Device, method and computer program product for medical image data annotation

Abstract

The present disclosure relates to an apparatus, method, and computer program product for annotating medical image data. The automated process for annotating medical image data includes: obtaining image data from an imaging sensor; segmenting the image data; determining an object of interest in the segmented image data; generating an initial contour having one or more control points about the object of interest; determining a manual adjustment for a control point; automatically adjusting the position of at least one other control point within a predetermined range of the manually adjusted control point to a new position, the new position of the at least one other control point and the manually adjusted control point defining a new contour; and generating an updated image using the new contour and the corresponding control points.

Description

Technical Field

[0001] Aspects of the disclosed embodiments relate generally to data annotation, and more particularly to automated contour adjustment for data annotation. Background Art

[0002] Data annotation is the process of labeling data for machine learning. Data can be obtained in various formats, such as text, video, or images. In the field of medical imaging, for example, artificial intelligence (AI) and machine learning offer advantages by making it easier to predict outcomes more accurately with greater accuracy and speed. However, in order to create such automated applications or machines, large training datasets are required. For supervised machine learning, labeled datasets are needed to enable the machine to learn input patterns and provide accurate predictions.

[0003] During the data annotation process, the initially given outline (also known as the boundary or edge) of the object of interest may not be satisfactory. This often requires the user to exhaustively adjust the control points to obtain a satisfactory outline annotation. This manual intervention is time-intensive, error-prone, and generally inefficient.

[0004] It would therefore be desirable to provide methods and apparatus that address at least some of the above-mentioned issues. Summary of the Invention

[0005] Aspects of the disclosed embodiments relate to automated control point adjustment in data annotation. This and other advantages of the disclosed embodiments are provided substantially as illustrated in and / or described in conjunction with at least one of the accompanying drawings, and as set forth in the independent claims. Further advantageous modifications may be found in the dependent claims.

[0006] According to a first aspect, disclosed embodiments relate to a device for automated control point adjustment in data annotation. The device includes a processor configured to automatically adjust the positioning of control points, which define the contours of a medical image used for a segmentation mask, based on user input and image segmentation. This reduces human labeling effort compared to conventional methods.

[0007] In a possible implementation, a device is configured to receive medical image data and determine the contours of one or more objects of interest in the medical image data. The device is configured to segment the medical image data in a region at least within an initial contour based on one or more characteristics of the medical image data. The device is further configured to generate one or more control points on the contours of the one or more objects of interest. An adjustment of a position of a first control point is detected. The device is configured to automatically adjust the position of at least one control point adjacent to the first control point based on the detected adjustment of the first control point.

[0008] In a possible implementation form, the original image, the control points and the contour are presented on a display of the user interface.

[0009] In a possible implementation form, at least one adjacent control point is within a predetermined distance from the first control point.

[0010] In a possible implementation form, the at least one adjacent control point includes all control points within a predetermined distance from the first control point.

[0011] In a possible embodiment, the distance and direction of the adjustment of the position of the first control point are determined.

[0012] In a possible embodiment, at least one adjacent control point is moved in the same direction as the movement of the first control point.

[0013] In a possible implementation form, the movement distance of at least one adjacent control point is proportional to the movement distance of the first control point.

[0014] In a possible implementation form, the movement distance of at least one adjacent control point is the same as the movement distance of the first control point.

[0015] In a possible implementation form, at least one adjacent control point is moved in the same direction as the movement of the first control point until a next segmentation line associated with the object of interest is detected.

[0016] In a possible embodiment, after automatically adjusting at least one adjacent control point, the initial contour is automatically updated to a next contour with respect to the object of interest.

[0017] In a possible implementation form, the adjustment of at least one adjacent control point is a pixel-based adjustment.

[0018] In a possible implementation form, the segmentation within the initial contour may be automatically generated based on geometric information of the object of interest.

[0019] In possible implementation forms, the output of the segmentation may be one or more of a coarse segmentation or a fine segmentation.

[0020] In a possible embodiment, the extent to which at least one adjacent control point is moved during the automatic adjustment of the control points and the contour depends on the granularity or fineness of the segmentation.

[0021] In a possible implementation form, control points are generated for the initial contour based on one or more properties of points along the initial contour.

[0022] In a possible implementation form, the number of control points generated for the contour may be adjustable based on annotation requirements or user input.

[0023] In possible implementation forms, the number of automatically adjusted control points may be specified by a user or automatically defined based on one or more of the size and / or shape of the object of interest and characteristics of the initial contour.

[0024] In a possible implementation form, the user can further adjust the adjusted control point.

[0025] In a possible embodiment, the medical image data are two-dimensional image medical image data.

[0026] In a possible embodiment, the medical image data are three-dimensional medical image data.

[0027] In a possible embodiment, the medical image data include data slice images, wherein the slice images are loaded and annotated one by one.

[0028] In a possible embodiment, a segmentation mask is output after the control points have been automatically adjusted.

[0029] The image data may also be one or more of an RGB image, a depth image, a thermal image, or a medical scan image.

[0030] According to a second aspect, disclosed embodiments relate to a method. In one embodiment, the method includes automatically adjusting the positioning of control points based on user input and image segmentation, the control points defining the contours of a medical image used for a segmentation mask. This reduces human labeling effort compared to conventional methods.

[0031] According to a third aspect, the disclosed embodiments relate to a computer program product embodied on a non-transitory computer-readable medium, the computer program product comprising computer instructions which, when executed on at least one processor of a system or device, are configured to perform possible implementation forms described herein.

[0032] According to a fourth aspect, the disclosed embodiments relate to an apparatus comprising means for performing the possible implementation forms described herein.

[0033] These and other aspects, embodiments, and advantages of the exemplary embodiments will become apparent from the embodiments described herein when considered in conjunction with the accompanying drawings. However, it should be understood that the description and drawings are designed for illustrative purposes only and are not intended as a definition of the limits of the disclosed invention, for which reference should be made to the appended claims. Additional aspects and advantages of the invention will be set forth in the following description and, in part, will be apparent from the description, or may be learned by practice of the invention. Furthermore, aspects and advantages of the invention may be realized and obtained by means of the devices and combinations particularly pointed out in the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In the following detailed portion of the present disclosure, the present invention will be described in more detail with reference to example embodiments shown in the accompanying drawings, in which:

[0035] Figure 1 is a block diagram of a device incorporating aspects of the disclosed embodiments.

[0036] Figure 2 Exemplary medical image data to be annotated by a device incorporating aspects of the disclosed embodiments is illustrated.

[0037] Figure 3 Illustrate generated by a device incorporating aspects of the disclosed embodiments Figure 2 Example segmentation masks for medical image data.

[0038] Figure 4A Illustrated Figure 2 Example segmentation of medical image data.

[0039] Figure 4B Illustrated Figure 4A An exploded view of the object of interest.

[0040] Figure 5 Illustrated Figure 4A An example of manual adjustment or movement of control points on segmented medical image data.

[0041] Figure 6 One example of automatic adjustment of adjacent control points according to aspects of the disclosed embodiment is illustrated.

[0042] Figure 7 is a flow chart illustrating an exemplary process flow incorporating aspects of the disclosed embodiments. DETAILED DESCRIPTION

[0043] The following detailed description illustrates exemplary aspects of the disclosed embodiments and how they may be implemented. Although some modes of implementing aspects of the disclosed embodiments have been disclosed, those skilled in the art will recognize that other embodiments for implementing or practicing aspects of the disclosed embodiments are also possible.

[0044] See also Figure 1 , illustrates a schematic block diagram of an exemplary apparatus 100 for automated contour adjustment for data annotation. Aspects of the disclosed embodiments generally relate to automatically adjusting the positions and initial contours of one or more control points based on user input and characteristics of an object or region of interest. This minimizes the need for manual intervention, which improves annotation efficiency.

[0045] In one embodiment, the apparatus 100 may be implemented as a tool in a medical image or medical image annotation apparatus or system 110. The apparatus 100 may be communicatively coupled to an imaging system 110, such as Figure 1 . In an alternative embodiment, the device 100 may be embodied in or be a part of an imaging system 110. Examples of such imaging systems may include, but are not limited to, X-ray imaging systems, medical resonance imaging (MRI) systems, and computed tomography (CT) systems. Although medical imaging systems are generally referred to herein, aspects of the disclosed embodiments are not limited thereto. In an alternative embodiment, aspects of the disclosed embodiments may be implemented in any imaging system in which it is desired to annotate the contours of an object or region of interest.

[0046] As used herein, the term "annotation" generally refers to defining the edges or boundaries of an object or region of interest in an image. Figure 2 , image 200 is a scan of a body organ (in this example, a scan of brain tissue). The object or region of interest 204 in this example is shown as a white or lighter area within or above a background of the overall shape of the organ 202. For the purposes of the description herein, the region or area 204 will be referred to as the "object of interest."

[0047] As an example, given a sequence of CT / MRI scan images of a patient with a tumor, the annotation tool of the disclosed embodiment provides the annotator with the utility of marking the tumor region in the scan image. In one embodiment, the output can be a binary mask of the same size as the scan image 200, where "1" indicates the tumor region and "0" indicates the normal or non-tumor region.

[0048] like Figure 2 As shown, the contour 206 determines the boundaries or edges of the object of interest 204. As is generally understood, determining the boundaries or edges of the object of interest 204 is necessary for accurate evaluation and labeling purposes.

[0049] In some imaging processes, the boundaries or edges of the object of interest 204 will be marked with lines or other suitable markers. For the purposes of the description herein, such markings or definitions of boundaries or edges will be referred to as "outlines 206."

[0050] As will be described further herein, control points or markers can generally be used to determine and annotate the outline 206 of the object of interest 204. When the outline 206 is not accurately marked by the lines or control points, the positions of the control points can be adjusted to more accurately define the outline 206. Aspects of the disclosed embodiments relate to automatic adjustment of the control point positions and definition of the outline 206.

[0051] See again Figure 1In one embodiment, the device or system includes at least one processor 102. The processor 102 is configured to receive image data 108 as input. As generally described herein, the image data 108 is such as Figure 2 The medical image data of the image 200 is shown.

[0052] In one embodiment, image data 108 is received from imaging system 110. Although Figure 1 The processor 102 and device 100 are shown as being external to the imaging system 110 , but aspects of the disclosed embodiments are not limited in this regard. In alternative embodiments, the device 100 and processor 102 may be components of the imaging system 110 .

[0053] In one embodiment, the processor 102 is configured to initially segment the input image data 108 based on a segmentation mask. Figure 3 An example of a segmentation mask is shown in FIG. In this example, the segmented image 300 includes a dark background region 302 and a bright or white region 304. The white region 304 typically contains Figure 2 The object of interest 204 is shown. Figure 4A As further described, the region 304 is segmented and control points are generated. The segmentation process described herein generally includes any suitable image segmentation process.

[0054] Figure 4A An example of segmented input image data 400 is shown in FIG. As can be seen from this example, Figure 2 The segmentation of the input image data 108 typically results in a segmented image 400 having a series of grid-like lines 402. Figure 4A In the example of FIG, the grid lines 402 are non-linear and are generated by a segmentation algorithm. The segmentation algorithm is typically configured based on the example shown as Figure 4A The grid lines 402 are generated based on pixel features of the input image 108 in the scanned image 404 . Figure 4A The white or lighter colored area 406 in the example of FIG is the object of interest. Other considerations when generating the grid lines 402 may include, but are not limited to, the geometry and size of the object of interest 406.

[0055] See also Figure 4B In one embodiment, the initial contour 410 determines the boundary or edge region of the object of interest 406. Figure 4B As shown, one or more control points or markers 412 are generated and used to determine or mark the contour 410 based on the segmentation. The control points 412 are generally configured to provide a visual determination of the contour 410 to the user.

[0056] Figure 4BThe number of control points 412 shown is merely exemplary. In alternative embodiments, the number of control points 412 can be any suitable number. For example, in one embodiment, the number of control points 412 is set by a user.

[0057] In one embodiment, a segmented input image 400 with control points 412 can be presented on a display 106 of the device 100. In one embodiment, the display 106 can be part of a user interface of the device 100 that allows an annotator to interact with and annotate the image 400, as generally described herein. In one embodiment, the device 100 can include an appropriate tool, such as a joystick, stylus, mouse, or other cursor device, that will allow the annotator to reposition the control points 412, as described herein. Aspects of the disclosed embodiments are configured to allow the annotator to click points in the image 400, draw lines on the image 400, and drag or move points and lines on or in the image 400. In one embodiment, the display or user interface 106 includes a touch screen or touch-sensitive device that allows the annotator to interact with the image 400, as generally described herein.

[0058] For example, in one embodiment, the input image data 108 includes CT / MRI scan images. The input image data 108 to be annotated will typically be in the form of a sequence of grayscale images. When a user or annotator begins annotating scanned images, these scanned images are loaded and displayed to the annotator via a computer screen or user interface 104, such as Figure 4A . The user can select an appropriate tool from a utility or toolbox provided by device 100 and use the tool to annotate image 400 as generally described herein. For example, in one embodiment, device 100 can provide a menu from which the annotator can select an appropriate tool or utility to annotate image 400, including grid lines 402, control points 412, and outlines 410.

[0059] The processor 102 is configured to set an initial contour 410 and control points 412 based on the segmentation. The segmentation process uses a suitable algorithm to determine the edges of the object of interest 406. The initial contour 410 and control points 412 are used to provide a visual demarcation of the edges as determined by the segmentation algorithm.

[0060] In one embodiment, you can change Figure 4A The granularity or fineness of the segmentation shown. The closer the grid lines are positioned, the finer the granularity of the segmentation. This granularity can be used to provide more definition to the outline 406. For example, finer granularity in the grid lines can enhance the segmentation algorithm's detection of the edges of the object of interest. In one embodiment, the granularity or fineness of the segmentation can be set or adjusted by the user.

[0061] like Figure 4A and Figure 4B As shown in the example of FIG, the processor 102 is configured to generate an initial outline 410 of the object of interest 406. Figure 4B In the example of FIG4 , an initial outline 410 is determined by one or more control points 412 positioned on or near grid lines 402. Aspects of the disclosed embodiments are configured to distinguish between light and dark areas and determine the edge of the object of interest 406. Thus, control points 412 are arranged on or in conjunction with one or more grid lines that form an edge or are closest to an edge, such as grid lines 402a and 402b.

[0062] The control points 412, also referred to as markers 412, are generally configured to provide defined points along the edge or boundary of the object of interest 406. In some cases, it may be necessary to manually adjust one or more of the control points 412 to more accurately define the edge. Figure 4B For example, one or more control points 412 (such as control points 412a and 412b) may not be accurately positioned relative to an edge or boundary of the object of interest. In this example, control points 412a and 412b are associated with or connected by grid line 402b. However, a more accurate placement or connection for control points 412a and 412b may be grid line 402c. Aspects of the disclosed embodiments are configured to enable an annotator to manually reposition one or more control points 412 and redefine initial outline 410 relative to grid line 402.

[0063] Figure 5 An example of an image 510 is illustrated, showing the manual repositioning of a control point 412. In this example, control point 412a is manually repositioned from initial position 502 to next position 504. This allows control point 412a to be more closely associated with grid line 402c. In this example, grid line 402c may be more accurately associated with the edge of object of interest 406 than grid line 402b. While only one control point 412a is shown as being manually adjusted, aspects of the disclosed embodiments are not limited thereto. In alternative embodiments, any suitable number of control points 412 may be adjusted.

[0064] Figure 6 The automatic adjustment of adjacent or nearby control points and contours according to aspects of the disclosed embodiments is illustrated. Figure 5 As shown, control point 412a is manually adjusted from position 502 to position 504. According to aspects of the disclosed embodiment, the positions of one or more control points adjacent to or within a predetermined distance from the manually adjusted control point are automatically adjusted and updated.

[0065] As used herein, the term "neighboring control point" generally refers to a control point 412 that is within a certain distance or range of a manually adjusted control point. Figure 5 and Figure 6 In the example shown, it is control point 412a. Although the term "adjacent" is used herein, aspects of the disclosed embodiments are intended to apply to any control point within a predetermined range or region of manually adjusted control points. In one embodiment, the predetermined distance, range, or region can be manually set or adjusted by the user / annotator.

[0066] exist Figure 6 In the example of FIG. 6 , region 602 is defined as the region of control points adjacent to manually adjusted control point 412a. In this example, control point 412b is automatically adjusted from Figure 5 Its position 504 is adjusted to Figure 6 The new position 604 is shown. The new position 604 in this example is associated with grid line 402c.

[0067] In one embodiment, the adjustment of control point 412b is based on the extent of the manually adjusted movement of control point 412a. For example, the determination of new position 604 of control point 412b may be proportional to the distance of movement of manually adjusted control point 412a.

[0068] In one embodiment, the movement of control point 412b relative to the manually adjusted movement of control point 412a will be to the next closest grid line. Figure 6 In the example of FIG. 4 , control point 412 b moves from grid line 402 b to grid line 402 c in a direction and distance relative to the adjustment direction and distance of control point 412 a.

[0069] like Figure 6 As shown, in addition to adjusting control point 412b, additional control points 412c and 412d may be generated. Additional control points 412c and 412d are arranged on the grid lines connecting the respective control points to the relocated control point 412a. For example, control point 412c is positioned on grid line 402c connecting control points 412a and 412c.

[0070] The automatic repositioning of adjacent control points results in the generation or definition of an adjusted or new contour line. Figure 6 608. As shown in this example, the manual adjustment relative to control point 412a is Figure 4B The initial contour line 410 is in terms of Figure 6 is modified or changed in area 602. Figure 4B In the example shown in FIG. 4 , the initial contour line 410 is associated with the grid line 402 b. Figure 6 As shown, updated contour line 608 is now associated with grid line 402c.

[0071] Figure 71 is a flow chart illustrating one embodiment of a process incorporating aspects of the disclosed embodiments. In one embodiment, image data is input 702. The image data may include an image sequence. A segmentation mask is generated 704. Segmentation is then applied 706. Control points are generated 706 on an initial outline of an object of interest.

[0072] In one embodiment, a manual adjustment of at least one control point is determined or detected 708. Adjacent or nearby control points are determined, and the positions of the adjacent control points are automatically adjusted 710 relative to the movement of the manually adjusted control point. Updated control points and contours are generated 712 for visualization by a user and / or annotator. This may include, for example, displaying an image with the updated control points and contours.

[0073] In one embodiment, Figure 1 The apparatus 100 shown generally comprises a computing device. The computing device may comprise or include any suitable computer or computing arrangement.

[0074] In one embodiment, processor 102 comprises a hardware processor. Although generally described herein as only one processor 102, aspects of the disclosed embodiments are not limited thereto. In alternative embodiments, device 100 may include any suitable number of processors 102.

[0075] See again Figure 1 , the apparatus 100 generally includes suitable logic, circuitry, interfaces and / or code configured to receive input image data 108 and process the image data 108 as generally described herein. In some embodiments, the processor 102 may be configured to receive a sequence of image frames (e.g., one or more videos) of a patient from an imaging system 110. The imaging system 110 will generally include a suitable image capture device or sensor.

[0076] The processor 102 generally includes appropriate logic, circuitry, interfaces, and / or code configured to process the image input data 108 as generally described herein. The processor 102 is configured to respond to and process instructions that drive the device 100. Examples of the processor 102 include, but are not limited to, a microprocessor, a microcontroller, a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or any other type of processing circuit. Alternatively, the processor 102 may be one or more separate processors, processing devices, and various elements associated with the processing devices that may be shared by other processing devices. In addition, the one or more separate processors, processing devices, and elements may be arranged in various architectures to respond to and process instructions that drive the system 100. The device 100 may include any suitable components or devices, such as memory or storage, required to perform the processes described herein.

[0077] In one embodiment, the apparatus 100 may include or be part of a standalone computing device that is in communication with or is part of the imaging system 110. In one embodiment, the apparatus 100 will include or be connected to the machine learning models required to perform aspects of the disclosed embodiments described herein.

[0078] exist Figure 1 In the example of , device 100 also includes or is communicatively coupled to memory 104. Although not shown, device 100 can be communicatively coupled to a network or network interface to enable communication with components and devices of device 100 and imaging system 110.

[0079] The memory 104 may include suitable logic, circuitry, interfaces, and / or code that may be configured to store instructions executable by the processor 102. The memory 104 is further configured to store image data 108. The memory 104 may also be configured to store an operating system and associated applications for the processor 102. Examples of implementations of the memory 104 may include, but are not limited to, random access memory (RAM), read-only memory (ROM), a hard disk drive (HDD), flash memory, and / or a secure digital (SD) card. The computer-readable storage medium for a computer program product providing non-transitory memory may include, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.

[0080] Aspects of the disclosed embodiments relate to an interactive contour refinement process for efficient data annotation. The position of one or more control points on a contour is automatically adjusted relative to manual adjustment of another control point on the contour. Implementation of aspects of the disclosed embodiments can be in the form of a portal or software installed on a computer that can read / load / store sensor data (e.g., CT / MRI scans), display images, and provide tools to annotators (users) for annotating images. The output can be a binary mask generated from the control points and contour lines.

[0081] The various embodiments and variations disclosed above with respect to the aforementioned system 100 apply mutatis mutandis to this method. The method described herein is computationally efficient and does not place a processing burden on the processor 102.

[0082] Modifications to the embodiments of the disclosed embodiments described above are possible without departing from the scope of the disclosed embodiments as defined by the appended claims. Expressions such as "comprising," "incorporating," "having," and "being" used to describe and claim aspects of the disclosed embodiments are intended to be interpreted in a non-exclusive manner, i.e., to allow for the presence of items, components, or elements not expressly described. References to the singular are also to be interpreted as relating to the plural.

[0083] Thus, although the basic novel features of the present invention as applied to exemplary embodiments of the present invention have been shown, described and pointed out, it will be understood that various omissions, substitutions and changes may be made by those skilled in the art in the form and details of the illustrated apparatus and methods and their operation without departing from the spirit and scope of the presently disclosed invention. Further, it is expressly contemplated that all combinations of those elements which perform substantially the same function in substantially the same manner to achieve the same results are within the scope of the present invention. Moreover, it will be appreciated that the structures and / or elements shown and / or described in conjunction with any disclosed form or embodiment of the present invention may be incorporated into any other disclosed or described or suggested form or embodiment as a general matter of design choice. Accordingly, the present invention is intended to be limited only as indicated by the scope of the appended claims.

Claims

1. A device for annotating medical image data, the device comprising at least one hardware processor configured to: obtaining image data from an imaging sensor; segmenting the image data; determining an object of interest in the segmented image data; generating an initial contour having one or more control points about the object of interest; determining a manual adjustment of said control point; automatically adjusting the position of at least one other control point within a predetermined range of the manually adjusted control point to a new position, wherein the automatic adjustment comprises: repositioning the at least one other control point to a next grid line, wherein the next grid line is closest to a current grid line associated with the at least one other control point and is in a direction of movement of the manually adjusted control point, and the new position of the at least one other control point and the manually adjusted control point define a new contour; and An updated image is generated using the new contour and corresponding control points.

2. The device according to claim 1, wherein The at least one hardware processor is configured to generate a segmentation mask from the obtained image data, the segmentation mask being configured for use in determining a location of the object of interest.

3. The device according to claim 1, wherein The at least one hardware processor is configured to adjust the position of the at least one other control point in the same direction as the direction of movement of the manually adjusted control point.

4. The device according to claim 1, wherein The at least one hardware processor is configured to automatically adjust the position of the at least one other control point by an amount proportional to the movement of the manually adjusted control point.

5. The apparatus according to claim 1, wherein The position of the at least one other control point is adjacent to the position of the manually adjusted control point.

6. The apparatus according to claim 1, wherein The hardware processor is further configured to determine a granularity of fineness of segmentation of the image data and to adjust the position of the at least one other control point based on the determined granularity.

7. The apparatus according to claim 1, wherein The adjustment of the position of the at least one other control point is a pixel-based adjustment, the image data is one or more of an RGB image, a depth image, a thermal image, or a medical scan image, wherein the segmentation of the image data includes generating nonlinear grid lines on the object of interest.

8. A method for annotating medical image data, comprising: obtaining image data from an imaging sensor; segmenting the image data; determining an object of interest in the segmented image data; generating an initial contour having one or more control points about the object of interest; determining a manual adjustment of said control point; automatically adjusting the position of at least one other control point within a predetermined range of the manually adjusted control point to a new position, wherein the automatic adjustment comprises: repositioning the at least one other control point to a next grid line, wherein the next grid line is closest to a current grid line associated with the at least one other control point and is in a direction of movement of the manually adjusted control point, and the new position of the at least one other control point and the manually adjusted control point define a new contour; and An updated image is generated using the new contour and corresponding control points.

9. A computer program product comprising a non-transitory computer-readable medium having machine-readable instructions stored thereon, the machine-readable instructions, when executed by a computer, causing the computer to perform the method according to claim 8.

Images