CT scanner and scanning method for performing brain scanning

Abstract

A CT scanner for performing brain scans performs low dose 3D scans with a first field of view. By analyzing low-dose 3D scans, such as detection using markers, a suggested field of view for full-dose 3D brain scans may be obtained. A representation of the suggested field of view is displayed and a final field of view is generated as an updated version of the suggested field of view or of the suggested field of view provided by the CT scanner user.

Description

Technical Field

[0001] The present invention relates to a CT scanner and a scanning method for performing brain scanning. Background Art

[0002] Medical imaging, such as CT imaging, is an important tool for the proper analysis and assessment of a subject / patient.

[0003] Proper patient positioning is one of the most important considerations in medical imaging, especially CT imaging, to ensure high quality scans. In CT imaging of the head ("head CT"), quality relates to diagnostic aspects of the scan, such as diagnostic usefulness, and dose exposure to the patient, which is preferably minimized or reduced.

[0004] When performing a CT scan of the human brain, the operator's task is to position the patient and plan the scan so that the entire brain is within the field of view of the scan, while the eyes are optimally shielded from direct radiation. Some CT scanners allow the gantry of the CT system to be tilted, which can support optimal scanning of the patient, especially if the patient's condition does not allow the head to be tilted in the desired direction.

[0005] Another task is to ensure that the scans are reformatted into standard views that make it easier for radiologists to read the examinations. Official standards for medical imaging techniques (such as head CT) provide guidance for appropriate medical scans, such as generating clinically useful and reproducible medical images. Failure to follow the guidelines can reduce the quality of images produced by the scanning procedure and may result in unnecessary radiation exposure to risk organs or anatomical features (such as the lens of the eye). Similarly, following the guidelines produces high-quality images because the guidelines generally ensure that the images contain diagnostically sensitive or useful structures.

[0006] These guidelines are usually defined with respect to a set of anatomical landmarks. For example, a set of guidelines for head CT published by the American College of Physicians defines the recommended scanning angle as "parallel to the line formed by the supraorbital ridge and the posterior rim of the inner plate of the foramen magnum." For example, Figure 1 The positions of the posterior occipital point (oo), which is the midpoint of the posterior edge of the foramen magnum, and points on the supraorbital ridges of the left (le) and right (re) eyes with respect to the skull are shown, along with the corresponding CT images of the subject.

[0007] Planning brain CT scans is particularly challenging because there is a small margin between cropping the lower brain and including the orbits in the field of view. Neither of these should occur in an optimal scan, but to ensure complete coverage of the brain, most technologists add a margin to the field of view during scan planning and accept partial or full exposure to the eyes (against guidelines). This is complicated by the fact that current scan planning is based on 2D unscanned scans, usually lateral projections.

[0008] There is therefore a need for an improved scanning procedure to reduce radiation exposure of the eye during brain scans. Summary of the invention

[0009] The invention is defined by the claims.

[0010] According to an example according to one aspect of the present invention, there is provided a CT scanner for performing a brain scan of a patient, comprising:

[0011] frame;

[0012] X-ray tube and detector in the gantry;

[0013] a camera system for detecting the position and orientation of the patient's head relative to the CT scanner;

[0014] Display; and

[0015] A controller, wherein the controller is configured to:

[0016] deriving a first field of view based on the detected head position and orientation;

[0017] controlling a CT scanner to perform a low-dose 3D scan using the derived first field of view;

[0018] Analyze the results of low-dose 3D scans to derive a recommended field of view for full-dose 3D brain scans;

[0019] displaying a representation of the proposed field of view; and

[0020] A final field of view is determined to be the suggested field of view for the CT scanner or an updated version of a suggested field of view provided by a user.

[0021] The scanner enables low-dose 3D scans (so-called scout or survey scans) to adjust the field of view of the final full-dose / conventional-dose scan. Low-dose 3D scans enable the field of view to be adjusted so as to cover the desired brain area without irradiating the eye.

[0022] The system performs an automatic analysis of the low-dose scan, providing a suggested field of view that can be accepted or modified by the operator. The suggested field of view is the result of a specific patient position, gantry position, and scan plan.

[0023] The availability of 3D low dose scans (to create 3D survey images) enables detection of anatomical landmarks in the low dose 3D images to determine head orientation. The full 3D orientation of the head can be determined, not just rotation about one axis as achieved using lateral 2D survey images.

[0024] For example, the low dose is achieved by an X-ray tube current of about 2 mAs and a corresponding radiation intensity CTDI-vol of about 0.2 mGy. For example, the full dose / conventional dose is achieved by an X-ray tube current of about 300 mAs and a corresponding radiation intensity CTDI-vol of about 50 to 60 mGy. Thus, the low dose has, for example, a tube current in the range of 0.2 to 5% of the full dose / conventional dose and a radiation intensity in the range of 0.1 to 2% of the full dose / conventional dose.

[0025] By incorporating a camera system for detecting the patient's head position and orientation relative to the CT scanner and configuring the controller appropriately, a first field for a low dose scan can be derived based on the detected head position and orientation. This first field of view can already exclude the eye for low dose. Thus, the camera system can implement a two-stage approach, first a rough estimate of the head orientation derived from the optical camera image, which is then refined based on the 3D CT survey image.

[0026] In addition to avoiding radiation to the eye, the analysis can also be used (eg based on artificial intelligence) to detect foreign metallic objects.The gantry can then be tilted or the plane of axial slice reconstruction adjusted in such a way that foreign objects are avoided in a given axial field of view.

[0027] The proposed field of view is part of the overall scan plan. The scan plan includes the definition of the scan length, which is defined by the initial and final table positions. It also includes the definition of a box around the target anatomy, which defines the field of view for reconstruction.

[0028] For example, the first field of view includes the brain and the eye, and the final field of view does not include the eye. Thus, a low dose plan is used when the eye (e.g., part) is within the field of view, but the final field of view does not include the eye.

[0029] For example, the controller is configured to determine the final field of view by setting an angle for the gantry tilt. For scanners with adjustable gantry angles, this is one of the parameters that can be selected based on the analysis of the low-dose scan. For the purposes of this application, in this case, the gantry tilt can also be considered as part of the scan plan.

[0030] If the scanner is not equipped with a gantry tilt feature, or the patient position (head orientation) does not allow the repair eye to remain in the field of view with the gantry tilted, there may be an alternative solution of reminding the operator to reposition the patient's head.

[0031] For example, the controller is configured to analyze the results of the low-dose 3D scan to identify landmarks on the skull. These landmarks can be used to identify the location of different areas of the head within the initial field of view and determine the orientation of the head. Image detection methods for detecting landmark points on the skull (using conventional doses) are known.

[0032] These landmarks may include the supraorbital rims of the left and right eyes, the occipital bone, and the top of the skull.

[0033] The supraorbital rim and occipital bone determine head orientation and position. The plane defined by these landmarks establishes the lowest axial slice of the optimally planned brain scan. The top of the skull enables determination of the desired limits of the field of view.

[0034] For example, the controller is configured to derive a suggested field of view in the form of a rectangular 3D box surrounding the brain. The purpose of this box is to avoid irradiating the eye.

[0035] The controller is configured to export a rectangular 3D box as follows:

[0036] a frame having axes parallel to the CT scanner coordinate system; or

[0037] A frame that is tilted about an axis compared to a frame whose axis is parallel to the CT scanner coordinate system.

[0038] These different frame definitions are suitable, for example, for different scanners, eg with or without the ability to provide gantry tilt.

[0039] The controller can be configured to freely rotate the frame in 3D space (compared to a frame with axes parallel to the CT scanner coordinate system) to generate a diagnostic standard view representation. For example, a frame that freely rotates in 3D space can be used to provide a corrected representation of the scan that is preferred by a radiologist in a diagnostic interpretation.

[0040] For example, the controller is further configured to generate images of axial upper and lower slices of the final field of view based on the low-dose 3D scan.

[0041] These images from low-dose scans will not typically show enough information for diagnostic purposes. However, they can show gross anatomical structures well due to the high contrast between bone and soft tissue. For example, a user can review these images to verify that anatomical structures are included as the radiologist intended. Axial views are a common representation used for diagnostic readings.

[0042] The present invention also provides a method for preparing a patient for a CT brain scan, comprising:

[0043] controlling a camera system to detect the position and orientation of the patient's head relative to the CT scanner;

[0044] deriving a first field of view based on the detected head position and orientation;

[0045] controlling a CT scanner to perform a low-dose 3D scan using the derived first field of view;

[0046] Analyze the results of low-dose 3D scans to derive a recommended field of view for full-dose 3D brain scans;

[0047] displaying a suggested field of view to a user of the CT system; and

[0048] The final field of view is determined to be the proposed field of view or an updated version of the proposed field of view provided by the user.

[0049] The method may also include analyzing the results of the low-dose 3D scan to identify landmarks on the skull, such as the supraorbital margins of the left and right eyes, the location of the occipital bone, and the top of the skull.

[0050] The method may comprise deriving a proposed field of view in the form of a rectangular 3D box enclosing the brain, and the method comprises deriving the rectangular 3D box as follows:

[0051] a frame having axes parallel to the CT scanner coordinate system; or

[0052] A frame that is tilted about an axis compared to a frame whose axis is parallel to the CT scanner coordinate system.

[0053] The method may include freely rotating the frame in 3D space (compared to a frame with axes parallel to the CT scanner coordinate system) to generate a diagnostic standard view representation.

[0054] The method may include controlling the camera system to detect a head position and orientation of the patient relative to the CT scanner, and the controller is further configured to derive the first field of view based on the detected head position and orientation.

[0055] The present invention also provides a computer program, which includes computer program code. When the program is run on a computer, the computer program code is suitable for implementing the above method.

[0056] Another concept is a CT scanner for performing brain scans, including:

[0057] frame;

[0058] X-ray tube and detector in the gantry;

[0059] a camera system for detecting the position and orientation of the patient's head relative to the CT scanner;

[0060] Display; and

[0061] A controller, wherein the controller is configured to:

[0062] Analyzing head position and orientation to determine possible desired field of view and / or patient head motion; and

[0063] A display is used to output to the user a representation of the desired field of view and / or desired patient head movement.

[0064] The camera system is, for example, an overhead 2D or 3D (depth) camera mounted above the patient couch, or one or more cameras integrated with the gantry. It is used to predict the required head tilt and guide the operator in optimal examination preparation. This can be achieved by tilting the CT gantry or guiding the operator to adjust the orientation of the patient's head according to the calculated tilt. In both cases, the brain plane and the CT gantry axis become parallel, and the field of view for the brain scan can be determined in such a way that it excludes the eye in both the low-dose scan and the actual scan, or even completely avoids the need for a low-dose scan.

[0065] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] For a better understanding of the invention, and to show more clearly how it may be put into practice, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0067] Figure 1 The position of the posterior occipital point is shown, as well as respective points on the left and right supraorbital ridges;

[0068] Figure 2 A known CT system is schematically illustrated;

[0069] Figure 3 shows an example using Figure 2 The system is a method of preparing a patient for a CT brain scan;

[0070] Figure 4a A field of view frame having axes parallel to the scanner coordinate system is shown;

[0071] Figure 4b shows a field of view box that has been tilted about one axis to optimize a tightly enclosed brain;

[0072] Figure 4c The field of view box is shown to be freely rotated in 3D space to best capture the brain anatomy;

[0073] Figure 5 A visual representation of the scan plan is shown;

[0074] FIG6 shows an example of an axial slice obtained from a 3D low-dose scan; and

[0075] Figure 7a and 7b An image of an example subject's head captured using an overhead camera is shown, with landmarks marking the eyebrows, eyes, nose, lips, and facial features. DETAILED DESCRIPTION

[0076] The present invention will be described with reference to the accompanying drawings.

[0077] It should be understood that the detailed description and specific examples, although indicating exemplary embodiments of the device, system and method, are intended only for illustrative purposes and are not intended to limit the scope of the invention. These and other features, aspects and advantages of the device, system and method of the present invention will be better understood based on the following description, the appended claims and the accompanying drawings. It should be understood that the drawings are only schematic and are not drawn to scale. It should also be understood that throughout the drawings, the same reference numerals are used to represent the same or similar parts.

[0078] The present invention provides a CT scanner and method for performing a brain scan. A low-dose 3D scan is performed using a first field of view. By analyzing the low-dose 3D scan, for example using landmark detection, a suggested field of view for a full (i.e., conventional) dose 3D brain scan can be obtained. A representation of the suggested field of view is displayed, and a final field of view is generated as an updated version of the suggested field of view or a suggested field of view provided by a user of the CT scanner.

[0079] Thus, the present invention can automatically analyze the head anatomy based on a low-dose 3D scan, resulting in an automatic scan plan, preferably based on a suggested field of view in the form of a 3D scan box. For CT scanners with gantry tilt control, the proposed scan plan includes guidance on the CT system gantry tilt angle to achieve an optimal scan.

[0080] Before explaining the modified functionality provided by the present invention, the basic known components of a CT scanner will first be described.

[0081] Figure 2 A known CT system 100 is schematically illustrated that includes a CT scanner 102. The CT scanner 102 includes a generally stationary gantry 104 and a rotating gantry 106 that is rotatably supported by the stationary gantry 104 and rotates about an examination region 108 about a z-axis. An object support 110, such as a couch, supports a target or object in the examination region 108.

[0082] A radiation source 112 , such as an X-ray tube, is rotatably supported by the rotating gantry 106 , rotates with the rotating gantry 106 , and emits radiation that traverses the examination region 108 .

[0083] The radiation sensitive detector array 114 subtends an angular arc opposite the radiation source 112 across the examination region 108. The radiation sensitive detector array 114 detects radiation that traverses the examination region 108 and generates electrical signal(s) (projection data) indicative thereof.

[0084] The detector array 114 may include single layer detectors, direct conversion photon counting detectors, and / or multi-layer detectors. Direct conversion photon counting detectors may include conversion materials such as CdTe, CdZnTe, Si, Ge, GaAs, or other direct conversion materials. Examples of multi-layer detectors include double layer detectors, such as described in U.S. Patent No. 7,968,853 B2, filed on April 10, 2006, entitled "Double Decker Detector for Spectral CT".

[0085] The reconstructor 116 of the imaging system 102 receives the projection data from the detector array 114 and reconstructs one or more CT images based on the projection data. The reconstructed CT images may include one or more 2D or 3D images. The mechanism of reconstructing one or more CT images based on the projection data is already very mature in the art.

[0086] The controller 118 is configured to process the CT images generated by the imaging system 102. The controller 118 may include a processor 120 (e.g., a microprocessor, a controller, a central processing unit, etc.) and a computer-readable storage medium 122, which excludes non-transitory media and includes transitory media such as physical memory devices and the like.

[0087] The computer-readable storage medium 122 may include instructions 124 for processing images (as described below), including detecting landmarks in low-dose (plain) scans and executing scan plans. The processor 120 is configured to execute the instructions 124. The processor(s) 120 may additionally be configured to execute one or more computer-readable instructions carried by carrier waves, signals, and / or other transient media.

[0088] However, instead of the processor 120 executing instructions to perform the methods described herein, fixed function circuits (eg, a suitably programmed FPGA, etc.) may instead be included to perform the described methods.

[0089] In some examples, the controller may also serve as an operator console. The console 118 includes human-readable output devices, such as a monitor and input devices such as a mouse, keyboard, etc. Software resident on the controller 118 allows an operator to interact with and / or operate the scanner 102 via a graphical user interface (GUI) or otherwise.

[0090] In one variation, a separate controller (not shown) may be used as an operator console and include the relevant operator console elements previously described.

[0091] Figure 2 Also shown are a camera 130 (described below) and a display 132 for presenting scan results and scan planning results (so-called survey or scout scan results) to an operator.

[0092] Figure 3 shows an example using Figure 2 The system is a method 200 of preparing a patient for a CT brain scan.

[0093] The first step in the application of the present invention is to acquire a low-dose (or ultra-low-dose) 3D scan of the patient in step 302. This is performed using a first field of view. Due to the ultra-low dose, the first field of view may be allowed to include irradiation of the patient's eye, but as described below, measures may be taken to prevent irradiation of the patient's eye even with a low-dose field of view.

[0094] For example, a low dose is achieved by an X-ray tube current of about 2 mAs and a corresponding radiation intensity CTDI-vol of about 0.2 mGy. For example, a full dose / conventional dose is achieved by an X-ray tube current of about 300 mAs and a corresponding radiation intensity CTDI-vol of about 50 to 60 mGy. Thus, a low dose has, for example, a tube current in the range of 0.2 to 5% of the full dose and a radiation intensity in the range of 0.1 to 2% of the full dose.

[0095] In step 304, the results of the low-dose 3D scan are analyzed, in particular a suggested field of view for a full-dose 3D brain scan is derived. Analysis of the low-dose scan enables analysis of the head anatomy. This can be accomplished by various known medical image analysis techniques. One example is detection of landmarks on the skull, such as shown in 1 (the supraorbital margins of the left and right eyes and the posterior convexity of the occipital bone).

[0096] For example, the analysis in step 304 involves the detection of these three landmarks. This is sufficient to analyze the three-dimensional rotation state of the head.

[0097] For example, an example of a method for landmark detection can be found in: Hrishikesh Deshpande, Axel Saalbach, Tim Harder, Stewart Young, Thomas Buelow, "Deep learning for the detection of landmarks in head CT images and automatic quality assessment," Proc. SPIE 11596, Medical Imaging 2021: Image Processing, 115960N (15 February 2021); https: / / doi.org / 10.1117 / 12.2581810. Landmark detection is also described in WO 2022 / 023228.

[0098] The plane defined by these three landmarks constitutes the lowest axial slice of the optimally planned brain scan. To determine the scan plan, additional landmarks on the top of the skull are preferably also detected in step 304 .

[0099] Based on the derived head anatomy, a rectangular 3D box enclosing the brain is then derived in step 306. This rectangular box can be considered to represent the proposed field of view for the final full-dose scan.

[0100] In step 306, the proposed field of view is displayed to the user of the CT system. In this way, based on the scan frame and the 3D low dose scan, the operator receives a visual representation of the scan plan in step 308. The visual representation illustrates the expected scan results when scanning using the current patient position, gantry tilt (if adjustable) and other parameters of the scan plan that define the field of view (in the form of the scan frame). The expected results are presented in real time, allowing the user to manipulate the scan frame. This manipulation can be achieved through a graphical user interface.

[0101] The final field of view is then derived in step 310. It may be the suggested field of view (if approved by the user) or it may be an updated version of the suggested field of view provided by the user.

[0102] As described above, a rectangular box is exported as the field of view. As shown in Figure 4, two different box types can be exported.

[0103] Figure 4a A frame 400 is shown with axes parallel to the scanner coordinate system. Such a frame may be advisable for systems that do not allow for gantry tilt.

[0104] Figure 4bA frame 402 is shown tilted about one axis to optimize a tight enclosure of the brain. A single degree of freedom is selected to correspond to the gantry tilt angle. It corresponds to the patient's nod angle. It can be adapted to systems that allow for gantry tilt. The tilt angle between scan frames 400 and 402 represents a recommended gantry tilt to be applied to the scanner hardware.

[0105] The rectangular box can also be manipulated to obtain standard views for diagnostic purposes. Figure 4c A rectangular box 404 is shown that has been freely rotated in 3D space to best capture the brain anatomy. The box represents a corrected representation of the final scan result as preferred by the radiologist for diagnostic reading.

[0106] Figure 5 A visual representation of the scan plan provided in step 308 is shown.

[0107] Frame 500 (untilted) and frame 502 (tilted) are superimposed on a 2D synthetic lateral scout scan. The synthetic scout scan is derived from a 3D low dose scan. Starting from these scan frame representations, the operator can decide whether a gantry tilt should be applied. Also, the user can adjust the start and end points (upper and lower positions of the scan frame).

[0108] Apart from Figure 5 In addition to the visual scan box representation shown in , 3D low dose scanning also allows the generation of axial slices representing upper and lower slices of the expected final scan results based on the operator's selection.

[0109] The results may accommodate the operator's selection of scanning without gantry tilt (block 400 ), scanning with gantry tilt (block 402 ), and may also be displayed and transmitted to the radiologist in a reformatted manner (block 404 ).

[0110] FIG6 shows an example of an axial slice derived from a 3D low-dose scan. Figure 6a shows a simulation result of a scan performed according to the scanning block 400, Figure 6b shows simulation results of a scan performed with a tilted gantry according to scan block 402, and Figure 6c Simulated results of a scan reformatted according to scan box 404 are shown.

[0111] Anatomically aligned reformatting of scans is of clinical importance for diagnostic reading. Measures such as the ASPECT score, used to quickly assess stroke severity, rely on symmetrical representation of brain anatomy. Manual reformatting wastes valuable radiologist time.

[0112] As mentioned above, the most accurate way to align the brain plane with the gantry is to tilt the gantry so that its axis is parallel to the head tilt measured during the 3D low-dose scan. However, not all CT devices have the ability to tilt the gantry. Without this feature, the only chance to align the brain plane with the gantry is to manually change the position of the head, such as by using cushions or different types of head supports.

[0113] This manual process is done primarily by eye, without direct feedback about correct alignment, and most of the time is done before a low-dose scan is acquired. To ensure that the new head tilt angle is indeed appropriate for the gantry axis, for example, an additional low-dose scan is required, which again exposes the entire head (including the eye) to radiation.

[0114] To prevent eye irradiation even for low-dose scans, we use an improved method to use a camera system to determine the field of view for low-dose scans. Specifically, an overhead depth camera ( Figure 2 130) to predict the required head tilt and guide the operator to optimally prepare for the examination.

[0115] This can be achieved by tilting the CT gantry, if possible, or by directing the operator to adjust the orientation of the patient's head according to the calculated tilt. In both cases, the brain plane and the CT gantry axis become parallel, and the field of view for the brain scan can be determined in a way that excludes the eyes in both the low-dose scan and the actual scan.

[0116] The depth map collected by the depth camera provides information about the height of individual parts of the body. By using landmark detection algorithms that usually work on color, grayscale, depth or infrared images, specific surface parts of the face can be detected, such as different points on the eyebrows, nose, lips, chin, etc.

[0117] Standard 3D cameras can usually acquire both color and depth images simultaneously. With proper calibration, surface landmarks detected from the color image can be mapped to the depth image, since the face is largely unoccluded for patient safety reasons, and therefore, can provide reliable information about the depth of these specific points. By comparing the depths of different landmark points with each other, the tilt of the entire head can be calculated.

[0118] Figure 7a and 7b An image of an example subject's head captured using an overhead camera is shown with landmarks marking the eyebrows, eyes, nose, lips, and facial features. For example, detected landmarks are visualized as colored overlays on top of the corresponding grayscale version of the image (in Figure 7a and 7b in black and white).

[0119] Figure 7a and 7b Different nodding angles are shown in Figure 1. For example, different colors are used to represent specific parts of the face. After mapping these points to their respective depth data, the depth values ​​of these fixed points can be compared with each other. Based on this depth information, a previously learned digital 3D head model can be matched to the landmarks, which is used as a basis for calculating the brain plane angle and deriving the optimal tilt of the head.

[0120] Once adapted to the depth data, the head model automatically allows calculation of the plane separating the eyes from the brain. For this purpose, the same landmarks above the eyes and below the skull obtained by low-dose scanning as described above can be used.

[0121] To reduce or avoid exposure of the eye lens, the scanning angle should be parallel to the line formed by the supraorbital ridge and the inner plate of the posterior edge of the foramen magnum. The tilt angle of the resulting plane can be compared with the axis of the gantry to determine the correction that needs to be made. If they are already parallel, the patient and gantry can remain in their actual positions. If a significant tilt angle is detected, there are two ways to correct for this. If the CT device requires a tilted gantry, the gantry can be moved in such a way that the tilt of the head and gantry are perfectly matched. If no gantry tilt is provided, the head needs to be moved manually, for example by changing the cushion or headrest.

[0122] With the aid of a computer display located near the gantry, the operator can be provided with visual guidance in real time on how to adjust the head to match the desired tilt. In an alternative arrangement, the head can be automatically repositioned until the correct position is achieved by moving an actuator that carries the headrest or by inflating or deflating an inflatable cushion under the patient's head.

[0123] Since facial feature detection and fitting of the digital head model is done in real time, the correctness of the position can be checked instantly. Thus, the head can be moved until the system indicates the appropriate tilt angle. In this way, real-time guidance can be provided to the operator. When the head orientation coincides with the gantry axis, the defined planes can be used to define the end of the scanning range, so that the entire brain (but not the eyes) becomes part of the field of view.

[0124] In a different example, the orientation of the head is derived directly from the relative positions of the landmarks in the color image without the need for depth data. This can be achieved, for example, by knowing the tilt of the brain plane in images taken with different head orientations. Due to the projective geometry of the camera, the relative positions of the facial landmarks in the color image depend on the current head orientation.

[0125] Therefore, the camera system enables a two-stage approach, first a rough estimate of the head orientation derived from the optical camera images, which is then refined based on the 3D CT survey images. The 3D survey images may be replaced by the 2D survey images.

[0126] The camera image is used to analyze the patient's head orientation. In many cases, this information can be used to reposition the patient. However, in many cases, such as for trauma patients, it is not possible to achieve optimal patient positioning. Therefore, based on the head orientation estimated by the camera system, the optimal gantry tilt angle is calculated and applied to the scanner. Based on the optical camera image, this first tilt is a rough approximation of the tilt angle used for clinical scans. Since this step does not involve the application of ionizing radiation, the operator can accompany the patient in the scanning room and ensure the patient's safety during the gantry tilt.

[0127] Only after applying the gantry tilt, a 3D low-dose scan can be performed. In addition, the scan range of the low-dose scan can be limited based on anatomical analysis of the optical camera image. This helps to avoid exposing the eye to ionizing radiation during the low-dose scan.

[0128] After acquiring the low-dose scan, the resulting 3D plain scan image can be analyzed to more accurately derive the anatomy of the patient's head. Based on the 3D view, the gantry tilt can be optimized before acquiring the clinical scan.

[0129] The camera system can also implement additional features to enable better 3D low-dose scanning:

[0130] (i) Body shape and size can be assessed to provide optimized dose levels and dose modulation profiles for 3D low-dose scanning;

[0131] (ii) The start and end positions of the 3D low-dose scan can be defined, not just the head region and tilt

[0132] (iii) The optimal couch height for 3D low-dose scanning can also be derived so as to place the target anatomy as close to the isocenter as possible for best image quality.

[0133] The above camera system example still utilizes an initial low-dose scan. However, another embodiment of the camera system uses only optical analysis to determine the desired field of view, thereby avoiding the low-dose scan entirely.

[0134] In an additional enhancement, the AI ​​model for detecting foreign metal objects can be applied to 3D low-dose scans. The results can be used to change the plane of axial image acquisition: For head anatomy and other anatomical structures, axial slices of patients are acquired in clinical routine. If a foreign object is detected in this case, acquiring 3D information using 3D planar imagery helps generate a 3D model of the foreign object present in the field of view. Taking this information into account, the gantry can be tilted or the plane of axial slice reconstruction adjusted in such a way that the foreign object is avoided in a given axial field of view. As a result, radiologists can observe the anatomical structures of interest with minimal foreign object artifacts.

[0135] In an additional enhancement, the correctness of scan planning and patient positioning can be aggregated across a large number of scans and collected into a database. This information can be displayed on a dashboard for educational and training purposes, giving technologists insight into the most common positioning and scan planning issues.

[0136] Those skilled in the art can understand and implement variations to the disclosed embodiments when practicing the claimed invention by studying the drawings, the disclosure and the claims. In the claims, the word "comprising" does not exclude other elements or steps, and the word "one" or "an" does not exclude a plurality.

[0137] The functions implemented by the controller may be implemented by a single processor or by multiple independent processing units, which may be collectively considered to constitute a “controller.” In some cases, these processing units may be remote from each other and communicate with each other in a wired or wireless manner.

[0138] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0139] The computer program may be stored / distributed on suitable media such as optical storage media or solid-state media provided together with or as part of other hardware, but may also be distributed in other forms such as via the Internet or other wired or wireless telecommunication systems.

[0140] If the term "suitable for" is used in the claims or the description, it should be noted that the term "suitable for" is intended to be equivalent to the term "configured to". If the term "arranged" is used in the claims or the description, it should be noted that the word "arranged" is intended to be equivalent to the term "system", and vice versa.

[0141] Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. A CT scanner (100) for performing a brain scan of a patient, comprising: Rack (104, 106); An X-ray tube (112) and a detector (114) within the gantry; a camera system for detecting the position and orientation of the patient's head relative to the CT scanner; Display (132); as well as A controller (118), wherein the controller is configured to: deriving a first field of view based on the detected head position and orientation; controlling the CT scanner to perform a low-dose 3D scan using the derived first field of view; analyzing the results of the low-dose 3D scan to derive a suggested field of view for a full-dose 3D brain scan; displaying a representation of the proposed field of view; and A final field of view is determined to be the proposed field of view or an updated version of the proposed field of view provided by a user of the CT scanner.

2. The scanner according to claim 1, wherein: The first field of view includes the brain and the eye, and the final field of view does not include the eye.

3. The scanner according to claim 1 or 2, wherein: The controller is configured to determine a final field of view by setting an angle for the gantry tilt.

4. The scanner according to any one of claims 1 to 3, wherein: The controller is configured to analyze the results of the low-dose 3D scan to identify landmarks on the skull.

5. The scanner according to any one of claims 1 to 4, wherein: The controller is configured to derive a proposed field of view in the form of a rectangular 3D box surrounding the brain.

6. The scanner according to claim 5, wherein: The controller is configured to export the rectangular 3D box as: A frame (400) having axes parallel to the CT scanner coordinate system; or A frame (402) is tilted about an axis compared to a frame having an axis parallel to a CT scanner coordinate system.

7. The scanner according to claim 6, wherein: The controller is configured to freely rotate the frame in 3D space compared to the frame having axes parallel to the CT scanner coordinate system to generate a diagnostic standard view representation.

8. The scanner according to any one of claims 1 to 7, wherein: The controller is further configured to generate images of axial upper and lower slices of the final field of view based on the low-dose 3D scan.

9. A method of preparing a patient for a CT brain scan, comprising: controlling a camera system to detect a position and orientation of the patient's head relative to a CT scanner; deriving a first field of view based on the detected head position and orientation; (302) controlling a CT scanner to perform a low-dose 3D scan using the derived first field of view; (304) analyzing the results of the low-dose 3D scan to derive a suggested field of view for a full-dose 3D brain scan; (308) displaying the suggested field of view to a user of the CT system; as well as (310) Determine a final field of view as the suggested field of view or an updated version of the suggested field of view provided by the user.

10. The method of claim 9, further comprising analyzing the results of the low-dose 3D scan to identify landmarks on the skull, such as the supraorbital margins of the left and right eyes, the location of the occipital bone, and the top of the skull.

11. A method according to claim 9 or 10, comprising deriving a proposed field of view in the form of a rectangular 3D box surrounding the brain, and the method comprises deriving the rectangular 3D box as: a frame having axes parallel to the CT scanner coordinate system; or A frame that is tilted about one axis compared to a frame having axes parallel to the CT scanner coordinate system.

12. A computer program comprising computer program code adapted to implement the method according to any one of claims 9 to 11 when said program is run on a computer.

Images