1.C₁:={y ∈ R³:y₁=cos s₁,y₂=sin s₁,y₃=0,0≦s₁≦s₁^mx},C₂:={y ∈ R³:y₁=cos s₂,y₂=0,y₃=sin s₂,0≦s₂≦s₂^mx},   (1)where

• • C₁ denotes the circle,
  • C₂ denotes the arc,
  • s₁,s₂ are real parameters, and
  • y=(y₁,y₂,y₃)=y(s) is a point on the trajectory representing the x-ray source position.

Without loss of generality we assume that both the circle and arc are of radius 1 and centered at the origin. Let ƒ be a function supported inside the cylinder x₁²+x₂²≦r², r<cos s₂^mx<1. The problem is to reconstruct ƒ inside a region of interest U. This set is the union of surfaces U=∪_{s2∈[0,s2mx)}S(s₂), and each surface S(s₂) is formed by line segments connecting the point (cos s₂,0,sin s₂) to all points (cos s₁, sin s₁,0)∈C₁ with 0≦s₁≦s₁^mx and cos s₁<cos s₂ (see FIG. 3).

Next, pick a reconstruction point x=(x₁,x₂,x₃)∈U,x₃≧0. We say that x admits a π-line L_π(x), if x belongs to a line segment one end-point of which is on C₁, y(s₁)∈C₁, and the other—on C₂,y(s₂)∈C₂ (see FIG. 3). It is shown in the manuscript by A. Katsevich “Image reconstruction for the circle-and-arc trajectory”, submitted for publication, that all x=(x₁,x₂,x₃)∈U,x₃≧0, admit unique π-lines. The π-line L_π(x) determines two parametric intervals. The first one I₁(x)⊂I₁ corresponds to the section of C₁ between y₀ and y(s₁)∈C₁. The second one I₂(x)⊂I₂ corresponds to the section of C₂ between y₀ and y(s₂)∈C₂.

We use the following notations in equations 2 and 3 as follows:

$\begin{array}{cc}{D}_f\left(y,\Theta \right):={\int }_0^{\infty }f\left(y+\Theta t\right)dt,\Theta \in S^2;& \left(2\right)\\ \beta \left(s,x\right):=\frac{x-y\left(s\right)}{x-y\left(s\right)},x\in U,s\in I_1\left(x\right)\bigcup I_2\left(x\right).& \left(3\right)\end{array}$ where

• • S² is the unit sphere,
  • ƒ is the function representing the distribution of the x-ray attenuation coefficient inside the object being scanned,
  • Θ is a unit vector,
  • D_ƒ(y,Θ) is the cone beam transform of ƒ,
  • β(s,x) is the unit vector from the focal point y(s) pointing towards the reconstruction point x.

Next, we suppose s∈I₁(x). Project x,C₁,C₂ onto the detector plane DP(s) as shown in FIG. 4. We assume that for each y(s)∈C₁∪C₂, DP(s) contains the origin O and is perpendicular to the line through y(s) and O. The shape of Ĉ₂ on DP(s) depends on s. If 0<s<π, then Ĉ₂ is bent to the right as shown in FIG. 5. If s=π, Ĉ₂ is a line segment. If π<s<s₂^mx, then Ĉ₂ is bent to the left. Here and everywhere below we use the convention that whenever a geometrical object is projected onto the detector plane (e.g., point x, circle C₁, arc C₂, etc.), the corresponding projection is denoted with hat (e.g., {circumflex over (x)}, Ĉ₁, Ĉ₂, etc.). For y(s)∈C₁ define the unit vector by equation 4.

$\begin{array}{cc}{u}_1\left(s,x\right):=\frac{\stackrel{.}{y}\left(s\right)\times \beta \left(s,x\right)}{\stackrel{.}{y}\left(s\right)\times \beta \left(s,x\right)},x\in U,s\in I_1\left(x\right).& \left(4\right)\end{array}$ u₁(s,x) is the unit vector perpendicular to the plane through x, y(s), and tangent to C₁ at y(s). Here {dot over (y)}(s) is the velocity vector of the source at the current position.

We now pick a source position y(s) on the arc y(s)∈C₂,s∈I₂(x). Find a plane through x and y(s), which is tangent to C₁ at some y(s_t(s,x)),s_t(s,x)∈I₁(x). This determines another unit vector by equation 5.

$\begin{array}{cc}{u}_2\left(s,x\right):=\frac{\left(y\left(s_t\left(s,x\right)\right)-y\left(s\right)\right)\times \beta \left(s,x\right)}{\left(y\left(s_t\left(s,x\right)\right)-y\left(s\right)\right)\times \beta \left(s,x\right)},x\in U,s\in I_2\left(x\right).& \left(5\right)\end{array}$ u₂(s,x) is the unit vector perpendicular to the plane through x, y(s), and tangent to C₁ at y(s_t(s,x)).

The detector plane DP(s) corresponding to the source y(s)∈C₂ with various points and lines projected onto it is shown in FIG. 6.

Using equations 4 and 5 we obtain the following reconstruction formula for ƒ∈C₀^∞(U) by equation 6:

$\begin{array}{cc}\begin{array}{c}f\left(x\right)=-\frac{1}{2{\pi }^2}\sum _{k=1}^2{\int }_{l_k\left(x\right)}\frac{{\delta }_k\left(s,x\right)}{x-y\left(s\right)}\\ {\int }_0^{2\pi }\frac{\partial }{\partial q}{D}_f\left(y\left(q\right),{\Theta }_k\left(s,x,\gamma \right)\right){|}_{q=s}\frac{d\gamma }{\sin \gamma }ds,\end{array}& \left(6\right)\end{array}$ where using equation 7Θ_k(s,x,γ):=cos γβ(s,x)+sin γe_k(s,x),e_k(s,x):=β(s,x)×u_k(s,x).   (7)

• • x is the cross-product of two vectors,
  • q is parameter along the trajectory,and δ_k is defined by equation 8 as follows:δ₁(s,x)=1,s∈I₁(x), δ₂(s,x)=−sgn(u₂(s,x)·{dot over (y)}(s)),s∈I₂(x);   (8)

If C₂ is parameterized as in equation 1 (with {dot over (y)}(s) as shown in FIG. 6), then δ₂(s,x)=−1,s∈I₂(x). If C₂ is parameterized in such a way that the source moves down along C₂ as s increases, then δ₂(s,x)=1,s∈I₂(x). In this case the motion of the source along C₁∪C₂ is continuous, so δ_k(s,x) is continuous as well.

We now describe an efficient (that is, of the convolution-based FBP type) implementation of inversion formula equation 6. Pick the source on the circle y(s)∈C₁. Then filtering is performed along the lines on the detector parallel to {dot over (y)}(s). The resulting family, which is one-parametric, is denoted L₁ in FIG. 7. Pick any l∈L₁.

As follows from FIG. 5, all x whose projection belongs to l and appears to the right of Ĉ₂ share l as their filtering line. Suppose now the source is on the arc y(s)∈C₂. For a point x∈U we have to find s_t∈I₁(x) (see FIG. 6). This determines the filtering line on the detector, which is tangent to Ĉ₁ at ŷ(s_t). The set of all such lines is a one-parametric family, which we denote L₂. The range of s_t values, s_t^mn≦s_t≦s_t^mx, depends on the region of interest (ROI) and is illustrated in FIG. 8. Pick s_t∈[s_t^mn,s_t^mx], and let l∈L₂ be the corresponding filtering line. Clearly, all x∈U which project onto l to the left of y(s_t) share l as their filtering line. Hence the proposed algorithm is of the convolution-based FBP type. One can perform filtering along lines on the detector (either L₁ or L₂, depending on the source position), and then perform backprojection.

Let us describe this in more detail. It is clear that s_t(s,x) actually depends only on s and β(s,x). Therefore, we can write below including equations 9 and 10.

$\begin{array}{cc}{u}_k\left(s,\beta \right):=u_k\left(s,x\right),e_k\left(s,\beta \right):=\beta \times u_k\left(s,\beta \right),k=1,2,\beta =\beta \left(s,x\right),& \left(9\right)\\ {\Psi }_k\left(s,\beta \right):={\int }_0^{2\pi }\frac{\partial }{\partial q}{D}_f\left(y\left(q\right),\cos \gamma \beta +\sin \gamma e_k\left(s,\beta \right)\right){|}_{q=s}\frac{1}{\sin \gamma }d\gamma ,& \\ f\left(x\right)=-\frac{1}{2{\pi }^2}\sum _{k=1}^2{\int }_{l_k\left(x\right)}\frac{{\delta }_k\left(s,x\right)}{x-y\left(s\right)}{\Psi }_k\left(s,\beta \left(s,x\right)\right)ds.& \left(10\right)\end{array}$

Any source position y(s) and a filtering line from the corresponding family (either L₁ or L₂, depending on whether y(s)∈C, or y(s)∈C₂), determine a plane. We call it a filtering plane. Since e(s,β)·β=0,|e(s,β)|=1, we can write equations 11 and 12.β=(cos θ,sin θ); e_k(s,β)=(−sin θ,cos θ)   (11)for all β, e_k(s,β) confined to a filtering plane. Here θ denotes polar angle within the filtering plane. Therefore,

$\begin{array}{cc}{\Psi }_k\left(s,\beta \right)={\int }_0^{2\pi }\frac{\partial }{\partial q}{D}_f\left(y\left(q\right),\left(\cos \left(\theta +\gamma \right),\sin \left(\theta +\gamma \right)\right)\right){|}_{q=s}\frac{1}{\sin \gamma }d\gamma & \left(12\right)\end{array}$ for all β confined to a filtering plane.

Equation (12) is of convolution type and one application of Fast Fourier Transform (FFT) gives values of Ψ_k(s,β) for all β confined to the filtering plane at once. Equations (10) and (12) would represent that the resulting algorithm is of the convolution-based FBP type. This means that processing of every CB projection consists of two steps. First, shift-invariant and x-independent filtering along a family of lines on the detector is performed. Second, the result is back-projected to update the image matrix. A property of the back-projection step is that for any point {circumflex over (x)} on the detector the value obtained by filtering at {circumflex over (x)} is used for all points x on the line segment connecting the current source position y(s) with {circumflex over (x)}. Since ∂/∂q in (12) is a local operation, each CB projection is stored in memory as soon as it has been acquired for a short period of time for computing this derivative at a few nearby points and is never used later.

Now we describe the algorithm in detail following the seven steps 10-70 shown in FIG. 2.

Step 10. We load the current CB (cone beam) projection into computer memory. Suppose that the mid point of the CB projections currently stored in memory is y(s₀). The detector plane corresponding to the x-ray source located at y(s₀) is denoted DP(s₀).

Step 20. Here we assume that the x-ray source is located on the circle C₁. Referring to FIG. 7, we form a set of lines parallel to the projection of the circle that covers the projection of the region of interest (ROI) inside the object being scanned.

Step 30. FIG. 9 is a three substep flow chart for identifying the set of lines, which corresponds to step 30 of FIG. 2. Here we assume that the x-ray source is located on the arc C₂. Referring to FIG. 8, the set of lines can be selected by the following substeps 31, 32, and 33.

Step 31. Choose a discrete set of values of the parameter s_t inside the interval [s_t^min,s₁^max].

Step 32. For each s_t chosen in Step 31 find a line tangent to the projected circle Ĉ₁.

Step 33. The collection of lines constructed in Step 32 is the required set of lines (see FIG. 8 which illustrates the family of lines used in the algorithm of the invention).

Step 40. Preparation for Filtering

FIG. 10 is a seven substep flow chart for preparation for filtering, which corresponds to step 40 of FIG. 2, which will now be described.

Step 41. If the x-ray source is located on the circle C₁, fix a filtering line l_ƒlt∈L₁ from the set of lines obtained in Step 20. If the x-ray source is located on the arc C₂, fix a filtering line l_ƒlt∈L₂ from the set of lines obtained in Step 30.

Step 42. Parameterize points on the said line by polar angle γ in the plane through y(s₀) and l_{71 lt}.

Step 43. Choose a discrete set of equidistant values γ_j that will be used later for discrete filtering in Step 50.

Step 44. For each γ_j find the unit vector β_j which points from y(s₀) towards the point on l_ƒlt that corresponds to γ_j.

Step 45. Using the CB projection data D₇₁(y(q),Θ) for a few values of q close to s₀ find numerically the derivative (∂/∂q)D_ƒ(y(q),Θ)|_q=s0 for all Θ=β_j.

Step 46. Store the computed values of the derivative in computer memory.

Step 47. Repeat Steps 41-46 for all lines l_ƒlt. This way we will create the processed CB data Ψ(s₀,β_j) corresponding to the x-ray source located at y(s₀).

Step 50. Filtering

FIG. 11 is a seven substep flow chart for filtering, which corresponds to step 50 of FIG. 2, which will now be described.

Step 51. Fix a filtering line l_ƒlt. If the x-ray source is located on the circle C₁, we take l_ƒt∈L₁. If the x-ray source is located on the arc C₂, we take l_ƒlt∈L₂.

Step 52. Compute FFT (Fast Fourier Transform) of the values of the said processed CB data computed in Step 40 along the said line.

Step 53. Compute FFT of the filter 1/sin γ

Step 54. Multiply FFT of the filter 1/sin γ (the result of Steps 53) and FFT of the values of the said processed CB data (the result of Steps 52).

Step 55. Take the inverse FFT of the result of Step 54.

Step 56. Store the result of Step 55 in computer memory.

Step 57. Repeat Steps 51-56 for all lines in the said family of lines. This will give the filtered CB data Φ(s₀,β_j).

By itself the filtering step can be well known in the field and can be implemented, for example, as shown and described in U.S. Pat. No. 5,881,123 to Tam, which is incorporated by reference.

Step 60. Back-Projection

FIG. 12 is an eight substep flow chart for backprojection, which corresponds to step 60 of FIG. 2, which will now be described.

Step 61. Fix a reconstruction point x, which represents a point inside the patient where it is required to reconstruct the image.

Step 62. If s₀ belongs to I₁(x)∪I₂(x), then the said filtered CB data affects the image at x and one performs Steps 63-68. If s₀ is not inside I₁(x)∪I₂(x), then the said filtered CB data is not used for image reconstruction at x. In this case go back to Step 61 and choose another reconstruction point.

Step 63. Find the projection {circumflex over (x)} of x onto the detector plane DP(s₀) and the unit vector β(s₀,x), which points from y(s₀) towards x.

Step 64. Identify filtering lines l_ƒlt∈L₁ or 1_ƒlt∈L₂ (depending on where the x-ray source is located) and points on the said lines that are close to the said projection {circumflex over (x)}. This will give a few values of Φ(s₀,β_j) for β_j close to β(s₀,x).

Step 65. With interpolation estimate the value of Φ(s₀,β(s₀,x)) from the said values of Φ(s₀,β_j) for β_j close to β(s₀,x).

Step 66. Compute the contribution from the said filtered CB data to the image being reconstructed at the point x by multiplying Φ(s₀,β(s₀,x)) by −δ_k(s₀,x) /(2π²|x−y(s₀)|). The quantity δ_k(s₀,x) is defined by equation 8).

Step 67. Add the said contribution to the image being reconstructed at the point x according to a pre-selected scheme (for example, the Trapezoidal scheme) for approximate evaluation of the integral in equation (10).

Step 68. Go to Step 61 and choose a different reconstruction point x.

Step 70. Go to Step 10 (FIG. 2) and load the next CB projection into computer memory. The image can be displayed at all reconstruction points x for which the image reconstruction process has been completed (that is, all the subsequent CB projections are not needed for reconstructing the image at those points). Discard from the computer memory all the CB projections that are not needed for image reconstruction at points where the image reconstruction process has not completed. The algorithm concludes when the scan is finished or the image reconstruction process has completed at all the required points.

The invention is not limited to an object that undergoes a scan consisting of a single circle and a single arc. The algorithm can be applied to trajectories consisting of several circles and arcs by applying it to various circle and arc pairs, and then combining the results. The algorithm can be applied to trajectories in which a planar curve is not necessarily a circle, but, for example, an ellipse, and the like.

Other Embodiments of the invention are possible. For example, one can integrate by parts in equation (6) as described in the inventor's previous U.S. patent application Ser. No. 10/143,160 filed May 10, 2002 now U.S. Pat. No. 6,574,299, now incorporated by reference, to get an exact convolution-based FBP-type inversion formula which requires keeping only one CB projection in computer memory. The algorithmic implementation of this alternative embodiment can be similar to and include the algorithmic implementation of Embodiment Two in the inventor's previous U.S. patent application Ser. No. 10/143,160 filed May 10, 2002 now U.S. Pat. No. 6,574,299, now incorporated by reference. The corresponding equations will now be described. Introduce the following notations in equation 13:

$\begin{array}{cc}{\Psi }_{1k}\left(s,\beta \right)={\int }_0^{2\pi }{D}_f\left(y\left(s\right),{\Theta }_k\left(s,\beta ,\gamma \right)\right)\frac{d\gamma }{\sin \gamma },{\Psi }_{2k}\left(s,\beta \right)={\int }_0^{2\pi }\left({\nabla }_{u_k\left(s,\beta \right)}{D}_f\right)\left(y\left(s\right),{\Theta }_k\left(s,\beta ,\gamma \right)\right)\cot \left(\gamma \right)d\gamma ,{\Psi }_{3k}\left(s,\beta \right)={\int }_0^{2\pi }\left({\nabla }_{u_k\left(s,\beta \right)}{D}_f\right)\left(y\left(s\right),{\Theta }_k\left(s,\beta ,\gamma \right)\right)d\gamma {\Psi }_{4k}\left(s,\beta \right)={\int }_0^{2\pi }\left(\frac{\partial }{\partial \gamma }{D}_f\left(y\left(s\right),{\Theta }_k\left(s,\beta ,\gamma \right)\right)\right)\frac{d\gamma }{\sin \gamma },k=0,1,2.& \left(13\right)\end{array}$ Here ∇_uD_ƒ denotes the derivative of D_ƒ with respect to the angular variables along the direction u by equations 14 and 15:

$\begin{array}{cc}\left({\nabla }_u{D}_f\right)\left(y\left(s\right),\Theta \right)=\frac{\partial }{\partial t}{D}_f\left(y\left(s\right),\sqrt{1-t^2}\Theta +\mathrm{tu}\right){|}_{t=0},\Theta \in u^{\perp }.& \left(14\right)\end{array}$ Here u^⊥ denotes the set of unit vectors perpendicular to u. Denote also

$\begin{array}{cc}{\mu }_{1k}\left(s,x\right)=\frac{\partial }{\partial s}\frac{1}{x-y\left(s\right)},{\mu }_{2k}\left(s,x\right)=\frac{{\beta }_s\left(s,x\right)·u_k\left(s,x\right)}{x-y\left(s\right)},{\mu }_{3k}\left(s,x\right)=\frac{{\left(e_k\right)}_s\left(s,x\right)·u_k\left(s,x\right)}{x-y\left(s\right)},{\mu }_{4k}\left(s,x\right)=\frac{{\beta }_s\left(s,x\right)·e_k\left(s,x\right)}{x-y\left(s\right)}.& \left(15\right)\end{array}$ Here β_s=∂β/∂s and e_s=∂e/∂s.Let s_bk(x) and s_tk(x) denote the end-points of I_k(x), k=1,2. More precisely, I_k(x)=[s_bk(x),s_tk(x)].

Integrating by parts with respect to s in equation (6) and using that δ_k(s, x) is a constant with respect to s within each I_k(x) (so it can be replaced by δ_k(x)) we get similarly to equation (17) in U.S. Pat. No. 6,574,299 to Katsevich and the same assignee of the subject invention which is incorporated by reference:

$\begin{array}{cc}\begin{array}{c}f\left(x\right)=-\frac{1}{2{\pi }^2}\sum _{k=1}^2{\delta }_k\left(x\right)\{\frac{{\Psi }_{1k}\left(s,\beta \right)}{x-y\left(s\right)}{|}_{s=s_{\mathrm{bk}}\left(x\right)}^{s=s_{\mathrm{tk}}\left(x\right)}-\\ {\int }_{I_k\left(x\right)}\sum _{j=1}^4{\mu }_{\mathrm{jk}}\left(s,x\right){\Psi }_{\mathrm{jk}}\left(s,\beta \right)ds\},\end{array}& \left(16\right)\end{array}$

While the invention has been described with rotating C-arm type devices, the invention can be used with rotating gantry devices.

Furthermore, the amount of rotating can include a single rotational curve of at least approximately 5 degrees up to approximately 360 degrees or greater. Theoretically, there is no limit on the minimum range of rotation. Under realistic practical circumstances, a minimum range of rotation is between approximately 10 and approximately 20 degrees.

The circle and arc scanning of an object can have an arc scanning before or after a single rotational curve scan as defined above.

Subsequent circle and arc scanning can occur as needed for image reconstruction.

The invention can allow for two scanning curves to be used which are transversal to one another. One of the scanning curves can be sized anywhere from an arc up to a full circle, and the other scanning curve can also be sized anywhere from an arc up to a full circle.

In medical applications, the length of one curve is generally shorter than the length of the other curve. The planes of the curves can cross each other at a tranversal angle of approximately 45 to approximately 135 degrees. Alternatively, the transversal angle can be in the range of approximately 80 to approximately 100 degrees. Still furthermore the second curve can be approximately perpendicular to the first curve.

Although the preferred embodiments describe applications of using x-ray sources for creating data for image reconstruction, the invention can be applicable with other sources that create line integral data for image reconstruction, such as but not limited to early arriving photons.

Other computational schemes for evaluating the integrals in equations 9, 10, 12, 13, and 16 are possible. For example, one can choose to perform filtering using native geometries, as explained in the paper by F. Noo, J. Pack, D. Heuscher “Exact helical reconstruction using native cone-beam geometries”, published in the journal “Physics in Medicine and Biology”, 2003, volume 48, pages 3787-3818.

Introduce the coordinates in the detector plane as follows. u is the horizontal coordinate, and w is the vertical coordinate. If the detector is flat, then computation of ƒ using equations 9-12 can be replaced by the following steps.

Step 1. Distance weighting.

$\begin{array}{cc}{g}_1\left(s,u,w\right)=\frac{1}{\sqrt{u^2+w^2+1}}\frac{\partial }{\partial q}{D}_f\left(q,u,w\right){❘}_{q=s}.& \left(17\right)\end{array}$ Step 2. Hilbert transform filtering of g₁ along filtering lines on the detector:

$\begin{array}{cc}{g}_2\left(s,u,\rho \right)={\int }_{-\infty }^{\infty }{K}_H\left(u-u^{\prime }\right)g_1\left(s,u^{\prime },w\left(\rho ,u^{\prime }\right)\right)d{u}^{\prime },& \left(18\right)\end{array}$ where

K_H is the Hilbert transform kernel,

ρ is the variable that parametrizes filtering lines, and

w(ρ,u^t) is the w-coordinate of the point, whose u-coordinate equals u^t, and

which belongs to the filtering line determined by the parameter ρ.

Step 3. Backward height rebinning.g₃(s,u,w)=g₂(s,u,ρ(u,w)),   (19)where

ρ(u,w) is the function that gives the value of the parameter ρ, corresponding to the filtering line that contains all x projected onto the point on the detector with coordinates (u,w).

Step 4. Backprojection. Evaluate the integrals

$\begin{array}{cc}{\int }_{I_k\left(x\right)}\frac{1}{V^{*}\left(s,x\right)}{g}_3\left(s,u^{*}\left(s,x\right),w^{*}\left(s,x\right)\right)ds,k=1,2,& \left(20\right)\end{array}$ where

u*(s,x),w*(s,x) are the coordinates of the point x projected onto the detector plane DP(s),

V*(s,x)=1−(x₁y₁(s)+x₂y₂(s)+x₃y₃(s)) if y(s) is on the arc,

V*(s,x)=1−(x₁y₁(s)+x₂y₂(s)) if y(s) is on the circle.

Step 5. Multiply the integrals in (20) by δ_k(s,x)/(2π),s∈I_k(x), and add them to obtain ƒ(x).

Integrating by parts with respect to s in equation 20 we can obtain another set of formulas, in which the derivative along the source trajectory is avoided. In this case the filtering step will involve computing the following quantities:

$\begin{array}{cc}{G}_1\left(s,u,\rho \right)=\int K_H\left(u-u^{\prime }\right)g_1\left(s,u^{\prime },w\left(\rho ,u^{\prime }\right)\right)d{u}^{\prime },G_2\left(s,u,\rho \right)=\int K_H^{\prime }\left(u-u^{\prime }\right)g_1\left(s,u^{\prime },w\left(\rho ,u^{\prime }\right)\right)d{u}^{\prime },G_3\left(s,\rho \right)=\int g_{1,w}^{\prime }\left(s,u^{\prime },w\left(\rho ,u^{\prime }\right)\right)d{u}^{\prime },& \left(21\right)\end{array}$

where

g^t_1,w is the derivative of g₁ with respect to w.

Also, equations analogous to 17 through 21 can be obtained if the detector is curved, and not flat.

While the invention has been described, disclosed, illustrated and shown in various terms of certain embodiments or modifications which it has presumed in practice, the scope of the invention is not intended to be, nor should it be deemed to be, limited thereby and such other modifications or embodiments as may be suggested by the teachings herein are particularly reserved especially as they fall within the breadth and scope of the claims here appended.

Claims

1. A method of reconstructing images from data provided by at least one detector, comprising the steps of: moving a scanner along a first curve within a plane relative to a stationary object while scanning the object;scanning the object along a second curve within a plane transversal to the plane of the first curve;storing at least one cone beam projection in memory at a time;using at least one family of lines for the step of reconstructing;applying a convolution based shift invariant Filtered Back Projection algorithm to the cone beam projections of the first single curve and the second single curve;back projection updating the image of the scanned object and reconstructing an exact image of the scanned object with the convolution based Filtered Back Projection algorithm for circle and are cone beam computed tomography.

2. The method of claim 1, wherein a single curve includes the step of: rotating a C-arm device about a portion of the object.

3. The method of claim 1, wherein a single curve includes the step of: rotating a gantry about a portion of the object.

4. The method of claim 1, wherein a single carve includes the step of: rotating between approximately 5 degrees up to approximately 360 degrees.

5. The method of claim 4, further conipiising the step of: rotating over approximately 360 degrees about the object.

6. The method of claim 1, wherein the step of reconstructing includes the steps of: storing approximately 2 to approximately 4 cone beam (CB) projections in memory at a time; andusing one family of lines for the step of reconstructing.

7. A method of reconstructing images from two planar curve scans of an object, being called curve scan A and curve scan B, the planes of these two curves being transverse to each other for use with circle and are cone beam computed tomography, comprising the steps of: (a) collecting cone beam (CB) data from a detector during two planar curve scans of the object;(b) identifying lines on a plane Π intersecting the cone beam, wherein the step (b) of identifying lines includes the steps of: (bi) when an x-ray source belongs to the curve scan A, project the curve scan B onto Π and choose a discrete set of lines tangent to that projection; and(bii) when the x-ray source belongs to the curve scan B, project the curve scan B onto Π and choose a discrete set of lines parallel to that projection;(c) preprocessing and shift invariant filtering said data along said lines, wherein the step (c) of preprocessing includes computing the derivative (∂/∂s)Dƒ(y(s), Θ), wherein s is parameter along the scan path, which determines point y(s) on the said path,Dƒ(y,Θ) is the cone beam transform of ƒ corresponding to the x-ray source located at the point y and the direction,ƒ is a function describing the object being scanned;(d) back projecting said filtered data to form a precursor of said image; and(e) repeating steps a, b, c, and d until an exact image of the object is reconstructed.

8. The method of claim 7, wherein shift-invariant filtering in step (c) includes convolving the derivative (∂/∂s)Dƒ(y(s),Θ) with kernel 1/sin(γ) within a filtering plane containing y(s) and a line, identified in step (b) above, where γ is polar angle in the plane.

9. The method of claim 7, wherein a planar curve scan includes: a complete circle about the object.

10. The method of claim 7, wherein a planar curve scan includes: less than complete circle about the object.

11. The method of claim 7, wherein the back-projection step (d) includes the steps of: (di) fixing a reconstruction point x, which represents a point inside the object being scanned, to reconstruct the image;(dii) when s belongs to I(x), then the said filtered CB data affects the image at x and one performs Steps (diii) to (dvii) and when s is not inside the interval I(x), then the said filtered CB data is not used for the image reconstruction at x and go back to step (di) and choose another reconstruction point, wherein I(x) is the parametric interval corresponding to the section of the scan path bounded by the PI-line of x;PI-line of x is the line segment containing x, one endpoint of which belongs to the first planar curve scan, and the other endpoint of which belongs to the second planar curve scan;(diii) find the projection {circumflex over (x)} of x onto a detector plane DP(s) and unit vector β(s,x), which points from y(s) towards x;(div) estimate a value of Φ(s,β(s.x)), where Φ(s,β(s,x)) is the filtered CB data corresponding to the source position located at the point y(s) and direction β(s,x);(dv) determine contribution from filtered CB data to the image being reconstructed at the point x by multiplying Φ(s,β(s,x)) by a weighting factor;(dvi) add the said contribution to the image being reconstructed at the point x according to a pre-selected scheme; and(dvii) go to step (di) and choose a different reconstruction point x.

12. The method of claim 7, further comprising the steps of: storing approximately 2 to approximately 4 cone beam (CB) projections in memory at a time; andusing one family of lines for each x-ray source position for the step of filtering.

13. The method of claim 7, wherein the step (c) of preprocessing includes distance weighting of the derivative (∂/∂s)Dƒ(y(s), Θ).

14. The method of claim 13, wherein shift-invariant filtering in step (c) includes convolving the distance weighted derivative (∂/∂s)Dƒ(y(s), Θ) with a Hubert transform kernel along said lines.

15. The method of claim 13, wherein the step (c) of preprocessing includes distance weighting of the data Dƒ(Y(s),Θ).

16. The method of claim 15, wherein shift-invariant filtering in step (c) includes convolving the distance weighted data Dƒ(Y(s),Θ) with the derivative of a Hilbert transform kernel along said lines.

17. A method of computing images derived from two planar curve scans, being called curve scan A and curve scan B, for circle and are cone beam computed tomography comprising the steps of: (a) collecting cone beam (CB) data from a detector during two planar curve scans of an object;(b) identifying lines on a plane Π intersecting the cone beam, wherein the step (b) of identifying lines includes the steps of: (bi) when an x-ray source belongs to the curve scan A, project the curve scan B onto Π and choose a set of lines tangent to that projection; and(bii) when the x-ray source belongs to the curve scan B, project the curve scan B onto Π and choose a set of lines parallel to that projection;(c) preprocessing and shift invariant filtering said data along said lines, wherein the step (c) of preprocessing includes computing the derivative of Dƒ(y(s),Θ) with respect to Θ along a direction non-parallel to the plane determined by y(s) and a filtering line, the said plane being a filtering plane, wherein s is parameter along the scan path, which determines point y(s) on the said path,Dƒ(Y(s),Θ) is the cone beam transform of ƒ corresponding to the x-ray source located at the point y and the direction Θ,ƒ is a function describing the object being scanned;(d) back projecting said filtered data to fonn a precursor of said image; and(e) repeating steps a, b, c, and d until an exact image of the object is reconstructed.

18. The method of claim 17, wherein shifi-invariant filtering in step (c) includes convolving the data Dƒ(Y(s),Θ) with kernel I/sin(γ) within a filtering plane, where γ is polar angle in the plane.

19. The method of claim 17, wherein shift-invariant filtering in step (c) includes convolving the data Dƒ(Y(s),Θ) with kernel

20. The method of claim 17, wherein shift-invariant filtering in step (c) includes convolving the derivative of Dƒ(Y(s),Θ) with a kernel within a filtering plane, the derivative of Dƒ(Y(s),Θ) is the derivative with respect to Θ along a direction non-parallel to the filtering plane.

21. The method of claim 20, wherein y(s) belongs to the curve scan A.

22. The method of claim 20, wherein y(s) belongs to the curve scan B.

23. The method of claim 17, wherein a planar curve scan includes: a complete circle about the object.

24. The method of claim 17, wherein a planar curve scan includes: less than complete circle about the object.

25. The method of claim 17, wherein the back-projection step (d) includes the steps of: (di) fix a reconstruction point x which represents a point inside the object being scanned, to reconstruct the image;(dii) when s belongs to I(x), then the said filtered CB data affects the image at x and one performs Steps (diii) to (dvii) and when s is not inside the interval I(x), then the said filtered CB data is not used for die image reconstruction at x and go back to step (di) and choose another reconstruction point, wherein I(x) is the parametric interval corresponding to the section of the scan path bounded by the PI-line of x;PI-line of x is the line segment containing x, one endpoint of which belongs to the curve scan A, and the other endpoint of which belongs to the curve scan B;(diii) find the projection {circumflex over (x)} of x onto a detector plane DP(s) and unit vector β(s,x), which points from y(s) towards x; (div) estimate a value of Φ(s,β(s,x)), where Φ(s,β(s,x)) is the filtered CB data corresponding to the source position located at the point y(s) and direction β(s,x);(dv) detenuine contribution from filtered CB data to the image being reconstructed at the point x by multiplying Φ(s,β(s,x)) by a weighting factor;(dvi) add the said contribution to the image being reconstructed at the point x according to a pre-selected scheme; and(dvii) go to step (di) and choose a different reconstruction point x.

26. The method of claim 17, further comprising the steps of: storing 1 cone beam(CB) projection in memory at a time; andusing one family of lines for each x-ray source position for the stop of filtering.

27. A method of reconstructing images from data provided by at least one detector, comprising the stops of: scanning an object with two planar curve scans by at least one detector;storing at least one cone beam projection in memory at a time;using at least one family of lines for the step of reconstructing;applying a convolution based shift invariant Filtered Back Projection algorithm to the two planar curve scans;back projection updating the image of the scanned object and reconstructing an exact image of the scanned object with a convolution based FBP (Filtered Back Projection) algorithm for circle and arc cone beam computed tomography.

28. The method of claim 27, wherein the scanning step includes the step of: scanning by the two planar curves being transversal to one another.

29. The method of claim 27, wherein the scanning step includes the step of: scanning by the object with a first curve followed by scanning the object with another curve being transversal to the first curve.

30. The method of claim 27, further comprising the step of: providing a C-arm device for the scanning of the object.

31. The method of claim 27, further comprising the step of: providing a gantry for the scanning of the object.

32. The method of claim 27, wherein at least one scan curve includes: at least a full circle scan about the object.

33. The method of claim 27, wherein at least one scan curve includes: less than a fall circle scan about the object.

34. The method of claim 27, further comprising the step of: consecutively scanning the object with another planar curve scan.

35. A method of reconstructing images from data provided by a detector, comprising the steps of: first scanning an object with a first curve moving along a first plane relative to the object;second scanning the object with a second curve moving within a second plane that is transversal to the first plane;detecting data from the first scanning and the second scanning steps with at least one detector; andreconstructing an exact image of the scanned object with the detected data with a convolution based FBP (Filtered Black Projection) algorithm for circle and arc cone beam computed tomography.

36. The method of claim 35, wherein the second plane crosses at a transversal angle of approximately 45 to approximately 135 degrees relative to the first plane.

37. The method of claim 35, wherein the second plane crosses at a transversal angle of approximately 80 to approximately 100 degrees relative to the first plane.

38. The method of claim 35, wherein the second plane crosses at a transversal angle of approximately perpendicular to the first plane.

39. The method of claim 35, wherein one of the curves includes a full circle.

40. The method of claim 35, wherein one of the curves includes less than a full circle.

41. The method of claim 35, wherein one of the curves is shorter than the other curve.

42. The method of claim 35, wherein the first scanning step and the second scanning step include the step of: providing one C-arm scanning device to provide both the first and the second scanning steps.