Computer based clinical decision support system and method for determining classification of lymphedema induced fluorescence pattern

Abstract

To provide a computer based clinical decision support system (CDSS) for enhancing determination of a classification of a lymphedema induced fluorescence pattern, and a computer implemented method of enhancing determination of a classification of a lymphedema induced fluorescence pattern.SOLUTION: Described herein are a computer based clinical decision support system (CDSS) 62 and a method for determining a classification of a lymphedema induced fluorescence pattern. A fluorescence image 7 is determined from a measurement of a fluorescence signal in a tissue of a body part 4, to which a fluorescent agent 8 has been added. The CDSS 62 includes: an input interface 64 through which the fluorescence image 7, which is specific to a patient 6, is provided as an input feature to an artificial intelligence (Al) model 68; a processor 66 that performs an inference operation in which the fluorescence image 7 is applied to the Al model 68 to generate a classification of a lymphedema induced fluorescence pattern; and a user interface (UI) 14 through which the classification of the lymphedema induced fluorescence pattern is communicated to a user.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a computer-based clinical decision support system (CDSS) configured to output a classification of a lymphoedema-induced fluorescence pattern based on a fluorescence image determined from measuring fluorescence signals in tissue of a body part to which a fluorescent agent has been added. Furthermore, the present invention relates to a computer-implemented method of determining a classification of a lymphoedema-induced fluorescence pattern using a CDSS, the classification being based on a fluorescence image determined by measuring fluorescence signals in tissue of a body part to which a fluorescent agent has been added.

[0002] Additionally, the present invention relates to methods for diagnosing lymphedema and methods for the long-term treatment of lymphedema. [Background technology]

[0003] Lymphedema is the accumulation of lymphatic fluid in the tissues of the body. Oxygenated blood is pumped from the heart to the tissues through the arteries, while deoxygenated blood returns to the heart through the veins. The pressure level on the arterial side is much higher than on the venous side, so the colorless fluid portion of the blood is forced into the spaces between the cells. Typically, more fluid is pushed out than is reabsorbed on the venous side. The excess fluid is transported by the lymphatic vessels. In addition, the fluid carries away localized foreign bodies such as larger proteins and cell debris. Once in the lymphatic system, this fluid, including the transported materials, is called lymph or lymphatic fluid.

[0004] The lymphatic system contains lymphatic vessels that have one-way valves similar to those in veins to transport lymph fluid to the next lymph node, which removes certain substances and cleanses the fluid before it is returned to the bloodstream.

[0005] When lymphatic system is blocked, such that lymph flow is hindered or not performed at the desired level, lymph fluid accumulates in the interstitial space between tissue cells. This accumulation due to impaired lymphatic transport is called lymphedema. The accumulation of lymph fluid can cause an inflammatory response that damages the cells surrounding the affected area. It can further cause fibrosis, which can transform into hardening of the affected tissue.

[0006] Since lymphedema is a lifelong disease with no cure or medical treatment, early diagnosis and appropriate early countermeasures to improve drainage and reduce fluid load are crucial to the health and recovery of the patient. Possible treatments, such as lymphatic massage and compression bandaging up to surgery, depend on the severity, which is a four-stage system defined by the World Health Organization (WHO) as follows: Stage 1: Normal flow in the lymphatic system. No signs or symptoms. Stage 2: Fluid accumulation with swelling. Stage 3: Persistent swelling that does not resolve with elevation of the affected limb or body part. Stage 4: Elephantiasis (severe limb deformity), thickening of the skin with "warty" growths, and extensive scarring.

[0007] A commonly used technique for diagnosing the function of the lymphatic system is manual examination of the affected limb or body part by a physician. A known imaging technique is lymphoscintigraphy, in which a radioactive tracer is injected into the tissues of the affected body part, followed by MRI (magnetic resonance imaging), CT (computed tomography), PET-CT scan (positron emission tomography) or ultrasound imaging.

[0008] A relatively new imaging technique is infrared fluorescence imaging, which uses a fluorescent dye, e.g., ICG (Indocyanine Green). ICG is a green medical dye that has been used for over 40 years. The dye fluoresces when excited with near-infrared light having a wavelength of 600 nm to 800 nm. This excitation causes ICG to emit fluorescence at 750 nm to 950 nm. The fluorescence of the ICG dye can be detected using a CCD or CMOS sensor or camera. The fluorescent dye is administered to the tissues of the affected limb or body part, and lymphatic concentration and flow can be tracked based on the detected fluorescence. Summary of the Invention [Problem to be solved by the invention]

[0009] It is an object of the present invention to provide a computer-based clinical decision support system (CDSS) for enhancing the determination of a classification of lymphoedema-induced fluorescence patterns, and a computer-implemented method for enhancing the determination of a classification of lymphoedema-induced fluorescence patterns.

[0010] It is a further object of the present invention to provide enhanced methods for diagnosing lymphedema and enhanced methods for the long-term treatment of lymphedema. [Means for solving the problem]

[0011] This object is solved by a computer-based clinical decision support system (hereinafter referred to as "CDSS") configured to output a classification of lymphedema-induced fluorescence patterns based on a fluorescence image determined from measurements of fluorescence signals in tissues of a body part to which a fluorescent agent has been added, the CDSS comprising: an input interface that provides a patient-specific fluorescence image as input features to an artificial intelligence (AI) model; a processor performing an inference operation in which the fluorescence image is applied to an AI model to generate a classification of lymphedema-induced fluorescence patterns; a user interface (UI) by which the classification of the lymphedema-induced fluorescence pattern is communicated to a user; Equipped with.

[0012] Early diagnosis and appropriate treatment are important for successful treatment and prevention of lymphedema. For this early diagnosis, fluorescence imaging is an appropriate approach. In fluorescence imaging, the distribution and progression of a fluorescent agent traveling with the lymphatic fluid is analyzed. Before the first start of measuring the fluorescence, the fluorescent agent is administered to the tissues of the body part of the patient undergoing the medical examination. In a healthy lymphatic system, the fluorescent agent is rapidly taken up by the lymphatic system (along with the lymphatic fluid) and transported rapidly within the lymphatic vessels. The normal rhythmic contraction of the lymphatic vessels can be observed in time-resolved fluorescence images. It can be seen how the fluorescent agent pushes up the limb while the one-way valves in the lymphatic vessels prevent the backflow of fluid.

[0013] In lymphedema, the blockage of lymphatic fluid leads to increased pressure within the lymphatic vessels, which causes lymphatic fluid to leak into the subcutaneous tissue. As lymphedema progresses, characteristic reflux patterns can be seen on fluoroscopy.

[0014] Fluorescent images are typically captured at two time points: a first image is captured in an early transient phase and at least a second image is captured in a late plateau phase. In this procedure, the patient first remains motionless for a few minutes and a certain amount of fluorescent agent is injected subcutaneously. Immediately after the injection, in the so-called "transient" phase, observable fluorescent images show lymphatic flow and can measure, for example, lymphatic pump function. These images are captured while the patient remains motionless. The patient is then allowed to move freely and lymphatic circulation is assessed based on observable patterns. These patterns allow for lymphedema severity staging, for example the severity staging described above as defined by the WHO. Depending on the degree of patient movement, the fluorescent agent can reach a plateau phase at about 2 hours after injection. Essentially, lymphatic circulation can be evaluated to some extent between 2 and 72 hours after injection.

[0015] The fluorescence images analyzed by the CDSS according to the embodiments are always those taken during the late plateau phase, when typical lymphatic reflux patterns can be observed.

[0016] Characteristic reflux patterns can be seen on the fluorescence images, which are clinically classified into several types, depending on the stage of lymphedema. When the lymphatic system is in a healthy state, a linear pattern can be observed, which indicates a reflux of lymph in the lymphatic vessels or veins. In early stages of lymphedema, a so-called "splashback" pattern can be observed. As lymphedema progresses further, a so-called "starburst" pattern occurs. In severe and advanced stages of lymphedema, lymph and with it the fluorescent agent flow throughout the skin and subcutaneous tissue, resulting in a so-called "diffuse" pattern.

[0017] Advantageously, the CDSS performs an automatic classification of lymphedema-induced fluorescence patterns and communicates the same to the user, which greatly assists medical professionals in finding the correct diagnosis regarding lymphedema.

[0018] According to an advantageous embodiment, the classification of the lymphoedema-induced fluorescence pattern is a stage of lymphoedema severity and / or a clinical type of the fluorescence pattern. The stage of severity can be, for example, a stage according to the WHO definition. The clinical types are described above. Possible clinical types that can be distinguished are, for example, a linear pattern, a splash pattern, a stardust pattern and a diffuse pattern.

[0019] In clinical practice, it is often difficult to distinguish between different types of fluorescence patterns. Furthermore, it can be difficult to assign a particular stage of lymphedema to a particular fluorescence image. Advantageously, a CDSS assists the clinician during this decision process.

[0020] According to a further advantageous embodiment, the fluorescence image and the corresponding visible light image are provided to the artificial intelligence model as input features through an input interface. In other words, the fluorescence image and / or a combination of the fluorescence image and the visible light image can be used as input to the artificial intelligence model. Since the clinical manifestations of lymphedema involve certain symptoms that are more or less visible on the surface of the body part, for example on the integument, information that can be derived from the visible light image can assist and thereby improve the process of finding a classification of the fluorescence pattern.

[0021] Within the context of this specification, the term "image" should not be interpreted as being limited to a 2D image. The image may also be a reconstructed 3D image. The image may be captured, for example, by a stereoscopic imaging unit, or may be calculated from a 2D image + additional depth information, which may be captured, for example, by a scanner or any other suitable device.

[0022] In other words, the terms "fluorescence image" and "visible light image" are not limited to 2D images. These images can also be 3D images or 2D images that contain additional information, such as depth information. The 2D images can be captured using a suitable camera. The 3D images can be captured, for example, using a stereo camera. The 2D images or 3D images can be line scans that can be captured using an image line scanner. The images can also be LIDAR scans (and additional color information, for example taken from the 2D images) that also contain depth information as well as 3D images. Data from the LIDAR scanner can be added to the 2D images as additional information. The combination of 2D image data and LIDAR scan data provides similar information as 3D images, since the 2D image data can be combined with the corresponding depth information.

[0023] According to another advantageous embodiment, the CDSS has an input interface configured to measure a fluorescent signal in tissue of the body part and a direct link to an image capture and processing device configured to image a surface of the body part, the tissue doped with a fluorescent agent forming part of the body part, the image capture and processing device comprising an image capture device, the image capture device comprising: an illumination unit configured to illuminate the tissue with excitation light having a wavelength suitable for generating emission light by excitation emission of a fluorescent agent; a fluorescence imaging unit configured to capture a fluorescence image by spatially resolved measurement of the emitted light to provide a fluorescence image; a visible light imaging unit configured to capture a corresponding visible light image of a portion of a surface of the body part; And, The fluorescence imaging unit and the visible light imaging unit are configured such that the viewing directions and / or viewpoints of the fluorescence images and the corresponding visible light images are linked via a known relationship. ,and It will be further strengthened in that respect.

[0024] Advantageously, the image capture and processing device may be part of an endoscope or laparoscope.

[0025] Advantageously, the fluorescent imaging unit and the visible light imaging unit are configured such that the viewing directions and / or viewpoints of the fluorescent image and the visible light image are linked via a known, fixed relationship, in that they can capture corresponding images. The images are linked via a known, fixed spatial relationship of the capture devices. This significantly improves the success rate when applying stitching algorithms to the images, and stitching results in larger fluorescent images and larger visible light images. The larger images are larger in that they show a larger portion of the body part than the portion of the body that can be captured by a single fluorescent image and a single visible light image, respectively.

[0026] According to one embodiment, the CDSS comprises: a) providing a large fluorescent image and a corresponding large visible light image as input features to an artificial intelligence model through an input interface; the fluorescence imaging unit and the visible light imaging unit are further configured to repeat capturing the fluorescence images and the visible light images to provide a series of fluorescence images and a series of visible light images; The image capture and processing device further comprises a processing device, the processing device comprising: To generate large visible light images of body parts, A set of stitching parameters is determined and applied Stitching algorithm For a series of visible light images Apply Rus Equipped with a teaching unit, The stitching unit is further configured to apply a stitching algorithm to the series of fluorescence images to generate a large fluorescence image; Apply to a series of fluorescent images Stitching Algorithm So , Large Visible light image Generate A set of stitching parameters determined when but Applicable Be This will be further strengthened in that

[0027] Within the context of this specification, the term "stitching" is not limited to the combination of two or more 2D images. Stitching can also be performed based on other image data, for example based on 3D image data. For example, further information such as depth information captured by a LIDAR scanner can be taken into account when performing the stitching algorithm.

[0028] Stitching can be performed based on two or more 2D images, resulting in a larger 2D image. When the stitching algorithm is performed based on multiple 3D images, the result is a large 3D image. It is also possible to perform the stitching process in that multiple 2D images and additional information (e.g., data from a LIDAR scan or similar device) are processed, and the result is a large 3D image. In this case, the stitching process includes reconstructing a 3D image from the 2D image data and the additional information. Image stitching includes identifying unique special features that correspond to each other and are visible in the two images that need to be stitched together. This requires that the fields of view of the two images to be combined in the stitching process overlap partially.

[0029] Taking into account this requirement for the stitching process, according to one embodiment, the step of repeating the capture of fluorescence images and visible light images to provide a series of fluorescence images and a series of visible light images is performed such that each subsequent image in the sequence is captured such that the fields of view of the subsequent images at least slightly overlap.

[0030] However, the aforementioned additional information that can be used for the stitching process is not limited to, for example, depth information assigned to each individual pixel in the 2D image. During image acquisition of the series of fluorescent and visible light images, it is also possible for the image capture device to acquire further data, i.e. additional information, data regarding the spatial orientation of the image capture device. The spatial orientation can be, for example, the spatial orientation of the image capture device in the coordinate system of the examination room. captureThe orientation may be the orientation of the device. This orientation may be characterized by three Cartesian coordinates x, y and z, together with the viewing direction (e.g., a vector in the coordinate system of the examination room) and the tilt angle of the image capture device around the viewing direction (e.g., the rotation angle of the image capture device around the viewing direction). This information may be captured for each single image (or image pair including a fluorescent image and a visible light image) of the series of visible light and fluorescent images. This additional information indicating the orientation of the image capture device in space may be used when performing the stitching process. In particular, the orientation of the image capture device, e.g. defined in the coordinate system of the examination room, may be recalculated to the orientation of the image capture device relative to the body part. This information may be used when performing a 3D reconstruction of the body part during the stitching process.

[0031] Advantageously, the stitching of the series of fluorescent images is performed based on a set of stitching parameters previously determined when performing the stitching of the visible light images. The fluorescent images typically offer rare special features suitable for performing the stitching process. Since the visible light images and the fluorescent images are linked by a known constant relationship regarding the viewing direction and / or viewpoint, the parameters of the stitching algorithm used for the visible light images can also be applied to the stitching of the fluorescent images. This significantly improves the stitching process, resulting in larger fluorescent images of better quality.

[0032] Fluorescent agents are, for example, ICG (indocyanine green) or methylene blue. Within the context of this specification, the term "fluorescent dye" or "dye" (also referred to as "fluorescent dye" or "fluorophore") refers to a moiety of a molecule that makes the molecule fluorescent. A moiety is a functional group in a molecule that absorbs energy at a specific wavelength and re-emits the energy at a different specific wavelength. In various embodiments, the fluorescent agent includes a fluorescent dye, an analog thereof, a derivative thereof, or a combination thereof. Suitable fluorescent dyes include, but are not limited to, indocyanine green (ICG), fluorescein, methylene blue, isosulfan blue, patent blue, cyanine 5 (Cy5), cyanine 5.5 (Cy5.5), cyanine 7 (Cy7), cyanine 7.5 (Cy7.5), cypate, silicon rhodamine, 5-ALA, IRDye 700, IRDye 800 CW, IRDye 800 RS, IRDye 800 BK, porphyrin derivatives, Illuminare-1, ALM-488, GCP-002, GCP-003, LUM-015, EMI-137, SGM-101, ASP-1929, AVB-620, OTL-38, VGT-309, BLZ-100, ONM-100, BEVA 800.

[0033] The stitching algorithm is, for example, a panoramic stitching algorithm, in which images are analyzed to extract special and characteristic features. These features are then linked together in multiple images and an image transformation (e.g., shift, rotation, stretching along one or more axes, or keystone correction) is performed. The positions of the linked features are used to determine image transformation parameters, also called stitching parameters. Following image orientation, the images are merged and thereby "stitched" together. This transformation can be performed on visible light images as well as on fluorescent images. Finally, the two images are output. This output may include, for example, displaying the images on a screen. For example, the two images are shown side by side or in an overlay image.

[0034] Often, the function of the lymphatic system is affected in the extremities of the body. However, the CDSS is not limited to the examination of extremities. The system can be applied to the examination of a general body part, which can be the torso, head, neck, back or any other part of the body, as well as the extremities of the patient. It is also possible that the body part is an organ. The CDSS can be applied during open surgery. As well as situations in which the surgery is a minimally invasive surgery performed using an endoscope or laparoscope.

[0035] Within the context of this specification, a "visible light image" is an image of a real-world situation. It reproduces an image impression similar to what the human eye can see. Unlike the human eye, the visible light image can be a color image, a grayscale image, or even a pseudocolor scale plot. The visible light image shows the surface of the body part including the tissue to which the fluorescent agent has been administered. If the tissue is located on the surface of the body part, imaging of the surface of the body part includes imaging of the surface of the tissue.

[0036] According to another advantageous embodiment, the viewing direction and the field of view of the fluorescent image and the visible light image are identical, in particular the fluorescent image and the visible light image are captured through one and the same objective lens, in particular through one single objective lens, which is similar to a camera lens and comprises one or more lenses.

[0037] Advantageously, the fluorescence image and the visible light image can be captured by an imaging device comprising a prism assembly and a number of image sensors assigned thereto. The fluorescence and visible light enter the prism assembly as a common light beam through the same entrance face of the prism assembly. The prism structure includes a filter to separate the visible wavelength range from the infrared wavelength range in which the excitation emission of the fluorescent agent typically occurs. The different wavelength bands, i.e. visible light (also abbreviated Vis) and infrared light (also abbreviated IR), are directed to different sensors. The capture of the visible light image and the fluorescence image through one single objective lens allows perfect alignment of the viewing direction and perspective of the two images. In particular, the viewing direction and viewpoint of the visible light image and the fluorescence image are identical.

[0038] In an advantageous embodiment, the capture of the fluorescent image and the capture of the visible light image are performed simultaneously, without any time switching between the fluorescent image signal and the visible light image signal.

[0039] Advantageously, this Computer implementation The method obviates the need for time switching of signals. This is advantageous because the infrared image, the fluorescence image, and the visible light image are captured exactly at the same time using separate image sensors. Images can therefore also be captured at high frame repetition rates of 60 fps or more. High frame rates cannot usually be achieved when time switching is applied. Also, when the fluorescence and visible light images are captured with separate sensors, the sensors can be positioned so that they are in focus. This improves the sharpness of the images. Furthermore, the sensor settings can be adjusted to the individual requirements for image acquisition of the visible light and fluorescence images. This concerns, for example, the adjustment of sensor gain, noise reduction, exposure time, etc.

[0040] According to yet another advantageous embodiment, the steps of capturing a fluorescent image, illuminating the tissue with excitation light, and simultaneously capturing a visible light image are performed by a single image capture device. When illumination and image capture are integrated in one device, the overall process of measuring the fluorescent signal and simultaneously capturing a visible image can be improved.

[0041] Furthermore, according to another advantageous embodiment, the CDSS includes an image capture device comprising a dichroic prism assembly configured to receive the fluorescent light and the visible light via an entrance surface, the prism assembly comprising a first prism, a second prism, a first compensator prism located between the first prism and the second prism, a further dichroic prism assembly for splitting the visible light into three light components, and a second compensator prism located between the second prism and the further dichroic prism assembly; The first prism and the second prism each have a cross section with at least five corners, each corner having an internal angle of at least 90 degrees. ,So Each of the incident surfaces has a respective exit surface, and the direction of the incident surface normal is parallel to the The first prism and the second prism each of An incident beam incident on the entrance surface is The first prism and the second prism It is reflected twice inside each prism and parallel to the normal of the corresponding exit surface. To Through the exit surface The first prism and the second prism Each Reka Each is designed to emit light from Each of the first prism and the second prism The normal to the incident surface and the normal to the exit surface are perpendicular to each other, When light enters the first prism through the input surface, the light is partially reflected toward the output surface of the first prism. , No. The light travels a first path length from the input face of the first prism to the output face of the first prism, partially enters the second prism through the first compensator prism, and is partially reflected toward the output face of the second prism. , No. travel a second path length from the input face of the first prism to the output face of the second prism; The first prism is further enhanced in that it is larger than the second prism such that the first path length and the second path length are the same.

[0042] Advantageously, the five prism assembly described above allows capturing two fluorescence imaging wavelengths and three colors for visible light imaging, e.g., red, blue, and green. The five prism assembly is advantageous in that the optical path of the light going from the entrance face to each one of the sensors has the same length. Thus, all sensors are in focus and furthermore, there are no timing gaps between the sensor signals. Advantageously, the device does not require time switching of the received signals.

[0043] According to yet another aspect, the image capture device defines first, second, and third optical paths for directing the fluorescent light and the visible light to the first, second, and third sensors, respectively, the image capture device further comprising a dichroic prism assembly configured to receive the fluorescent light and the visible light through an input surface, the dichroic prism assembly comprising a first prism, a second prism, and a third prism, each prism having a respective first, second, and third output surface, the first output surface carrying a first sensor, and the second output surface carrying a second sensor. and a third sensor is provided on the third exit surface, and in particular, a first filter is provided in the first optical path, a second filter is provided in the second optical path, and a third filter is provided in the third optical path, the first, second, and third filters being, in any order, a green filter, an infrared filter, and a red / blue patterned filter with red and blue filters in an alternating pattern such that half of the light received by the red / blue patterned filter passes through the blue filter and half of the light received by the red / blue patterned filter passes through the red filter.

[0044] Further, according to another aspect, the first, second, and third filters are, in any order, a red / green / blue patterned filter (RGB filter), a first infrared filter, and a second infrared filter, and in particular, the first and second infrared filters have different transmission wavelengths.

[0045] In other words, the first and second infrared filters are for filtering IR light of different IR wavelength intervals, for example a first IR band in which a typical fluorescent dye emits a first fluorescence peak, and a second IR band in which a typical fluorescent dye emits a second fluorescence peak. Typically, the second IR band is located at a higher wavelength compared to the first IR band. The first and second infrared filters can also be tuned to the emission bands of different fluorescent agents. Thus, for example, the emission of a first fluorescent agent can pass through the first filter (particularly blocked by the second filter) and be detected on a corresponding first sensor, and the emission of a second fluorescent agent can pass through the second filter (particularly blocked by the first filter) and be detected on a corresponding second sensor. For example, the first filter can be configured to measure the fluorescence emission of methylene blue, and the second filter can be configured to measure the fluorescence emission of ICG.

[0046] According to an advantageous aspect of the present invention, the CDSS is further enhanced in that the illumination unit, the fluorescence imaging unit and the visible light imaging unit are arranged in a single image capture device, which further comprises a measurement unit configured to measure the distance between the surfaces of the body parts captured in the visible light image. Furthermore, the CDSS may be enhanced in that the image capture device is further configured to output a signal indicative of the measured distance. Measurements at different distances may be performed to optimize the illumination and image capture to find the best image acquisition conditions. This best-fit distance may then be stored in the imaging system as a target distance for subsequent measurements.

[0047] For example, a visible and / or audio signal may be output by the image capture device. This signal may guide the operator when manipulating the image capture device such that image acquisition is performed at at least approximately a constant distance relative to the surface of the body part. The integration of a distance sensor in the image capture device with a user-assistance output (visual or optical signal) allows the operator to capture images with more uniform illumination. This improves the quality of the measurement of the fluorescence signal.

[0048] For the determination of the best match distance, the image capture device can repeatedly capture fluorescent and visible light images of the same portion of the surface of the body part while measuring the distance. Multiple sets of fluorescent and visible light images can be captured at different distances. An analysis of the set of images can then be performed taking into account the imaging quality, and the best match distance resulting in the highest quality image can be determined. With this in mind, the output signal, which can be an audio or optical signal, can also indicate the deviation of the measured distance from the best match distance. This can directly inform the operator whether optimal capture conditions, especially with regard to lighting, have been applied during the measurement.

[0049] According to yet another advantageous embodiment of the invention, the CDSS, the input interface of which is furthermore a direct link to an electronic patient record, is capable of inputting patient-related data, in particular the patient's age, sex, height, weight, BMI (Body Mass Index), fat mass, muscle mass, daily physical activity, whether they are employed, skin colour, medication status, presence or absence of vascular diseases, in particular varicose veins or venous oedema, presence or absence of diseases, in particular dialysis or diabetes, blood albumin level, renal function, liver function, cardiac function, blood haemoglobin (Hb) level, blood estimates, lipid metabolism rate, blood glucose level, urea nitrogen (blood BUN, UN level), ABI (Ankle / Brachial Index), the same position before the onset of lymphoedema, PlaceThe method is further enhanced in that the measurement data of lymphatic function, endocrine information, data on hormone levels are provided through the input interface as further input features to the artificial intelligence model. By further taking this additional input information into account, the output, i.e. the classification of lymphedema induced for the fluorescence pattern, can be enhanced. For example, the input interface can comprise a piece of hardware that allows the user to manually input data, such as the patient's age, sex, height, weight, BMI (body mass index), fat mass, muscle mass, daily physical activity, whether or not they work, skin color, medication status, presence or absence of vascular diseases, in particular varicose veins or venous edema, presence or absence of diseases, in particular dialysis or diabetes, amount of albumin in the blood, renal function, liver function, cardiac function, blood hemoglobin (Hb) concentration, blood estimate, lipid metabolism rate, blood glucose concentration, urea nitrogen (blood BUN, amount of UN), ABI (ankle / brachial blood pressure ratio), measurement data of lymphatic function at the same position (40A) before the onset of lymphedema, endocrine information, data on hormone levels. Downloading this information from the electronic patient record, if available, will speed up this process.

[0050] As mentioned above, the input features of the AI ​​model can be either 2D or 3D images, and the fluorescent image can be a combination of a visible light image and a fluorescent image. Furthermore, it is possible to use more than one fluorescent agent. For example, fluorescent agents such as ICG (indocyanine green) and methylene blue can be used. The image to be analyzed can be a single dye image, a combined fluorescent image, i.e., two or more dyes that can be further combined with a visible light image. It can be a stitched image or a reconstructed image. It is also possible to process an overlay image, for example including a visible light image and a fluorescent image.

[0051] Based on this information, the AI ​​model can perform image segmentation using algorithms such as watershed, thresholding, clustering, histograms, etc. It is also possible to define user-defined regions of interest and / or segment lymphatic details within the image.

[0052] The software module for performing the interference operation based on the AI ​​model may be a pre-trained AI model for classifying the fluorescence pattern. The AI ​​may also include multiple pre-trained AI models. The user may then select between different AI models or systems and select the one that best suits the user's needs.

[0053] According to yet another advantageous embodiment, the input interface may include a user interface for correcting the automatically generated classification of the fluorescence pattern. This applies in particular to the automatic generation of stages of severity and clinical type of the fluorescence pattern. The corrected classification may be stored or assigned in association with the fluorescence image and the patient-related data in addition to the automatic result for the classification that is output by the AI ​​model. It is also possible to overwrite the AI ​​predicted data.

[0054] According to yet another advantageous embodiment, the AI ​​model can be customized by user-specific training. The user can select the training mode and input his own data, meaning his own fluorescence images as ground truth. The user-trained AI model is then stored in addition to the pre-trained model. The user can select which AI model to use. A user-trained AI model can be particularly advantageous when a proprietary staging system is used that deviates from common staging systems for lymphedema, such as the WHO definition.

[0055] This object is further solved by a computer-implemented method for determining a classification of lymphedema-induced fluorescence patterns using a computer-based clinical decision support system (CDSS), the classification being based on a fluorescence image determined by measuring the fluorescence signal in tissue of a body part to which a fluorescent agent has been added; Computer implementation The method is: receiving, via an input interface, a patient-specific fluorescence image as input features for an artificial intelligence (AI) model; The processor performs the inference operations death , Fluorescence images but Applied to an AI model to generate a classification of lymphedema-induced fluorescence patterns Ruko And, communicating the classification of the lymphedema-induced fluorescence pattern to a user through a user interface (UI); Includes.

[0056] The same or similar advantages mentioned with respect to the CDSS also apply in the same or similar manner to the computer-implemented method and therefore will not be repeated.

[0057] Furthermore, Computer implementation The method may be advantageously enhanced in that the classification of lymphedema-induced fluorescence patterns is the stage of lymphedema severity and / or the clinical type of fluorescence pattern.

[0058] Furthermore, Computer implementation The method is advantageously enhanced in that the fluorescent image and the corresponding visible light image are provided as input features to an artificial intelligence (AI) model through an input interface.

[0059] According to yet another advantageous embodiment, the Computer implementationThe method includes an input interface configured to measure a fluorescence signal in tissue of the body part and receiving the fluorescence image and the corresponding visible light image via a direct link to an image capture processing device configured to image a surface of the body part, the tissue doped with the fluorescent agent forming part of the body part, the image capture processing device comprising an image capture device comprising an illumination unit, a fluorescence imaging unit and a visible light imaging unit; Computer implementation The method is illuminating the tissue with an illumination unit with excitation light having a wavelength suitable for generating emission light by excited emission of a fluorescent agent; capturing a fluorescence image with a fluorescence imaging unit by spatially resolved measurement of the emitted light to provide a fluorescence image; capturing a visible light image with a visible light imaging unit by capturing a corresponding visible light image of a portion of a surface of the body part. and The fluorescence imaging unit and the visible light imaging unit are configured such that the viewing directions and / or viewpoints of the fluorescence images and the corresponding visible light images are linked via a known relationship. R, It is strengthened in that respect.

[0060] Furthermore, Computer implementation The method includes providing the large fluorescent image and the corresponding large visible light image as input features to an artificial intelligence (AI) model through an input interface; The fluorescent imaging unit and the visible light imaging unit repeatedly capture the fluorescent images and the visible light images to provide a series of fluorescent images and a series of visible light images. the image capture processing device further comprises a processing device comprising a stitching unit; Computer implementation The method is By stitching unit to generate a large visible light image of the body part A set of stitching parameters is determined and applied Stitching algorithm A series of visible light images Applicable works As , Further applying a stitching algorithm to the series of fluorescent images by a stitching unit to generate a large fluorescent image. and Apply to a series of fluorescent images Stitching Algorithm So , Large Visible light image Generate A set of stitching parameters determined when but Applicable Be R ,and In this respect, it can be advantageously strengthened.

[0061] It is further advantageous if the viewing direction and field of view of the fluorescent and visible light images are the same, in particular the fluorescent and visible light images are captured through one of the same objective lenses.

[0062] Furthermore, Computer implementation The method may be enhanced in that the capture of the fluorescence image and the capture of the visible light image are performed simultaneously, in particular without any time switching between the signals for the fluorescence image and the visible light image.

[0063] Furthermore, in accordance with an advantageous embodiment, the steps of capturing a fluorescence image, illuminating the tissue with excitation light, and simultaneously capturing a visible light image are performed by a single image capture device.

[0064] According to another embodiment, Computer implementation Advantageously, the method further comprises measuring the distance between the surface of the body part captured in the visible light image and the capture device. Furthermore, a signal indicative of the measured distance may be output by the imaging device. The method may further comprise repeatedly capturing fluorescent and visible light images of the same portion of the surface of the body part while measuring the distance. Multiple sets of fluorescent and visible light images may be captured at different distances. The set of images may be analysed taking into account the imaging quality and a best matching distance may be determined to obtain the highest quality image. The output signal may thus indicate the deviation of the measured distance from the best matching distance.

[0065] According to yet another advantageous embodiment, the method for determining a classification of a lymphedema-induced fluorescence pattern comprises the steps of: measuring the fluorescence signal on tissue to which at least a first and a second fluorescent agent have been added, and capturing the fluorescence image comprises: capturing a first fluorescence image in a first wavelength range produced by illuminating the tissue with a first excitation light having a first wavelength suitable for producing emission light by a first excitation emission of a first fluorescent agent; capturing a second fluorescence image within a second wavelength range produced by illuminating the tissue with a second excitation light having a second wavelength suitable for producing emission light by a second excitation emission of a second fluorescent agent; Including, providing the first and second fluorescent images as input features to an artificial intelligence (AI) model through an input interface; An input interface receives the first and second fluorescent images as input features for an artificial intelligence (AI) model; This can be further enhanced in that the processor performs an inference operation by applying the first and second fluorescence images to an AI model to generate a classification of the lymphedema-induced fluorescence pattern.

[0066] Furthermore, Computer implementation The method may be enhanced in that patient-related data, in particular data relating to the patient's gender, age or BMI, is provided through the input interface as further input features to the artificial intelligence (AI) model via a direct link to an electronic patient record.

[0067] The object is to provide a method for diagnosing lymphedema, comprising: administering a fluorescent agent to a body part of a patient; determining a classification of lymphedema-induced fluorescence pattern using a computer-based clinical decision support system (CDSS), the classification being based on a fluorescence image determined by measuring the fluorescence signal in tissue of the body part to which the fluorescent agent has been added; wherein the method comprises: receiving, via an input interface, a patient-specific fluorescence image as input features for an artificial intelligence (AI) model; performing an inference operation by a processor, the fluorescence image being applied to an AI model to generate a classification of lymphedema-induced fluorescence patterns; Deriving diagnostic results, particularly regarding the stage of lymphedema, from the classification of lymphedema-induced fluorescence patterns; communicating the lymphedema-induced fluorescence pattern classification and diagnosis to a user through a user interface (UI); The problem is further solved by a method further comprising:

[0068] The method of diagnosing lymphedema can advantageously be performed with a higher level of accuracy and reliability. The method advantageously does not rely on the decisions of medical professionals to interpret fluorescence patterns, which decisions are necessarily contaminated by the personal experience of medical personnel. The method is more objective and therefore accurate.

[0069] The object is to provide a method for the long-term treatment of lymphedema, comprising: diagnosing the severity of lymphoedema by carrying out on the patient a method for diagnosing lymphoedema according to one or more of the above embodiments; administering a treatment to the patient, the treatment being customized to the diagnosis regarding the severity of lymphedema; repeating the steps of diagnosing the severity of lymphedema and administering treatment to the patient, where in each iteration the treatment is adjusted to the detected stage of lymphedema, in particular to the detected stage of lymphedema; The problem is further solved by a method comprising the steps of:

[0070] Methods of long-term treatment according to aspects of the invention are particularly useful because the diagnosis of lymphedema, in contrast to conventional methods, provides an objective result regarding the severity of the disease. The success of long-term treatment can be analyzed from an objective perspective.

[0071] CDSS applies artificial intelligence models. Machine learning (ML) is a branch of artificial intelligence. Machine learning algorithms build models based on example data, known as training data, to make predictions or decisions without being explicitly programmed to do so.

[0072] There are two common modes of machine learning (ML): supervised ML and unsupervised ML. Supervised ML uses prior knowledge (e.g., examples correlating inputs to outputs or outcomes) to learn relationships between inputs and outputs. The goal of supervised ML is to learn a function that best approximates the relationship between training inputs and outputs given some training data, so that an ML model can implement that same relationship when given the input to produce a corresponding output. Unsupervised ML is the training of an ML algorithm using information that is not classified or labeled, allowing the algorithm to act on that information without guidance. Unsupervised ML is useful in exploratory analysis because it can automatically identify structures in the data.

[0073] Common tasks in supervised ML are classification and regression problems. Classification problems, also called categorization problems, aim to classify an item into one of multiple categorical values ​​(e.g., is this object an apple or an orange?). Regression algorithms aim to quantify some item (e.g., by providing a score for the value of some input). Some examples of commonly used supervised ML algorithms are logistic regression (LR), naive Bayes, random forest (RF), neural networks (NN), deep neural networks (DNN), matrix factorization, and support vector machines (SVM).

[0074] Some common tasks for unsupervised ML include clustering, representation learning, and density estimation. Some examples of commonly used unsupervised ML algorithms are K-means clustering, principal component analysis, and autoencoders.

[0075] Another type of ML is federated learning (also known as collaborative learning), which trains algorithms across multiple distributed devices that hold local data without exchanging data. This approach contrasts with traditional centralized machine learning techniques, where all local datasets are uploaded to one server, as well as more classical distributed approaches that often assume that local data samples are identically distributed. Federated learning allows multiple actors to build a common, robust machine learning model without sharing data, thus addressing critical issues such as data privacy, data security, data access rights, and access to heterogeneous data.

[0076] In some examples, the AI ​​model may be trained continuously or periodically prior to performance of the inference operation by the processor. Then, during the inference operation, the patient-specific input features provided to the AI ​​model may be propagated from an input layer, through one or more hidden layers, and finally to an output layer corresponding to a classification of the lymphedema-induced fluorescence pattern, such as a stage of lymphedema severity or level, and / or a clinical type of the fluorescence pattern, such as one of a linear pattern, a splash pattern, a stardust pattern, or a diffuse pattern.

[0077] During and / or following the inference operation, a classification of the lymphedema-induced fluorescence pattern may be communicated to a user via a user interface (UI). For example, a fluorescence image or an overlay image may be displayed on a display along with information about the lymphedema-induced fluorescence pattern. This may be, for example, a report indicating the level of lymphedema or clinical type of the fluorescence pattern corresponding to the AI-generated confidence level. The report may also include suggested diagnostic and / or treatment options.

[0078] Further features of the invention will become apparent from the description of embodiments according to the invention in conjunction with the claims and the accompanying drawings, which may embody individual features or a combination of several features.

[0079] The present invention is described below on the basis of exemplary embodiments, without limiting the general spirit of the invention, and explicit reference is made to the drawings for the disclosure of all details according to the invention that are not described in more detail herein. [Brief description of the drawings]

[0080] [Figure 1] 1 shows a schematic diagram of an image capture and processing device forming part of the CDSS. [Diagram 2] 2 shows a schematic diagram of an image capture device and a processing device of an image capture and processing device; [Diagram 3] 1 shows an example of a visible light image and a fluorescent image corresponding to the visible light image. [Figure 4] FIG. 4 shows a large overlay image generated in part from the visible light and fluorescent images shown in FIGS. 3(a) and 3(b). [Diagram 5] 1 shows the internal prism assembly of the image capture device. [Figure 6] 1 illustrates a schematic representation of an endoscope or laparoscope including an image capture device. [Figure 7] 1 shows a flowchart of the stitching algorithm. [Figure 8] 1 illustrates a schematic of a computer-based clinical decision support system (CDSS). [Figure 9A] An example of a fluorescence pattern is shown. [Figure 9B] An example of a fluorescence pattern is shown. [Figure 9C] An example of a fluorescence pattern is shown. [Figure 9D] An example of a fluorescence pattern is shown. [Figure 10] 13A and 13B illustrate schematic diagrams of another internal prism assembly of an image capture device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0081] In the drawings, elements of the same or similar type or respectively corresponding parts are provided with the same reference numbers in order not to have to reintroduce items.

[0082] Figure 1 shows an image capture and processing device 2 that can act as an input interface to a computer-based clinical decision support system (CDSS), which will be described with reference to Figure 8. Focus is first given to data acquisition by the image capture and processing device 2.

[0083] The image capture processing device 2 is configured to measure fluorescent signals in tissue of a body part 4 of a patient 6. Merely by way of example, the examined body part 4 of the patient 6 is an arm. The measurement of the fluorescent signals can also be performed on other body parts 4 of the patient 6, for example a leg, a part of the head, the neck, the back or any other part of the body. The measurement can also be performed during an open surgery. In this application scenario, the body part 4 can be, for example, an internal organ of the patient 6. The measurement of the fluorescent signals can also be performed during a minimally invasive surgery. For this application scenario, the image capture processing device 2 is, for example, at least partially integrated into an endoscope or laparoscope. For example, the endoscope or laparoscope comprises the image capture device 10.

[0084] Prior to commencing measurements for the first time, a fluorescent agent 8 is administered, i.e. injected, into the tissue of the patient's body part 4. The capture of the fluorescent signal by the image capture and processing device 2 within the tissue of the body part 4 excludes the step of administering the fluorescent agent 8.

[0085] The fluorescent agent 8 is, for example, ICG. ICG (indocyanine green) is a green medical dye that has been used for over 40 years. ICG emits fluorescence when excited with near-infrared light having a wavelength of 600 nm to 800 nm. The emitted fluorescence is 750 nm to 950 nm. The fluorescent agent 8 may also contain two different medical dyes. For example, the fluorescent agent 8 may be a mixture of methylene blue and ICG.

[0086] Following administration of the fluorescent agent 8, the patient's body part 4 may be examined using an image capture device 10 forming part of the image capture and processing device 2, as indicated by the arrow in FIG.

[0087] During fluorescence imaging of lymphedema, images are typically captured at two time points: the first is captured during an early transient phase when the fluorescent agent is rapidly taken up by the lymphatic system, and the second is captured during a late plateau phase. All fluorescence images processed by the CDSS are captured during this late plateau phase.

[0088] The image capture device 10 is configured to image the surface 11 of the body part 4 and to detect a fluorescence signal resulting from the illumination of the fluorescent agent 8 by the excitation light. If the image capture device 10 is applied in surgery, the surface 11 of the body part 4 is for example the surface of an internal organ. In this case, the surface 11 of the body part 4 is identical to the surface of the tissue to which the fluorescent agent 8 has been administered. For the emission of light with a suitable excitation wavelength, the image capture device 10 comprises an illumination unit 16 (not shown in FIG. 1 ).

[0089] The captured images are communicated to a processing device 12, which may implement a CDSS. The CDSS is configured to output a classification of lymphedema-induced fluorescence patterns based on the fluorescence images captured by the image capture device 10. According to this embodiment, the image capture and processing device 2 Image capture device 10 serves as the input interface to the artificial intelligence model.

[0090] It should be noted that the input interface does not necessarily have to be provided by the image capture processing device 2. This will be described later with reference to Figure 8. In the embodiment shown in Figure 1, the input interface is provided by the image capture device 10, as previously described. It provides a fluorescence image specific to the patient 6 to an artificial intelligence (AI) model implemented in the processing device 12. A processor in the processing device 12 performs an interferometric operation in which the fluorescence image is applied to the AI ​​model to generate a classification of a lymphoedema-induced fluorescence pattern. A typical lymphoedema-induced fluorescence pattern is shown in Figure 9, which will be described in more detail below.

[0091] The classification of the lymphedema-induced fluorescence pattern is communicated to a user, in a given embodiment to the physician 3. Output is performed via a user interface, which may be a display 14. The fluorescence image and the visible light image may be output on the display 14 along with the output of the AI ​​model, i.e., the classification of the lymphedema-induced fluorescence pattern. The image capture device 10 may be handled by the physician 3.

[0092] 2 shows in more detail the image capture device 10 and the processing device 12 of the image capture and processing device 2. The image capture device 10 comprises an illumination unit 16 configured to illuminate the tissue with excitation light having a wavelength suitable for exciting the emission of the fluorescent agent 8 and thereby generating fluorescence. The illumination unit 16 is provided with, for example, a number of LEDs.

[0093] The image capture device 10 further comprises an objective lens 18 for capturing the visible light and the fluorescent light. The light is directed through the objective lens 18 to a prism assembly 20. The prism assembly 20 is configured to separate the fluorescent light, particularly in the wavelength range of 750 nm to 950 nm, from the visible light resulting in a visible light image. The fluorescent light is directed to a fluorescence imaging unit 22, which may be, for example, a CCD or CMOS sensor, optionally with additional wavelength filters and electronics. The fluorescence imaging unit 22 is configured to capture the fluorescent image by spatially resolved measurement of the emitted light, i.e., the excitation emission of the fluorescent agent 8, to provide a fluorescent image. Furthermore, there is a visible light imaging unit 24, which may be another CCD or CMOS sensor, optionally with additional different wavelength filters and electronics. The prism assembly 20 is configured to direct the visible light to the visible light imaging unit 24, enabling it to capture a visible light image of a portion of the surface 11 of the patient's body part 4. Similarly, the prism assembly 20 is configured to direct the fluorescent light to the fluorescence imaging unit 22. The prism assembly 20, the fluorescent light imaging unit 22, and the visible light imaging unit 24 will be described in detail later.

[0094] Image capture device 10 may be a scanning unit, such as an image line scan unit or a LIDAR scanning unit. Image capture device 10 may also be a 3D camera suitable for capturing a pair of stereoscopic images from which a 3D image including depth information can be calculated. Of course, image capture device 10 may also be a combination of these devices.

[0095] Image data is communicated from image capture device 10 to processing device 12 via a suitable data link 26, which may be a wireless data link or a wired data link, such as a data cable.

[0096] The image capture device 10 is configured to operate the fluorescence imaging unit 22 and the visible light imaging unit 24 to capture visible light images and fluorescence images simultaneously. In particular, the image capture device 10 does not perform time switching between the signals of the fluorescence image and the signals of the visible light image. In other words, the sensors of the fluorescence imaging unit 22 and the visible light imaging unit 24 are exclusively used to capture images in the respective wavelength ranges, which means that the sensors of the imaging units 22, 24 are used to capture fluorescence images in the IR spectrum or to capture visible light images in the visible spectrum. The imaging units 22, 24 are not used to capture images in both wavelength ranges. This brings important advantages. For example, the sensors can be positioned to be precisely in focus, which is not possible when an image sensor is used for both purposes, i.e. to capture visible and infrared light, since the positions of the focus for these different wavelengths are usually different. Furthermore, the sensor parameters can be individually adjusted, for example with regard to the required exposure time or the sensor gain. An individual setting is advantageous since the IR signal is typically lower than the visible light signal.

[0097] The fluorescent imaging unit 22 and the visible light imaging unit 24 have a fixed spatial relationship to each other because the units are located in one single mounting structure or frame of the image capture device 10. Furthermore, the fluorescent imaging unit 22 and the visible light imaging unit 24 use the same objective lens 18 and prism assembly 20 for capturing the fluorescent and visible light images, respectively. By these means, the fluorescent imaging unit 22 and the visible light imaging unit 24 are configured such that the viewing directions and viewpoints of the fluorescent and visible light images are linked via a known and fixed relationship. In the given embodiment, the viewing directions of the two images are the same and both units 22, 24 image through the same objective lens 18.

[0098] The image capture device 10 can be configured to operate the fluorescence imaging unit 22 and the visible light imaging unit 24 to repeatedly capture the fluorescence images and the visible light images to provide a series of fluorescence images and a series of visible light images. This operation can be performed by the processing device 12 operating the image sensor of the fluorescence imaging unit 22 and the image sensor of the visible light imaging unit 24. The series of images is typically captured while an operator or physician 3 (see FIG. 1 ) moves the image capture device 10 along the longitudinal direction L of the body part 4 of the patient 6. This movement can be performed such that subsequent images of the series of images include overlapping portions. In other words, details shown in a first image of the series of images are also shown in a subsequent second image of the series of images. This is important for a subsequent stitching process if desired. To ensure that corresponding features can be found in the subsequent images, the frequency of image acquisition can be set to a sufficiently high value. The capture of images can be initiated manually, for example by the physician 3, or the capture of images can be controlled by the image capture device 10 such that the described prerequisites are met.

[0099] Image capture device 10 may be further configured to acquire the position and orientation of image capture device 10 during this movement. For example, the position and orientation of image capture device 10 in the frame of reference of the examination room or the frame of reference of patient 6 may be determined for each image or image pair that is captured. This information may be stored and communicated with the images or image pairs, including visible and fluorescent images. This information may be useful for subsequent image reconstruction, such as to generate a 3D image from a series of 2D images.

[0100] Two series of images (i.e., a first series of visible light images and a second series of fluorescent images) or series of image pairs (each image pair including a fluorescent image and a visible light image) are imageOnce captured by the capture device 10 and received into the processing device 12, the series of visible light images may be processed by a stitching unit 28 (see FIG. 2 ). The stitching unit 28 is configured to apply a stitching algorithm to the series of visible light images to generate a large visible light image of the body part 4. The large image is "larger" in that it shows a larger portion of the body part 4 of the patient 6 compared to a single image.

[0101] Within the context of this specification, the term "stitching" should not be understood as limiting the stitching process to the combination of two or more 2D images. Stitching can also be performed on the basis of 3D images, the result of this process being a larger 3D image. Stitching can also be performed on the basis of 2D images and additional information regarding the direction of view from which the 2D images were captured. Further information regarding the position of the image capture device 10 can also be taken into account. As mentioned above, on the basis of these data sets a larger 3D image can be generated, i.e. a larger 3D image can be stitched together from a series of 2D images plus information regarding the position and orientation of the image capture device 10. It is also possible to combine 3D scan data, for example from a LIDAR sensor, with 2D image information. Again, the result of the stitching process is a larger 3D image. The term stitching particularly encompasses the process of reconstructing a 3D image from a data set.

[0102] Regardless of the specific type of data processed during stitching, the stitching algorithm starts with stitching a visible light image. The stitching algorithm generates and applies a set of stitching parameters when performing the stitching operation. The detailed operation of the stitching unit 28 is described below. The stitching unit 28 is configured to apply the stitching algorithm to a series of visible light images as well as a series of fluorescence images to generate a large fluorescence image. Again, the stitching process is not limited to the combination of two or more 2D images. It is also possible to generate a 3D fluorescence image in a similar manner as described above for the visible light image.

[0103] The stitching algorithm applied to stitch the fluorescence images is the same algorithm used for stitching the visible light images. Moreover, the stitching of the fluorescence images is performed using the same set of stitching parameters determined when performing the stitching of the visible light images. This is possible because there is a fixed relationship between the viewing direction and field of view of the visible light images and the fluorescence images. Of course, if the viewing direction and viewpoint of the visible light images and the fluorescence images are not the same, a fixed offset or shift of the stitching parameters needs to be applied. This takes into account the known fixed spatial relationship between the IR and Vis image sensors and the corresponding optical systems.

[0104] Following stitching, the large visible light image and the large fluorescence image are available and can serve as input to the AI ​​model, and the large visible light image and the fluorescence image can be output. For example, the images are displayed side-by-side on the display 14 along with the results of the AI ​​model, i.e., classification of lymphedema-induced fluorescence patterns.

[0105] Unlike conventional examination systems, the display 14 can show the visible light image and the fluorescence image, which correspond to each other, together with the results of the AI ​​model. Details visible in the fluorescence image, e.g. high fluorescence intensity indicating a characteristic pattern of lymphatic fluid accumulation or lymphatic reflux, can be found exactly in the patient's body part 4 at the corresponding location shown in the visible light image. This allows the physician 3 to precisely locate, e.g., areas where lymphatic fluid accumulation is found. This is highly useful information for a tailored specific treatment of the patient 6, for example.

[0106] The visible light image and the fluorescent image, in particular the large visible light image and the large fluorescent image, can also be superimposed to provide an overlay image, in particular a large overlay image, of the body part 4. This can be performed by a superimposition unit 30 of the processing device 12. The overlay image can also be output via the display 14.

[0107] Figure 3 (a) 1 shows an example of a visible light image 5 in which a portion of a surface 11 of a body part 4 of a patient 6 is visible. 6 A portion of the leg of the (b) Patient 6 FIG. 1 shows a corresponding fluorescence image 7 determined by measuring the fluorescence signal of a fluorescent agent 8 applied to the leg tissue of a subject. High intensity spots or areas of the fluorescence signal are visible. This type of fluorescence image 7 The fluorescent patterns observed are typically classified as diffuse, which indicates a later or severe stage of lymphedema. The different types of lymphedema-induced fluorescent patterns are described in more detail below. 6 The image strongly indicates slow lymphatic transport in the leg and lymphatic accumulation due to lymphedema. Advantageously, by comparing the fluorescent image 7 with the visible light image 5, the physician 3 can identify the location of the area where slow lymphatic transport is performed. Furthermore, the physician 3 is provided with the output of the AI ​​model, e.g., stage of lymphedema severity and / or clinical type of fluorescence pattern.

[0108] In Fig. 4, there is an overlay image 9, and in Fig. 3 (a) and Figure 3 (b) In addition to the images shown in FIG. 3, stitching of a visible light image 5 and a fluorescent image 7 has been performed. Exemplary single visible light image 5 and fluorescent image 7 can also be seen in FIG. 4, each projected between the straight dashed lines shown in a large overlay image 9. By stitching the visible light image 5 and the fluorescent image 7, a large overlay image 9 can be provided that shows almost the entire body part 4 of the patient 6. The fluorescent signal can be shown in pseudocolor to clearly distinguish it from the features of the visible light image 5. Also, the overlay image 9 can be used to provide a large overlay image 9 that shows almost the entire body part 4 of the patient 6. The fluorescent signal can be shown in pseudocolor to clearly distinguish it from the features of the visible light image 5. 9 can serve as input to the AI ​​model of the CDSS.

[0109] FIG. 5 shows a prism assembly 20 that can be implemented in the image capture device 10. The first prism P1 is a pentaprism. An incident light beam A, which is visible light and fluorescent light, enters the first prism P1 through an incident surface S1 and is partially reflected at a surface S2, which is one of the two surfaces not adjacent to the incident surface S1. The reflected beam B is then reflected against a first surface adjacent to the incident surface S1. The reflection angle can be less than the critical angle so that the reflection is not internal (the adjacent surface can be coated to avoid light leakage and reflect the required wavelengths of interest). The reflected beam C then intersects with the incident light beam A and exits the first prism P1 through a second surface adjacent to the incident surface S1 toward the sensor D1. A portion of the beam A passes through the surface S2 and enters the compensation prism P2. Two non-internal reflections can be used to direct the incident beam A to the sensor D1 via beams B and C. Furthermore, there may be no gap between prisms P1 and P2, no gap between prisms P3 and P4, and no gap between prisms P2 and P3. Prism P2 is a compensator prism for adjusting the individual lengths of the optical paths from input surface S1 to sensors D1-D5.

[0110] prismFrom P2, beam D enters a second pentaprism P3. As with prism P1, internal reflections are used to make the beam cross itself. For brevity, the description of the beams will not be repeated except to note that in prism P3, beam portions E, F, and G correspond to beam portions A, B, and C, respectively, in prism P1. Prism P3 also cannot use internal reflections to reflect the incident beam toward sensor D2. Two non-internal reflections can be used to direct incident beam E to sensor D2 via beams F and G.

[0111] After the prism P3 there is another compensation prism P4. Finally, the beam H enters a dichroic prism assembly including prisms P5, P6 and P7 with sensors D3, D4 and D5, respectively. The dichroic prism assembly is for splitting the visible light into red, green and blue components towards the respective sensors D3, D4 and D5. The light enters the prism assembly through the beam I. An optical coating C1 is placed between the prisms P5 and P6, and another optical coating C2 is placed between the prisms P6 and P7. Each optical coating C1 and C2 has a different reflectivity and wavelength sensitivity. Optical Coatings At C1, the incident beam I is partially reflected (beam J) off the same face of the prism from which it entered. At that same face, a beam, now labeled K, is reflected once again towards sensor D3. beam From J beam The reflection to K is an internal reflection. Thus, sensor D3 receives light reflected by coating C1, and in an analogous manner, sensor D4 receives light from beam L reflected by coating S2 (beams M and N), and sensor D5 receives light from beam O that traversed the prism unimpeded.

[0112] There is an air gap between prisms P4 and P5. In prism assembly 20, the following total path lengths can be defined for each end channel (defined with respect to the sensor at the end of the channel): Sensor D1 (e.g., first near-infrared) path: A+B+C Sensor D2 (e.g., second near-infrared) path: A+D+E+F+G Sensor D3 (e.g., red) path: A+D+E+H+I+J+K Sensor D4 (e.g., blue) Path: A+D+E+H+I+ L+M+N Sensor D5 (e.g., green) Path: A+D+E+H+I+ L+O

[0113] The path length is A+B+C=A+D+E+F+G=A+D+E+H+l+J+K=A+D+E+H+l+ L+M+N =A+D+E+H+I+ L+O The equation is adjusted so that

[0114] Matching the path lengths may include adjusting the focal plane focus position difference of the wavelengths detected at the sensors D1-D5. That is, for example, the path length to a sensor for blue (B) light may not be exactly the same as the path length to a sensor for red (R) light. This is because the ideal distance to produce a sharply focused image depends to some extent on the wavelength of the light. The prisms can be configured to allow for these dependencies. The length of D+H can be adjusted by lateral displacement of the compensator prisms P2, P4, which can act as focus compensators due to wavelength shifts.

[0115] The larger air gap in path I can be used for additional filters or can be filled with glass compensators for focus shift and compensation. Due to internal reflections in the path from beam J to beam K, an air gap needs to exist at that particular bottom surface of the red prism. Space can be reserved between the prism output surface and each of the sensors D1-D5 to provide additional filters or should be filled by glass compensators accordingly.

[0116] The sensors D1 and D2 are IR sensors configured to capture the fluorescence image 7. As an example, the sensors D1 and D2 plus suitable electronics are part of the fluorescence imaging unit 22. The sensors D3, D4 and D5 are for capturing the three components of the visible light image 5. As an example, the sensors D3, D4 and D5 plus suitable electronics are part of the visible light imaging unit 24. It is also possible to consider corresponding prisms that direct the light beams to the sensors that are part of the respective units, i.e. the fluorescence imaging unit 22 and the visible light imaging unit 24.

[0117] 6 shows a schematic representation of an endoscope 50 or laparoscope. In view of the aspects of the present invention, the difference between a laparoscope and an endoscope is relatively small. Thus, where the description refers to an endoscope, a laparoscopic configuration is generally also possible. By way of example only, in the following reference will be made to the endoscope 50.

[0118] The endoscope 50 can serve as an input interface for the CDSS. The endoscope 50 includes an image capture device 10, described in more detail above, that captures fluorescence images that can be input features for the AI ​​model. The image capture device 10 includes an objective lens 18 that captures a fluorescence image 7 and a visible light image 5. The objective lens 18 focuses light incident on an entrance face S1 of a prism assembly 20 onto sensors D1-D5. The objective lens 18 can also be integrated into the last part of the endoscope section to match the back focal length of the prism.

[0119] The endoscope 50 comprises an optical fiber 52 connected to a light source 54 that couples light into the endoscope 50. The light source 54 can provide white light for illumination of the surface 11 of the body part 4 and capture of the visible light image 5. The light source 54 can also be configured to emit excitation light suitable for exciting a fluorescent dye applied as a fluorescent agent to fluoresce. In other words, the light source 54 can be configured to emit both visible light and light in the IR spectrum.

[0120] Inside the shaft 56 of the endoscope 50, the optical fiber 52 splits into multiple fibers 51. The endoscope 50 can have a flexible shaft 56 or a rigid shaft 56. In the rigid shaft 56, a lens system consisting of lens elements and / or relay rod lenses can be used to guide the light through the shaft 56. If the endoscope 50 has a flexible shaft 56, the fiber bundle 51 can be used to guide the light of the light source 54 to the tip of the endoscope shaft 56. A fiber bundle 58 is disposed within the shaft 56 of the endoscope 50 to guide the light coming from the inspection area from the distal tip of the endoscope shaft 56 (not shown in FIG. 6 ) to the image capture device 10 at the proximal end of the shaft 56. In another embodiment not shown, the entire image capture device 10 can be miniaturized and disposed at the distal tip or end of the endoscope shaft 56.

[0121] FIG. 7 shows a flow chart of a stitching algorithm that can be used for stitching visible light and fluorescent images. The flow chart is more or less self-explanatory and will be explained very simply. First, the acquired sequence of images (S1) is transferred to the stitching unit 28 of the processing device 12. Then, the algorithm performs a frame pre-selection (step S2). In this pre-selection step, frames suitable for stitching are selected. S3 represents the selected images to be stitched, which are then subjected to pre-processing (step S4). On the pre-processed images (S5), feature extraction is performed (step S6). Once the image features are extracted (S7), image matching (step S8) is performed using the known image from S3 and the features extracted from step S7. Based on the selected images (S9), the transformation of the image is estimated (step S10). This image transformation estimate (S11), also called stitching parameters, is applied (step S12). Application of the transformation results in a transformed image (S13). Further image corrections, such as exposure corrections, can be performed (step S14). The transformed and corrected images (S15) are stitched together (step S16) by locating the seams (i.e. the lines along which the images are joined). Data indicating the seam locations (S17) are used together with the transformed and corrected images (S12) to create a composite of the images (step S18). In a given embodiment, this results in a large visible light image or a large fluorescent image as a result of stitching (S19).

[0122] Furthermore, the image capture device 10 applied for capturing the visible light image 5 and the fluorescence image 7 may further comprise a measurement unit 32 configured to measure a distance d (see FIG. 1 ) between the surface 11 of the patient's body part 4 captured in the visible light image 5 and the image capture device 10. A distance sensor 33 in communication with the measurement unit 32 is, for example, an ultrasonic sensor, a laser distance sensor, or any other suitable distance measuring device. Furthermore, the image capture device 10 is configured to output a signal indicative of the measured distance d. For example, the image capture device 10 outputs an optical or acoustic signal that gives the operator of the device 10 information about the best distance d for performing the measurement. Performing the measurement at a constant distance d significantly improves the measurement result, especially due to the presence of a uniform illumination.

[0123] In addition to the distance sensor 33, the image capture device 10 may include an internal measurement unit (IMU) 35, which may be used to collect data regarding rotation in pitch, yaw, and roll, as well as acceleration data in three spatial axes (x, y, and z). This IMU information may be used as additional data to improve the performance of the stitching algorithm, or to provide feedback to the operator for positioning and rotating the camera to provide a better image for the stitching algorithm.

[0124] The processing device 12 may further comprise an AI unit 60 implementing a CDSS. Figure 8 shows a schematic representation of a computer-based clinical decision support system (CDSS) 62. The CDSS comprises an input interface 64, a processor 66 implementing an AI model 68, and an output interface 70. The input interface 64 is coupled to the image capture processing device 2, which serves as a direct data link providing input for the AI ​​model 68. The input interface 64 is further coupled to a user input interface 72, which may be a keyboard, a touchpad, or any other device suitable for that purpose. The input interface 64 is further coupled to a database 74, which holds patient-related data, in particular electronic patient records. The output interface 70 is coupled to a display 14, which may be identical to the display 14 of the image capture processing device 2 shown in Figure 1.

[0125] An exemplary CDSS 62 is configured to provide a classification of a lymphoedema-induced fluorescence pattern based on the fluorescence image 7. In various embodiments also outlined above, the CDSS 62 includes an image capture processing device 2, in particular the image capture device 10, as an input interface 64 where the patient 6-specific fluorescence image 7 is provided as input features to an artificial intelligence (AI) model 68. A processor 66 performs an inference operation where the fluorescence image 7, large fluorescence image or overlay image 9 is applied to the AI ​​model 68 to generate a classification of the lymphoedema-induced fluorescence pattern. A user interface (UI) where the classification of the lymphoedema-induced fluorescence pattern, e.g. the stage of lymphoedema severity and the clinical type of the fluorescence pattern, is communicated to a user, e.g. a clinician, may be the display 14.

[0126] In some embodiments, the input interface 64 may be a direct data link between the CDSS 62 and one or more medical devices that generate at least some of the input features. For example, the input interface 64 may directly transmit fluorescence images to the CDSS 62 during a therapeutic and / or diagnostic medical procedure. This may be performed by the image capture and processing device 2. Additionally or alternatively, the input interface 64 may be a classic user interface that facilitates interaction between a user and the CDSS 62. For example, the input interface 64 may facilitate a user input interface 72 through which a user can manually input further patient-related data, such as, for example, gender, age or body mass index. Additionally or alternatively, the input interface 64 may provide the CDSS 62 with access to an electronic patient record from which one or more input features may be extracted. This may be performed by a direct link to a database 74 that holds the respective information. In any of these cases, the input interface 64 is configured to collect, at or before the time the CDSS 62 is used to assess the level or stage of lymphedema, one or more of the following input features associated with a particular patient 6: a fluorescence image 7, a large fluorescence image, an overlay image 9, and optionally additional information from the patient record as described above.

[0127] Based on one or more of the above input features, the processor 66 performs an inference operation using the AI ​​model 68 to generate a classification of lymphedema-induced fluorescence patterns. For example, the input interface 64 may deliver the fluorescence images 7 into an input layer of the AI ​​model 68, which propagates these input features through the AI ​​model 68 to an output layer. The AI ​​model 68 can provide a computer system with the ability to perform tasks without being explicitly programmed by performing inferences based on patterns found in the analysis of data. The AI ​​model 68 explores research and building algorithms (e.g., machine learning algorithms) that can learn from existing data and make predictions on new data. Such algorithms operate by building an AI model from example training data to make data-driven predictions or decisions, which are represented as outputs or evaluations. Training and training of the AI ​​model can also be performed by manual user input, for example correcting the automatically generated stage of lymphedema severity or clinical type of fluorescence pattern.

[0128] Training of the AI ​​model can also be performed in the cloud using data from various locations, and then the trained network can be downloaded and used for the respective treatment. Using the cloud for training is particularly advantageous due to its higher performance compared to local systems.

[0129] In Figure 9, there are examples of clinical types of fluorescent patterns. Figure 9A shows a typical linear pattern. This pattern can be observed when the lymphatic system is in good condition and the fluorescent agent 8 is transported up the lymphatic vessels with lymph fluid. The stage classification of this finding is typically stage 1, meaning there is normal lymph flow and no signs or symptoms. In Figure 9B, there is a splash pattern that can be observed if the lymph flow is partially blocked. Typically, stage 2 is assigned to this finding, meaning there is an accumulation of lymph fluid with possible swelling. Figure 9C shows a stardust pattern that can be observed if lymphedema has progressed further. This pattern is typically assigned to stages 2 or 3, where permanent swelling of the body part 4 can also be observed. In stage 3 of lymphedema, this swelling typically does not dissipate with the elevation of the affected body part 4. Finally, Figure 9D shows a diffuse pattern that indicates a severe degree of lymphedema. The fluorescent agent 8, along with lymphatic fluid, is poured throughout the skin and subcutaneous tissue of the patient 6. This clinical appearance is typically identified as stage 3 or 4, meaning that there is thickening of the skin accompanied by growth and deformation of the affected limb 4. The illustrated patterns A-D can be found anywhere on the body or limb 4 of the patient 6, as indicated on the right side of FIG. 9 by the dashed rectangle.

[0130] 10 shows another embodiment of the prism assembly 20 of the image capture device 10. The prism assembly 20 comprises prisms P5, P6, and P7, which are configured to split the light into red, green, and blue components, for example, towards the respective sensors D3, D4, and D5. According to a further embodiment, the prism assembly 20 is configured to split the incident light into a green, a red / blue, and an infrared component and direct them to the respective sensors D3, D4, and D5. According to yet another embodiment, the prism assembly 20 is configured to split the incident light into a visible light component directed to a red / green / blue sensor (RGB sensor), a first infrared component of a first wavelength or wavelength interval, and a second infrared component of a second wavelength or wavelength interval and direct them to the respective sensors D3, D4, and D5.

[0131] Light enters the prism assembly 20 through the arrows shown. prism An optical coating C1 is disposed between prisms P5 and P6, and an optical coating C2 is disposed between prisms P6 and P7, with each optical coating C1 and C2 having a different reflectance and wavelength sensitivity. coating At C1, the incident beam I is partially reflected (beam J) by the same face of prism P5 from which it entered, where a beam, now labeled K, is reflected once again towards filter F3 and sensor D3. beam From J beam The reflection to K is an internal reflection. Thus, filter F3 and sensor D3 receive light reflected by coating C1, and similarly, filter F4 and sensor D4 receive light reflected by coating C2. C Filter F5 and sensor D5 receive light from beam L that is reflected by prism 2 (beams M and N). Filter F5 and sensor D5 receive light from beam O that passes through the prism unimpeded.

[0132] When referring to an embodiment in which the incident light is split into red, green and blue components, the coatings and filters are selected accordingly.

[0133] In an embodiment where the incident light is separated into green, red / blue and infrared components, filter F3 can be a patterned filter (red / blue). In particular, there is an array of red and blue filters in an alternating pattern. The pattern can consist of a group of 2x2 pixels that are filtered for one particular color. Filter F4 can be a green filter, meaning that the filter only contains green filters. There is a single pixel grid, and the light received at each pixel is filtered with a green filter. Filter F5 can be an IR filter. Each pixel is filtered with an IR filter.

[0134] In general, coatings C1, C2 should be matched to filters F3, F4, F5. For example, the first coating C1 can transmit visible light while reflecting IR light such that IR light is directed towards IR filter F3. The second coating C2 may reflect red and blue light while being transparent to green light such that filter F4 should be a red / blue patterned filter and F5 should be a green filter 23.

[0135] According to a further embodiment in which the incident light is split into a visible light component (RGB), a first infrared component and a second infrared component, the coatings C1, C2 and the filters F3, F4, F5 are configured, for example, such that the sensor D4 is a color sensor (RGB sensor) for detecting all three visible light images. Further, the sensor D3 can be configured to detect fluorescent light at a first wavelength and the sensor D5 is configured to detect fluorescent light at a second wavelength.

[0136] Similarly, referring to the prism assembly 20 of FIG. 5, the coatings S1, S2, S3, S4, C1, and C2 and filters F1, F2, F3, F4, and F5 disposed in front of each one of the sensors D1, D2, D3, D4, and D5 can be configured to detect up to four fluorescence wavelengths. For example, the sensor D4 is a color sensor that detects visible light images of all three colors. The sensor D3 is for detecting fluorescence at a first wavelength, the sensor D5 is for detecting fluorescence at a second wavelength, the sensor D1 is for detecting fluorescence at a third wavelength, and the sensor D2 is for detecting fluorescence at a fourth wavelength.

[0137] All specified features, including those obtained only from the drawings, and individual features disclosed in combination with other features, are considered to be important to the invention both alone and in combination. The embodiments according to the invention can be realized by individual features or combinations of features. Features combined with the expressions "in particular" or "especially" should be treated as preferred embodiments. [Explanation of symbols]

[0138] 2. Image capture and processing device 3. Doctor 4 body parts 5 Visible light images 6 patients 7. Fluorescence Images 8 Fluorescent Agents 9. Overlay Images 10 Image capture devices 11 Surface 12 Processing Device 14 Display 16 Lighting Unit 18 Objective Lens 20 Prism Assembly 22 Fluorescence imaging unit 24 Visible light imaging unit 26 Data Links 28 Stitching Unit 30 Superposition Unit 32 Measurement Units 33 Distance Sensor 35 Internal Measuring Unit 50 Endoscopy 52 Optical Fiber 51 Fiber 54 Light source 56 Shaft 58 Fiber Bundle 60 AI units 62 CDSS 64 Input Interfaces 66 processors 68 AI models 70 Output Interface 72 User Input Interface 74 Database P1 First pentaprism P2,P4 compensation prism P3 Second Pentaprism P5, P6, P7 Dichroic Prism Assembly A Incident light beam B Light beam S1 entrance plane D1 ~ D5 Sensor C1, C2 Coating F1 ~ F5 Filter L Longitudinal direction d Distance

Claims

1. 1. A computer-based clinical decision support system configured to output a classification of lymphedema-induced fluorescence patterns based on a fluorescence image determined from measurements of fluorescence signals in tissue of a body part to which a fluorescent agent has been added, the computer-based clinical decision support system comprising: an input interface for providing said patient-specific fluorescence image as input features to an artificial intelligence model; a processor performing an inference operation in which the fluorescence image is applied to the artificial intelligence model to generate the classification of the lymphedema-induced fluorescence pattern; a user interface through which the classification of the lymphedema-induced fluorescence pattern is communicated to a user; A clinical decision support system.

2. A clinical decision support system as described in claim 1, wherein the classification of the lymphedema-induced fluorescence pattern is a stage of lymphedema severity and / or a clinical type of the lymphedema-induced fluorescence pattern.

3. The clinical decision support system of claim 1 , wherein the fluorescence image and the corresponding visible light image are provided as input features to the artificial intelligence model through the input interface.

4. the input interface is configured to measure the fluorescent signal in the tissue of the body part and is a direct link to an image capture processing device configured to image a surface of the body part, the tissue to which the fluorescent agent is added forming part of the body part, the image capture processing device comprising an image capture device; The image capture device includes: an illumination unit configured to illuminate the tissue with excitation light having a wavelength suitable for generating emission light by excitation emission of the fluorescent agent; a fluorescence imaging unit configured to capture the fluorescence image by spatially resolved measurement of the emitted light to provide the fluorescence image; a visible light imaging unit configured to capture the corresponding visible light image of a portion of a surface of the body part; 4. A clinical decision support system according to claim 3, wherein the fluorescence imaging unit and the visible light imaging unit are configured such that viewing directions and / or viewpoints of the fluorescence images and the corresponding visible light images are linked via a known relationship.

5. providing the large fluorescence image and the corresponding large visible light image as input features to the artificial intelligence model through the input interface; the fluorescence imaging unit and the visible light imaging unit are further configured to repeat capturing the fluorescence images and the visible light images to provide a series of fluorescence images and a series of visible light images; The image capture and processing device further comprises a processing device; The processing device comprises: a stitching unit for applying a stitching algorithm to the sequence of visible light images, wherein a set of stitching parameters is determined and applied to generate the large visible light image of the body part; the stitching unit is further configured to apply the stitching algorithm to the series of fluorescence images to generate the large fluorescence image; The clinical decision support system of claim 4 , wherein the stitching algorithm applied to the series of fluorescence images applies the set of stitching parameters determined when generating the large visible light image.

6. the image capture device comprises a dichroic prism assembly configured to receive fluorescent light and visible light via an input surface; the dichroic prism assembly comprises a first prism, a second prism, a first compensator prism located between the first prism and the second prism, a further dichroic prism assembly for splitting the visible light into three light components, and a second compensator prism located between the second prism and the further dichroic prism assembly; the first prism and the second prism each have a cross section with at least five corners, each corner having an internal angle of at least 90 degrees, a respective entrance face and a respective exit face, each designed such that an incident beam incident on the entrance face of each of the first prism and the second prism in a direction parallel to a normal to the entrance face is reflected twice within each of the first prism and the second prism, and exits each of the first prism and the second prism through an exit face parallel to the normal to the exit face; the normal to the entrance surface of each of the first prism and the second prism and the normal to the exit surface of each of the first prism and the second prism are perpendicular to each other; When light enters the first prism through the entrance surface, the light is partially reflected towards the exit surface of the first prism and travels a first path length from the entrance surface of the first prism to the exit surface of the first prism, the light partially enters the second prism through the first compensator prism and is partially reflected towards the exit surface of the second prism and travels a second path length from the entrance surface of the first prism to the exit surface of the second prism, The clinical decision support system of claim 3 , wherein the first prism is larger than the second prism such that the first path length and the second path length are the same.

7. 2. The clinical decision support system of claim 1, wherein the input interface is further a direct link to an electronic patient record, and wherein data regarding the patient's age, sex, height, weight, BMI (body mass index), fat mass, muscle mass, daily physical activity, employment status, skin color, medication status, vascular disease, presence or absence of varicose veins or venous edema, disease, dialysis or diabetes, blood albumin amount, renal function, liver function, cardiac function, blood hemoglobin (Hb) concentration, blood estimate, lipid metabolism rate, blood glucose concentration, urea nitrogen, ABI (ankle-brachial index) value, lymphatic function measurement data at the same location before the onset of lymphedema, endocrine information, hormone levels are provided through the input interface as further input features to the artificial intelligence model.

8. 1. A computer-implemented method for determining a classification of a lymphedema-induced fluorescence pattern using a computer-based clinical decision support system, comprising: the classification is based on a fluorescence image determined by measuring a fluorescence signal in tissue of a body part to which a fluorescent agent has been added; The computer-implemented method comprises: receiving said patient-specific fluorescence image through an input interface as input features for an artificial intelligence model; performing an inference operation by a processor, the fluorescence image being applied to the artificial intelligence model to generate the classification of the lymphedema-induced fluorescence pattern; communicating the classification of the lymphedema-induced fluorescence pattern to a user through a user interface; 4. A computer-implemented method comprising:

9. The computer-implemented method of claim 8, wherein the classification of the lymphedema-induced fluorescence pattern is a stage of lymphedema severity and / or a clinical type of the fluorescence pattern.

10. 9. The computer-implemented method of claim 8, wherein the fluorescent image and the corresponding visible light image are provided as input features to the artificial intelligence model through the input interface.

11. the input interface is configured to measure the fluorescence signal in the tissue of the body part and to receive the fluorescence image and the corresponding visible light image via a direct link to an image capture processing device configured to image a surface of the body part, the tissue doped with the fluorescent agent forming part of the body part, the image capture processing device comprising an image capture device comprising an illumination unit, a fluorescence imaging unit and a visible light imaging unit; The computer-implemented method comprises: illuminating the tissue with excitation light having a wavelength suitable for generating emission light by excitation emission of the fluorescent agent by the illumination unit; capturing the fluorescence image with the fluorescence imaging unit by spatially resolved measurement of the emitted light to provide the fluorescence image; capturing the visible light image with the visible light imaging unit by capturing the corresponding visible light image of a portion of a surface of the body part; 11. The computer-implemented method of claim 10, wherein the fluorescence imaging unit and the visible light imaging unit are configured such that viewing directions and / or viewpoints of the fluorescence images and the corresponding visible light images are linked via a known relationship.

12. providing the large fluorescence image and the corresponding large visible light image as input features to the artificial intelligence model through the input interface; the fluorescence imaging unit and the visible light imaging unit repeatedly capture the fluorescence images and the visible light images to provide a series of fluorescence images and a series of visible light images; the image capture processing device further comprises a processing device comprising a stitching unit; The computer-implemented method comprises: applying a stitching algorithm to the sequence of visible light images, a set of stitching parameters being determined and applied by the stitching unit to generate the large visible light image of the body part; and further applying the stitching algorithm to the series of fluorescence images by the stitching unit to generate the large fluorescence image. The computer-implemented method of claim 11 , wherein the stitching algorithm applied to the series of fluorescence images applies the set of stitching parameters determined when generating the large visible light image.

13. said measuring said fluorescent signal is performed on tissue to which at least a first and a second fluorescent agent have been added; The step of capturing the fluorescent image comprises: capturing a first fluorescence image in a first wavelength range produced by illuminating the tissue with a first excitation light having a first wavelength suitable for generating emission light by a first excitation emission of the first fluorescent agent; capturing a second fluorescence image within a second wavelength range produced by illuminating the tissue with a second excitation light having a second wavelength suitable for generating emission light by a second excitation emission of the second fluorescent agent; the first and second fluorescent images are provided to the artificial intelligence model as input features through the input interface; the input interface receives the first and second fluorescent images as input features for the artificial intelligence model; 12. The computer-implemented method of claim 11, wherein the processor performs the inference operation by applying the first and second fluorescence images to the artificial intelligence model to generate the classification of the lymphedema-induced fluorescence pattern.

14. The computer-implemented method of claim 8, wherein data regarding the patient's age, sex, height, weight, BMI (body mass index), fat mass, muscle mass, daily exercise, employment, skin color, medication status, vascular disease, presence or absence of varicose veins or venous edema, disease, dialysis or diabetes, blood albumin level, renal function, liver function, cardiac function, blood hemoglobin (Hb) concentration, blood estimates, lipid metabolism rate, blood glucose concentration, urea nitrogen, ABI (ankle-brachial index) value, measurement data of lymphatic function at the same location before the onset of lymphedema, endocrine information, and hormone levels are provided through the input interface as further input features to the artificial intelligence model via a direct link to an electronic patient record.

Images