Patent Grant
8787634
This application is a 371 national stage entry of PCT/IB2007/055102 filed Dec. 14, 2007 which claims the benefit of EP 06 126451 filed Dec. 19, 2006.
The present invention relates to computer aided detection systems and methods, and relates particularly, but not exclusively, to an apparatus and method for suppressing and indicating computer-detected features of interest in anatomical regions with a significant likelihood of containing computer-detected false positives.
The detection and diagnosis of anatomical features of interest such as pulmonary nodules is a standard procedure in radiological practice. The speed and sophistication of current computer-related systems support development of faster and more sophisticated medical imaging systems. The consequential increase in amounts of data generated for processing, and post-processing, has led to the creation of numerous application programs to automatically analyze the medical image date. Thus, various data processing software and systems have been developed in order to assist physicians, clinicians and radiologists, etc., in evaluating medical images to identify and/or diagnose and evaluate medical images. For example, computer aided detection (CAD) systems have been developed to automatically identify “suspicious” regions such as lesions or pulmonary nodules, which could be an indication for lung cancer or may be metastases from other types of cancer.
CAD systems are known which automatically detect (identify and delineate) morphologically interesting regions (e.g. lesions, nodules, calcification) and other structurally detectable regions that might be of clinical relevance. The CAD software automatically marks or highlights medical anomalies in medical images provided from e.g. computer tomography, magnetic resonance, ultrasound or x-ray scanner. For example, in the analysis of a lung image seeking possibly cancerous nodules, the CAD system will mark the nodules detected.
However, false positive markings often occur close to the beating heart due to image artefacts caused by cardiac motion. The heart beat causes erroneous displacements of the image slices, which in turn leads to an apparent discontinuity in, for example, vessel structures, airways and other anatomical features, which can be misinterpreted by the CAD software and thus lead to false positive markings during the computer aided detection process.
False positive markings may also occur due to artefacts caused by breathing motion, photon or energy starvation areas or streak artefacts caused by high-density objects such as metal prostheses, surgical clips, or dental fillings. This can make it more difficult for a radiologist, physician or clinician to classify and interpret the detected features of interest. This could result in an increased period of time needed to assess the CAD results and lead to assessment errors by the physician due to degraded concentration.
Preferred embodiments of the present invention seek to overcome one or more of the above disadvantages of the prior art.
According to an aspect of the present invention, there is provided a data structure for use by a computer system for processing medical image data representing at least one first region of a patient including at least one computer-detected feature of interest, the data structure comprising:
first computer code executable to detect first data representing at least one second region, included within a respective said first region, wherein at least one said feature of interest in said second region has a significant likelihood of representing a computer-detected false positive; and
second computer code executable to provide second data for enabling at least one said first region to be displayed on a display device, such that at least one said second region is displayed on the display apparatus differently from part of said first region not containing features of interest having a significant likelihood of representing computer-detected false positives.
By differentiating the region with a significant likelihood of representing computer-detected false positives from the image data containing all computer-detected features of interest, this provides the advantage of simplifying and therefore accelerating the radiologist's process of evaluating the medical image data including the computer-detected features of interest. For example, in the case of medical images of the lung, false positive markings are most likely to occur in the vicinity of the heart due to cardiac motion artefacts. In order to simplify the evaluation process of the radiologist, those areas are detected automatically and indicated by, for example, visual cues to the radiologist, improving his or her awareness that in those areas the computer aided detection may have limited sensitivity, therefore requiring a visual inspection with extra diligence.
Said first computer code may be executable to detect data representing at least one said second region on the basis of motion, streak artefact and/or photon-starvation of said second region.
This provides the advantage that physiological regions, where computer-detected false positive markings are likely, are identified within the medical image.
Said second computer code may be adapted to cause said second region to be displayed by means of a first set of graphical markers, such as different colored boundary lines.
This provides the advantage that the regions with a significant likelihood of representing false positive markings can be more easily identified by the radiologist, saving time for evaluating the medical image and therefore improving the quality of the assessment by the radiologist.
Said second computer code may also be adapted to cause at least some of said features of interest in at least one said second region to be displayed by means of a second set of graphical markers, such as different colored boundary lines encircling said features of interest. Said second computer code may be adapted to cause display of at least one said feature of interest in at least one said second region to be suppressed.
This provides the advantage of enabling easier differentiation between the feature of interests detected within said first region and the features of interest detected within said second region.
According to another aspect of the present invention, there is provided a computer readable medium carrying a data structure as defined above thereon.
According to a further aspect of the present invention, there is provided a medical data processing apparatus for processing medical image data representing at least one first region of a patient including at least one computer-detected feature of interest, the apparatus comprising at least one processor adapted to process a data structure as defined above.
This provides the advantage of providing medical image data including computer-detected features of interest that are indicated by different graphical markers depending on whether the feature of interest was detected in a region of significant likelihood of false positive markings or not, therefore, improving the effectiveness and quality of the medical image assessment by the radiologist.
According to a further aspect of the invention, there is provided a medical imaging apparatus comprising:
at least one imaging device for forming medical image data representing at least one first region of a patient including at least one computer-detected feature of interest;
a medical data processing apparatus as defined above; and
at least one display device for displaying at least one said first region and at least one said second region.
According to a further aspect of the invention, there is provided a method of processing medical image data representing at least one first region of a patient including at least one computer-detected feature of interest, the method comprising:
providing first data representing at least one second region, included within a respective said first region, wherein at least one said feature of interest in said second region has a significant likelihood of representing a computer-detected false positive; and
providing second data for enabling at least one said first region to be displayed on a display device, such that at least one said second region is displayed on the display apparatus differently from part of said first region not containing features of interest having a significant likelihood of representing computer-detected false positives.
A preferred embodiment of the present invention will now be described, by way of example only and not in any limitative sense, with reference to the accompanying drawings, in which:
Referring to
A patient 8 is supported on a platform 9 which can be moved in the longitudinal axis of the platform 9 relative to the frame 4 by means of a motor (not shown) under the control of a control unit (not shown). Data detected by the detectors 3 is input to the processor 5. The processor 5 processes the data received using first computer code 6a and second computer code 6b to provide imaging data output 7a, 7b or 7c. The image data output 7a, 7b or 7c is displayed by a display unit (not shown) in order to assist a physician.
The first computer code 6a may include an image-processing algorithm that is able to identify image areas with strong cardiac motion. These areas may be identified as shown in the following example:
After a lung segmentation and thresholding with a threshold of e.g. −400 HU, the image is subjected to a slice-wise 2D distance transformation (a transformation which assigns to each foreground voxel its Euclidean distance to the nearest background voxel). For each voxel at the position (x, y, z), which has a distance value of D(x, y, z)<25 mm, the second derivative L(x, y, z) in the z-direction (i.e. through-slice direction) of the distance map values D, is calculated as L(x, y, z)=2*D(x, y, z)−D(x, y, z−1)−D(x, y, z+1). This value indicates the amount of discontinuous inter-slice-displacement.
To achieve robustness against the noise and spurious inter-slice-displacement-discontinuities L(x, y, z), a range of k=±10 slices is considered around each voxel. Inside this range, the occurring displacement-discontinuity values are ordered by magnitude, and the third largest is taken as the robust displacement discontinuity value L′ for this voxel. If this value L′ is larger than 7 mm, then the voxel is considered to be part of an image area affected by cardiac motion artefacts. The influence region of this voxel is assumed to be L′ multiplied with a range influence factor f=5, with an upper bound of 75 mm. Each nodule candidate which is within this influence region of radius R=f×L′ is then suppressed.
Referring to
A second computer code 6b is used by the processor 5 at step S40 to indicate the regions where a significant likelihood of representing false positives exists with graphical cues such as marking the boundaries of the region with a colored line or by coloring that region in a faint color in order to distinguish the region with a significant likelihood of representing false positives within the displayed imaging data.
Furthermore, at step S50 the lung nodules detected within the identified region with a significant likelihood of representing false positives are displayed with a different set of markers from the lung nodules detected in regions with no a significant likelihood of representing false positives. For example, the different sets of detected lung nodules are displayed in different colors or have graphical cues of different shape.
In order to raise the awareness of the physician that the detected regions of cardiac motion, breathing movement, streak artefacts or photon starvation may include more false positives and have only limited detection sensitivity, detected lung nodules are not displayed within the imaging data forcing the physician to inspect that region with extra care.
At step S60 the imaging data output produced by the processor 5 is displayed at a display unit (not shown).
The algorithm mentioned in the above example describes only one way of finding cardiac motion artefacts in CT images. Other embodiments of image processing algorithms, which mark artefact-suffering image regions, may be algorithms finding cardiac motion artefacts (i.e. heart beat) in MR images, such as cardiac motion affected areas in MR-breast images, algorithms finding respiratory motion artefacts (breathing), which are likely to occur near the diaphragm, algorithms finding metal artefacts in CT images (streak artefacts) and algorithms finding photon-starvation areas in CT images (typical near the shoulder and hip-bone areas).
It will be appointed to persons skilled in the art that the above embodiment has been described by way of example only, and not in any limitative sense, and that various alterations and modifications are possible without departure from the scope of the invention as defined by the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 06126451 | Dec 2006 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/IB2007/055102 | 12/14/2007 | WO | 00 | 6/18/2009 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2008/075272 | 6/26/2008 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 20020076091 | Wang | Jun 2002 | A1 |
| 20030076988 | Liang et al. | Apr 2003 | A1 |
| 20030095697 | Wood et al. | May 2003 | A1 |
| 20030144598 | Zhang et al. | Jul 2003 | A1 |
| 20040184647 | Reeves et al. | Sep 2004 | A1 |
| 20050171409 | Arimura et al. | Aug 2005 | A1 |
| 20050207630 | Chan et al. | Sep 2005 | A1 |
| 20050240882 | Morita et al. | Oct 2005 | A1 |
| 20060177115 | Fujita et al. | Aug 2006 | A1 |
| 20060209063 | Liang et al. | Sep 2006 | A1 |
| 20070177780 | Chui | Aug 2007 | A1 |
| Number | Date | Country |
|---|---|---|
| 2004029851 | Apr 2004 | WO |
| 2004046995 | Jun 2004 | WO |
| 2006054269 | May 2006 | WO |
| 2006054272 | May 2006 | WO |
| Entry |
|---|
| Rivière—“Reduction of noise-induced streak artifacts in x-ray computed tomography through penalized-likelihood sinogram smoothing”—IEEE, 2004, pp. 3239-3243. |
| Dagli, M. S., et al.; Localization of Cardiac-Induced Signal change in fMRI; 1999; NeuroImage; 9:407-415. |
| Ritchie, C. J., et al.; Correction of Computed Tomography Motion Artifacts Using Pixel-Specific Back-Projection; 1996; IEEE Trans. on Medical Imaging; 15(3)333-342. |
| Strickland, R. N.; Tumor Detection in Nonstationary Backgrounds; 1994; IEEE Trans. on Medical Imaging; 13(3) 491-499. |
| Wiemker, R., et al.; Computer aided lung nodule detection on high resolution CT data; 2002; Medical Imaging; SPIE vol. 4684:677-688. |
| Xu, S., et al.; Lung Deformation Estimation and Four-dimensional CT Lung Reconstruction; 2006; Acad. Radiol.; 13:1082-1092. |
| Number | Date | Country | |
|---|---|---|---|
| 20100002922 A1 | Jan 2010 | US |
