WO2009101577A2

WO2009101577A2 - Interactive selection of a region of interest and segmentation of image data

Info

Publication number: WO2009101577A2
Application number: PCT/IB2009/050544
Authority: WO
Inventors: Roland Opfer; Rafael Wiemker; Thomas Blaffert
Original assignee: Koninklijke Philips Electronics N.V.; Philips Intellectual Property & Standards Gmbh
Priority date: 2008-02-15
Filing date: 2009-02-10
Publication date: 2009-08-20
Also published as: WO2009101577A3

Abstract

The invention relates to a system (100) for interactive definition of a region of interest in an image data space, the system (100) comprising a point unit (110) for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs and a boundary unit (120) for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary unit further comprises a domain unit (122) for determining a domain space for a parameterization of the boundary, a projection unit (124) for projecting each point of the plurality of points onto the domain space and an approximation unit (126) for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map. Only points necessary for defining the ROI need to be selected. For a simple-shape structure of interest, or for a structure of interest which is at a fair distance from other non- interesting structures, the number of points for defining a ROI comprising said structure of interest can be quite low. For a complex-shape structure of interest, a sufficient number of points can be selected to define a ROI that comprises said structure of interest but does not comprise, for example, a view-occluding structure. The intensities of voxels comprised in the structure of interest do not affect the definition of the ROI, because the ROI is defined on the basis of the selected plurality of points and is not affected by said intensities.

Description

Interactive selection of a region of interest and segmentation of image data

FIELD OF THE INVENTION

The invention relates to interactive selection of a region of interest and segmentation of image data.

BACKGROUND OF THE INVENTION

In many medical image analysis applications there is a need to define a region of interest (ROI). In the case of volume rendering, for instance, a user often wants to exclude parts of the image which occlude a view of a structure of interest such as the heart. In the case of tumor volumetry, the user may wish to quickly obtain a rough volume estimate of a breast nodule, for example. Most medical image analysis applications provide tools for an interactive definition of a ROI. For example, WO 2007/107907 entitled "Systems and methods for interactive definition of regions and volumes of interest" describes a method based on thresholding. In an embodiment of the invention described in WO 2007/107907, the user may select a seed point and define a threshold, and the system is arranged to perform region growing. The threshold is lower than image intensities typical for said structure of interest. Thus, the region obtained via region growing is smaller than the region representing the structure of interest. Nevertheless, the grown region is assumed to describe the shape of said structure. Hence, in the next step the user may adjust a scaling factor to expand the region of interest to include the structure of interest. A drawback of this method lies in that it is difficult to use it when the structure of interest shows a wide range of intensities.

Other methods of interactive ROI definition include positioning of predefined objects like spheres or boxes, or extracting 3D objects using 2D contours delineated in 2D slices. An embodiment of the latter method is described in US2001/0033283. However, the first technique often results in a region that is too large, while the latter one is quite tedious.

SUMMARY OF THE INVENTION

It would be advantageous to provide an alternative solution for ROI definition that is suitable for structures of interest that exhibit a wide range of intensities and shapes and are easy to use and relatively fast. Thus, in an aspect of the invention, a system for interactive definition of a region of interest in an image data space is provided, the system comprising: a point unit for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and - a boundary unit for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary unit further comprises: a domain unit for determining a domain space for a parameterization of the boundary, - a projection unit for projecting each point of the plurality of points onto the domain space, and an approximation unit for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

The condition may be that the distance between each point of the plurality of points and a point obtained by mapping the projection of said point, using the computed map, into the image data space, also referred to in the professional literature as the image data volume, is less than a threshold, for example. Hence, the method allows defining a ROI in a flexible and intuitive way. Only points necessary for defining the ROI need to be selected.

For a simple-shape structure of interest, or for a structure of interest which is at a fair distance from other non-interesting structures, the number of points for defining a ROI comprising said structure of interest can be quite low. For a complex-shape structure of interest, a sufficient number of points can be selected to define a ROI that comprises said structure of interest but does not comprise, for example, a view-occluding structure. The intensities of voxels comprised in the structure of interest do not affect the definition of the ROI, because the ROI is defined on the basis of the selected plurality of points and is not affected by said intensities.

In an embodiment of the invention, the domain space is a sphere. The projection of each point of the plurality of points may be defined by the crossing of the sphere and a ray extending from the center of the sphere towards said each point.

In an embodiment of the invention, the condition is that the composition of said projection and said map maps each point of the plurality of points into said each point. In other words, the map is a map interpolating the plurality of points. In an embodiment of the invention, the domain space is a sphere and said map is a linear combination of radial basis functions, and the approximation unit is arranged to compute a coefficient of each radial basis function from the set of radial basis functions. This approach is computationally very effective because in this approach the interpolation problem becomes a simple linear algebra problem.

In an embodiment of the invention, the system further comprises a segmentation unit for delineating a structure of interest described by a portion of image data comprised in the region of interest, the segmentation unit comprising: a box unit for defining a box comprising the portion of the image data, - a watershed unit for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, and a portion unit for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest.

The definition of the ROI can be seen as a high-level, low-resolution segmentation of the image data, while the decomposition of the image into meaningful parts may be seen as a low-level, high-resolution segmentation of the image data. Combining the results of the low- and high-level segmentation of the image data allows for a more accurate delineation of the whole structure of interest. Advantageously, the method is useful for delineating irregular structures of interest such as tumors.

In a further aspect of the invention, the system according to the invention is comprised in an image acquisition apparatus.

In a further aspect of the invention, the system according to the invention is comprised in a workstation.

In a further aspect of the invention, a method of interactive definition of a region of interest in an image data space is provided, the method comprising: a point step for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and - boundary steps for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary steps further comprise: a domain step for determining a domain space for a parameterization of the boundary, a projection step for projecting each point of the plurality of points onto the domain space, and an approximation step for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

In an embodiment of the invention, the method further comprises segmentation steps for delineating a structure of interest described by a portion of image data comprised in the region of interest, the segmentation steps comprising: - a box step for defining a box comprising the portion of the image data, a watershed step for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, a portion step for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest.

In a further aspect of the invention, a computer program product to be loaded by a computer arrangement is provided, the computer program product comprising instructions for interactive definition of a region of interest in an image data space, the computer arrangement comprising a processing unit and a memory, the computer program product, after being loaded, providing said processing unit with the capability to carry out the tasks of: selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and - determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein determining the boundary further comprises: determining a domain space for a parameterization of the boundary, projecting each point of the plurality of points onto the domain space, and - computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

In an embodiment of the invention, the computer program product further comprises instructions for delineating a structure of interest described by a portion of image data comprised in the region of interest, thereby providing the processing unit with the capability to carry out the further tasks of: defining a box comprising the portion of the image data, computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest. It will be appreciated by those skilled in the art that two or more of the above- mentioned embodiments, implementations, and/or aspects of the invention may be combined in any way deemed useful.

Modifications and variations of the image acquisition apparatus, of the workstation, of the method, and/or of the computer program product, which correspond to the described modifications and variations of the system, can be carried out by a person skilled in the art on the basis of the present description.

A person skilled in the art will appreciate that the method may be applied to multidimensional image data, e.g., to 3-dimensional (3-D) or 4-dimensional (4-D) images, acquired by various acquisition modalities such as, but not limited to, standard X-ray Imaging, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound (US), Positron Emission Tomography (PET), Single Photon Emission Computed Tomography (SPECT), and Nuclear Medicine (NM).

BRIEF DESCRIPTION OF THE DRAWINGS These and other aspects of the invention will become apparent from and will be elucidated with respect to the implementations and embodiments described hereinafter and with reference to the accompanying drawings, wherein:

Fig. 1 schematically shows a block diagram of an exemplary embodiment of the system, Fig. 2 shows two exemplary planar views of a ROI defined using an embodiment of the system of the invention,

Fig. 3 illustrates segmentation of a tumor using the region of interest definition according to the invention and watershed decomposition, Fig. 4A shows a flowchart of a first exemplary implementation of the method according to the invention,

Fig. 4B shows a flowchart of a second exemplary implementation of the method according to the invention, Fig. 5 schematically shows an exemplary embodiment of the image acquisition apparatus, and

Fig. 6 schematically shows an exemplary embodiment of the workstation. Identical reference numerals are used to denote similar parts throughout the Figures.

DETAILED DESCRIPTION OF EMBODIMENTS

Fig. 1 schematically shows a block diagram of an exemplary embodiment of the system 100 for interactive definition of a region of interest in an image data space, the system 100 comprising: - a point unit 110 for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and a boundary unit 120 for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary unit further comprises: - a domain unit 122 for determining a domain space for a parameterization of the boundary, a projection unit 124 for projecting each point of the plurality of points onto the domain space, and an approximation unit 126 for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

The exemplary embodiment of the system 100 further comprises an optional segmentation unit 130 for delineating a structure of interest described by a portion of image data comprised in the region of interest, said segmentation unit 130 comprising: a box unit 132 for defining a box comprising the portion of the image data, a watershed unit 134 for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, and a portion unit 136 for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest. The exemplary embodiment of the system 100 further comprises the following optional units: a control unit 160 for controlling the workflow in the system 100, a user interface 165 for communicating with a user of the system 100, and a memory unit 170 for storing data. In an embodiment of the system 100, there are three input connectors 181, 182 and 183 for the incoming data. The first input connector 181 is arranged to receive data coming in from a data storage means such as, but not limited to, a hard disk, a magnetic tape, a flash memory, or an optical disk. The second input connector 182 is arranged to receive data coming in from a user input device such as, but not limited to, a mouse or a touch screen. The third input connector 183 is arranged to receive data coming in from a user input device such as a keyboard. The input connectors 181, 182 and 183 are connected to an input control unit 180.

In an embodiment of the system 100, there are two output connectors 191 and 192 for the outgoing data. The first output connector 191 is arranged to output the data to a data storage means such as a hard disk, a magnetic tape, a flash memory, or an optical disk. The second output connector 192 is arranged to output the data to a display device. The output connectors 191 and 192 receive the respective data via an output control unit 190.

A person skilled in the art will understand that there are many ways to connect input devices to the input connectors 181, 182 and 183 and the output devices to the output connectors 191 and 192 of the system 100. These ways comprise, but are not limited to, a wired and a wireless connection, a digital network such as, but not limited to, a Local Area Network (LAN) and a Wide Area Network (WAN), the Internet, a digital telephone network, and an analog telephone network.

In an embodiment of the system 100, the system 100 comprises a memory unit 170. The system 100 is arranged to receive input data from external devices via any of the input connectors 181, 182, and 183 and to store the received input data in the memory unit 170. Loading the input data into the memory unit 170 allows quick access to relevant data portions by the units of the system 100. The input data may comprise, for example, the image data. The memory unit 170 may be implemented by devices such as, but not limited to, a Random Access Memory (RAM) chip, a Read Only Memory (ROM) chip, and/or a hard disk drive and a hard disk. The memory unit 170 may be further arranged to store the output data. The output data may comprise, for example, the ROI definition, i.e. the map for mapping the domain space into the image data space and onto the boundary of the ROI. In an embodiment of the system 100, the output data may further comprise the portion of the image data describing the structure of interest. The memory unit 170 may be also arranged to receive data from and/or deliver data to the units of the system 100 comprising the point unit 110, the boundary unit 120, the domain unit 122, the projection unit 124, the approximation unit 126, the segmentation unit 130, the box unit 132, the watershed unit 134, the portion unit 136, the control unit 160, and the user interface 165, via a memory bus 175. The memory unit 170 is further arranged to make the output data available to external devices via any of the output connectors 191 and 192. Storing data from the units of the system 100 in the memory unit 170 may advantageously improve performance of the units of the system 100 as well as the rate of transfer of the output data from the units of the system 100 to external devices. Alternatively, the system 100 may comprise no memory unit 170 and no memory bus 175. The input data used by the system 100 may be supplied by at least one external device, such as an external memory or a processor, connected to the units of the system 100. Similarly, the output data produced by the system 100 may be supplied to at least one external device, such as an external memory or a processor, connected to the units of the system 100. The units of the system 100 may be arranged to receive the data from each other via internal connections or via a data bus.

In an embodiment, the system 100 comprises a control unit 160 for controlling the workflow in the system 100. The control unit may be arranged to receive control data from and provide control data to the units of the system 100. For example, after selecting all points for defining the boundary of the ROI, the point unit 110 may be arranged to provide control data "the points are selected" to the control unit 160 and the control unit 160 may be arranged to provide control data "determine the boundary of the ROI" to the boundary unit 120. Alternatively, a control function may be implemented in another unit of the system 100, e.g., the point unit 110 may be arranged to communicate directly with the boundary unit 120. In an embodiment, the system 100 comprises a user interface 165 for communicating with the user of the system 100. The user interface 165 may be arranged to receive user inputs for selecting the plurality of points for defining the ROI boundary. The user interface may be also arranged to receive further user inputs, e.g. inputs for defining the box comprising the portion of image data describing a structure of interest to the user. The user interface may further provide means for presenting a view of the ROI boundary computed by the system. Optionally, the user interface may receive a user input for determining a domain space for parametrizing the boundary surface, e.g., the center of a unit sphere. A person skilled in the art will understand that more functions may be advantageously implemented in the user interface 165 of the system 100.

The point unit 110 is adapted for selecting a plurality of points (X₁ , X₂, • • ., X_N} , N> 1, for defining a boundary of the region of interest on the basis of user inputs. The user inputs may be obtained from a user input device such as a mouse or a trackball. Any number of points requested by the user may be selected. The user interface 165 may be adapted for assisting the user in navigating through the volumetric data, e.g., through a stack of image data slices.

The boundary unit 120 is adapted for determining the boundary B on the basis of the plurality of points, thereby defining the region of interest. The boundary B is defined by a parametric function. A parametric function, i.e. map, is a function of a parametrizing variable and the values of the parametrizing function are points of the parametrized boundary. Parametrizing functions are typically used for describing curves and surfaces. The boundary B of a ROI is substantially a closed surface bounding the ROI.

The domain unit 122 is arranged for determining a domain space D for a parameterization of the boundary B. The domain space D is typically given by a surface topologically equivalent to (e.g., isomorphic, homeomorphic or diffeomorphic with) the boundary. In an embodiment of the system 100, the domain space D is a unit sphere. The center x# of the sphere may be based on a user input. Alternatively, the center x# of the sphere D may be computed by the domain unit 122 as the center of mass of the plurality of points, for example. In an embodiment of the system 100, each point X₁ has the same mass. The projection unit 124 is adapted for projecting each point X₁ of the plurality of points {x_{l s} X₂, ..., X_N) onto the domain space D. For example, the projection of each point of the plurality of points onto a sphere may by defined by the point where the ray cast from the center of the sphere towards said each point crosses the surface of the sphere. This construction assumes that each point of the plurality of points defines a different ray. If this is not the case, the projection unit may be arranged to give the user a warning. The user may change the domain space, for example, by moving the center of the sphere to another location.

The approximation unit 126 is adapted for computing a map/for mapping the domain space D into the image data space S, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection pr and said map/satisfies a condition for defining the map/ The map/maps a point x of the domain space D into a mapped point flx) (i.e., a value of the map) in the image data space S. The plurality of mapped points {βx) : x e D) defines the boundary of the ROI. In order for the defined boundary to be close to the boundary envisaged by the user, i.e., the boundary defined on the basis of the plurality of points {xi, X₂, • • ., X_N} , the projection Pr(X₁) of each point X₁ of the plurality of points {xi, X₂, ..., X_N} must be mapped by the map/near said each point X₁. Thus, the condition to be satisfied by the map may be, for example, that the distance Wf(Pr(X₁)) - X₁W is less than a threshold δ, or that the sum of all distances \\f(pr(xι)) -

+ Wf(p^r(^χ2)) ^~ + ■ ■ ■ + Wf(p^r(^χN)) -

is less than a threshold M⁾.

In an embodiment of the system 100, the condition to be satisfied by the map/ is that/is an interpolation of the plurality of points {xi, X₂, ... , X_N} , i.e., \\f(pr(x\)) - xi|| + |J/(/?r(x₂)) - x₂|| + ... + \\f{pr(x_N)) - x_N\\ = 0, i.e.,/(/?r(x,)) = x, for each x, of {x_u x₂, ..., x_N}.

In an embodiment of the system 100, the domain space D is a sphere and said map/is a linear combination of radial basis functions. The approximation unit 126 is arranged to compute a coefficient C₁ of each radial basis function from the set of radial basis functions. For example, for a radial basis function r, the map/may be defined for all points x of the unit sphere D as

/(X) = I l + r(|| x -x₇ II) L.

The coefficients CC₇ are solutions to the system of linear equations CL₁ = || X₁ - x_D Il -\ , i = 1, 2, ... N,

where

A_g = r(\\ x_l - x_J \\) .

Radial basis functions are described, for example, in R. Schaback and H. Wendland, "Kernel techniques: From machine learning to meshless methods", Acta Numerica, 15:543-639, 2006. A person skilled in the art will understand that there exist other interpolation or, more generally, approximation methods suitable for defining the map for mapping the domain space into the image data space. The scope of the claims should not be construed limited by the described exemplary embodiments of the invention. Fig. 2 shows two exemplary planar views of a ROI defined using an embodiment of the system 100 of the invention. Both views show points of the plurality of points selected based on user inputs and a contour of the boundary of the ROI defined by the plurality of points. The first view 21 is an axial view and the second view 22 is a coronal view.

Delineation of tumors is an important clinical task in medical practice. It is needed for instance for tumor volumetry in therapy response assessment or for target dose definition and computation in radiotherapy. However, manual tumor delineation typically is a tedious and time-consuming task. Since tumors can have a very different shape or contrast to the background, it is very difficult to design a completely automatic segmentation algorithm workable for all tumor types. To address this problem, in an embodiment, the system 100 further comprises a segmentation unit 130 for delineating a structure of interest described by a portion of image data comprised in the region of interest. In an embodiment of the system 100, the segmentation unit 130 comprises a box unit 132, a watershed unit 134 and a portion unit 136.

The box unit 132 of the segmentation unit 130 is adapted for defining a box comprising the portion of the image data. In an embodiment of the system 100, a rectangular box is defined based on user inputs.

The watershed unit 134 is adapted for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm. Watershed algorithms are described in an article by Jos B. T. M. Roerdink and Arnold Meijster, "The Watershed Transform: Definitions, Algorithms and Parallelization Strategies", Fundamenta Informatica 41 (2001) 187-228. In an embodiment of the system 100, the watershed algorithm is applied to the magnitude of the gradient. The term "meaningful part" means a part that is visually meaningful. For example, a meaningful part may be defined as a connected set of voxels comprising voxels having values within a predefined range of intensities and surrounded by voxels with values outside said predefined range of intensities. The portion unit 136 is adapted for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest, i.e., if said part and the ROI have common data. The data element depends on the way of describing the data in the image data set. Typically, a data element is a voxel.

Fig. 3 illustrates segmentation of a tumor using the region of interest definition according to the invention and watershed decomposition. Each column comprises different planar views of the tumor. The fist column 31 shows axial views, the second column 32 shows coronal views and the third column 33 shows sagittal views of the tumor. Images in the first row 34 show a user-defined rectangular box comprising a portion of image data describing the structure of interest. Images in the second row 35 illustrate the results of the watershed decomposition based on the image gradient magnitude. Images in the third row 36 illustrate the region of interest defined by the system 100 on the basis of the plurality of points selected based on user inputs. The fourth row 37 shows the results of the tumor segmentation according to the invention.

A person skilled in the art will appreciate that the system 100 may be a valuable tool for assisting a physician in many aspects of her/his job. Those skilled in the art will further understand that other embodiments of the system 100 are also possible. It is possible, among other things, to redefine the units of the system and to redistribute their functions. Although the described embodiments apply to medical images, other applications of the system, not related to medical applications, are also possible. The units of the system 100 may be implemented using a processor. Normally, their functions are performed under the control of a software program product. During execution, the software program product is normally loaded into a memory, like a RAM, and executed from there. The program may be loaded from a background memory, such as a ROM, hard disk, or magnetic and/or optical storage, or may be loaded via a network like the Internet. Optionally, an application-specific integrated circuit may provide the described functionality.

Fig. 4A shows a flowchart of a first exemplary implementation 400A of the method of interactive definition of a ROI in an image data space. The method 400A begins with a point step 410 for selecting a plurality of points for defining a boundary of the ROI on the basis of user inputs. After the point step 410, the method 400A continues with boundary steps 420 for determining the boundary on the basis of the plurality of points, thereby defining the ROI. The boundary steps 420 comprise a domain step 422 for determining a domain space for a parameterization of the boundary, a projection step 424 for projecting each point of the plurality of points onto the domain space and an approximation step 426 for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the ROI, such that the composition of said projection and said map satisfies a condition for defining the map. After the boundary steps 420, the method 400A continues with an optional decision step 428 A. If the user is satisfied with the defined ROI, the method 400A terminates. If the user wants to improve/correct the ROI definition, the method 400A returns to the point step 410. In the point step 410, more points may be selected and added to the plurality of points. Optionally, the point step 410 may be adapted for removing a previously selected point of the plurality of points based on user inputs. Fig. 4B shows a flowchart of a second exemplary implementation 400B of the method of interactive definition of a ROI in an image data space. The method 400B begins with a box step 432 for defining a box comprising the portion of the image data. After the box step 432, the method 400B continues to a watershed step 434 for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm. After the watershed step 434, the method 400B continues to the point step 410 for selecting a plurality of points for defining a boundary of the ROI on the basis of user inputs. After the point step 410, the method 400B continues with the boundary steps 420 for determining the boundary on the basis of the plurality of points, thereby defining the ROI. The boundary steps 420 comprise the domain step 422, the projection step 424 and the approximation step 426, and are identical to the boundary steps 420 of the first exemplary implementation 400A of the method. After the boundary steps 420, the method continues with a portion step 436 for determining the portion of the image data based on the defined ROI and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined ROI. After the portion step 436, the method 400B continues to an optional decision step 428B. If the user is satisfied with the segmentation result, the method 400B terminates. If the user wants to improve/correct the segmentation result, the interactively defined ROI may be improved/corrected. To this end, the method 400B returns to the point step 410. In the point step 410, more points may be selected and added to the plurality of points. Optionally, the point step 410 may be adapted for removing a previously selected point of the plurality of points based on user inputs.

A person skilled in the art may change the order of some steps or perform some steps concurrently using threading models, multi-processor systems or multiple processes without departing from the concept as intended by the present invention. Optionally, two or more steps of the method of the current invention may be combined into one step. Optionally, a step of the method of the current invention may be split into a plurality of steps.

Fig. 5 schematically shows an exemplary embodiment of the image acquisition apparatus 500 employing the system 100, said image acquisition apparatus 500 comprising a CT image acquisition unit 510 connected via an internal connection with the system 100, an input connector 501, and an output connector 502. This arrangement advantageously increases the capabilities of the image acquisition apparatus 500, providing said image acquisition apparatus 500 with advantageous capabilities of the system 100. Fig. 6 schematically shows an exemplary embodiment of the workstation 600.

The workstation comprises a system bus 601. A processor 610, a memory 620, a disk input/output (I/O) adapter 630, and a user interface (UI) 640 are operatively connected to the system bus 601. A disk storage device 631 is operatively coupled to the disk I/O adapter 630. A keyboard 641, a mouse 642, and a display 643 are operatively coupled to the UI 640. The system 100 of the invention, implemented as a computer program, is stored in the disk storage device 631. The workstation 600 is arranged to load the program and input data into memory 620 and execute the program on the processor 610. The user can input information to the workstation 600, using the keyboard 641 and/or the mouse 642. The workstation is arranged to output information to the display device 643 and/or to the disk 631. A person skilled in the art will understand that there are numerous other embodiments of the workstation 600 known in the art and that the present embodiment serves the purpose of illustrating the invention and must not be interpreted as limiting the invention to this particular embodiment.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim or in the description. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements and by means of a programmed computer. In the system claims enumerating several units, several of these units can be embodied by one and the same item of hardware or software. The usage of the words first, second, third, etc., does not indicate any ordering. These words are to be interpreted as names.

Claims

CLAIMS:

1. A system (100) for interactive definition of a region of interest in an image data space, the system (100) comprising: a point unit (110) for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and - a boundary unit (120) for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary unit further comprises: a domain unit (122) for determining a domain space for a parameterization of the boundary, - a projection unit (124) for projecting each point of the plurality of points onto the domain space, and an approximation unit (126) for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

2. A system (100) as claimed in claim 1, wherein the domain space is a sphere.

3. A system (100) as claimed in claim 1, wherein the condition is that the composition of said projection and said map maps each point of the plurality of points into said each point.

4. A system (100) as claimed in claim 3, wherein the domain space is a sphere and said map is a linear combination of radial basis functions, and wherein the approximation unit (126) is arranged to compute a coefficient of each radial basis function from the set of radial basis functions.

5. A system (100) as claimed in claim 1, further comprising a segmentation unit (130) for delineating a structure of interest described by a portion of image data comprised in the region of interest, the segmentation unit (130) comprising: a box unit (132) for defining a box comprising the portion of the image data, a watershed unit (134) for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, and - a portion unit (136) for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest.

6. An image acquisition apparatus (500) comprising a system (100) as claimed in claim 1.

7. A workstation (600) comprising a system (100) as claimed in claim 1.

8. A method (400A, 400B) of interactive definition of a region of interest in an image data space, the method (400) comprising: a point step (410) for selecting a plurality of points for defining a boundary of the region of interest on the basis of user inputs, and boundary steps (420) for determining the boundary on the basis of the plurality of points, thereby defining the region of interest, wherein the boundary steps further comprise: a domain step (422) for determining a domain space for a parameterization of the boundary, a projection step (424) for projecting each point of the plurality of points onto the domain space, and an approximation step (426) for computing a map for mapping the domain space into the image data space, wherein values of the map are points defining the boundary of the region of interest, such that the composition of said projection and said map satisfies a condition for defining the map.

9. A method (400B) as claimed in claim 1, further comprising segmentation steps

(430) for delineating a structure of interest described by a portion of image data comprised in the region of interest, the segmentation steps comprising: a box step (432) for defining a box comprising the portion of the image data, a watershed step (434) for computing a decomposition of the image data comprised in the box into meaningful parts using a watershed algorithm, a portion step (436) for determining the portion of the image data based on the defined region of interest and the decomposition of the image data, wherein the portion of the image data comprises every part of the decomposition such that at least one data element of said part is also a data element of the defined region of interest.

10. A computer program product to be loaded by a computer arrangement, comprising instructions for carrying out the steps of a method as claimed in claim 8 or 9.