Image processing

Abstract

The invention is concerned with converting a bitmap image obtained for example by X-ray, Ultrasound or CT scanning of a human or animal body part to a vector line image for the purpose of detecting an abnormality. According to a first aspect of the present invention, there is provided a computer implemented method for processing a bitmap image of an object, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity value. A user input is received at a user interface of a computing device and a number of threshold groups is defined with each threshold group having an upper and a lower intensity threshold. Each pixel of the bitmap image is then assigned to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within the upper and lower intensity thresholds of the threshold group to which said pixel is assigned. A path is then generated at each boundary between pixels assigned to different threshold groups to generate a vector line image from which areas of interest can be identified. The organ of interest may be the brain and the vector line image may be displayed superimposed on a functional MRI image of the brain.

Description

Technical Field The present invention relates to image processing. Background Images may be created using either a raster graphics or vector elements. Raster images are comprised of a matrix or grid of individual pixels of colour. 3-dimensional (3D) raster images are made up of voxels, which are 3D pixel units. Raster images may be represented by a bitmap. The raster is defined by the number of pixels and the dimensions of the image. Digital images are most commonly stored in raster graphics formats, including Graphic Interchange Format (GIF), Joint Photographic Experts Group (JPEG), and Portable Network Graphics (PNG) images. Each pixel is assigned one or more pixel values, depending on whether the image is greyscale or colour. Pixels of colour images are represented by three pixel values, indicating a red intensity, a green intensity, and a blue intensity to be rendered by the pixel on a display. Greyscale images are comprised of pixels defined by a single pixel intensity. The clarity of the image is dependent on the parameters defining the raster, i.e. the dimensions of the image and the number of pixels. For example, a small image with a large number of pixels, i.e. a high pixel density, is clearer than a larger image with the same or lower number of pixels, i.e. a lower pixel density. When a raster image is enlarged, individual pixels may become visible, resulting in a blurry image. Vector images are images comprising paths, the paths being defined by a set of 2-dimensional (2D) points which are connected by lines and curves to form polygons and other shapes. The paths may have various properties, such as colour, shape, thickness, and fill. Vector graphics are commonly found in Scalable Vector Graphics (SVG), Encapsulated PostScript (EPS), Portable Document Format (PDF), or Adobe Illustrator Artwork (AI) graphic file formats. One advantage of vector images over raster images is that the images can be infinitely scaled without any loss to image quality. That is, there is no blurring effect when vector images are enlarged. However, modem displays and printing apparatus are devices that output in raster form. Therefore, vector graphics have to be converted to raster format before they can be displayed or printed. Due to the way in which the vector image is defined, the resolution requirements can be met by generating a bitmap file of a certain size. Rasterization is the process of converting a vector image into a raster image, whereby a scene geometry defined by the vector image is mapped to pixels. In some rasterization techniques, prior to rasterization, individual polygons defined in the vector image are broken down into triangles. Rasterization rules are established which are used to guarantee that no holes are left in images and no pixel is rasterized more than once. An example rule is referred to as the top-left rule, which states that a pixel is rasterized if and only if its centre lies completely inside a triangle or exactly on a triangle edge. Scan imaging techniques can be used to observe components which are not visible from the exterior of an object. For example, scan imaging techniques can be used to generate digital images such as X-rays, magnetic resonance imaging (MRI) and ultrasound, and are used in industries such as medicine and security. The digital images are formed of pixels, or voxels in the case of 3D images, and may be represented by a bitmap. Images defined by a bitmap may be blurry such that sufficient clear information cannot be derived from them. Summary Images generated from scan data are used to identify areas of interest of an obj ect which cannot necessarily be seen from outside the object. An area of interest may be an abnormality in human or animal tissue, or a dangerous or prohibited object within a container, for example. The generated images are raster images. They, therefore, have the disadvantage of appearing blurry when enlarged. Raster images also require more time and power for computer systems to interpret compared to vector images due to the way in which the information is stored and displayed. The present disclosure provides a method for converting raster or analogue images generated from scan data into vector images, allowing information gathered by the scan to be better and more easily interpreted by both humans and computer systems. According to a first aspect of the present invention, there is provided a computer implemented method for processing a bitmap image of an object, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity value; the method comprising: receiving at a user interface of a computer device a user input; defining a number of threshold groups, each threshold group having an upper and a lower intensity threshold; assigning each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within the upper and lower intensity thresholds of the threshold group to which said pixel is assigned; and generating a path at each boundary between pixels assigned to different threshold groups to generate a vector line image. The vector line image may be rendered on one or more displays. For each pixel a modified intensity value may be generated wherein each pixel assigned to the same threshold group is assigned the same modified intensity value to form a threshold image. At least one of the bitmap image and a threshold image may be rendered on a display with the vector line image. The display of the bitmap image and / or the threshold image may be inhibited such that only the vector line image is displayed. The modified intensity value generated for the pixels of the same threshold group may fall within the upper and lower intensity threshold of said threshold group. Each threshold group may have an intensity range of equal size. The bitmap image may be generated from scan data of a scanned object. The scan data may correspond to a three-dimensional scan model. The scan model may be generated as a result of an MRI scan of the scanned object. The scanned object may be scanned using one of one X-ray, CT scan and ultrasound, the bitmap image being generated from an output of the scan. The scanned object may be one or more parts of a human or animal body. The scanned object may be a brain. The scan data may be video scan data, the video scan data comprising two or more frames, wherein a bitmap image is generated for each frame of the video scan data. A vector line video may be generated comprising the vector line images generated from the bitmap images corresponding to each frame of the video scan data. The generated bitmap image may be a greyscale bitmap image. An area of interest of the scanned object may be identified by the computing device using the vector line image. The area of interest may be an abnormality in the brain. The vector line image of the brain may be rendered on one or more displays superimposed with a functional MRI scan of a brain. According to a second aspect of the present invention, there is provided a computer program comprising computer executable instructions configured, when executed, to implement any of the method of any preceding claim. According to a third aspect of the present invention, there is provided a computer device comprising: a memory for storing the computer program according to claim 20; one or more processors for executing the computer program; and a display for rendering the vector line image to a user of the computer device. According to a fourth aspect of the present invention, there is provided a computer apparatus for detecting an abnormality in a human or animal body part, the apparatus comprising: a scanning system for scanning the human or animal body part to generate scan data from which a bitmap image is generated, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity; and an image processing system configured to: use a number of threshold groups, each threshold group having an upper and a lower intensity threshold; assign each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within an upper and lower intensity thresholds of the threshold group to which the pixel is assigned; render on a display a path at each boundary between pixels assign to different threshold groups to generate on the display a vector line image in which the abnormality is observable. The scanning system may be one of an X-ray, CT scan, ultrasound and MRI scanning apparatus. According to a fifth aspect of the present invention, there is provided a method for detecting an abnormality in a human or animal body part, the method comprising: scanning the human or animal body part to generate scan data from which a bitmap image is generated, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity; defining a number of threshold groups, each threshold group having an upper and a lower intensity threshold; assigning each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within an upper and lower intensity thresholds of the threshold group to which the pixel is assigned; and rendering on a display a path at each boundary between pixels assign to different threshold groups to generate on the display a vector line image in which the abnormality is observable. Brief Description of the Drawings For a better understanding of the present invention and to show how the same may be carried into effect reference will now be made by way of example to the following drawings, in which: Figure lisa flow diagram of an example method for generating a vector line image; Figures 2A-2E are examples of images generated at each step of the method of Figure 1; Figure 3 is an example of a vector line image with an enlarged section; Figure 4 shows two vector line images generated from the same scan image, each vector line image having a different number of threshold groups; Figures 5 A-5E illustrate possible medical uses for the present invention; Figure 6 is an example of a vector line image superimposed on a functional MRI (fMRI) scan; Figure 7 shows an example security use for the present invention; Figure 8 is an example MRI scan of the brain with an enlarged section; and Figure 9 is a schematic diagram of a computer system for implementing the present invention. Detailed Description Images generated from an output of a scan are stored in raster form. The present disclosure provides a method of generating a vector image from a bitmap image generated from scan data. Vector images have many advantages over raster images. In the context of scans, where the purpose of a scan is to discover information about the scanned object which is not observable from the exterior of the object, vector images provide a clearer and more easily interpretable image from which information can be extracted. Figure 8 shows an example MRI scan image of a brain with a section enlarged. It can be seen that the enlarged section appears blurry and therefore appears to be of lower quality. The enlarged section is therefore not useful for seeing an area more clearly or in greater detail. In the present disclosure, the method is exemplified using imaging of the brain, the brain scan being taken by an MRI machine. It will be appreciated that other objects may be imaged and other imaging techniques used, while applying the invention described herein. Some alternatives are described below. An MRI scan of the brain is comprised of “voxels”, which are 3D units of graphics information that define a point in 3D space similar to pixels units of 2D images. The volume of a voxel may be as small as one tenth of a cubic millimetre. The MRI scan provides a single value for each voxel that corresponds to a hydrogen density of the area of the brain represented by the voxel. This value can be mapped onto a greyscale or colour scale to generate a greyscale or colour 3D model of the brain respectively. Hydrogen density of tissues depends on the amount of fat and water present in the tissue, so can be used to identify abnormalities, such as tumours, and also differentiate between different regions of the brain. Figure 1 shows an example method for generating a vector line image. Figures 2A-2E show example images generated at different stages of the process described in Figure 1. At step SI00, scan data is generated. For example, an MRI of a patient’s brain is taken using an MRI scanner. The MRI scanner generates a stream of data in the form of a matrix of values, which may be stored in the form of a Neuroimaging Informatics Technology Initiative (NUTT) file. The NlfTI file contains information that allows a 3D scan model to be generated and viewed using software such as MRIcron. At step 102, a bitmap image is generated from the NlfTI data. Since a bitmap image is 2D, a plane of the 3D scan model is first chosen from which the bitmap image is generated. Techniques for generating bitmap images from a NlfTI file are known. The resultant bitmap image corresponds to a “slice” of the 3D model generated from the MRI scan data. Figure 2A shows raster images generated from an MRI scan of the brain. As discussed above, the MRI scan provides data to generate a 3D model 200. Slices 202a, 202b, 202c are shown of the sagittal plane, coronal plane, and horizontal plane respectively. These slices 202a, 202b, 202c are 2D bitmap images. The bitmap image generated is a traditional raster image. That is, it is defined by its dimensions and the number of pixels it comprises. The bitmap image may be stored as a PNG file, for example, or any other raster file type. It will be appreciated that, in some cases, the 2D bitmap image may be generated directly from the scan data. In such cases, the 3D scan model need not be generated. In some cases, for example when an X-ray is taken, the scan data may comprise only 2D image information, such that a 3D scan model cannot be generated. At step S104, the bitmap image is used to generate a “threshold image”. Thresholding is a type of image processing known in the art. In the simplest thresholding methods, each pixel of an image is replaced with either a black pixel if the pixel intensity of said pixel is less than a predefined threshold intensity, or a white pixel if the pixel intensity of said pixel is greater than the predefined threshold intensity. The pixels are referred to herein as being divided into “threshold groups”. In this case there are two threshold groups, one for pixels with an intensity below the threshold and one for pixels with an intensity above the threshold. The resultant threshold image is, therefore, a black and white image. Figure 2B shows an example of a black and white threshold image 204 generated from the bitmap image of the coronal plane slice 202b. A size of the resultant vector line image may be defined at step SI04. It will be appreciated that the vector line image can be resized once it has been generated without compromising the quality of the image. Parameters associated with the threshold image may be adjusted. For example, a different number of threshold groups may be defined. A higher number of threshold groups results in a larger number of different shades of grey being used to define the threshold image, and therefore allows for a higher level of detail to be shown. The resultant image is referred to as a greyscale threshold image. In the case of more than two threshold groups being defined, the threshold groups are defined by an upper intensity threshold and a lower intensity threshold, such that each pixel is assigned to the threshold group between who’s upper and lower intensity thresholds the pixel threshold falls. The intensity thresholds of the threshold groups correspond to the type of bitmap image generated. For example, in the case of a greyscale bitmap image, the intensity thresholds are defined as values of the greyscale. However, for a colour bitmap image, the intensity thresholds are defined by the colour scale used for the bitmap image. The threshold group to which the pixel is assigned determines the shade of grey the pixel displays in the greyscale threshold image. The shade of grey, or pixel intensity, assigned to the pixels of the threshold image is referred to as a modified intensity value. For example, the number of threshold groups may be set to 10. That is, 10 different shades of grey are used to represent the image (including white and black). The pixels of the image are assigned as being a member of one threshold group, each group comprising the pixels falling within the threshold limits. The threshold groups may each be associated with intensity ranges of the same size. For example, with 10 threshold groups, each group comprises a different 10% of the possible intensity range, i.e. 1-10%, 11-20%, 21-30%, etc. of the possible intensity values. Most commonly, greyscale intensity values range from 0 (black) to 255 (white). The pixels which fall within the same threshold group are shown as the same shade of grey, i.e. they have the same modified intensity value. In some embodiments, the modified intensity value is the midway intensity of the threshold group. It will be appreciated that the pixels of each group may be displayed as some other intensity value falling within the threshold limits, for example the upper or lower threshold intensity. Figure 2C shows two examples of possible threshold numbers with the corresponding greyscale threshold image 206a, 206b of the coronal plane slice 202b. The greyscale threshold image 206a has 34 greyscale threshold groups. The 34 different shades of grey are shown in greyscale threshold slide 208a. The greyscale threshold image 206b, however, only has 5 greyscale threshold groups, as represented by the greyscale threshold slide 208b. It can be seen that the greyscale threshold image 206a with the larger number of threshold groups shows more detail than the greyscale threshold image 206b with fewer threshold groups. This is because a larger number of threshold groups allows for smaller differences between pixel intensity, and so amount of hydrogen, to be shown. That is, each change in shade is the result of a smaller change in amount of hydrogen in the brain tissue. In some embodiments, the black and white threshold image 204 is generated prior to defining the number of threshold groups, such that the greyscale threshold image 206 is generated after the black and white threshold image 204. At step SI06, the vector line image is generated. Paths are generated between neighbouring pixels of different threshold groups. That is, the generated paths represent boundaries between pixels assigned different shades of grey such that the pixels contained within the same paths fall within the same threshold group. Image tracing, also referred to as raster-to-vector conversion or vectorisation, is the conversion of raster graphics into vector graphics and is known in the art. This may be done manually, where, for example, a person writes an output file by hand from looking at an image, or automatically using a computer program. Such computer programs have mechanisms by which users can adjust conversion settings. In the method described herein, the step of generating the vector line image is implemented by a computer program. The number of threshold groups defined at step S104 also defines the level of detail possible in the resultant vector image. Figure 2D is an example of a vector line image overlaying a greyscale threshold image of the sagittal slice 202a. The paths show the boundaries between the different greyscale threshold groups. An optional step SI08 is shown in Figure 1. At step SI08, the greyscale threshold image is removed such that only the vector line image remains. Methods for removing or hiding one or more layers of an image are known in the art and may be implemented in the process of Figure 1 to remove the greyscale threshold image. Figure 2E shows the vector image 210a of the sagittal plane 202a with the greyscale image removed. In some areas, the paths appear to be much thicker than in other areas. This is not the case. The paths are shown with equal thickness. This effect is the result of multiple paths being very close together. That is, the gradient of greyscale in that area is large such that there are multiple threshold boundaries close together, resulting in multiple paths being drawn close together. This represents a large change in hydrogen content over a small distance within the brain tissue. The paths of the vector line image 210 are synonymous to, for example, contour lines on a topographical map. In topographical maps, each contour line represents an increase or decrease in elevation by a predefined amount. Contour lines close together correspond to steep areas of ground, i.e. areas with a large gradient, whereas contour lines which are far apart correspond to gently sloping areas, i.e. areas with a small gradient. In vector line images 210, if the threshold groups are defined such that the intensity range falling within each threshold group is equal as exemplified above, each path represents an increase or decease in hydrogen by a predefined amount. Figure 3 shows a vector line image 210b of the coronal plane, with a section enlarged. The enlarged section shows more clearly that the appearance of thicker paths in the vector line image 210 is the result of paths rendered close together. As discussed above, vector images do not lose quality as they are enlarged. That is, because of the way the images are defined, the vector line image 210 can be enlarged without blurring, allowing more intricate features to be visualised more clearly, as shown by the enlarged section of Figure 3. Figure 4 shows the effect of the number of threshold groups on the vector line image 210. The left-hand vector line image 210c is formed from a greyscale threshold image 206 with 20 threshold groups, that is the threshold values are set at every 5% of the maximum possible intensity value. The right-hand vector line image 210d shows the same sagittal slice 202a as the left-hand vector line image 210c, but with only 5 threshold groups, i.e. thresholds sets every 20% of the maximum possible intensity value. It can be seen from the images in Figure 4 that a higher number of threshold groups, i.e. thresholds at smaller intervals, shows a higher level of detail in the image. In some instances, it may not always be beneficial to use a large number of threshold groups. For example, a lower number of threshold groups may be sufficient for identifying an abnormality within a large area. It may be easier to spot an abnormality if the whole image is displayed. A higher level of detail may make abnormalities less obvious when an image of a large area is displayed on a small display. Once the abnormality has been identified, a higher level of detail, meaning more threshold groups, is desirable to inspect the abnormality and surrounding tissue in more detail, and most likely at an enlarged size. It will be appreciated that the level of detail required, and correspondingly the number of threshold groups used, may vary depending on the use of the present technique. Some examples of where the present technique may be used are presented below. Referring to Figure 1, it will be appreciated that the process shown need not be started again in order to generate an image with more or less threshold groups. A first vector line image 210 can be generated with a first number of threshold groups by completing steps SI 00 to SI 08. If a second vector line image 210 with a second number of threshold groups is then required, the process of Figure 1 need only start at step SI 04, where the second number of threshold groups is defined, such that a second greyscale threshold image 206, corresponding to the second number of threshold groups, is generated from the same bitmap image as used to generate the first vector line image 210. The threshold groups are defined such that they cover the full range of possible intensities. That is, a pixel of the bitmap image can have any possible intensity and will always be able to be assigned to one of the threshold groups. The threshold images described above are greyscale images. However, it will be appreciated that colour threshold images may be generated instead. Colour threshold images are generated in a similar way, the difference being that the modified pixel intensity is a pixel value of a colour scale, where different colours or colour intensities correspond to the different threshold groups. The paths are generated at the boundaries between the different threshold groups, as described above. Figures 2A-2E have been provided in order to illustrate the different steps of the method of Figure 1. It will be appreciated that the images may not be rendered at each step of the process. In some embodiments, for example, once the plane of the MRI image has been chosen, the method steps may be performed without rendering any images until the vector line image 210 is generated and rendered. In embodiments in which the threshold image is not rendered visible, the pixels of the bitmap image are assigned to threshold groups, but they need not be assigned an intensity associated with the threshold group. The paths are drawn at the boundaries of the threshold groups in the same way as described with reference to step SI06. In such an embodiment, the vector line image may be rendered superimposed with the scan model or bitmap image. Alternatively, the vector line image may be rendered independently of any other image without the latter image having to be removed, i.e. no step S108 is required. It will also be appreciated that, although the scan data and bitmap image must be generated, one or both may be generated prior to deciding to generate a vector line image such that there is a significant time gap between generating the scan image and the vector line image. For example, the scan data may be generated and used to generate the scan model or bitmap image, which may be used without processing if possible. If not clear enough, then the vector line image is generated. That is, steps SI 00 and / or SI 02 may not be a direct step of the process for generating the vector line image. The scan data and / or bitmap image may be generated by a different computer system to that for generating the vector scan image. That is, the scan data or bitmap image may be an input to a computer system for implementing the method of generating the vector line images. As discussed above, the technique disclosed herein may be used within the field of medicine for the analysis of the brain. It may also be used for processing scan data of other parts of the human or animal body. While in the examples above, the scan has been an MRI scan, any imaging technique may be used which results in an image comprising pixels of different pixel intensities. For example, X-ray or CT scan, ultrasound scan, or radioactive tracer imaging may be used to generate the scan image of step SI 00 of Figure 1. Figures 5A-5E show some example medical uses. Figure 5 A shows an MRI scan in raster image form of the horizontal plane of the brain on the left-hand side and the corresponding vector line image on the right-hand side. There is a tumour 500 in the scanned brain. The tumour 500 is visible in the MRI image as a white area of the scan. In the vector line image, the tumour is an empty space due to the tumour having a constant hydrogen density and so no change in pixel intensity in the greyscale threshold image. It can be seen from the two images in Figure 5A that the MRI scan does not show clearly the boundary between the tumour 500 and the healthy brain tissue. This would not be improved by enlarging the image, as discussed above. However, the boundary is clear in the vector line image as there is a path drawn between the tumour 500 and the healthy tissue. Figure 5B shows an MRI scan (left) and corresponding vector line image (right) of the horizontal plane of an adult brain from which an entire hemisphere was removed during childhood (the right-hand hemisphere in the images of Figure 5B). The area from which brain tissue has been removed, now filled with cerebrospinal fluid, is shown in a constant grey in the MRI image and as a path-free space in the vector line image. Figure 5C shows an MRI scan (left) and corresponding vector line image (right) of the horizontal plane of a brain which has cerebral oedema. Cerebral oedema is when fluid builds up in or around the brain and is indicative of intracranial brain swelling. The build up of fluid 502 is shown in the MRI scan as the white region and the vector line image as the area with no paths. The vector line image shows more clearly which areas are filled with fluid. Figure 5D shows a mammogram for detecting breast cancer. A mammogram is an X-ray of the breast. The mammogram bitmap image is shown on the left-hand side and the corresponding vector line image on the right-hand side. Figure 5E shows an ultrasound (top) and the corresponding vector line image (bottom) of a foetus. The method described above can be applied to both still images and video. In order to convert a video, for example an ultrasound video of a foetus, into a vector line video, the method of converting scan data to a vector line image as set our with reference to Figure 1 is applied to each frame of the video scan data, where each frame corresponds to a separate bitmap image. The vector line images generated from the bitmap images of each frame can be used to generate a vector line video, where each frame of the vector line video is one of the vector line images. The vector line images of the vector line video are ordered to match the order of the frames of the scan data video. The present image processing technique may also be used for medical research purposes. Figure 6 shows a functional MRI (fMRI) scan 600. This differs from a conventional MRI scan in that it shows areas of activity in the brain by detecting changes associated with blood flow. When an area of the brain is in use, the blood flow to that area increases. fMRI images indicate areas of use by displaying different areas of the brain in different colours. In order to identify more easily the areas of the brain associated with the activity, a vector line image of the same plane of the brain is superimposed with the fMRI scan image 600, as shown in the combined fMRI-vector line image 602. The vector line image may be generated from a conventional MRI of the same brain, or a generic brain vector line image may be used. The generic brain vector line image may be generated from a conventional MRI of a different brain. Alternatively, the generic brain vector line image may be based on a combination of the structure of multiple neurotypical brains. In such an embodiment, the generic brain vector image may be generated by combining the vector line images of the multiple neurotypical brains, or by combining the scan data or bitmap images of the multiple neurotypical brains to generated a single bitmap image from which the generic brain vector image is generated. As can be seen in Figure 6, the superimposed vector line image shows the structure of the brain, allowing for the areas of activity to be more easily identified. This can be used in medical research for understanding the functioning of the brain. It may also be used in medical diagnosis or detecting functional abnormalities. The above described medical uses of the present technology may be used to identify abnormalities in the human or animal body. It will be appreciated that an abnormality represents a deviation from the norm or expected structure and / or behaviour of the body part. Although the identification of an abnormality may be used to diagnose a patient, either by a human or a computer diagnostic program, the identification of the abnormality itself is not a diagnosis. The present invention may also be used in computing. Vector line images could be used to assist automated artificial intelligence (AI) analysis of scan images, for example for detecting and highlighting brain abnormalities. The vector line images make detecting and highlighting abnormalities automatically more accurate and efficient as less time and power is required to scan a vector line image than a traditional MRI scan by the automated system. The present invention may be used for detection of objects which are not part of the human or animal body. Figure 7 shows an X-ray image (left) and corresponding vector line image (right) of the contents of a bag. Such an X-ray scan may be taken for security reasons, for example at an airport. The vector line image could assist automated AI analysis in detecting and highlighting objects more accurately and efficiently since less time and power is required to process the vector line image than X-ray, or other scan, images. The method described herein may be implemented in a computer system. A user of the computer system is able to define parameters of the vector line image 210 via at least one input device of the computer system. Alternatively or additionally, some or all of the parameters may be predefined for a given scan technique, scan area, and / or predicted area of interest. For example, the number of threshold groups for a scan of a brain for which it is predicted that there is a tumour, the parameters may be predefined and automatically applied. The user may be able to alter these via the input device. One example of a parameter is the number of threshold groups, as discussed in detail above. Another is the size of the vector line image. The size of the vector line image is only used for initially rending the vector line image to the user. As discussed, the vector line image can be enlarged without losing quality. Other parameters may also be selected by the user. For example, the user may select a line thickness for the paths of the vector line image. The user may also be able to alter the colour of the paths. In the examples given herein, the lines are shown as white against a black background. The user can select any combination of path and background colour in which the vector line image is to be rendered. The method disclosed above may be implemented in a computer system. Figure 9 shows a schematic block diagram of such a computer system. The computer system comprises a computer device 1 and a scan device 3. The computer device 1 comprises a memory 5 and one or more processing units 7. It may also comprise an I / O port 8, and / or a network interface 9. The memory 5 stores one or more computer programs 6 which, when implemented by the processing unit(s) 7, processes the bitmap image to generate the vector line image 210. The computer device 1 may also comprise one or more input devices, such as a mouse or keyboard, via which a user of the computer device 1 can choose the parameters such as the threshold group number. The computer device 1 may also comprise one or more displays 4 for rending at least the vector line image 210 to the user. It will be appreciated that the input device 2 and display 4 may be comprised within a single device, such as a touch screen. The computer device 1 may be connected, via the I / O port 8, to one or more scan devices 3. Examples of scan devices 3 which may be connected to the computer device 1 are MRI scanner, X-ray scanners, CT scanners and ultrasound scanners. In some embodiments, the scan device 3 may comprise a processing unit which is responsible for generating the scan model or bitmap image from the scan data. In other embodiments, the scan data collected by the scan device 3 is passed to the computer device 1 where it is processed by the processing unit 7 to generate the bitmap image. The scan device 3 may also be referred to as a scanning system. In some embodiments, the scan devices 3 may communicate with the computer device 1 via a network, such as the Internet, via the network interface 9. In some embodiments, the computer device 1 is not in direct communication with the scan device 3. Instead, the scan data, scan model, or bitmap image is supplied to the computer device 1 by the user. The scan data, scan model, or bitmap image may be downloaded from the internet via the network interface 9, for example, or uploaded to the computer device 1 from an external memory device. There has been described herein a technique for processing scan data or a scan model to generate a vector line graphic of the scanned object which is useful for identifying areas of interest in the scanned object. While particular embodiments have been described, other applications and variants of the disclosed techniques may become apparent to a person skilled in the art once given the disclosure herein. The scope is not limited by the described embodiments but only by the following claims.

Claims

1. A computer implemented method for processing a bitmap image of an object, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity value; the method comprising:receiving at a user interface of a computer device a user input;defining a number of threshold groups by defining an upper and a lower intensity threshold for each threshold group;assigning each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within the upper and lower intensity thresholds of the threshold group to which said pixel is assigned; andgenerating a path at each boundary between pixels assigned to different threshold groups to generate a vector line image.

2. The method according to claim 1, further comprising the step of rendering the vector line image on one or more displays.

3. The method according to claim 1 or claim 2, comprising generating for each pixel a modified intensity value wherein each pixel assigned to the same threshold group is assigned the same modified intensity value to form a threshold image.

4. The method according to claim 1 or claim 2, wherein at least one of the bitmap image and a threshold image are rendered on a display with the vector line image.

5. The method according to claim 2 and claim 3, wherein the method further comprises the step of inhibiting the display of the bitmap image and / or the threshold image such that only the vector line image is displayed.

6. The method according to claim 2, wherein the modified intensity value generated for the pixels of the same threshold group falls within the upper and lower intensity threshold of said threshold group.

7. The method according to any preceding claim, wherein each threshold group has an intensity range of equal size.28 08 248. The method according to any preceding claim, wherein the bitmap image is generated from scan data of a scanned object.

9. The method according to claim 8, wherein the scan data corresponds to a three-dimensional scan model.

10. The method according to claim 9, wherein the scan model is generated as a result of an MRI scan of the scanned object.

11. The method according to any of claims 1 to 8, wherein the scanned object is scanned using one of one X-ray, CT scan and ultrasound, the bitmap image being generated from an output of the scan.

12. The method according to claim 8, 9, 10 or 11, wherein the scanned object is one or more parts of a human or animal body.

13. The method according to any of claims 8 to 12, wherein the scanned object is a brain.

14. The method according to any of claims 8 to 13, wherein the scan data is video scan data, the video scan data comprising two or more frames, wherein a bitmap image is generated for each frame of the video scan data.

15. The method according to claim 14, wherein a vector line video is generated comprising the vector line images generated from the bitmap images corresponding to each frame of the video scan data.

16. The method according to any preceding claim, wherein the generated bitmap image is a greyscale bitmap image.

17. The method according to any preceding claim, wherein the method further comprises identifying, by the computing device, an area of interest of the scanned object using the vector line image.28 08 2418. The method according to claims 13 and 17, wherein the area of interest is an abnormality in the brain.

19. The method according to claim 13, wherein the vector line image of the brain is rendered on one or more displays superimposed with a functional MRI scan of a brain.

20. A computer program comprising computer executable instructions configured, when executed, to implement any of the method of any preceding claim.

21. A computer device comprising:a memory for storing the computer program according to claim 20;one or more processors for executing the computer program; anda display for rendering the vector line image to a user of the computer device.

22. A computer apparatus for detecting an abnormality in a human or animal body part, the apparatus comprising:a scanning system for scanning the human or animal body part to generate scan data from which a bitmap image is generated, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity; andan image processing system configured to:use a number of threshold groups, each threshold group defined by an upper and a lower intensity threshold;assign each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within an upper and lower intensity thresholds of the threshold group to which the pixel is assigned;render on a display a path at each boundary between pixels assign to different threshold groups to generate on the display a vector line image in which the abnormality is observable.

23. The computer apparatus according to claim 22 wherein the scanning system is one of an X-ray, CT scan, ultrasound and MRI scanning apparatus.

24. A method for detecting an abnormality in a human or animal body part, the method comprising:scanning the human or animal body part to generate scan data from which a bitmap image is generated, the bitmap image comprising a plurality of pixels, each pixel having a corresponding intensity;defining a number of threshold groups be defining an upper and a lower intensity threshold for each threshold group;assigning each pixel of the bitmap image to one of the threshold groups, such that the intensity of the pixel of the bitmap image falls within an upper and lower intensity thresholds of the threshold group to which the pixel is assigned; andrendering on a display a path at each boundary between pixels assign to different threshold groups to generate on the display a vector line image in which the abnormality is observable.28 08 24

Images