MEDICAL IMAGING METHOD AND MEDICAL IMAGING DEVICE

Description

State of the art

[0001] The invention relates to a method for medical imaging according to the preamble of claim 1 and to a medical imaging device for carrying out the method according to claim 12.

[0002] A medical imaging method is already known from EP2851662A2, in which a hyperspectral spectral image of an examination area is acquired in a single acquisition process by an endoscopic and / or exoscopic hyperspectral spectral camera through time-shifted spatial scanning of the examination area in several line steps. In each line step, one spatial dimension and one spectral dimension are acquired. By concatenating the various line steps, the spectral image, comprising one spectral dimension and two spatial dimensions, can then be generated. However, since the line steps are sequentially staggered, motion artifacts can occur due to movements of the spectral camera and / or movement of the examination area, which can reduce the quality of the image or even render it unusable.These motion artifacts arise because the movement causes parts of the area under investigation to be scanned multiple times or even skipped in different line steps. This results in uneven exposures during imaging and thus incorrect intensities of the respective spectral information for each line step. The same problem occurs when acquiring multispectral images or using multispectral spectral cameras, although it may be less pronounced there due to the lower spectral resolution requirements.

[0003] Furthermore, document US 2020 / 0400570 A1 discloses a method comprising actuating an emitter to emit pulses of electromagnetic radiation and measuring reflected electromagnetic radiation with a pixel array of an image sensor. The method also includes detecting motion between two or more successive exposure images, compensating for the motion, and combining the two or more successive exposure images to generate an image.

[0004] Furthermore, document EP 3 626 160 A1 discloses an imaging element for providing image data, such as for LSCI / LASCA measurements, wherein the imaging element comprises a camera, a laser, a motion detector, and a processing device. The processing device is configured to correct an image based on detected motion.

[0005] The object of the invention is, in particular, to provide a generic method or device with improved imaging quality, especially one that takes motion artifacts into account. This object is achieved according to the invention by the features of claim 1, while advantageous embodiments and further developments of the invention can be found in the dependent claims. Advantages of the invention

[0006] The invention relates to a method for medical imaging in which, in a single recording process, at least one multi- and / or hyperspectral spectral image of an examination area is recorded by an endoscopic and / or exoscopic multi- and / or hyperspectral spectral camera.

[0007] It is proposed that the method comprises at least one collation process in which motion artifacts in the spectral image, which occur due to movement of the spectral camera and the area under investigation relative to each other during the recording process, are at least partially, preferably at least to a large extent and particularly preferably completely, reduced and / or avoided by taking into account at least one reference of the spectral camera and / or the area under investigation, wherein movement of the spectral camera relative to the area under investigation is indicated and / or detected on the basis of the reference.

[0008] This allows for the advantageous detection of movement between the spectral camera and the area under investigation by taking the reference into account. This movement can then be used to warn the user of a decline in spectral image quality. Alternatively or additionally, the spectral image can be adjusted and / or corrected based on the reference. This helps to avoid and / or reduce motion artifacts. Overall, this significantly improves the quality of the spectral image.

[0009] Medical imaging is defined as imaging that allows conclusions to be drawn about the physiological properties of an area under investigation, such as tissue type and / or tissue characteristics, such as fat content, water content, oxygenation, the presence of a pigment, or the like. Preferably, the imaging uses spectral analysis to determine these physiological properties. The area under investigation is, in particular, an area containing physiological components, such as tissue, blood, or the like. The area under investigation is located, for example, within a natural or artificially created cavity. Such cavities include, for instance, the abdominal cavity, the intestine, the bladder, the kidney, or the like. However, open tissue could also serve as the area under investigation.The term "acquisition process" is understood to mean, in particular, a process for recording data that comprises at least one or more procedural steps. In the acquisition process of medical imaging, a spectral image or parts thereof of the spectral image of an examination area are recorded using an endoscopic and / or exoscopic multi- and / or hyperspectral spectral camera. A "multi- and / or hyperspectral spectral camera" is understood to mean, in particular, a spectral camera that is configured for recording multi- and / or hyperspectral images. "Configured" is understood to mean, in particular, specifically programmed, designed, adapted, and / or equipped. The fact that an object is configured for a specific function is understood to mean, in particular, that the object fulfills and / or performs this specific function in at least one application and / or operating state.For acquiring multi- and / or hyperspectral spectral imaging, the technique of spatial scanning is preferably used. The investigation area is generated sequentially in several acquisition steps from individual sub-sections, particularly by scanning the investigation area line by line at different times. The sub-sections can then provide information about a spatial dimension and a spectral dimension through spectral dispersion. A spectral image can then be assembled from the sub-sections, whereby these sub-sections are juxtaposed along a further spatial dimension to construct the image. The spectral image is formed, in particular, by a data cube that includes the spatial dimensions, the further spatial dimension, and the spectral dimension.However, other techniques for acquiring multi- and / or spectral images are also known, such as spectral scanning, spatio-spectral scanning, and / or non-scanning or snapshots. These techniques can also produce motion artifacts, as they, for example, capture different spectral bands at different times and construct them into a spectral image. Depending on the application, hybrid forms of these methods are also known. It is conceivable that the spectral camera could additionally enable stereoscopic functionality, allowing the acquisition of additional three-dimensional spatial information. In this case, the spectral image is formed by a four-dimensional tesseract, which includes the spatial dimensions, the additional spatial dimension, the spectral dimension, and a further spatial dimension.This can be achieved, for example, by combining two spectral cameras arranged offset from each other in their viewing direction. A "multi- and / or hyperspectral spectral image" is understood to mean, in particular, an image that includes spectral information from more than three, preferably more than ten, and most preferably more than one hundred different color channels and can preferably be used for a spectral analysis based on these color channels. The color channels have, in particular, a spectral half-width of at most 80 nm, preferably at most 40 nm, and most preferably at most 20 nm. The color channels may at least partially overlap spectrally, but are preferably designed to be at least substantially distinct from one another.When using more than three and fewer than ten color channels, the term multispectral imaging is generally used, while when using more than ten color channels, the term hyperspectral imaging is used. In comparison, a white light image typically acquired with an R,G,B sensor, for example, only contains spectral information from three color channels, such as red, green, and blue, which usually have a full width at half maximum (FWHM) of more than 80 nm. An "endoscopic spectral camera," used to acquire such multi- and / or hyperspectral images, is understood to be, in particular, a spectral camera that interacts with, is coupled to, or can be coupled to an endoscope, or is preferably integrally formed with it.An "exoscopic spectral camera" is understood to mean, in particular, a spectral camera that interacts with, is coupled to, or can be coupled to an exoscope, or is preferably integrally formed with it. An "exoscope" can also be understood to be, in particular, a surgical microscope. Preferably, however, an exoscope is understood to be a surgical microscope that is free of an eyepiece and instead preferably comprises a camera assembly, preferably with at least one white-light camera.

[0010] A "collation process" is understood to mean, in particular, a process for collating images, which comprises at least one or more procedural steps. The collation process can occur at least partially simultaneously with the acquisition process. In the collation process, at least one comparison with at least one reference or several references takes place. The reference serves, in particular, to detect movement of the investigation area and / or the spectral camera and preferably to establish a relationship between movement of the investigation area and the spectral camera. In particular, motion artifacts in the spectral image are corrected taking into account the reference and, in particular, the movement detected based on the reference.Alternatively, based on the reference and especially the motion detected based on the reference, a user can be alerted to motion artifacts in the multi- and / or hyperspectral image, for example by a marker on a screen, or even a spectral image can be automatically deleted.

[0011] Furthermore, it is proposed that at least one target position for the spectral camera relative to the study area, as well as an actual position of the spectral camera relative to the study area, be recorded and output as a reference, whereby motion artifacts are avoided by congruently arranging the target position and the actual position. Advantageously, a targeting device is provided to the user, which allows them to avoid or reduce movements of the study area and the hyperspectral camera during the acquisition of a multi- and / or hyperspectral image. The target position is, in particular, a designated position of the study area from which a spectral image is to be acquired. The actual position corresponds, in particular, to the position of the spectral camera relative to the study area, especially also with respect to a distance between the spectral camera and the study area.The target position and / or the actual position can be projected onto a white light image and / or the area under investigation itself. Image artifacts can be avoided by aligning the target and actual positions congruently. The projections and the white light image are displayed to a user on a screen or on the area under investigation, preferably in real time. The target and / or actual position can be determined, for example, by image tracking based on the white light image, preferably in conjunction with stereoscopic triangulation, and / or by sensors.

[0012] In a preferred embodiment, it is proposed that at least one reference be provided by a sensor system that detects movement of the multi- and / or hyperspectral camera and / or the area under investigation. This enables particularly reliable motion detection and thus the avoidance and / or determination of motion artifacts. The sensor system can be, in particular, an optical or electromagnetic sensor system, such as an optical and / or electromagnetic tracking system. Such a tracking system includes markers arranged on the spectral camera and / or the area under investigation. Alternatively or additionally, the sensor system can include inertial sensors, accelerometers, Hall effect sensors, or the like, arranged on an endoscope / exoscope connected to the spectral camera, on the spectral camera itself, and / or on the area under investigation.By monitoring the sensors, it is possible to infer movement of the study area or the multi- and / or hyperspectral camera, which can then be used to correct the multi- and / or hyperspectral imaging. Alternatively or additionally, the sensors can be used to determine the current and / or target position. Furthermore, the sensors can be used to determine the distance between the multi- and / or hyperspectral camera and the study area. It is advantageous to specify the movement of the multi- and / or hyperspectral camera and the study area more precisely. The distance can be determined using various methods, but preferably via point projection, ultrasound, time-of-flight, stereo reconstruction, and / or time-shifted perspective reconstruction.

[0013] According to the invention, it is proposed that, in the collation process, at least one white-light image of the investigation area, calibrated to the multi- and / or hyperspectral image, is provided as at least one reference. Advantageously, any movement of the multi- and / or hyperspectral camera and the investigation area relative to each other can be easily detected, based on which the hyperspectral image can be post-processed to compensate for motion artifacts. The white-light image and the spectral image are superimposed to detect deviations between specific structures in the white-light image and the spectral image. For this purpose, feature tracking, similar to motion capture, can be used, in which specific structures in the white-light image can be compared with structures in the multi- and / or hyperspectral image.Furthermore, if movement is detected during the comparison, the user could be notified that motion artifacts have been detected. According to the invention, the motion artifacts in the spectral image are corrected based on the motion detection using white light imaging. A camera whose position relative to the area of ​​investigation and / or spectral camera is known can be used to provide the white light image. This relationship is preferably stored in a control electronics unit that can be recalled. It is particularly preferred that the spectral camera and a white light camera, which serves to provide the white light image, are arranged along a common optical axis. Furthermore, the white light camera could be attached to or integrated into the multi- and / or hyperspectral camera, so that their relative positions are always fixed, thus eliminating the need for recalibration.Particularly preferred are the spectral camera and the white light camera as part of a common camera system which is connected to the endoscope and / or the exoscope, a medical imaging device.

[0014] Furthermore, it is proposed that during the collation process, at least one white-light image is calculated from the multi- and / or hyperspectral image and compared with the reference white-light image. This simplifies the comparison of the spectral image with the white-light image, since any deviation of the white-light image calculated from the multi- and / or hyperspectral image from the reference white-light image directly indicates an imaging error due to motion. In particular, if the spectral image has a large number of color channels and if the quantum efficiency of these color channels and that of an R,G,B sensor used to acquire the white-light image are known, the acquired hyperspectral image can be advantageously downsampled to a white-light image.

[0015] It is further proposed that at least one white-light image of the study area be acquired during the spectral imaging process, serving as a reference. This ensures that any detected movement actually occurred during a multi- and / or hyperspectral imaging acquisition and not before or after. Preferably, the white-light image(s) are acquired at least at the start of a spectral imaging acquisition, and the second white-light image is acquired at the end of the spectral imaging acquisition. The difference between the two white-light images can then be used, for example, with a tracking algorithm, to detect whether movement occurred between the start and end of the acquisition.

[0016] Image quality can be further improved, in particular, by capturing multiple white light images during a single spectral imaging scan. Preferably, these multiple white light images are acquired at a frame rate of at least 20 Hz. This allows for a reliable determination of when a movement that could distort the spectral image occurred, thus enabling the identification of which parts of the spectral image are still potentially usable and which are not.

[0017] In a preferred embodiment, it is proposed that several acquisition steps be performed successively during the acquisition process, in which sub-sections of the spectral image are recorded and from which the image is then composed, with at least one white light image being recorded in each acquisition step. Advantageously, each sub-section can be examined for its quality with regard to any potential motion artifacts. Preferably, a sub-section is acquired at an acquisition rate of at least 100 Hz. Most preferably, several white light images are acquired in each acquisition step.

[0018] In a further aspect of the invention, a medical imaging device is claimed, comprising at least one endoscopic and / or exoscopic multi- and / or hyperspectral spectral camera configured to acquire at least one multi- and / or hyperspectral spectral image of an examination area, and at least one control electronics configured to carry out the above method. This advantageously provides a device that improves the quality of the imaging.

[0019] Further advantages become apparent from the following drawing description. The drawing illustrates an embodiment of the invention. The drawing, the description, and the claims contain numerous features in combination. Drawings

[0020] Further advantages will become apparent from the following description of the drawings. The drawings illustrate an embodiment of the invention. The drawings, the description, and the claims contain numerous features in combination. A person skilled in the art will expediently consider the features individually and combine them into meaningful further combinations.

[0021] They show: Fig. 1 a schematic representation of a medical imaging device in a perspective view, Fig. 2 a schematic setup of a camera assembly of the medical imaging device with a spectral camera and a white light camera, Fig. 3 a schematic flow chart of an exemplary procedure for operating the medical imaging device, Fig. 4 an exemplary white light image of an examination area taken by the camera assembly with differing actual position and target position as reference, Fig. 5 an exemplary white light image of the examination area taken by the camera assembly with congruent actual position and target position as reference, Fig. 6 a white light image of the examination area taken by the camera assembly, Fig. 7 a spectral image taken by the camera assembly, after the white light image has been taken Fig. 6with motion artifacts, Fig. 8 a spectral image of the image taken by means of the camera device Fig. 7 with correction of motion artifacts. Description of the exemplary implementations

[0022] Fig. 1 Figure 1 shows a schematic representation of a medical imaging device in a perspective view. In this case, the medical imaging device forms a medical imaging system. Alternatively, however, the medical imaging device can also be just one part of the system.

[0023] The medical imaging device comprises a control electronics unit 10. In this case, the control electronics unit 10 is part of a separately designed control unit 12 of the medical imaging device. The control electronics unit 10 is configured to control other components of the medical imaging device. The control electronics unit 10 is connected to these other components of the medical imaging device. The control electronics unit 10 is configured to carry out a medical imaging procedure. The control electronics unit 10 has at least one processor (not shown). Furthermore, the control electronics unit 10 has at least one memory (not shown). An operating program is stored in the memory. The operating program comprises the medical imaging procedure. The operating program is executable by the processor.

[0024] The medical imaging device includes an endoscope 14. In this case, the endoscope 14 is designed as a rigid endoscope. However, the endoscope 14 could also be a flexible endoscope. Alternatively or additionally to the endoscope, the medical imaging device could include an exoscope.

[0025] The endoscope 14 has a shaft 16. The shaft 16 is designed as an elongated component. In this case, the shaft 16 is rigid. The shaft 16 is designed to be inserted into a natural and / or artificially created cavity during imaging using the medical imaging device.

[0026] The shaft 16 is configured to transmit light towards a study area 18. For clarity, the study area 18 is represented in this case as a hand. Alternatively, the study area could also be a natural and / or artificial cavity or another physiological component. The shaft 16 includes an illumination light guide (not shown). This illumination light guide could be a fiber optic cable, a lens relay, or the like, and in particular, a combination thereof.

[0027] Furthermore, the shaft 16 is configured to transmit light reflected, transmitted, and / or emitted to the investigation area 18, for example, by means of fluorescence or phosphorescence. For this purpose, the shaft 16 has at least one investigation area optical waveguide (not shown). The investigation area optical waveguide can be a fiber optic cable, a lens relay, or the like, and in particular, a combination thereof. The shaft 16 can also include other components, such as a working channel, a fluid channel, or the like.

[0028] The medical imaging device has at least one illumination device 20. The illumination device 20 is connected to the control electronics 10 for control purposes. The illumination device 20 is part of the control unit 12. The illumination device 20 has at least one light source 22. The light source 22 is configured to emit light with a spectral distribution suitable for desired multi- and / or hyperspectral imaging. In this case, the light source 22 is a hyperspectral light source. It emits a broadband spectrum that extends over the entire visible spectral range. Furthermore, it extends at least partially over the NIR spectral range. To generate the emitted light, the light source 22 has at least one luminaire (not shown).The light source can be an LED, an OLED, a white light lamp, a laser, or the like. In this case, the light source is a broadband fluorescent LED. The light source 22 can conceivably have at least two or more light sources, which can differ from one another, particularly in their spectral emission characteristics. It is conceivable that the multiple light sources can be alternately pulsed or at least partially activated simultaneously in order to provide and / or adapt a desired spectral emission characteristic of the light for illuminating the area under investigation. In the present case of hyperspectral imaging, a spectrally broadband light source, such as a white light source, is / would be suitable. For use in multispectral imaging, the light source can preferably comprise several narrowband LEDs.

[0029] The imaging device includes a camera unit 24. The camera unit 24 is connected to the control electronics 10 for control purposes. The camera unit 24 can be coupled to the endoscope 14. In an operating state, the camera unit 24 is coupled to the endoscope 14.

[0030] In Fig. 2 Figure 24 shows a schematic diagram of the camera assembly 24 of the medical imaging device. The camera assembly 24 has an input lens 26. Furthermore, the camera assembly 24 has a beam splitter 28, which is arranged upstream of the input lens 26. The beam splitter 28 divides the light incident through the input lens 26, which is emitted from the examination area 18, into a first beam path 30 and a second beam path 32.

[0031] The camera system 24 includes a spectral camera 34. In this case, the spectral camera 34 is configured as a hyperspectral camera. Alternatively, it can also be configured as a multispectral camera. The spectral camera 34 is arranged upstream of the beam splitter 28. The spectral camera 34 is located in the first beam path 30. In this case, the spectral camera 34 is configured as a spectral-scanning hyperspectral camera. This camera uses a movable slit 36 ​​and a dispersive element 38 to image the investigation area 18 section by section onto an image sensor 40. The dispersive element 38 splits the light into its spectral components. By moving the spectral camera 34 or the slit 36 ​​relative to the first beam path 30, a spectral image can be assembled from partial sections.In other words, the spectral camera 34 scans the investigation area 18 step by step, i.e., line by line, with a time delay and uses this to create a spectral image 44 (cf. . Fig. 4 ) together.

[0032] Furthermore, the camera device 24 includes at least one white light camera 42. The white light camera 42 is arranged upstream of the beam splitter 28. The white light camera 42 is located in the second beam path 32. The white light camera 42 serves to generate a white light image 46. The white light camera 42 and the spectral camera 34 are arranged in a fixed relationship to each other, so that the relationship between the spectral image 44 recorded by the spectral camera 34 and the white light image 46 recorded by the white light camera 42 is known. They can therefore be mapped onto each other.

[0033] The medical imaging device also includes at least one sensor 48 (see Fig. 1The sensor 48 is designed to detect movement of the camera device 24 or the endoscope 14, and thus of the spectral camera 34 or 42, relative to the area of ​​investigation 18. For monitoring purposes, the sensor 48 is connected to the control electronics 10. In this case, the sensor 48 is configured as an optical tracking system. Alternatively, it could be an electromagnetic tracking system.

[0034] The sensor system 48 comprises at least one tracking marker 50. The tracking marker 50 is located on the area under investigation 18. The sensor system 48 also comprises another tracking marker 52. This additional tracking marker 52 is located on the camera assembly 24. Furthermore, the sensor system 48 comprises a tracking camera 54. The tracking camera 54 detects the tracking markers 50 and 52, thereby enabling the determination of the positions and movements of the area under investigation 18 and the camera assembly 24. For example, if tracking markers cannot be located on the area under investigation and / or the camera assembly, the sensor system could alternatively or additionally include an accelerometer or similar device located on the spectral camera to determine its orientation. Alternatively or additionally, the sensor system could also include a distance sensor.This could be done using time-of-flight, ultrasound and / or stereo reconstruction to determine the distance between the area under investigation and the camera device.

[0035] The medical imaging device comprises at least one display means 56. In this case, the display means 56 is a screen. The display means 56 is connected to the control electronics 10. The display means 56 is configured to output information, such as a white light image, a spectral image, an overlay, or the like.

[0036] In Fig. 3 Figure 1 shows a schematic flowchart of an exemplary procedure for operating the medical imaging device. The procedure is part of the operating program stored on the control electronics 10. The procedure is executed by the control electronics 10.

[0037] The procedure comprises at least one acquisition process 100. Furthermore, the procedure comprises at least one collation process 120. The acquisition process 100 and the collation process 120 can run simultaneously, at least temporarily.

[0038] The recording process 100 comprises a recording step 102. In recording step 102, the position of the camera device is determined using sensor 48. Furthermore, the position of the investigation area 18 is determined using sensor 48. This serves as a reference in collation step 120. Recording step 102 is performed continuously during the recording process 100.

[0039] The recording process 100 comprises a further recording step. In this further recording step, at least one white light image 46 is recorded with the white light camera 42 of the camera device 24. In the present case, several white light images 46 are recorded. One such white light image 46 is shown by way of example in Fig. 4 This is shown. This is done with a frame rate of at least 20 Hz. In the present case, this is even done with a frame rate of 100 Hz. The control electronics 10 simultaneously queries the information from the sensor 48 and the white light images 46, so that one of the white light images 46 can always be assigned to a position of the spectral camera 34 and the investigation area 18 determined by the sensor 48. The white light images serve as a reference in the collation process 120. The acquisition step 104 is carried out continuously during the acquisition process 100.

[0040] The collation process 120 comprises at least one collation step 122. In collation step 122, the control electronics 10 queries, as a reference, the position of the camera device 24 relative to the investigation area 18 as detected by the sensor 48. Furthermore, the control device 10 queries, as a further reference, the white light images 46 recorded by the camera device 24. The control electronics 10 assigns an actual position 64 of the spectral camera 34 relative to the investigation area 18 to each white light image 46 (see figure). Fig. 4 and Fig. 5 Thus, the actual position of the camera device 24 relative to the investigation area 18 is known for each white light image.

[0041] In Figs. 4 and 5A white light image 46 of the investigation area 18 is displayed together with an overlay 62 consisting of an actual position 64 and a target position 66. In the collation step, the actual position 64 of the spectral camera 34 is displayed on the display 56 as an overlay 62, along with a white light image 46. Furthermore, the user is prompted to define a target position 66 within the investigation area 18 in the white light image 46, for example, by using gestures or touch controls. The target position 66 then serves as a reference in the overlay 62 and is displayed together with the actual position 64. The actual position 64 is determined by the data acquired by the sensor 48. The operator can align the target position 66 and the actual position 64 congruently by moving the spectral camera 34. If target position 66 and actual position 64 are congruent, a recording of a spectral image 44 can be started.

[0042] The recording process 100 comprises a further recording step 106. In recording step 106, a spectral image 44 is recorded. Figs. 6 and 7 For example, a spectral image 44 is shown together with a corresponding white light image 46. The spectral image 44 is acquired using the spectral scanning method. The investigation area 18 is scanned step by step in staggered intervals, each step comprising individual segments of the spectral image 44. These segments are then combined to form the spectral image 44. Each step is performed at a frequency of 100 Hz. Since the white light images 46 are also acquired at this frequency, a white light image 46 is available for each of these steps, which can be used as a reference for correcting the spectral image 44.

[0043] The collation process 120 includes a further collation step 124. In this collation step, another white light image is calculated from the spectral image. This can be used as an additional reference. The additional white light image is compared with the previously recorded white light image. If they differ, movement occurred during the recording. If they are the same, no movement occurred. The user is informed whether movement occurred. Alternatively or additionally, the sensor data during the recording can be used as a reference to determine such movement. If movement has occurred and motion artifacts have been generated, as is the case, for example, in Fig. 7If an error has occurred, the user is notified. The spectral image can then be discarded. Alternatively, in a further collation step, an attempt can be made to correct the spectral image based on the reference of the white light images captured during the recording and / or sensor data. This is used, for example, in the Fig. 8 shown.

[0044] The collation process 120 comprises at least one further collation step 126. In collation step 126, the sub-sections of the spectral image 46 are compared with their respective temporally assigned white light images 46, since the white light image is acquired at the same frame rate. Thus, it can be determined for each sub-section whether movement has occurred and a corresponding correction can be made. Based on this, a corrected spectral image is then generated from the spectral image. Fig. 8The white light image 46 is shown together with the spectral image 68 corrected from the spectral image 44. 10 Control electronics 12 Control unit 14 Endoscope 16 Shaft 18 Examination area 20 Lighting device 22 Light source 24 Camera device 26 Input lens 28 Beam splitter 30 First beam path 32 Second beam path 34 Spectral camera 36 Slit 38 Dispersive element 40 Image sensor 42 White light camera 44 Spectral imaging 46 White light imaging 48 Sensors 50 Tracking marker 52 Additional tracking marker 54 Tracking camera 56 Display device 62 Overlay 64 Actual position 66 Target position 68 Corrected spectral image 100 Acquisition process 102 Acquisition step 104 Acquisition step 106 Acquisition step 120 Collation process 122 Collation step 124 Collation step 126 Collation step

Claims

1. Method for medical imaging, in which, in a capture process (100), at least one multi- and / or hyperspectral spectral image (44) of an examination region (18) is captured by an endoscopic and / or exoscopic multi- and / or hyperspectral spectral camera (34), the examination region (18) being generated successively during the capture process (100) in a plurality of capture steps from individual subsections by temporally offset line-by-line scanning of the examination region (18), and the spectral image (44) being composed of the subsections, the subsections providing information about a spatial dimension and a spectral dimension by spectral fanning-out, the method comprising at least one collation process (120) in which motion artifacts in the spectral image (44) caused by movement of the spectral camera (34) and the examination region (18) relative to one another during the capture process (100) are at least partially reduced and / or eliminated taking into account at least one reference of the spectral camera (34) and / or of the examination region (18), a movement of the spectral camera (34) relative to the examination region (18) being identified and / or recognized on the basis of the reference, and at least one comparison with the reference taking place in the collation process (120), characterized in that, in the collation process (120), at least one white-light image (46) of the examination region (18) calibrated to the spectral image (44) is provided as at least one reference, the white-light image (46) and the spectral image (44) being imaged onto one another in order to recognize discrepancies between certain structures in the white-light image (46) and the spectral image (44), and the motion artifacts in the spectral image (44) being corrected based on movement recognition by means of the white-light image (46).

2. Method according to claim 1, characterized in that at least one target position (66) for the spectral camera (34) relative to the examination region (18) and an actual position (64) of the spectral camera (34) relative to the examination region (18) are detected and output as at least one reference, motion artifacts being eliminated by arranging the target position (66) and the actual position (64) congruently.

3. Method according to either one of claims 1 and 2, characterized in that, in the collation process (120), at least one reference is provided by a sensor system (48) which detects a movement of the spectral camera (34) and / or of the examination region (18).

4. Method according to any one of the preceding claims, characterized in that, in the collation process (120), at least one white-light image (46) is computed from the spectral image (44) and is compared with the white-light image (46) serving as a reference.

5. Method according to any one of the preceding claims, characterized in that, during the capture process (100) for capturing the spectral image (44), the at least one white-light image (46) of the examination region (18) serving as a reference is captured.

6. Method according to claim 5, characterized in that, during the capture process (100), a plurality of white-light images (46) of the examination region (18) serving as reference are captured.

7. Method according to claim 6, characterized in that the plurality of white-light images (46) are detected at an image frequency of at least 20 Hz.

8. Method according to claim 6 or 7, characterized in that, during the capture process (100), a plurality of capture steps are carried out one after the other, in which subsections of the spectral image (44) are captured, of which subsections said spectral image is composed, at least one white-light images (46) being captured per capture step.

9. Method according to claim 8, characterized in that, during the capture process (100), the capture steps take place at a capture rate of at least 100 Hz.

10. Method according to claim 8 or 9, characterized in that a plurality of white-light images (46) are captured per capture step of capturing the spectral image (44).

11. Medical imaging device having at least one endoscopic and / or exoscopic multi- and / or hyperspectral spectral camera (34) which is configured to capture at least one multi- and / or hyperspectral spectral image (44) of an examination region (18), and having at least one control electronics unit (10) which is configured to carry out a method according to any one of claims 1 to 10.

12. Medical imaging device according to claim 11, characterized in that it comprises an endoscope and / or an exoscope, which is connectable to, connected to and / or integrally formed with the multi- and / or hyperspectral spectral camera (34).