System and method for automatic perfusion measurement

Abstract

To continuously measure and assess hemodynamics in medical procedures using fluorescence imaging.SOLUTION: One aspect relates to a method of automatic perfusion assessment of an anatomical structure of a subject, the method comprising administration into a vein of a bolus corresponding to less than 0.01 mg ICG / kg body weight of a first fluorescence imaging agent. Another aspect relates to a system for automatic perfusion assessment of an anatomical structure during a medical procedure of a subject comprising a controllable injection pump for holding at least one first fluorescence imaging agent, the injection pump being configured for injecting a predefined amount of the first fluorescence imaging agent into blood of the subject, wherein the system is configured for receiving and analyzing a time series of fluorescence images of a tissue of the anatomical structure following the injection of the first fluorescence imaging agent, and identifying at least one perfusion parameter of the anatomical structure on the basis of the analysis.SELECTED DRAWING: Figure 1

Description

[Background technology]

[0001] The present disclosure relates to systems and methods for automatically measuring and assessing hemodynamics in tissue of an anatomical structure of interest, particularly to continuous measurement and assessment of hemodynamics during medical procedures using fluorescence imaging, where administration of a fluorescent agent is controlled and automated.

[0002] Injection of fluorescent imaging agents, also known as fluorescent contrast agents, e.g., indocyanine green (ICG), to visualize blood flow and perfusion in anatomical structures was introduced many years ago, but clinical use of this technique is limited. Currently, the dose of fluorescent agent must be sufficient to ensure a strong visual signal that can be easily detected by the surgeon. Thus, for example, assessment of perfusion in tissues based on fluorescent agents is based on the surgeon's visual inspection. This is highly subjective and therefore may vary between surgeons. An improved quantitative analysis approach to perfusion assessment in the gastrointestinal tract is disclosed in pending application PCT / EP2017 / 082204, entitled "System and method for assessing perfusion in an anatomical structure," by the same inventors and published as WO2018 / 104552. This application is incorporated herein by reference in its entirety.

[0003] Existing fluorescence imaging methods are typically based on a few, perhaps single, perfusion measurements taken at key points during a medical procedure, such as before bowel resection and after the creation of a bowel anastomosis. To create a fluorescent signal visible to the surgeon (or other medical professional), a significant amount of fluorescent agent is required. This significant amount not only creates a visible burst of fluorescence, but also a washout period of perhaps 20–30 minutes during which the fluorescent agent is still present in the patient's blood, creating a visible background fluorescence signal. This visible fluorescence signal during this washout period, during which the fluorescent agent is removed from the blood, often prevents medical personnel from initiating a new fluorescence measurement during this washout period.

[0004] Currently, fluorescent imaging measurements involve many manual steps that "pause" the entire operating room for several minutes. Typically, it is the surgeon who decides to perform a fluorescent imaging measurement, such as assessing perfusion in tissues of an anatomical structure. First, the surgeon correctly positions the anatomical region of interest on the image received from a white-light camera, e.g., an endoscopic camera. The surgeon then switches from the standard white light to another camera capable of capturing the fluorescence emitted from the region of interest, and the surgeon has an assistant inject a fluorescent agent into a peripheral vein. After waiting approximately 30 seconds, the first fluorescent emission signal appears, and the surgeon waits several minutes until the visible fluorescent signal is deemed sufficient for evaluation. Summary of the Invention [Means for solving the problem]

[0005] Manual administration and evaluation of fluorescence imaging combined with only discrete measurements with long washout times constitute a significant practical limitation for the use of fluorescence image perfusion analysis in both elective and emergency procedures. During emergency procedures, it is important not to perform more extensive surgical intervention than necessary, as this increases operative time and postoperative morbidity. At the same time, surgeons cannot afford to leave poorly perfused organs or portions of the intestine, which can lead to tissue ischemia, necrosis, infection, anastomotic leakage, and even death [Lioit et al. 2018]. Therefore, one objective of the present invention is to make fluorescence imaging more easily integrated into emergency procedures.

[0006] In a first embodiment, the present disclosure thus relates to a system for automated perfusion assessment of an anatomical structure of a subject during a medical procedure, e.g., a surgical procedure. The system may be configured to control the injection of a predetermined amount of a fluorescent contrast agent into the subject's blood. The injection may be performed by a controllable infusion pump, which may be under the control of the system. The infusion pump may be part of a system including at least one controllable infusion pump for holding at least one fluorescent contrast agent. That is, the infusion pump may be configured to inject a predetermined amount of the fluorescent contrast agent into the subject's blood. Preferably, the system is configured to inject the predetermined amount of the fluorescent contrast agent repeatedly at regular and / or predetermined intervals. The system may further be configured to receive and analyze a time series of fluorescent images, also known as videos, of the tissue of the anatomical structure after injection of the fluorescent contrast agent. Based on this analysis, at least one perfusion parameter of the anatomical structure may be determined.

[0007] Automating fluorescence perfusion assessment reduces the amount of time the surgeon and others in the operating room spend standing around doing nothing. Automating the administration of the fluorescence agent allows for even greater use of fluorescence imaging, making perfusion assessment much easier. This may lead to changes in surgical technique, such as creating a larger resection or performing an unplanned resection, or even not performing a resection at all, even if the plan was to start with a resection.

[0008] In emergency situations, operative time is a limiting factor. Also, compared to elective surgery, the surgical planning is necessarily more spontaneous. During such emergency procedures, many perfusion assessments of the same or different tissues may be required in a short period of time. Although one perfusion measurement can be performed within a reasonable time, multiple measurements quickly lengthen the operative time and become impractical. This is an obstacle and the reason why these measurements are not currently routinely used.

[0009] The disclosed system may therefore further be configured to control an infusion pump to inject a small initial bolus of fluorescent contrast agent and then analyze the resulting fluorescence emission. This small initial bolus, preferably in combination with a saline flush, should be selected depending on the situation, i.e., the type of fluorescent agent used, and the patient's morphology, e.g., age, weight, height, etc. However, an amount of less than 0.01 mg / kg of subject weight, i.e., less than 0.5 mg for a 50 kg patient and less than 1 mg for a 100 kg patient, is typically a desirable starting point, particularly when ICG is used.

[0010] The inventors have realized that quantifiable fluorescence emissions from tissues of anatomical structures are much smaller than those visible to the human eye, for example, by using computer image analysis. That is, the fluorescence signal intensity only needs to be large enough to be measurable by computer image recognition and analysis, rather than by visual inspection by a surgeon, allowing for much smaller doses, e.g., microdoses. That is, microdoses of fluorescent agents can be administered to a subject, and perfusion parameters, for example, can be determined therefrom.

[0011] The minimum bolus dose that results in quantifiable fluorescence can be estimated based on the circumstances. However, to find a more accurate minimum effective bolus dose, the system of the present disclosure may be further configured to identify a subject-specific minimum effective bolus dose of a fluorescent contrast agent by controlling an infusion pump to inject a series of boluses of the fluorescent contrast agent in varying amounts according to predetermined criteria, e.g., gradually increasing or decreasing with predetermined intervals between each bolus, and analyzing the fluorescence emission from the anatomical structure after each bolus injection to identify the size of the minimum effective bolus dose that results in quantifiable fluorescence emission from the anatomical structure.

[0012] By applying microdoses of fluorescent agents, the minimum time between successive measurements can be significantly reduced. Identifying the actual subject-specific minimum bolus dose that results in quantifiable fluorescence also ensures the minimum washout time, which can minimize the period between successive fluorescence measurements, since smaller doses are cleared from the blood more quickly.

[0013] After finding an appropriate bolus size that results in a quantifiable fluorescent signal and a short washout time that allows repeated injections of the fluorescent contrast agent and repeated fluorescent measurements, the system can be configured to perform automated measurements of perfusion parameters at a predetermined frequency, determined by, for example, the washout time. The disclosed system can therefore further be configured to automatically: 1) control an infusion pump to inject a series of predetermined bolus doses of the fluorescent contrast agent, i.e., predetermined bolus doses, such as a minimum effective bolus dose, with predetermined intervals between each bolus; and 2) determine at least one perfusion parameter of the anatomical structure after injection of each bolus.

[0014] Surgeons and other medical professionals are currently provided with continuous perfusion assessment of relevant anatomical structures in the operating room. The disclosed system and method therefore open up the possibility of providing continuous fluorescence imaging measurements that can provide a wide range of perfusion parameters as a kind of background information during the course of a medical procedure. That is, even if the surgeon returns to white-light imaging during the actual surgical procedure, fluorescence measurements can be performed automatically in the background without manual intervention. In this way, repeated perfusion measurements offer the possibility of tracking the evolution of perfusion parameters over time, thereby providing medical personnel with useful information about perfusion continuously and also from a temporal perspective. The use of automation and microdosing in fluorescence perfusion measurements therefore opens up an entirely new range of applications, including routine use for the continuous assessment of the viability of tissues and organs, e.g., thyroid and parathyroid glands, hepatic and bile ducts, reproductive organs and bladder, tumors and their localizations, including lymph nodes and possible metastases, in both emergency and elective surgery, as well as the assessment of skin / tissue / vascular perfusion in various medical procedures, e.g., wound and wound healing assessment.

[0015] When measuring fluorescence perfusion using traditional methods, i.e., performing a single measurement at one or more different time points during surgery, the measurement is always subject to measurement "noise." Both the automation of fluorescence perfusion measurements and the reduction of the minimum interval between successive measurements make it feasible to perform multiple measurements of the same region of interest over the course of a medical procedure. Providing multiple measurements is a significant advantage because it reduces the effects of random noise that can arise from stochastic diffusion of the fluorescent agent, physiological fluctuations in blood flow, and the distribution of the microcirculation in any anatomical region. Overall, multiple measurements of the same region of interest lead to better and more physiologically accurate perfusion measurements.

[0016] If the surgeon wishes, he or she can still take and store "normal" / full-dose fluoroscopic perfusion measurements that provide a visible signal at key points in the surgical procedure. These can be used, for example, as documentation of the quality of the procedure in the electronic patient record. After scheduling the initial measurements, the surgeon can continue the medical procedure with only minimal interruptions to interpret incoming perfusion values ​​or to change regions under continuous evaluation.

[0017] The inventors further realized that measurement and analysis of repeatable bolus injections can be further extended from interpretation and quantification of single inflow and / or outflow phases to analysis of oscillatory fluorescence dynamics, which may reveal physical perfusion characteristics previously unattainable without invasive procedures.

[0018] The systems and methods of the present disclosure are directed to small boluses, such as the minimum boluses disclosed herein. The bolus may be configured for repeated injections of a given bolus dose at regular intervals. These boluses may, for example, ... lead to periodic variations when measured depending on the injection interval, resulting in a regular oscillating curve approximating an oscillating curve, e.g., a sinusoidal curve. In such a curve, the measured intensity signal is expected to increase with the influx of fluorescent contrast agent from a given bolus, then decrease during the washout phase of the bolus until increasing again upon administration of a subsequent bolus, resulting in a periodic (sinusoidal) pattern.

[0019] Thus, the present disclosure further relates to a computer-implemented method for detecting perfusion changes in an anatomical region of interest of a subject by hemodynamic imaging of at least a portion of the anatomical region of interest in video acquired from the subject. In one embodiment, the method includes performing image analysis of at least one video sequence acquired during and / or after delivery of multiple boluses containing a fluorescent contrast agent to the subject. In this case, it is advantageous if the multiple boluses are delivered according to a predetermined pattern, e.g., in terms of frequency and / or dose, as also described in further detail in this disclosure. Based on the image analysis, subsequent perfusion parameters in one or more regions of interest can be calculated. That is, perfusion parameters can be calculated continuously, simultaneously with bolus delivery, while multiple doses are being administered to the subject. The multiple perfusion parameters provided as a function of time (and bolus administration) allow subsequent perfusion parameters to be monitored to identify changes in perfusion in the region(s) of interest. These changes in perfusion may indicate an abnormality.

[0020] Preferably, the disclosed system is configured to recognize vibration intensity curves, e.g., parameters of frequency, phase, and / or amplitude. This trained system can then predict both the direction and regularity of the dynamics of subsequent signals. The system preferably uses measurements to recognize vibration patterns, so that the system can then detect discrepancies between measured and expected values. The system does not necessarily need to continuously measure the anatomical region of interest; instead, it may only measure at sporadic intervals, e.g., in the case of an anatomical region of interest that drifts in and out of focus of the recorded image. In these situations, the expected phase of the vibration pattern at the measured interval may be compared with the measured phase. The measurements may also be continuously used to update the detected pattern, i.e., the expected value. Alternatively or additionally, infusion parameters, e.g., bolus frequency, dose, and flow rate, may be used to identify the expected value, i.e., the vibration pattern.

[0021] Deviations from the expected sinusoidal pattern may be caused, for example, by the onset of ischemia in at least a portion of the anatomical structure visible on the image, or by local changes in perfusion to a given region. An illustration showing this change in dynamics due to the onset of ischemia in a human subject is shown in FIG. 12A, with a more focused, zoomed-in view in FIG. 12B. As can be seen, it is possible to detect a transition from a regular oscillating fluorescent signal to an ischemic flat line. However, it should be noted that other measurement patterns may arise in addition to the ischemic flat line as perfusion of the anatomical structure of interest changes. One example is venous occlusion, in which blood outflow from the anatomical region is blocked or reduced, leading to changes in oscillatory dynamics due to overcrowding or pooling of fluorescent agent in a given region. As seen in FIG. 13C, the periodic oscillations cease, but the result is not a flat line.

[0022] Preferably, the system of the present disclosure includes tracking means and is capable of running independently in the background, during which the surgeon merely touches the visible white light signal during detection of the onset of ischemia, which may be defined, for example, by an extended time of the ischemic state, and is therefore only interrupted / notified by a warning signal.

[0023] Following on from the above disclosure, the present disclosure further relates to a method for automated perfusion assessment of a subject's anatomical structure, the method comprising intravenously administering a bolus dose of approximately 1 / 10 the usual dose used for perfusion assessment. For indocyanine green (ICG), a typical bolus dose is 0.1-0.3 mg / kg body weight. In accordance with the present disclosure, a bolus dose of a first fluorescent contrast agent of less than 0.01 mg / kg body weight may be used. For other fluorescent contrast agents described herein, the bolus dose may be similarly reduced in accordance with the present disclosure.

[0024] The disclosed systems and methods can be used, for example, in the field of visceral surgery, where quantifying circulation through tissues can be crucial during surgery, i.e., in left-sided colon and rectal resections, gastric transposition after esophagectomy, free small bowel transplants for interposition, anastomoses, etc. The disclosed approach can also be suitable for detecting secondary perfusion defects in cases of strangulated hernia or adhesive ileus. In cardiac surgery, the disclosed systems and methods can be used to examine the efficiency of coronary artery bypass and measure perfusion during the procedure. In the field of plastic surgery, it is possible to, for example, continuously monitor the perfusion of transplanted skin flaps, as well as to assess tissue damage in cases of trauma and wound healing, such as chronic wounds.

[0025] Another aspect of the present disclosure relates to continuous perfusion assessment, which involves repeated injections of a fluorescently activated agent and monitoring the resulting oscillation curve. In addition to detecting unexpected changes in perfusion, the system can be used to assess the perfusion area of ​​an artery. As an example, a surgeon may consider cutting an artery as part of a surgical procedure. Before cutting the artery, the surgeon may temporarily restrict perfusion through that artery, and the disclosed approach may enable visualization of the perfusion area of ​​that artery in a short period of time, e.g., less than one minute. This may be useful information for the surgeon during the ongoing surgical procedure. Similarly, the system may be used to assess the drainage area of ​​a vein or group of veins, lymphatic vessels, lymph nodes, or other portions of the circulatory and / or lymphatic pathways. By temporarily restricting blood flow through a blood vessel, the blood pools in the anatomical area normally drained by that vessel or group of vessels. This allows visualization of the anatomical area drained by the vessel in a relatively short period of time, e.g., less than two minutes. This can provide important information to surgeons, for example, during ongoing surgical procedures, in areas such as general surgery and plastic surgery, including wound and reconstructive surgery.

[0026] A further aspect of the present disclosure relates to a computer program, e.g., a computer program recorded on a storage medium that is loaded into the memory of a computer, or a system as disclosed herein that causes a computer / system to perform any of the steps of the methods disclosed herein.

[0027] A further aspect of the present disclosure relates to an imaging system, i.e., an endoscopic imaging system, including a processing device configured to perform any of the steps of the methods disclosed herein. The present invention provides, for example, the following. (Item 1) 1. A system for automated perfusion assessment of an anatomical structure during a medical procedure on a subject, comprising: a controllable infusion pump for holding at least one first fluorescent contrast agent, the infusion pump configured to inject a bolus dose corresponding to a predetermined amount of the first fluorescent contrast agent into a vein of the subject; the system configured to receive and analyze time-series fluorescent images of tissue of the anatomical structure after the injection of the first fluorescent contrast agent, and to identify at least one perfusion parameter of the anatomical structure based on the analysis. (Item 2) 2. The system of claim 1, wherein the bolus dose corresponds to less than 0.01 mg of ICG / kg body weight of the first fluorescent contrast agent. (Item 3) 2. The system of claim 1, wherein the bolus dose corresponds to less than 0.5 mg of the first fluorescent contrast agent of ICG. (Item 4) Item 10. The system of item 1, wherein the fluorescent agent is ICG and the amount of ICG in the bolus dose is less than 0.01 mg / kg body weight. (Item 5) Item 10. The system of item 1, wherein the fluorescent agent is ICG and the amount of ICG in the bolus dose is less than 1 mg of ICG or less than 0.5 mg of ICG. (Item 6) 10. The system of any preceding claim, wherein the system is configured to inject the bolus at intervals between 5 and 600 seconds, e.g., between 15 and 300 seconds, e.g., between 45 and 210 seconds, e.g., between 90 and 120 seconds. (Item 7) The system comprises administering a subject-specific minimum effective bolus dose of a fluorescent imaging agent, - controlling the infusion pump to inject a series of boluses of varying amounts of fluorescent contrast agent with predetermined intervals between each bolus administration; - analyzing the fluorescence emission from said anatomical structure after injection of each said bolus; - identifying the minimum bolus dose size that provides a quantifiable fluorescence emission from the anatomical structure; Item 10. The system of any preceding item, configured to identify by: (Item 8) 8. The system of claim 7, wherein the quantifiable fluorescent emission from the anatomical structure corresponds to a fluorescent emission that allows for identification of a perfusion gradient. (Item 9) The system of any of the preceding items, wherein the system is configured to identify a subject-specific interruption interval defined as the period from the rise in the fluorescence slope until the fluorescence emission falls below a number of standard deviations, e.g., 20, 10, or 5 standard deviations, below background noise. (Item 10) The system automatically: 1) controlling the infusion pump to inject a series of predetermined bolus doses of fluorescent contrast agent, i.e., predetermined bolus doses such as the minimum effective bolus dose, with predetermined intervals between each bolus dose; 2) The system of any of the preceding items, configured to determine at least one perfusion parameter of the anatomical structure after injection of each bolus. (Item 11) The system automatically: 1) controlling the infusion pump to inject a series of boluses of varying amounts of fluorescent contrast agent with predetermined intervals between each bolus administration; 2) A system according to any of the preceding items, configured to determine at least one perfusion parameter of the anatomical structure after injection of each bolus dose. (Item 12) 13. The system of claim 1, further comprising: a second fluorescent contrast agent different from the first fluorescent contrast agent; and a second fluorescent contrast agent configured to inject one or more boluses having a predetermined amount of the second fluorescent contrast agent into the subject's blood. the system is configured to identify the at least one perfusion parameter in one or more regions of interest located in the anatomical structure and, optionally, in adjacent anatomical structures, the regions of interest being optionally selectable by a user of the system. A system described in any of the line items. (Item 14) 10. The system of claim 1, further comprising: at least one light source configured to provide excitation light to stimulate fluorescent emission from the first and / or second fluorescent agents in the anatomical structure; and an imaging device configured to record the fluorescent emission from the anatomical structure in the time series. (Item 15) 10. The system of claim 1, further configured to track movement of at least a subsection of the anatomical structure in the time series of images and to correlate the movement so that at least the first region of interest corresponds to the same subsection of the anatomical structure in the images, wherein movement tracking is provided by free image tracking and / or object-based tracking. (Item 16) 1. A method for automated perfusion assessment of an anatomical structure of a subject, the method comprising: intravenously administering a bolus dose of a first fluorescent contrast agent equivalent to less than 0.01 mg ICG / kg body weight; acquiring and analyzing time-series fluorescent images of tissue of the anatomical structure after injection of the first fluorescent contrast agent; and determining at least one perfusion parameter of the anatomical structure based on the analysis. (Item 17) 17. The method of claim 16, wherein the contrast agent is injected by a controllable injection pump. (Item 18) 18. The method of any of the preceding items 16-17, wherein the contrast agent is injected as a series of bolus doses with predetermined intervals between subsequent bolus doses. (Item 19) 19. The method of any of the preceding items 16-18, wherein the fluorescence emission from the anatomical structure is measured after injection of each bolus. (Item 20) 20. The method of any of the preceding items 16 to 19, wherein the bolus comprises gradually increasing or gradually decreasing amounts of the contrast agent. (Item 21) 21. The method of claim 20, wherein the amount increases or decreases in 10% increments from one bolus dose to a subsequent bolus dose. (Item 22) 22. The method of any of the preceding items 16-21, wherein a minimum bolus dose that results in a quantifiable fluorescence emission indicative of perfusion of the anatomical structure is identified after administering a series of increasing or decreasing bolus doses. (Item 23) 23. The method according to any of the preceding items 16 to 22, wherein the interval between the bolus administrations is between 5 and 600 seconds, for example, between 15 and 300 seconds, for example, between 45 and 210 seconds, for example, between 90 and 120 seconds. (Item 24) 24. The method of any of the preceding items 16-23, wherein the interval between bolus administrations is long enough to allow measurement of the perfusion gradient for each bolus administration in the anatomical structure, preferably the perfusion gradient includes a gradient onset and a washout gradient. (Item 25) 25. The method of any of the preceding items 16 to 24, wherein a volume of isotonic fluid (such as saline) is injected immediately after injection of the bolus of fluorescent contrast agent, for example, the volume is 1 to 20 mL, for example, 2.5 to 15 mL, for example, 5 to 10 mL. (Item 26) The amount of fluorescent contrast agent per bolus is between 0.001 and 0.01 mg ICG / kg body weight. 26. The method of any of the preceding items 16 to 25, wherein the bolus corresponds to a dose of ICG per kg body weight, for example, between 0.001 and 0.01 mg. (Item 27) 27. The method according to any one of the preceding items 16 to 26, wherein the initial amount of the fluorescent contrast agent corresponds to at least 0.001 mg ICG / kg body weight. (Item 28) 28. The method of item 27, wherein the amount of subsequent bolus doses increases or decreases from one bolus dose to the next by at least 0.001 mg of ICG / kg body weight. (Item 29) 29. The method according to any one of the preceding items 16 to 28, wherein the bolus dose is between 0.5 μL and 10 mL of liquid volume, for example, 0.5 to 5 mL. (Item 30) 30. The method of any of the preceding items 16-29, wherein a second fluorescent imaging agent is administered, said second fluorescent imaging agent having an emission maximum that differs from the emission maximum of said first fluorescent imaging agent by at least 50 nm. (Item 31) 1. A computer-implemented method for detecting changes in perfusion of an anatomical region of interest of a subject by hemodynamic imaging of at least a portion of the region of interest in an image acquired from the subject, the method comprising: - performing image analysis of at least one video sequence acquired during and / or after a plurality of boluses comprising a fluorescent contrast agent are delivered to the subject, the plurality of boluses being delivered according to a predetermined pattern, e.g., with respect to frequency and / or dosage; - calculating subsequent perfusion parameters in one or more regions of interest based on said image analysis; - monitoring the subsequent perfusion parameters to identify changes in perfusion in the region(s) of interest; The method comprising: (Item 32) Item 32. The method according to Item 31, comprising the steps of any one of Items 16 to 30. [Brief explanation of the drawings]

[0028] [Figure 1] A, C, and E show example intensity curves after a subject received a bolus of ICG, and B, D, and F show the corresponding intensity curves, where the hemodynamic parameters perfusion gradient, perfusion onset, gradient end maximum intensity, washout gradient, washout onset, and washout gradient end were calculated and shown on these graphs. [Figure 2] A-F show three examples illustrating the approach disclosed herein for identifying the point at which the perfusion gradient begins, i.e., gradient onset. B, D, and F are enlarged views of the gradient onset areas in A, C, and E, respectively. [Figure 3] AF show three examples illustrating the approach disclosed herein for identifying perfusion gradients based on histogram data. [Figure 4] A-F show three examples illustrating the approach to defining and identifying maximum gradient strengths disclosed herein. B, D, and F are enlarged views of the maximum strengths of the curves in A, C, and E, respectively. [Figure 5] A to F show three examples illustrating the approach to analyzing washout of a fluorescent contrast agent disclosed herein. B, D, and F are enlarged views of the areas of A, C, and E where ICG is washed out, respectively. [Figure 6] A-D show the analysis of two additional fluorescence measurements using ICG, demonstrating the robustness of the analytical approach of this disclosure. [Figure 7] Shown are output video frames acquired during bowel surgery, four different regions of interest, and their analysis. [Figure 8A] 1 shows a still image from a conventional image sequence acquired prior to resection of a patient's colon, showing the small intestine (lower portion) and colon (upper portion). [Figure 8B]8A shows a fluorescent image of substantially the same subsection of the gastrointestinal tract as FIG. 8A, but acquired at a later time, i.e., after the patient has been injected with a bolus of fluorescent contrast agent (ICQ). [Figure 9A] The intensity curves obtained in the ROIs of FIG. 8B and the perfusion gradients calculated according to the approach disclosed herein, i.e., the perfusion gradients of the colon and small intestine before resection, are shown. [Figure 9B] 9A shows the perfusion gradients of the small intestine (left) and colon (right) from FIG. 9A, where the gradients have been normalized to the perfusion gradient of the small intestine. [Figure 10A] 8B shows substantially the same subsection of the gastrointestinal tract as FIG. 8A, but an orthogonal view acquired after resection of the colon and before anastomosis. [Figure 10B] FIG. 10B is a fluorescence image corresponding to the image in FIG. 10A after a bolus of ICG was injected. Five ROIs are shown in this image. [Figure 11A] 10A and 10B show the intensity curves obtained from the measurements shown. [Figure 11B] Shown are perfusion gradients of the small intestine (left) and colon (blue, green, and yellow on the right) from FIG. 11A, where the gradients have been normalized to the perfusion gradient of the small intestine. [Figure 12A] 1 shows an oscillatory time intensity fluorescence curve, where oscillations are interrupted by the onset of ischemia in a human subject. [Figure 12B] The graph shows an expanded time interval around t=3800 seconds before the onset of ischemia occurs. [Figure 12C] Idealized data for ischemic and non-ischemic cases are shown. [Figure 12D] Idealized data is shown where only a portion of the oscillating time-intensity fluorescence curve can be detected. [Figure 13A] Serial measurements of a human subject infused with a microbolus dose are shown. [Figure 13B] 13B shows an enlargement of the interval shown in FIG. 13A. [Figure 13C] Measurements are shown for a person with a possible venous occlusion where blood flow is only partially restricted. DETAILED DESCRIPTION OF THE INVENTION

[0029] To identify a subject-specific minimum effective bolus dose, a background level of the setting can be advantageously identified so that a minimum quantifiable fluorescent signal can be identified. Thus, the system of the present disclosure is advantageously configured to: 1) receive time-series images of tissue of the anatomical structure of interest before injection of a fluorescent agent, and 2) identify a background noise level therefrom. One or more regions of interest (ROIs) can be selected before or after the time-series images. This ROI selection can be manual by the user, automatically provided by the system, or semi-automatically provided, with the system proposing multiple ROIs and allowing the user to edit and / or move the proposed ROIs. A background noise level can be identified for each ROI.

[0030] As mentioned above, a minimum effective bolus dose can be identified, and in particular, a subject-specific minimum effective bolus dose can be identified. The size of the minimum effective bolus dose can be determined, for example, based on the background noise level, in particular the standard deviation of the background noise level. For example, the minimum effective bolus dose can be determined as the bolus dose that results in a maximum intensity that is a predetermined factor multiplied by the standard deviation of the background noise level. Note that the intensity of the background noise level may or may not be close to zero. The background noise may also vary (e.g., as a result of a long series of microboluses). However, the change in background noise is preferably on a much larger time scale, e.g., at least two times, more preferably at least four times, even more preferably at least six times, and most preferably at least ten times, the intensity change observed with a single injection.

[0031] Values ​​from several ROIs may be provided, and the intensities, especially the maximum intensities, between different ROIs may vary significantly. The minimum effective bolus dose can be identified as the bolus dose that results in a maximum intensity multiplied by a predetermined factor and the standard deviation of the background noise level. In one embodiment, this should be applied to all ROIs. However, there may be situations where a selected ROI does not provide a viable signal, for example, if there is no perfusion in that area. In such situations, one or more of the ROIs can be left out of the minimum effective bolus dose evaluation.

[0032] The predetermined factor may be at least 5, more preferably at least 10, even more preferably at least 25, and most preferably at least 50.

[0033] Once a minimum effective bolus dose, e.g., a subject-specific minimum effective bolus dose, is identified, the actual bolus dose eventually used in the sequential and repeated measurement procedures may be this minimum effective bolus dose. However, to ensure that a usable signal is provided, the actual bolus dose used may be determined to be a percentage of the minimum effective bolus dose. The actual bolus dose used may be less than the identified feasible bolus dose, but in particular, the actual bolus dose used may be greater than the minimum effective bolus dose, such as at least 125%, more preferably at least 150%, even more preferably at least 200%, and most preferably at least 300% of the minimum effective bolus dose. However, it is important to note that the actual bolus dose may change over time and need not be set to a percentage of the minimum effective bolus dose. Instead, the actual bolus dose may change over time. This may be the case, for example, if a large actual bolus dose is initially infused, followed by a smaller actual bolus dose. Thus, an initial large actual bolus dose, which may be between 125% and 375% of the minimum effective bolus dose, more preferably between 150% and 350% of the minimum effective bolus dose, even more preferably between 175% and 325% of the minimum effective bolus dose, and most preferably between 200% and 300% of the minimum effective bolus dose, may be used to saturate the intensity signal. The initial large actual bolus dose may then be followed by repeated injections of smaller actual bolus doses at a percentage of the minimum effective bolus dose, such as near 100% of the minimum effective bolus dose.

[0034] When identifying a subject-specific minimum effective bolus dose, the boluses are injected with an interval between each bolus, preferably a predetermined interval, but optionally adjustable based on the measured fluorescence readings. The interval can also be customized to suit the specific situation, e.g., the subject. At least initially, the interval between injections is typically 20-60 seconds, and optionally 20-40 seconds, or even 20-30 seconds. In other examples, the interval between injections, at least when using ICG, is typically 5-600 seconds, and optionally 30-300 seconds, or even 90-120 seconds, since this is the typical time it takes for the gradient to rise and the intensity to decrease sufficiently again.

[0035] The time from injection of the fluorescent agent to when the fluorescent emission can be detected will vary and may generally depend, for example, on the perfusion and blood flow characteristics of the tissue, but may also depend on individual patient factors, such as the anatomy, tissue components and interaction dynamics, or possibly other factors, such as the fluorescent agent. While it is possible to estimate this time, it is advantageous to know the specific time. Thus, the systems of the present disclosure may be further configured to identify a subject-specific conversion time, defined as the time from injection of a bolus dose of fluorescent contrast agent to the rise of a fluorescence gradient in the fluorescent emission from the anatomy.

[0036] The systems of the present disclosure may further be configured to identify a subject-specific cut-off interval, i.e., a subject-specific time during which detectable fluorescence emission is present, defined as the time from the rise in the fluorescence slope until the fluorescence emission is comparable to background noise, or until the fluorescence emission falls below a number of standard deviations, e.g., 20, 10, or 5 times the SD, below background noise.

[0037] The systems of the present disclosure may further be configured to identify a subject-specific rise+fall interval, defined as the time from the rise of the fluorescence slope, passing through maximum intensity, until the fluorescence emission falls below 50% of maximum intensity, or more preferably, below 25%, even more preferably, below 10%, even more preferably, below 5%, and most preferably, below 1%.

[0038] The disclosed system may further be configured to identify a subject-specific injection interval, defined as the time from injection of the fluorescent agent through the rise of the fluorescence gradient, passing through maximum intensity, and until the fluorescence emission falls below 50% of maximum intensity, or more preferably, until it falls below 25%, even more preferably, until it falls below 10%, even more preferably, until it falls below 5%, and most preferably, until it falls below 1%. Once the fluorescence emission falls below a certain intensity, a new bolus can be detected to quantify the fluorescence emission. That is, the subject-specific injection interval can be considered the waiting time required before the next injection of a microdose of fluorescent agent. However, because it takes time for the subsequent bolus injection to reach the anatomical structure, it may also be the subject-specific rise and fall interval defined above that can be considered the waiting time before the next injection of a microdose of fluorescent agent.

[0039] The properties of existing fluorescent agents such as ICG are fairly well known, and the waiting time until the next injection can also be predetermined. The rise and fall interval defined above is typically about 20 to 60 seconds, and may be 20 to 40 seconds or 20 to 30 seconds. In another embodiment, the rise and fall interval defined above is typically about 5 to 600 seconds, and may be 30 to 300 seconds or 90 to 120 seconds.

[0040] Instead of waiting for the fluorescent agent to wash out of the blood, it may be possible to determine perfusion parameters after the injection of a new bolus of fluorescent agent before the previous bolus is removed / washed out. This may be possible, in particular, if the next bolus dose is greater than the previous bolus dose, ensuring that incremental amounts of fluorescent agent are also administered. Thus, the disclosed system may be configured to automatically: 1) control an infusion pump to inject a series of increasing or decreasing amounts of fluorescent contrast agent boluses, with predetermined intervals between each bolus dose, and 2) determine at least one perfusion parameter of the anatomical structure after each bolus injection. The incremental amounts may, for example, start at 100% and linearly increase by 10%, e.g., 110%, 120%, 130%, 140%, etc. Alternatively, the incremental amounts may be 25% increments, i.e., 100%, 125%, 150%, 175%, etc. Alternatively, a 50% increase, i.e., 100%, 150%, 200%, 250%, etc. Alternatively, a 100% increase, i.e., 100%, 200%, 300%, 400%, etc. Alternatively, an exponential increase, e.g., 100%, 200%, 400%, 800%, etc.

[0041] The amount of gradual decrease may, for example, start at 200% and decrease linearly by 10%, e.g., 190%, 180%, 170%, 160%, etc. Alternatively, decrease by 25%, i.e., 200%, 175%, 150%, 125%, etc. Alternatively, decrease by 50%, i.e., 250%, 200%, 150%, 100%, etc. Alternatively, decrease by 100%, i.e., 400%, 300%, 200%, 100%, etc. Alternatively, decrease exponentially, e.g., 800%, 400%, 200%, 100%, etc. .

[0042] Automated Systems The systems of the present disclosure may be configured to identify the at least one perfusion parameter in one or more regions of interest located in the anatomical structure and, optionally, in adjacent anatomical structures, and the systems may be configured such that these regions of interest can be selected by a user of the system.

[0043] The system of the present disclosure may further be configured to transfer the at least one perfusion parameter for display on a display, which may be a white light image with the parameter(s) superimposed thereon, allowing medical personnel to follow the progress of the perfusion assessment during a medical procedure.

[0044] The systems of the present disclosure may further include at least one processor and memory having instructions stored thereon that, when executed by the one or more processors, cause the system to perform the operations disclosed herein.

[0045] An infusion pump, also known as a dosing pump, can be part of the automated perfusion evaluation system of the present disclosure. The infusion pump can be configured to deliver a time-controlled, scheduled infusion of a fluorescent agent to a target, e.g., a subject. Control of the infusion pump can be provided by a programmable and / or controllable controller. In this manner, the controller can be configured to automatically operate the infusion pump to schedule bolus infusions of the fluorescent agent at controllable and / or predetermined amounts of the fluorescent agent at controllable and / or predetermined infusion intervals in accordance with the bolus doses disclosed herein. That is, the system may simply be configured to control the infusion pump, e.g., by a controller configured to control the infusion pump. Controllable infusion pumps are commercially available, e.g., the KDS single-syringe pump (series 100) from Sigma-Aldrich, the Legato 212 two-syringe pump from World Precision Instruments, or the Fusion controllable syringe pump from Chemyx.

[0046] The automated system of the present disclosure may be external to an existing fluorescence imaging system, i.e., it may simply be configured to control an infusion pump and receive time-series fluorescence images from an existing system for analysis, and may be an external system, such as an endoscopic and / or laparoscopic setup, such as the Novadaq Pinpoint endoscopic fluorescence imaging system or the Novadaq Spy-Phi portable handheld imaging system, with the imaging device and optionally a light source integrated into the system. Existing laparoscopic systems, for example, from Olympus, Stryker, or Karl Storz, or surgical robots from Intuitive are also options.

[0047] However, the system of the present disclosure may also be a more complete fluorescence imaging system, for example, an endoscopic and / or laparoscopic system that may include one or more light sources for fluorescence excitation and their control. That is, in a further embodiment, the system further includes at least one light source configured to provide excitation light for inducing fluorescence emission from the first and / or second fluorescent agents in the anatomical structure. For example, a near-infrared light source (e.g., for ICG) that can be directly attached to a camera. Similarly to the imaging device, the system may include an imaging device configured to record at least one time series of fluorescence emission from the anatomical structure. For example, a digital video camera that allows for real-time recording of the emission of the fluorescent agents. This means that perfusion can be assessed and documented in real time. The imaging device may further receive and / or display a normal image of the anatomical structure. The imaging device may also be configured for white light imaging, i.e., while simultaneously recording the fluorescent signal with another camera. This may be provided by an additional camera within the imaging device.

[0048] One embodiment of the present disclosure therefore relates to a system for automated perfusion assessment of an anatomical structure during a medical procedure on a subject, including: a controllable infusion pump for holding at least one fluorescent contrast agent; at least one light source configured to provide excitation light to stimulate fluorescence emission from said fluorescent agent in said anatomical structure; an imaging device configured to record at least one motion image sequence of fluorescent emissions from an anatomical structure; wherein the system automatically controls the infusion pump, the light source, and the imaging device to: - injecting a predetermined amount of said fluorescent contrast agent into the blood of a subject; - to induce and analyze fluorescence emission from the anatomical structure after injection of a fluorescent contrast agent, - configured to determine at least one perfusion parameter of the anatomical structure based on the analysis.

[0049] Intraoperative fluorescence imaging By using near-infrared light from a surgical microscope and acquiring images of near-infrared fluorescence excited by a fluorescent angiographic agent administered intravenously as a tracer, perfusion (blood flow, etc.) can be imaged and evaluated in real time during surgery. This allows the perfusion status during surgery to be confirmed in real time.

[0050] The systems and methods of the present disclosure can enhance information on tissue properties, including the location of superficial and deep blood vessels, particularly when different fluorescent agents are used, as selective selection of different fluorescent agents provides the option of having perfusion information from different depths of the tissue.

[0051] During medical procedures, such as diagnostic, screening, examination, and / or surgical procedures involving fluorescence imaging, a solution containing a fluorescent contrast agent such as ICG is injected intravenously, and the molecules are excited by an infrared light source, e.g., a laser in the infrared wavelength range, e.g., a wavelength near 780 nm. Fluorescence light with a wavelength near 830 nm is then emitted from the excited contrast agent molecules and can be recorded by an imaging device, e.g., a camera. Because the excitation intensity is typically much greater than the fluorescence intensity, a filter can be installed to block the excitation light. The excitation intensity can be approximately 1 W per angular radiance, while the fluorescence power per pixel can be approximately 0.15 pW. Despite the difference of several orders of magnitude, a good signal-to-noise ratio (SNR) can be achieved. The recorded fluorescence provides an image of the perfusion of the imaged tissue, allowing for the visualization of deeper blood vessels due to the ICG's penetration depth of 5-10 mm. Because ICG molecules bind to proteins in the blood, the images contain information about the level of perfusion, but that information can be difficult for surgeons to quantify during surgery if they only see the acquired images.

[0052] In the systems and methods of the present disclosure, the fluorescent imaging agent is selected from the group consisting of indocyanine green (ICG) and fluorescein isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, o-phthalaldehyde, fluorescamine, rose bengal, trypan blue, fluorogold, green fluorescent protein, flavin, methylene blue, porphysomes, cyanine dyes, IRDDye800CW, CLR1502 in combination with a targeting ligand, OTL38 in combination with a targeting ligand, or combinations thereof.

[0053] Indocyanine green (ICG) is a cyanine dye used in medical diagnostics and It is the most common dye used for blood flow assessment. It has a spectral absorption peak near 800 nm. These infrared frequencies penetrate retinal layers, allowing ICG angiography to image circulation patterns deeper than fluorescein angiography. ICG binds tightly to plasma proteins and becomes confined to the vasculature. When administered intravenously, it is excreted from the body by the liver into the bile with a half-life of approximately 3–4 minutes, depending on liver performance. ICG sodium salt is usually available in powder form and can be dissolved in various solvents. 5% (or less than 5% depending on the batch) sodium iodide is usually added to increase solubility. Sterile lyophilized water-ICG solutions have been approved as diagnostic agents for intravenous use in many European countries and the United States under the names ICG-Pulsion, IC-Green, and VERDYE.

[0054] The absorption and fluorescence spectra of ICG lie in the near-infrared region. Lasers with wavelengths around 780 nm are usually used for excitation. At this wavelength, it is possible to detect the fluorescence of ICG by filtering out scattered light from the excitation beam.

[0055] Although ICG is classified as low-toxicity, its administration is not without risk, for example, during pregnancy. ICG is known to decompose under the influence of UV light into toxic waste products, creating several unknown substances. Thus, it is in the patient's best interest that the dose of ICG used in the process of fluorescence imaging be minimized, as demonstrated herein.

[0056] Fluorescein is another dye that is widely used as a fluorescent tracer in many applications. Fluorescein has an absorption maximum at 494 nm and an emission maximum (in water) at 512 nm. Thus, the absorption and emission wavelengths of these two dyes are several hundred nanometers apart, making them suitable for use in combination with ICG.

[0057] Automated perfusion assessment method As previously mentioned, the present disclosure further relates to a method for automated perfusion assessment of a subject's anatomical structure, comprising intravenously administering a bolus dose approximately 1 / 10 the usual dose used for perfusion assessment. For indocyanine green (ICG), a typical bolus dose is 0.1-0.3 mg / kg body weight. According to the present disclosure, a bolus dose of less than 0.01 mg / kg body weight of the first fluorescent contrast agent may be used. For other fluorescent contrast agents described herein, the bolus dose may be similarly reduced according to the present disclosure. As noted above, the agent may be infused by a controllable infusion pump, e.g., as a series of bolus doses with predetermined intervals between subsequent bolus doses. After each bolus dose, fluorescence emission from the anatomical structure may be measured.

[0058] The minimum bolus dose that results in a quantifiable fluorescence emission indicative of perfusion of an anatomical structure can be identified after administering a series of increasing boluses, which include gradually increasing or decreasing amounts of agent, for example, the amount can increase or decrease in 10% increments from one bolus to a subsequent bolus.

[0059] The interval between bolus administrations can be between 5 and 600 seconds, e.g., between 5 and 300 seconds, e.g., between 10 and 180 seconds, e.g., between 10 and 140 seconds, e.g., between 10 and 90 seconds, e.g., between 15 and 80 seconds, e.g., between 20 and 70 seconds, e.g., between 30 and 60 seconds. In another embodiment, the interval between bolus administrations can be between 5 and 600 seconds, e.g., between 10 and 600 seconds, e.g., between 15 and 600 seconds, e.g., between 15 and 300 seconds, e.g., between 30 and 240 seconds, e.g., between 45 and 240 seconds, e.g., between 90 and 240 seconds, e.g., between 90 and 120 seconds. The interval between bolus administrations is preferably long enough to allow measurement of the perfusion gradient for each bolus administration in the anatomical structure, preferably The perfusion gradient includes a gradient start and a washout gradient.

[0060] In the case of ICG, the amount of fluorescent contrast agent is preferably 0.0001-0.01 mg / kg body weight per bolus, e.g., 0.001-0.01 mg / kg body weight per bolus. The initial amount of fluorescent contrast agent is advantageously at least 0.001 mg / kg body weight. Subsequent bolus doses may increase by at least 0.001 mg / kg body weight from one bolus to the next. For other types of fluorescent contrast agents, the dose is preferably selected relative to ICG based on its fluorescence. Thus, a fluorescent contrast agent with a higher emission rate is preferably administered at a correspondingly lower dose. The dose may, for example, be substantially inversely linear with respect to the quantum yield of the fluorescent contrast agent. The dose may further be based on its absorption and emission spectra relative to ICG.

[0061] The bolus injection is preferably a liquid volume between 0.5 μL and 10 mL, e.g., 0.5 to 5 mL. In a preferred embodiment of the present disclosure, a volume of isotonic liquid (such as saline) is injected immediately after the injection of the bolus of fluorescent contrast agent, e.g., the volume of the isotonic liquid is 1 to 20 mL, e.g., 2.5 to 15 mL, e.g., 5 to 10 mL.

[0062] In a further embodiment of the present disclosure, a second fluorescent imaging agent is administered, the second fluorescent imaging agent having an emission maximum that differs from that of the first fluorescent imaging agent by at least 50 nm, or at least 100 nm. The first and second fluorescent imaging agents are preferably administered alternately. Advantageously, the interval between administrations of different fluorescent imaging agents is half the interval to subsequent administrations of the same fluorescent imaging agent.

[0063] In a further embodiment of the disclosed method, a series of fluorescent images of the anatomical structure are formed for assessment of perfusion. Fluorescence may be automatically detected by illuminating the anatomical structure with a light source capable of exciting a fluorescent contrast agent, and luminescence is quantified across the series of fluorescent images of the anatomical structure.

[0064] The bolus interval is determined by a computer configured to detect the perfusion gradient caused by each bolus. Furthermore, the amount of fluorescent contrast agent during the bolus administration can be controlled by a computer configured to identify a minimum bolus dose corresponding to a minimum fluorescence emission indicative of perfusion of the anatomical structure. This computer can be part of the system of the present disclosure.

[0065] In further embodiments, the perfusion assessment includes locating perfusion complications in an anatomical structure. Thus, the perfusion assessment may be used in connection with diagnostic or surgical procedures, for example, methods including diagnostic laparoscopy, exploratory laparoscopy, surgical laparoscopy with conventional laparoscopy, robotic surgery, and open surgery. The method may alternatively include anastomosis, for example, intestinal anastomosis, wound, plastic surgery, cardiac surgery, or cancer.

[0066] Further embodiments of the present disclosure relate to fluorescent imaging agents for use in the methods disclosed herein. Yet further embodiments relate to the use of fluorescent imaging agents in the preparation of medicaments for use in the automated perfusion assessment methods disclosed herein.

[0067] In a further embodiment of the present disclosure, the fluorescent contrast agent is repeatedly injected. In certain cases, it may be necessary to prolong the phase without injecting the fluorescent contrast agent, for example, at least 2 minutes, preferably at least 3 minutes, even more preferably at least 4 minutes, even more preferably at least 5 minutes, most preferably at least 8 minutes, and most preferably at least 10 minutes, so that the fluorescent contrast agent is washed out and the background level decreases. When the background level reaches an acceptable level, for example, the maximum fluorescence intensity, After the fluorescence level drops below a certain percentage of the normal range, or until the fluorescence is substantially unmeasurable, the injection of the fluorescent contrast agent may continue.

[0068] The system of the present disclosure may be configured to perform the above-disclosed method for automated perfusion assessment of an anatomical structure, which may be provided by a system having at least one processor and a memory having instructions stored thereon, which, when executed by the one or more processors, cause the system to perform the method for automated perfusion assessment of an anatomical structure of the present disclosure.

[0069] Perfusion parameters Various parameters can be determined based on image analysis of the fluorescence emission. Extracted from the image analysis are typically intensity values ​​versus time, which can be used to generate several time-course curves, the shape of which can be analyzed. From this analysis, relative and / or quantitative data on perfusion, blood volume, and / or blood flow can be determined, i.e., based on the results of image analysis of a time series of fluorescence images, also known as a video sequence of fluorescence images. In particular, the perfusion gradient of the flow of fluorescent contrast agent through at least one region of interest can be determined. The perfusion gradient is an important parameter because it directly indicates the perfusion of the imaged tissue.

[0070] Perfusion parameters can be determined from fluorescence intensity values ​​extracted from one or more regions of interest, typically comprising tissue. The configuration of the regions of interest, e.g., region size, number of regions, location within the image, etc., can be automatically, semi-automatically, or manually provided by a user, e.g., a physician / surgeon. With at least some manual intervention, the user may be able to select additional regions of interest or remove existing regions of interest. Preferably, prior to capturing the image sequence, one or more of the regions of interest are also moved within the image so that the regions of interest are located in the relevant regions of the image.

[0071] The perfusion gradient can be determined from the fluorescence intensity values ​​integrated over a region of interest containing tissue. Before the initial injection of contrast agent, the curve is a substantially flat line. After the contrast agent is injected, the region of interest begins to fluoresce as soon as the bolus of contrast agent molecules reaches the region of interest, resulting in a substantially linearly increasing line. Once the bolus of contrast agent molecules levels off, the fluorescence intensity of the region of interest begins to wash out, and the amount of contrast agent molecules decreases (substantially linearly) to zero.

[0072] However, this is an ideal scenario and curves can vary from time to time between individual patients, so it is important to reliably define perfusion parameters so that they can be identified in an automated, ad hoc, repeatable and comparable manner.

[0073] The perfusion gradient can be defined by the slope of the extracted intensity values ​​from the gradient start to the gradient end. The perfusion gradient can be simply determined as a linear fit to a curve. The challenge, especially in real-time situations, is to identify the start (gradient start) and end (gradient end) of the fit. The gradient start is the most important of the two and can be defined as the point at which the gradient exceeds a predetermined first threshold. The first threshold can be determined, for example, by three parameters: a predetermined constant k, and the mean and standard deviation (std) of the intensity values ​​before the gradient start or before the delivery of the fluorescent contrast agent. The gradient start can be defined as the point at which the gradient exceeds the mean by k*std. The gradient end can accordingly be defined as the point at which the gradient decreases beyond a predetermined second threshold after the gradient start. The constant k can be determined based on the setup, but typically k is in the range of 3 to 10.

[0074] However, advantageously, the perfusion gradient can be determined from a histogram in parameter space that bins all gradients after gradient onset, and the perfusion gradient is determined by the most frequent gradient in the histogram. The gradient is specified as a value. That is, after gradient initiation, gradient values ​​are calculated for all subsequent intensity points based on the gradient initiation. The gradient end can then be estimated therefrom. Because it is certain that the perfusion gradient begins after the gradient initiation, gradient values ​​calculated immediately after the gradient initiation can be assigned a greater weight in the histogram than later gradient values. For example, the first 100 calculated gradient values ​​can be assigned weights of 100, 99, 98, etc., respectively, in the histogram. If a higher constant k is selected, the initial value of the perfusion gradient can be assigned an even greater weight. This histogram-centric approach is highly accurate and can be advantageously used in real-time or near-real-time situations.

[0075] Another parameter that can be determined is, for example, a washout gradient, which indicates the disappearance flow of contrast agent through at least one region of interest. While the perfusion gradient is typically positive due to increased contrast agent flow, the washout gradient is opposite (in sign) to the perfusion gradient, i.e., typically negative. The washout gradient can add information about perfusion within tissue. However, the washout gradient may also be relevant as an indicator of organ function, such as the liver. Similar to the perfusion gradient, the washout gradient can be defined by the gradient of the intensity values ​​from washout onset to washout end. Washout onset occurs after the gradient ends. The washout gradient can be determined from a histogram in parameter space that bins all gradients after washout onset, and the washout gradient is determined as the mode of the histogram. As described above for the perfusion gradient, some of the calculated washout gradient values, particularly the initial values ​​of the washout gradient after washout gradient onset, may be assigned a greater weight in the histogram than others.

[0076] The maximum intensity reached can be easily identified, for example, for each ROI. However, a more relevant parameter may be the maximum gradient intensity, which is the intensity at which the intensity value begins to plateau. The maximum gradient intensity can be defined as the intensity value at the end of the gradient. A more accurate definition may be the intensity value at which the distance to a line that has the perfusion gradient as the gradient and intersects with the curve point determined by the gradient start exceeds a predetermined limit, for example, a limit based on the standard deviation of the perfusion gradient. For example, the maximum gradient intensity may be the intensity level that differs from the perfusion gradient by a predetermined coefficient multiplied by the standard deviation of the perfusion gradient.

[0077] The gradient rise time may also be relevant and can be defined as the difference between the maximum gradient strength and the time of gradient onset, i.e., the time it takes for the contrast agent to pass through or accumulate in the tissue, which may be an indicator of the velocity of blood flow.

[0078] The relative perfusion gradient can then be defined as the inverse of the gradient rise time. The subject-specific relative perfusion gradient can then be defined as the relative perfusion gradient multiplied by the maximum intensity in the region of interest where perfusion is at a local (or global) extreme, i.e., the normalized perfusion parameter to become the patient-specific perfusion gradient parameter.

[0079] Tracking Further embodiments provide motion tracking in time-series images, e.g., videos, of an anatomical structure, e.g., the gastrointestinal tract. Thus, the systems and methods of the present disclosure can use tracking, as exemplified below, so that perfusion parameters can be extracted from the same anatomical structure(s) and the same portions of the anatomical structure(s) in repeated / successive measurements. This motion tracking of the anatomical structure, or at least portions thereof, can be used to ensure that at least one of the regions of interest corresponds to the same portion of the anatomical structure in the videos.

[0080] Especially in open surgery, the camera may be moved frequently during surgery. Thus, the selected ROI may completely disappear from the image during surgery and during continuous perfusion measurements. The tracking disclosed herein also relates to recognizing when relevant regions and / or anatomical structures are present within an image and when they are not, so that perfusion parameters measured from completely different regions are not erroneously compared.

[0081] The purpose of tracking is primarily to ensure that data, e.g., pixel intensity values, are sampled from the same tissue region. Thus, if an anatomical structure moves within an image, tracking should ensure that any region of interest defined herein moves accordingly, so that the sampled data for that region of interest is understandable. In that regard, it does not matter whether it is the anatomical structure that physically moves, e.g., due to the subject's breathing and / or peristaltic movement, or the imaging device capturing the image that moves relative to the anatomical structure. What matters is whether the imaged object moves within the captured image.

[0082] A further aspect of the present disclosure therefore relates more generally to an imaging system for imaging the movement / dynamics of at least a part of an anatomical structure (e.g., during a medical procedure) from a video representing at least the exterior of said anatomical structure, - selecting one or more regions of interest in at least one of said images, at least a first of said regions of interest corresponding to a part / subsection of an anatomical structure; - tracking the movement of an anatomical structure in said image; - correlating said movement of the anatomical structure so that at least said first region of interest corresponds to the same portion of the anatomical structure in said image; The present invention relates to a computer-implemented method including:

[0083] Tracking of objects in an image sequence, e.g., a video image sequence, can be provided in a variety of ways. Broadly speaking, there are at least two different approaches: free image tracking (FIT), which is based solely on the input video feed, and object-based tracking (OBT), in which a predetermined and / or recognizable object is coupled to the object to be tracked in the images.

[0084] Free image tracking may be provided, for example, by a classifier: based on an input image, a classifier algorithm calculates a classifier of the most recognizable features in the area surrounding a given ROI (for further ROIs, each ROI is assigned a sensitive region within which tracking works for the given ROI). In one embodiment of the present disclosure, movement tracking of anatomical structures is provided by free image tracking, for example, in the form of classifier-based tracking that includes identifying classifiers of one or more recognizable features within the video, preferably in an area adjacent to or surrounding at least one region of interest.

[0085] Free-form image tracking can also be based on color-based tracking: prior to a medical procedure, e.g., a surgical procedure, at least one ROI of an object, e.g., an anatomical structure, e.g., the intestine, is marked with a color and / or tattoo, preferably a predetermined color or tattoo. The marking can be provided, for example, by a surgical procedure. If it is an actual ROI that is marked, the color-based algorithm can obtain the form of the marking and use this form as a specific region of interest. The color-based algorithm can be configured to first perform color filtering, followed by object recognition. Based on the characteristics of the marker (mainly color), a target RGB or HSV index can be provided for filtering. Then, filtering can be provided, for example, in the form of HSV thresholding, to obtain a Boolean map of the input image pixels, which only includes pixels that cover the marker. Next, object recognition can be provided by removing noise from the Boolean map, for example, by noise filtering, for example, by opening or closing based on erosion / dilation. These noise filter(s) allow for " An improved Boolean map can be obtained with "filled" ROIs, i.e., the resulting Boolean map is filled with zeros except for patches filled with ones (or vice versa), where each patch corresponds to an ROI.

[0086] Another example of free-form image tracking is based on cross-correlation. After ROI image selection, typically, a "normal" white-light image of the region within each ROI is saved as an initial reference for each ROI. These initial ROI references are then used as a tracking function for each ROI, e.g., as a template for a cross-correlation function that is applied continuously and / or in real time. A cross-correlation function can be a form of pattern recognition and can be considered a measure of similarity as a function of the displacement of two images relative to each other, and thus may be highly suitable as a tracking function. Since actual movement is almost always cyclical, the actual tracking function may, for example, be limited, at least initially, to the region adjacent to each ROI. As additional ROI images may be acquired and saved during the course of a medical procedure, the initial template can be improved, and potentially continuously improved, if these additional ROI image acquisitions are used as the averaging basis for a new, possibly final, ROI template used for tracking. Such template improvement, including information from several ROI images, therefore includes a time aspect. This can significantly improve and / or further streamline tracking. For example, if a bowel movement is tracked and the bowel rotates back and forth during the procedure, it may be more difficult to track this ROI at all positions of the rotational movement if only an initial template ROI is used in the tracking, but it may be easier to track the ROI at all positions of the rotational movement if an average template is used, averaging several templates from the rotational movement.

[0087] In object-based tracking, one or more objects are physically coupled to a target that needs to be tracked, e.g., the intestine. The object(s) are typically predefined, e.g., with respect to size, shape, and color, so that a classifier can be trained prior to tracking. That is, the tracking system used can be configured to automatically recognize (and thereby track) this predetermined object. In one embodiment of the present disclosure, movement tracking of an anatomical structure is provided by object-based tracking, e.g., by tracking the movement of one or more predetermined objects coupled to the anatomical structure.

[0088] As an example of object-based tracking, two (or more) spheres (or other geometrically well-defined objects) may be attached to the "top" portion of an anatomical structure and one (or more) spheres may be attached to the lower / bottom portion (as viewed from the imaging device) of the anatomical structure. If the object attached to the top portion is different from the object(s) attached to the bottom portion, it is easy to distinguish between the top and bottom portions. If these spheres emit trackers, they are even more easily recognizable and therefore trackable. They may contain, for example, a fluorescent agent so that they become visible when excited. They may then be identified in the image by, for example, Hough circle recognition (or another feature extraction). They may also be colored and identified by the color recognition methods described above. Because the "top" / "bottom" objects are predefined and therefore known, it is easy to train classifiers for both types of objects. To train a classifier for an object, a large database of images of the object in question may be used to train the classifier. The locations of "top" and "bottom" objects in the image can therefore be very accurately identified using the classifier.

[0089] As the objects are fixed to targets, e.g., intestinal tissue, the ROI's can be defined based on these objects (e.g., "top" and "bottom"). For example, if four objects are used, the corners of the ROI can simply correspond to the positions of the four tracked objects. In the case of two objects, the ROI can be defined between the positions of the two objects, e.g., a parallelogram expanding to half height in the middle, which determines the angle.

[0090] Furthermore, it should be noted that tracking is not limited to determining the two-dimensional position / coordinates of the region of interest. Instead, tracking can be performed in a manner that determines the position / coordinates of the region of interest with respect to all three dimensions of Euclidean space. Several imaging methods are known for three-dimensional reconstruction of objects. These include, for example, methods based on oblique illumination, in which the object is illuminated from the side; microscopy techniques, such as confocal microscopy, light-sheet fluorescence microscopy, 3D deconvolution microscopy, and other methods that use known object properties to obtain depth information. Additional methods for obtaining depth information are known to those skilled in the art and may be used in conjunction with the system of the present disclosure to accurately track the region of interest in three dimensions. The obtained depth information is preferably used by the system in evaluating measured perfusion metrics, for example, by normalizing the measured fluorescence intensity based on the distance to the region of interest.

[0091] Perfusion assessment The above perfusion parameters can provide much useful information. However, in order to qualify the perfusion parameters, some reference may be required.

[0092] In one embodiment, image sequences acquired from different portions of the anatomy can be used to calculate perfusion parameters associated with each portion, and these perfusion parameters can be compared so that perfusion in different portions of the anatomy can be compared, i.e., perfusion parameters from one of the image sequences can be used as a reference so that a quantitative assessment of perfusion can be provided between image sequences associated with different portions of the anatomy.

[0093] In another embodiment, different regions of interest from the same video sequence can be selected such that perfusion parameters associated with one of the regions of interest are used as a reference for the other regions of interest, thereby providing a quantitative assessment of perfusion between different regions of interest in the same video sequence. These different regions of interest can be selected so that they represent different portions of the anatomical structure or different nearby anatomical structures. For example, if the anatomical structure is the gastrointestinal tract, the different portions of the gastrointestinal tract can be the colon and the small intestine. For example, surgery may be performed on the colon, but very often the small intestine is located very close to the colon, so a comparison with the small intestine, which can be imaged during video acquisition, can provide a reference unaffected by the surgery. Another example is a skin flap candidate compared to healthy, well-perfused skin in plastic surgery.

[0094] That is, identifying at least a first perfusion parameter, e.g., of the flow of fluorescent contrast agent through at least a first region of interest, the first perfusion parameter being selected from the group consisting of a perfusion gradient, a washout gradient, a maximum gradient strength, a relative perfusion gradient, and a subject-specific relative perfusion gradient, and identifying at least a second perfusion parameter, e.g., of the flow of fluorescent contrast agent through at least a second region of interest, the second perfusion parameter being selected from the group consisting of a perfusion gradient, a washout gradient, a maximum gradient strength, a relative perfusion gradient, and a subject-specific relative perfusion, wherein the first and second regions of interest represent different portions of an anatomical structure or different anatomical structures, such that perfusion of one of the different portions of an anatomical structure can be assessed by comparing it with perfusion of at least one other of the different portions.

[0095] A further embodiment therefore comprises the following steps: - performing image analysis of at least two of the following image sequences, each acquired after a fluorescent contrast agent has been delivered to the subject: a first image representing at least a first portion of an anatomical structure or at least a first anatomical structure; and a second image depicting at least a second different portion of the anatomical structure or at least a second different anatomical structure; - calculating intensity values ​​in one or more regions of interest based on image analysis of the first image and the second image; and - determining a perfusion gradient of the flow of fluorescent contrast agent through at least a first region of interest selected in the first sequence of images and at least a second region of interest selected in the second sequence of images;

[0096] Further embodiments relate more specifically to anastomosis processes where perfusion assessment can be an important indicator of where resections should be made and whether the final anastomosis will have sufficient perfusion. Further embodiments therefore further comprise the following steps: - performing image analysis of two or more of the following image sequences, each image sequence acquired after a fluorescent contrast agent is delivered to the subject: a) Images acquired prior to intestinal resection, e.g., bowel resection; b) Images acquired after resection and before anastomosis, and c) Images obtained after anastomosis. - calculating intensity values ​​of one or more regions of interest based on image analysis, where at least a first of said regions of interest is the same region in said two or more video sequences; and - determining a perfusion gradient of the flow of the fluorescent contrast agent through at least a first region of interest based on the two or more image sequences.

[0097] Based on these two or more video sequences, one or more of the following parameters can be determined based on the two or more video sequences: washout gradient, maximum gradient strength, relative perfusion gradient, and subject-specific relative perfusion gradient.

[0098] After acquiring parameters from two (or more) video sequences at different times during a medical procedure, parameter(s) extracted from one video sequence can be used as reference parameter(s). In this manner, quantitative data regarding perfusion in at least one of the regions of interest based on gradient parameters can be determined from the at least two video sequences. As a result, quantitative and qualitative evaluation parameters can be provided to a surgeon during or after a medical procedure, such as gastric surgery, to aid in assessing whether an intestinal, e.g., bowel, resection is promising. Both during and after surgery, results can be evaluated almost instantly, e.g., whether an anastomosis has sufficient perfusion. This can be done, for example, by comparing perfusion parameters obtained pre-, intra-, and / or post-operatively, or continuously during the medical procedure, to quantify perfusion changes. Mobile tracking can be key to accurately quantifying perfusion changes, as this is one way to ensure that the same region of interest is evaluated for perfusion pre-, intra-, and / or post-operatively, or continuously during the medical procedure.

[0099] Perfusion parameter-specific thresholds may be provided. Similarly, uncertainties may be associated with given thresholds. A comparison of the thresholds may indicate, for example, whether the surgery went smoothly or whether the perfusion according to the parameter(s) in question fell below a critical value. Also, for some perfusion parameters, a "weighted average answer" may be provided.

[0100] In one embodiment of the present disclosure, the perfusion gradient (and / or other perfusion parameters as described) is calculated from a video sequence acquired before resection and a video sequence acquired after resection but before anastomosis. The relationship between these two perfusion gradients is a measure of the difference in perfusion before and after resection. If perfusion falls below a predetermined threshold after resection, a warning may be issued. If perfusion gradients are calculated before and after resection for two, three, or more regions of interest, and these regions of interest were not resected before resection, a warning may be issued. If the same tissue region is imaged later, more information can be extracted.

[0101] vibration dynamics The inventors further realized that measurement and analysis of repeatable bolus injections can be further extended from interpretation and quantification of single inflow and / or outflow phases to analysis of oscillatory fluorescence dynamics, which may reveal physical perfusion characteristics previously unattainable without invasive procedures.

[0102] The disclosed systems and methods can be configured for repeated injections of small bolus doses, e.g., minimum bolus doses, at regular intervals. These bolus doses may lead to periodic variations when measured, e.g., depending on the injection interval, and approximate a sinusoidal curve. In such a curve, the measured intensity signal is expected to increase with the influx of fluorescent contrast agent from a given bolus, then decrease during the washout phase of the bolus, before increasing again during subsequent bolus doses, resulting in a periodic (sinusoidal) pattern.

[0103] Preferably, the system is configured to recognize vibration intensity curves, e.g., parameters of frequency and / or amplitude. This trained system can then predict both the direction and regularity of the dynamics of subsequent signals. The system preferably uses measurements to recognize vibration patterns, so that the system can then detect discrepancies between measurements and expectations. Measurements may be used continuously to further refine pattern recognition, i.e., expectations. Alternatively or additionally, infusion parameters, e.g., bolus frequency, dose, and flow rate, may be used to identify expectations, i.e., vibration patterns.

[0104] By predicting expected values, the system is able to detect and alert early, ideally instantaneously, to the onset of an ischemic condition. Detection of an ischemic condition can be a function of expected value(s) and detected value(s), e.g., thresholds.

[0105] Deviations from the expected sinusoidal pattern may be caused, for example, by the onset of ischemia in at least a portion of the anatomical structure visible on the image, or by local changes in perfusion to a given region. An illustration showing this change in dynamics due to the onset of ischemia in a human subject is shown in Figure 12A, with a more focused, zoomed-in view in Figure 12B. As can be seen, it is possible to detect a transition from a regular oscillating fluorescent signal to an ischemic flat line. However, it should be noted that changes in perfusion of the anatomical structure of interest can result in other measurement patterns in addition to the ischemic flat line. One example is venous occlusion, in which blood outflow from the anatomical region is blocked or reduced, leading to changes in oscillatory dynamics due to overcrowding or pooling of fluorescent agent in a given region. As seen in Figure 13C, the periodic oscillations cease, but the result is not a flat line.

[0106] The system described herein can observe and detect changes in the perfusion level of a given region within a video image within seconds. This can be detected in regions that have been observed for extended periods of time, e.g., minutes, where the dynamics are continuously visualized and the phase is therefore known. An illustration highlighting the differences between signals one might expect to observe in ischemic and healthy tissue regions is shown in Figure 12C. However, because the described system is trained to predict and detect certain phases of the described oscillatory dynamic signal, consisting of regular rises and falls in the temporal intensity signal at a given time point in the tissue, this can also be identified in anatomical regions that are only visualized for a short time interval, e.g., 10-20 seconds. See Figure 12D, which shows what this might look like when an anatomical region of interest drifts in and out of focus in the recorded image.

[0107] Preferably, the system includes tracking means and is capable of running independently in the background, during which the surgeon only needs to touch the visible white light signal, and is only interrupted / notified by a warning signal, e.g., during the detection of the onset of an ischemic condition.

[0108] Another aspect of the present disclosure relates to continuous perfusion assessment, which involves repeated injections of a fluorescently activated agent and monitoring the resulting oscillation curve. In addition to detecting unexpected changes in perfusion, the system can be used to assess the perfusion area of ​​an artery. As an example, a surgeon may consider cutting an artery as part of a surgical procedure. Before cutting the artery, the surgeon may temporarily restrict perfusion through that artery, and the disclosed approach may enable visualization of the perfusion area of ​​that artery in a short period of time, e.g., less than one minute. This may be useful information for the surgeon during the ongoing surgical procedure. Similarly, the system may be used to assess the drainage area of ​​a vein or group of veins, lymphatic vessels, lymph nodes, or other portions of the circulatory and / or lymphatic pathways. By temporarily restricting blood flow through a blood vessel, the blood pools in the anatomical area normally drained by that vessel or group of vessels. This allows visualization of the anatomical area drained by the vessel in a relatively short period of time, e.g., less than two minutes. This can provide important information to surgeons, for example, during ongoing surgical procedures, in areas such as general surgery and plastic surgery, including wound and reconstructive surgery.

[0109] anatomical structure The anatomical structure of the disclosed systems and methods may be an internal organ of the subject, and perfusion is then typically assessed in tissue of the external portion of the organ. Alternatively, the anatomical structure may be (part of) the skin of the subject, and perfusion is then typically assessed in skin tissue.

[0110] Wound perfusion assessment is also highly relevant. Thus, the anatomical structure may include at least one wound that is the subject of perfusion assessment.

[0111] The anatomical structure may be the gastrointestinal tract, preferably including the oral cavity; the pharynx; the small intestine, including the duodenum, jejunum, and ileum; the stomach, including the esophagus, cardia, and pylorus; and the large intestine, including the cecum, colon, rectum, and anal canal.

[0112] gastrointestinal tract Complications related to the gastrointestinal tract are often related to local hemodynamics. That is, alterations in normal hemodynamic status can be an indicator of an increased risk of complications. Therefore, perfusion assessment of tissues in the gastrointestinal tract, particularly within and near its surface, e.g., the gastrointestinal wall, can be an important diagnostic tool when examining the gastrointestinal tract, e.g., during diagnostic laparoscopy, exploratory laparoscopy, or surgical laparoscopy involving conventional or robotic laparoscopy, as well as during open surgery, e.g., for diagnosing or localizing complications. Perfusion assessment is also important during surgical procedures to create an anastomosis, which can be used to establish communication between two previously separate portions of the gastrointestinal tract. As an example, an intestinal anastomosis establishes communication between two previously separate portions of the intestine, typically restoring intestinal continuity after removal of a pathology affecting the intestine. Intestinal anastomosis may provide, for example, 1) restoration of intestinal, e.g., bowel, continuity after resection of diseased intestine, and 2) bypass of unresectable diseased intestine, e.g., bowel. Certain pediatric conditions may also require intestinal anastomosis.[6]

[0113] Resection of the diseased bowel may be performed in the following situations: Intestinal gangrene due to vascular compromise caused by mesenteric vascular disease, prolonged bowel obstruction, intussusception, or volvulus Malignant tumors Benign conditions (e.g., intestinal polyps, intussusception, roundworm infestation with intestinal obstruction) Infections (e.g., tuberculosis with stricture or perforation) ·Traumatic perforation Large perforations (traumatic) not suitable for primary wound closure Radiation enteritis with bleeding, stricture, or perforation Inflammatory bowel disease, ulcerative colitis, or Crohn's disease that is refractory to medical treatment or has complications (e.g., bleeding, perforation, toxic megacolon, dysplasia / carcinoma) Chronic constipation, idiopathic atonic constipation, or Hirschsprung's disease: If medical treatment for the disease is ineffective, a subtotal colectomy may be performed.

[0114] Bypass of unresectable diseased bowel may be performed in the following situations: Locally advanced tumors causing luminal obstruction Metastatic disease causing bowel obstruction Poor general condition or conditions that preclude major resection

[0115] Pediatric conditions that may require an intestinal anastomosis include: Congenital anomalies (e.g., Meckel's diverticulum, intestinal atresia, malrotation with volvulus leading to gangrene, meconium ileus, duplication cyst, Hirschsprung's disease) Inflammatory conditions (e.g., necrotizing enterocolitis, enteritis, tuberculosis, intestinal perforation) Other conditions (e.g., intussusception, angiodysplasia, polypoid disease, ascariasis) As part of other surgical procedures (e.g., Kasai portoenterostomy, choledochal cyst, urinary diversion, pancreatic tumor)

[0116] Postoperative complications associated with gastrointestinal anastomoses are unfortunately frequent and often result from insufficient perfusion (capillary blood supply) at the anastomosis, i.e., the joining of two sections of the tract. Inadequate perfusion can lead to anastomotic leak, a serious and frequent complication, for example, in colorectal surgery, where more than 10% of surgeries result in complications. In colon cancer surgery, more than 30% of patients with anastomotic leak die from postoperative complications, and approximately 25% of the remaining patients develop a lifelong stoma. Risk factors associated with leak include anastomotic tension, tissue damage, and, particularly, reduced blood perfusion.

[0117] The present disclosure thus relates, in one embodiment, to performing image analysis of one or more video image sequences representing at least a portion of the gastrointestinal tract before, during, and / or after surgery, for example, particularly in surgery involving the gastrointestinal tract. This may be particularly applicable to gastrointestinal surgery. The video image sequences may thus include at least some outer portions of the gastrointestinal tract, preferably so that perfusion in at least a portion of the gastrointestinal wall can be measured and assessed.

[0118] The gastrointestinal tract is the organ system in humans and other animals that ingests food, digests it to extract and absorb energy and nutrients, and excretes remaining waste products as feces and urine. The gastrointestinal tract can be viewed as a tube that transports food through the digestive system. As used herein, the term gastrointestinal tract therefore includes the oral cavity; the pharynx; the small intestine, including the duodenum, jejunum, and ileum; the stomach, including the esophagus, cardia, and pylorus; and the large intestine, including the cecum, colon, rectum, and anal canal. [Example]

[0119] The intensity curves shown in the examples are the result of injection of a bolus of a regular amount of fluorescent agent, in these cases ICG. The amount of ICG in each bolus was selected so that the fluorescent emission was visible to the human eye. These examples are provided to illustrate various perfusion parameters that can be calculated after fluorescent imaging. These same parameters can also be determined to a large extent after injection of much smaller doses, i.e., the micro-dose approach disclosed herein, possibly with repeated serial measurements and associated perfusion assessment.

[0120] Figures 1A, 1C, and 1E show example intensity curves obtained from tissue after a bolus of ICG has been administered to a subject, e.g., from a region of interest in a video sequence. The same type of data would be obtained using a different contrast agent. The intensity is essentially zero until a sharp increase in intensity indicates the passage of ICG molecules through the imaged tissue, which are excited and fluoresce. The intensity peak is followed by a gradual washout of the ICG molecules. This intensity is shown in arbitrary units. Figures 1B, 1D, and 1F show the corresponding intensity curves, where hemodynamic parameters such as perfusion gradient, perfusion onset, gradient end maximum intensity, washout gradient, washout onset, and washout gradient end are calculated and displayed on these graphs.

[0121] Figures 2A-2F show three examples illustrating the approach disclosed herein for identifying the point at which the perfusion gradient begins, i.e., gradient onset. Figures 2B, 2D, and 2F are enlarged views of the gradient onset portion of Figures 2A, 2C, and 2E, respectively; in other words, the graph on the right shows an enlarged view of the curve on the left, showing the gradient onset in more detail. It can be seen that gradient onset is defined as the point at which the gradient exceeds the mean by k*std, where k is a predetermined constant and std is the standard deviation of the intensity values ​​before the gradient onset. The gradient onset is indicated by a circle in Figure 2B.

[0122] Figures 3A-3F show three examples illustrating the approach to identifying perfusion gradients based on histogram data disclosed herein. The graphs on the left show intensity curves, with Figure 3A corresponding to Figure 2A and Figure 3E corresponding to Figure 2E. The gradient onset is indicated by an arrow in Figure 3A and a circle in Figures 3C and 3E. All possible gradients of the intensity curve were calculated from the gradient onset to the end of the intensity curve. All calculated gradients were compiled and binned into the histogram shown on the right. The perfusion gradient is defined as the mode of the histogram, i.e., the highest histogram bin. The calculated perfusion gradients for each of the intensity curves in Figures 3A, 3C, and 3E, i.e., the highest histogram bin in Figures 3B, 3D, and 3F, respectively, are characterized by straight lines in Figures 3A, 3C, and 3E.

[0123] Figures 4A-4F show three examples illustrating the approach to defining and identifying the maximum gradient strength disclosed herein. Figures 4B, 4D, and 4F are magnified views of the portions of the curves in Figures 4A, 4C, and 4E, respectively, where the maximum intensity is found. The maximum intensity of the curve is indicated by a star in Figure 4B and a square in Figures 4D and 4F, and the maximum gradient strength is indicated by a diamond in these figures. The maximum gradient strength is defined as the strength value at which the distance to the perfusion gradient exceeds a predetermined limit, e.g., a limit based on a constant (k2) multiplied by the standard deviation of the perfusion gradient. As can be seen in Figure 4, there can be a significant difference in time and strength between the maximum intensity and maximum gradient strength of the curve. The gradient rise time can be defined as the difference between the peak (maximum) intensity of the curve and the onset of the gradient. However, as shown here, the gradient rise time, defined as the difference between the maximum gradient strength and the onset of the gradient, provides a more relevant definition of the gradient rise time.

[0124] Figures 5A-5F show three examples illustrating the approach to analyzing washout of fluorescent contrast agents disclosed herein. These intensity curves are the same as those in Figure 4. Figures 5B, 5D, and 5F are enlarged views of the ICG washout sections of 5A, 5C, and 5E, respectively. In the left graph, the maximum intensity is indicated by a star in Figure 5A and a square in Figures 5C and 5E. Enlarged views of the washout sections are shown in the right graph. The washout data were analyzed in the same manner as the perfusion gradients, and all possible washout gradients were calculated. Similar to the above example identification of perfusion gradients, the washout gradients can be binned and sorted into a histogram (not shown) to select the most frequent washout gradient. The onset of washout typically occurs after the maximum intensity of the curve. In this example, the onset of washout is defined as symmetric with respect to the maximum gradient intensity near the maximum curve intensity. In this example, the washout end is determined in the same way as the maximum gradient strength determination in the example above, i.e., the point in time when the strength differs from the washout gradient by a predetermined constant multiplied by the standard deviation of the washout gradient.

[0125] Figures 6A-6D show the analysis of two additional fluorescence measurements using ICG, demonstrating the robustness of the disclosed analytical approach. First, Figure 6A shows the intensity data, gradient onset, calculated perfusion gradient (dotted line), maximum gradient strength, and maximum curve strength. The graph on the right shows a histogram containing binned perfusion gradient data. This intensity data appears less stable than the other intensity curves disclosed herein, with many local fluctuations and no clear decrease in intensity after the perfusion gradient. While there may be washout of ICG molecules, the data shown here does not include this portion. Figures 6A and 6B demonstrate that the exemplary approach disclosed herein is an extremely robust procedure and can be used for automatic, real-time determination of perfusion gradient and other perfusion parameters derived therefrom. Figure 6A also shows the large difference between the time points of maximum gradient strength and maximum curve strength. The gradient rise time, derived from this maximum gradient strength, proves to be a further relevant parameter characterizing the passage of the ICG bolus.

[0126] The second, in Figure 6C, similarly exhibits unstable intensity data, with all calculated perfusion gradients spread over a wide interval, as shown in the corresponding histogram in Figure 6D. However, by selecting the most frequent histogram bins, relevant and accurate perfusion gradient parameters can still be extracted from this data, providing another example of the robustness of the disclosed approach.

[0127] Figure 7 shows output video frames from a video sequence acquired during an intestinal surgery. The top right corner shows an unedited video (i.e., one frame) of the surgery acquired during the course of the passage of an ICG bolus. The top left corner shows the same video frame after image processing; tissue perfusion is now much more clearly visible. As indicated in these figures, four regions of interest (ROIs) are shown in this video frame. The graph below shows the average pixel intensity of the four ROIs plotted as a function of time (seconds) versus normalized intensity. The perfusion gradient was calculated for these four ROIs (ROIs 1, 2, 3, and 4) and is shown in this graph as a straight line.

[0128] Looking at only the top two video frames, it is impossible for a surgeon to determine whether ROIs 1, 2, and 3 are all equally well perfused, e.g., whether regions 1, 2, and 3 are equally suitable for creating an anastomosis. This is also seen after about 70 seconds in the bottom graph, where the pixel intensities of ROIs 1, 2, and 3 are similar. However, by applying the approach disclosed herein to identifying perfusion gradients of different ROIs, an objective perfusion measure can be instantly provided to the surgeon. In the example of FIG. 7 , the calculated perfusion gradient indicates decreased perfusion in ROI 3 compared to ROIs 1 and 2. This information provides the surgeon with objective perfusion parameters on which to base their surgical decisions, ultimately increasing the chances of a successful surgical outcome.

[0129] 8A shows a still image from a typical video sequence acquired prior to resection of a patient's bowel. The image shows the small intestine (lower portion) and colon (upper portion). Although it is the colon that is resected, including the small intestine in this image analysis can provide an additional, potentially unbiased, high-perfusion reference measurement of the patient's perfusion for comparison with subsequent perfusion measurements.

[0130] Figure 8B shows a fluorescence image of substantially the same subsection of the gastrointestinal tract as Figure 8A, but acquired later, i.e., after the patient had been injected with a bolus of fluorescent contrast agent (ICG). This image shows three ROIs used for image analysis: an upper left blue box located in the colon, a lower red box located in the small intestine (high perfusion reference), and an upper right box located in a reference location with essentially no blood perfusion in this image (no / low perfusion reference).

[0131] FIG. 9A shows the intensity curve obtained in the ROI of FIG. 8B and the approach disclosed herein. Figure 9B shows the perfusion gradients calculated according to (1) and (2), i.e., the perfusion gradients of the colon and small intestine before resection. Although these intensity curves appear quite different, the calculated perfusion gradients for the colon and small intestine are comparable; however, the perfusion gradient of the small intestine is steeper (higher perfusion level) than that of the colon. This is also summarized in Figure 9B, where the perfusion gradients of the small intestine (left) and colon (right) have been normalized to the perfusion gradient of the small intestine.

[0132] FIG. 10A shows essentially the same subsection of the gastrointestinal tract as FIG. 8A, but taken after resection of the intestine and before the anastomosis. This is a critical part of the procedure in which the surgeon must assess whether perfusion at both ends of the remaining intestine after resection is adequate for anastomosis, or whether additional intestine needs to be resected to ensure the anastomosis is created in an area with optimal perfusion, ultimately increasing the chances of a successful outcome. The surgeon is therefore interested in obtaining measurements of perfusion in various regions of the intestine surrounding the resection. The small intestine is marked at the bottom of the image, and the resected intestine (colon) is marked at the top of the image.

[0133] Figure 10B is a fluorescence image corresponding to the image in Figure 10A after a bolus of ICG was injected. Five ROIs are shown in this image: one (red) in the small intestine as a high perfusion reference, one (black) located in a reference location with virtually no blood perfusion in this image (no / low perfusion reference), and three (blue, green, and yellow) in the excised intestine (colon).

[0134] Figure 11A shows the intensity curves obtained from the measurements shown in Figures 10A and 10B. The red ROI corresponds to the small intestine, which provides the steepest perfusion gradient, while the black reference ROI necessarily provides the lowest perfusion gradient. The blue, green, and yellow ROIs, corresponding to the three ROIs located in the intestine, provide comparable perfusion gradients, as also summarized in Figure 11B. In Figure 11B, the perfusion gradients of the small intestine (left, red) and intestine (right, blue, green, and yellow) are normalized to the perfusion gradient of the small intestine. A striking difference is observed when compared to Figure 9B. In Figure 9B (before resection), intestinal perfusion was comparable to that of the small intestine, whereas after resection, perfusion in the resected intestine is much lower than that of the small intestine. Furthermore, when comparing Figure 9A (before resection) with Figure 11A (after resection), it is noted that the absolute perfusion gradient is much larger after resection, as it was for the small intestine. This indicates that the absolute value of this perfusion gradient (and other perfusion parameters) is less important than the relative value, as also shown in Figures 9B and 11B. That is, it is important to include one or more reference ROIs in the image analysis so that calculated perfusion parameters can be compared with equivalent perfusion parameters acquired from the same images. In this example, the perfusion determination approach disclosed herein detects a significant decrease in intestinal (colonic) perfusion relative to small bowel perfusion. This important information can guide the surgeon when selecting the optimal location for the anastomosis.

[0135] Figure 12A shows actual measurement data from a human subject. The subject was repeatedly injected with microbolus doses of ICG at regular intervals (approximately 2 minutes in this example). The time-intensity curve exhibits a substantially sinusoidal pattern that increases linearly over time. This increase in intensity over time is related to the ratio between the fluorescent agent dose and the washout time during which the fluorescent intensity decreases. At a specific time point, approximately t = 3800 s in Figure 12B, perfusion becomes limited, resulting in the onset of ischemia. This can be seen by the lack of oscillations after this time point, forming what can be described as an ischemic flat line.

[0136] FIG. 12C shows idealized data showing a sinusoidal time-intensity curve. The measured ROI intensity increases upon injection of the fluorescent contrast agent and decreases during the washout phase. At approximately t=3750 seconds, the measurement data shows the measured ROI intensity value unchanged due to the onset of ischemia. Alternatively, if there was no ischemia, the measurements would instead be expected to follow the dashed line, and as a result, the measured ROI values ​​would continuously follow this sinusoidal pattern.

[0137] FIG. 12D shows ideal data showing a sinusoidal time-intensity curve in a non-ischemic state where the anatomical region of interest drifts in and out of focus. The dashed line indicates the expected measurement value if the ROI is continuously observable. If this is not possible, for example, due to the anatomical region of interest drifting in and out of focus in the recorded image, the measurement data may not be complete; instead, there may be gaps, i.e., time intervals during which no measurement data is obtained for the anatomical region of interest. Thus, the system is preferably able to recognize the sinusoidal pattern even when the recorded data is incomplete. If the system is able to correctly recognize the sinusoidal pattern, an expected intensity value for the ROI at each time point is provided, which can then be used for comparison with the measured values. If the measured value(s) differ from the expected value(s), the system may be configured to provide an alert to the surgeon. Thus, the system may be configured to recognize the phase of the oscillation / sinusoidal pattern at a measured time point or interval and then compare this to the expected phase for that time point or interval, which is preferably based on the recognized oscillation pattern and / or the injection of a bolus repeated at a known frequency. As a result, the system does not necessarily require continuous measurements, but may instead be based on the expected phase of the vibration pattern combined with time information of the measured time points or intervals, such that a particular phase of the vibration pattern is expected to be present in the measured interval.

[0138] Figure 13A shows fluorescence intensity measurements taken over a longer time interval, i.e., approximately 40 minutes, in a human subject who was repeatedly injected with microbolus doses of ICG. Intensity measurements were taken at seven separate ROIs, each assigned a different color in the graph. The measured fluorescence intensity exhibited a periodic sinusoidal pattern, the frequency of which corresponded to the injection frequency (approximately 120 seconds). This pattern increased substantially linearly due to the accumulation of the fluorescent contrast agent, due to the relatively short injection time compared to the dose size. Cessation of the repeated injection of the fluorescent contrast agent at approximately t = 2000 seconds resulted in a near-exponential decay of the fluorescence intensity.

[0139] Figure 13B shows an enlargement of the marked area in Figure 13A. Here, smaller oscillations within the same ROI and between different ROIs can be seen. At the same time, these periodic intensity patterns are distinguishable, and the patterns in each ROI have the same period.

[0140] Figure 13C shows a time-intensity plot of measurements performed on a human subject by repeatable injection of a microbolus dose of fluorescent contrast agent. This graph illustrates the results of venous occlusion, where perfusion is limited but not completely prevented from approximately t = 62–78 min. In this case, the oscillatory dynamics of the measured fluorescence intensity ceases, and these measurements show an irregular increase during venous occlusion. Note, therefore, that the decrease in perfusion does not necessarily result in a flat line, as typically occurs during ischemic conditions.

[0141] References References [1]C.Toens et al:Validation of IC-VIEW fluorescence videography in a rabbit model of mesentereic ischaemia and reperfusion.Int J Colorectal Dis 2006;21:332-338.[2]N.Nerup et al:Quantification of fluorescence angiography in a porcine model.Langenbecks Arch Surg, published online 15.11.2016. [3]L.Boni et al:Indocyanine green-enhanced fluorescence to assess bowel perfusio n during laparoscopic colorectal resection.Surg Endosc(2016)30:2736-2742 [4]R.Uitert et al:A stable optic-flow based method for tracking colonoscopy images.Conference Paper,July 2008 [5]US2016 / 262638 [6] D. Stein et al.: Colon Resection. http: / / emedicine.medscape.com / article / 1891505-overview,Sep.2015

[0142] Terms 1. A method for automated perfusion assessment of an anatomical structure of a subject, the method comprising: intravenously administering a bolus dose of a first fluorescent contrast agent equivalent to less than 0.01 mg ICG / kg body weight; acquiring and analyzing time-series fluorescent images of tissue of the anatomical structure after injection of the first fluorescent contrast agent; and determining at least one perfusion parameter of the anatomical structure based on the analysis.

[0143] 2. The method of clause 1, wherein the contrast agent is injected by a controllable injection pump.

[0144] 3. The method of any of the preceding clauses, wherein the contrast agent is injected as a series of bolus doses with predetermined intervals between subsequent bolus doses.

[0145] 4. The method of any of the preceding clauses, wherein fluorescence emission from said anatomical structure is measured after injection of each bolus.

[0146] 5. The method of any of the preceding clauses, wherein the bolus comprises gradually increasing or gradually decreasing amounts of the contrast agent.

[0147] 6. The method of clause 5, wherein the amount increases or decreases in 10% increments from one bolus administration to a subsequent bolus administration.

[0148] 7. The method of any of the preceding clauses, wherein a minimum bolus dose that results in a quantifiable fluorescence emission indicative of perfusion of said anatomical structure is identified after administering a series of increasing or decreasing bolus doses.

[0149] 8. The method according to any of the preceding clauses, wherein the interval between bolus administrations is between 5 and 600 seconds, such as between 15 and 300 seconds, such as between 45 and 210 seconds, such as between 90 and 120 seconds.

[0150] 9. The method of any of the preceding clauses, wherein the interval between bolus administrations is long enough to allow measurement of the perfusion gradient for each bolus administration in the anatomical structure, preferably wherein the perfusion gradient includes a gradient onset and a washout gradient.

[0151] 10. The method of any of the preceding clauses, wherein a volume of isotonic fluid (such as saline) is injected immediately after injection of the bolus of fluorescent contrast agent, e.g., the volume is 1 to 20 mL, e.g., 2.5 to 15 mL, e.g., 5 to 10 mL.

[0152] 11. The method according to any of the preceding clauses, wherein the amount of the fluorescent contrast agent corresponds to between 0.0001 and 0.01 mg of ICG / kg body weight per bolus, for example 0.0001 to 0.01 mg of ICG / kg body weight per bolus.

[0153] 12. The method according to any of the preceding clauses, wherein the initial amount of fluorescent contrast agent corresponds to at least 0.001 mg ICG / kg body weight.

[0154] 13. The method of clause 12, wherein the amount of subsequent bolus doses increases or decreases from one bolus dose to the next by at least 0.001 mg ICG / kg body weight.

[0155] 14. The method of any of the preceding clauses, wherein the bolus dose is between 0.5 μL and 10 mL of liquid volume, for example, 0.5 to 5 mL.

[0156] 15. The method of any of the preceding clauses, wherein a second fluorescent imaging agent is administered, said second fluorescent imaging agent having an emission maximum that differs from the emission maximum of said first fluorescent imaging agent by at least 50 nm.

[0157] 16. The method of any of the preceding clauses, wherein a third, fourth, fifth or further fluorescent imaging agent is administered.

[0158] 17. The method of clause 15 or 16, wherein the first and subsequent fluorescent contrast agents are administered alternately.

[0159] 18. The method of clause 15, 16 or 17, wherein the interval between administrations of the different fluorescent imaging agents is half the interval between subsequent administrations of the same fluorescent imaging agent.

[0160] 19. The method of any of the preceding clauses, wherein the fluorescence is automatically detected by illuminating the anatomical structure with a light source capable of exciting the fluorescent contrast agent, and the emission is quantified through a series of fluorescence images of the anatomical structure.

[0161] 20. The method of any of the preceding clauses, wherein the bolus interval is determined by a computer configured to detect the perfusion gradient caused by each bolus administration.

[0162] 21. The method according to any of the preceding clauses, wherein the dose of the fluorescent contrast agent and / or the bolus administration interval are selected to receive an oscillation pattern of the mean intensity of the time series of the ROI.

[0163] 22. The method of any of the preceding clauses, wherein a longer, e.g., at least 1 minute, break is provided between boluses, for example, every 20 boluses, more preferably every 40 boluses, and most preferably every 60 boluses, so as to reduce background fluorescence levels.

[0164] 23. The method of any of the preceding clauses, wherein the amount of fluorescent contrast agent in the bolus is controlled by a computer configured to identify a minimum bolus dose capable of identifying a minimum fluorescence emission indicative of perfusion of the anatomical structure.

[0165] 24. The method of any of the preceding clauses, wherein the anatomical structure is the gastrointestinal tract, preferably including the oral cavity; the pharynx; the small intestine, including the duodenum, jejunum, and ileum; the stomach, including the esophagus, cardia, and pylorus; and the large intestine, including the cecum, colon, rectum, and anal canal.

[0166] 25. The method of any preceding clause, wherein the anatomical structure is an internal organ of the subject. method.

[0167] 26. The method of any of the preceding clauses, wherein the anatomical structure is the skin of the subject.

[0168] 27. The method of any preceding clause, wherein the anatomical structure comprises a wound of the subject of the perfusion assessment.

[0169] 28. The method of any preceding clause, wherein the fluorescent imaging agent comprises indocyanine green (ICG), fluorescein isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, o-phthalaldehyde, fluorescamine, rose bengal, trypan blue, fluorogold, green fluorescent protein, flavin, methylene blue, porphysomes, cyanine dyes, IRDDye800CW, CLR1502 in combination with a targeting ligand, OTL38 in combination with a targeting ligand, or a combination thereof.

[0170] 29. The method of any preceding clause, further comprising tracking movement of the anatomical structure in the images.

[0171] 30. A method according to any of the preceding clauses, further comprising the steps of tracking movement of at least a portion of the anatomical structure in the image and relating the movement so that at least the first region of interest corresponds to the same subsection of the anatomical structure in the image.

[0172] 31. The method of any of the preceding clauses, wherein the movement tracking is provided by free image tracking.

[0173] 32. A method according to any of the preceding clauses, wherein the mobile tracking is provided in the form of classifier-based tracking, including a step of identifying classifiers of one or more recognizable features within the video, preferably in an area adjacent to or surrounding at least one of the regions of interest, by free image tracking.

[0174] 33. The method of any of the preceding clauses, wherein the movement tracking is provided by free image tracking in the form of color-based tracking.

[0175] 34. The method of any of the preceding clauses, wherein movement tracking is based on color tracking of one or more color markers applied to the gastrointestinal tract.

[0176] 35. A method according to any of the preceding clauses, wherein movement tracking comprises steps of colour filtering and thresholding to obtain a Boolean map of pixels in the image.

[0177] 36. The method of any of the preceding clauses, further comprising a step of noise filtering to improve the Boolean map.

[0178] 37. The method of any of the preceding clauses, wherein the movement tracking is provided by object-based tracking.

[0179] 38. The method of any of the preceding clauses, wherein movement tracking is provided by tracking the movement of one or more predetermined objects coupled to said anatomical structure.

[0180] 39. A method according to any of the preceding clauses, comprising the step of creating a template of ROIs by first saving an image of each ROI, wherein motion tracking is provided by applying cross-correlation to the template of each ROI.

[0181] 40. The method of any of the preceding clauses, wherein the tracking is performed such that three-dimensional information of at least a portion of an anatomical structure is obtained.

[0182] 41. The method of any preceding clause, wherein the perfusion assessment includes locating perfusion complications in the anatomical structure.

[0183] 42. The method of any of the preceding clauses, wherein the perfusion assessment is used in connection with a diagnostic or surgical procedure.

[0184] 43. The method of clause 42, wherein the procedure includes diagnostic laparoscopy, exploratory laparoscopy, surgical laparoscopy with conventional laparoscopy, robotic surgery, and open surgery.

[0185] 44. The method of clause 42, wherein the procedure comprises an anastomosis, for example, an intestinal anastomosis.

[0186] 45. A fluorescent contrast agent for use in a method according to any of the preceding clauses.

[0187] 46. ​​Use of a fluorescent contrast agent in the preparation of a drug for use in the automated perfusion assessment method described in any of the preceding clauses 1 to 44.

[0188] 47. A system for automated perfusion assessment of an anatomical structure during a medical procedure on a subject, comprising: a controllable infusion pump for holding at least one first fluorescent contrast agent, the infusion pump configured to inject a predetermined amount of the first fluorescent contrast agent into the blood of the subject, the system configured to receive and analyze time-series fluorescent images of tissue of the anatomical structure after the injection of the first fluorescent contrast agent, and to identify at least one perfusion parameter of the anatomical structure based on the analysis.

[0189] 48. A system according to any preceding clause, wherein the system is configured to control the infusion pump to inject a small initial bolus of fluorescent contrast agent, preferably an amount equivalent to less than 0.01 mg ICG / kg body weight, and then analyze the fluorescence emission resulting from the initial bolus.

[0190] 49. A system according to any preceding clause, wherein the system is configured to control the infusion pump to inject a small initial bolus of fluorescent contrast agent, preferably an amount equivalent to less than 1 mg ICG, or less than 0.8 mg ICG, or less than 0.6 mg ICG, or less than 0.4 mg ICG, or less than 0.2 mg ICG, and then analyze the fluorescence emission resulting from the initial bolus.

[0191] 50. The system comprises: - controlling said infusion pump to inject a series of boluses of gradually increasing or decreasing predetermined amounts of fluorescent contrast agent with predetermined intervals between each bolus administration; - analyzing the fluorescence emission from said anatomical structure after injection of each said bolus; - identifying the minimum bolus dose size that provides a quantifiable fluorescence emission from the anatomical structure; 10. The system of any preceding clause, configured to identify

[0192] 51. The system: 1) time-series recording of the tissue of the anatomical structure prior to injection of the fluorescent agent. 1. The system of any preceding clause, configured to receive a sequence image and 2) identify a background noise level therefrom.

[0193] 52. The system of any of the preceding clauses, wherein the system is configured to identify a subject-specific conversion time defined as the time from injection of a bolus of the fluorescent contrast agent to an increase in a fluorescence gradient in the fluorescence emission.

[0194] 53. A system according to any of the preceding clauses, wherein the system is configured to identify a subject-specific interruption interval defined as the period from the rise of the fluorescence gradient until the fluorescence emission becomes equivalent to the background noise.

[0195] 54. A system according to any preceding clause, wherein the system is configured to automatically: 1) control the infusion pump to inject a series of predetermined bolus doses of fluorescent contrast agent, i.e., predetermined bolus doses such as the minimum effective bolus dose, with predetermined intervals between each bolus dose; and 2) determine at least one perfusion parameter of the anatomical structure after injection of each bolus.

[0196] 55. The system of any of the preceding clauses, wherein the system is configured to identify the at least one perfusion parameter in one or more regions of interest located in the anatomical structure and, optionally, adjacent anatomical structures.

[0197] 56. The system of any of the preceding clauses, wherein the system is configured to automatically: 1) control the infusion pump to inject a series of boluses of gradually increasing or decreasing amounts of fluorescent contrast agent with predetermined intervals between each bolus administration; and 2) determine at least one perfusion parameter of the anatomical structure after injection of each bolus.

[0198] 57. A system according to any preceding clause, including at least a second controllable infusion pump for holding at least a second fluorescent contrast agent different from the first fluorescent agent, wherein the second infusion pump is configured to inject a predetermined amount of the second fluorescent contrast agent into the blood of the subject.

[0199] 58. The system of any of the preceding clauses, wherein the system is configured to identify the at least one perfusion parameter in one or more regions of interest located in the anatomical structure and, optionally, adjacent anatomical structures.

[0200] 59. A system according to any preceding clause 58, configured so that the region of interest can be selected by a user of the system.

[0201] 60. The system of any preceding clause, further comprising at least one light source configured to provide excitation light for inducing fluorescence emission from the first and / or second fluorescent agents in the anatomical structure.

[0202] 61. The system of any preceding clause, further comprising an imaging device configured to record at least one time series of the fluorescent emissions from the anatomical structure.

[0203] 62. The system of any preceding clause, wherein the imaging device is configured for white light imaging.

[0204] 63. A system according to any preceding clause, configured to transfer the at least one perfusion parameter for display on a display.

[0205] 64. The system of any preceding clause, wherein the anatomical structure is the gastrointestinal tract, preferably including the oral cavity; the pharynx; the small intestine, including the duodenum, jejunum, and ileum; the stomach, including the esophagus, cardia, and pylorus; and the large intestine, including the cecum, colon, rectum, and anal canal.

[0206] 65. The system of any preceding clause, wherein the anatomical structure is an internal organ of the subject.

[0207] 66. The system of any preceding clause, wherein the anatomical structure is the skin of the subject.

[0208] 67. The system of any preceding clause, wherein the anatomical structure includes a wound that is the subject of the perfusion assessment.

[0209] 68. A system according to any preceding clause, wherein the system is part of a laparoscopic setup, and the imaging device and the light source are integrated into a laparoscopic device.

[0210] 69. A system according to any preceding clause, wherein the system is part of an open surgery setup, and the imaging device and the light source are integrated into an open surgery operating room.

[0211] 70. The system of any preceding clause, wherein the fluorescent imaging agent comprises indocyanine green (ICG), fluorescein isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, o-phthalaldehyde, fluorescamine, rose bengal, trypan blue, fluorogold, green fluorescent protein, flavin, methylene blue, porphysomes, cyanine dyes, IRDDye800CW, CLR1502 in combination with a targeting ligand, OTL38 in combination with a targeting ligand, or a combination thereof.

[0212] 71. The system of any of the preceding clauses, wherein the system is configured to evaluate the drainage area of ​​a vein or group of veins, a lymphatic vessel, a lymph node or other part of the circulatory and / or lymphatic pathway.

[0213] 72. A system according to any of the preceding clauses, configured to carry out a method according to any of clauses 1 to 44.

[0214] 73. A computer-implemented method for detecting changes in perfusion of at least a portion of an anatomical region, wherein repeatable boluses comprising a fluorescent contrast agent are continuously injected into a subject, said method comprising: i. measuring a time series of fluorescence intensity in a region of interest of at least a portion of an anatomical structure; ii. recognizing a measurement pattern in the measured time series; iii. creating a predicted pattern based on the measured pattern that the fluorescence intensity of at least a portion of the anatomical region of interest is expected to follow; iv. measuring the fluorescence intensity of the anatomical region of interest; v. analyzing the discrepancies between the expected patterns and the measurements; vi. repeating steps iv. and v. to continuously assess the perfusion of the at least part of the anatomical region of interest; The method comprising:

[0215] 74. The user may select the predicted pattern and the measured value according to a predetermined function, such as a predetermined threshold. 74. The computer-implemented method of claim 73, wherein the alert is based on a difference in values.

[0216] 75. The computer-implemented method of any of clauses 73-74, wherein the recognized vibration pattern is continuously updated between steps iv. and v.

[0217] 76. The computer-implemented method of any of clauses 73 to 75, wherein the vibration pattern is recognized based on frequency, amplitude, phase and / or background intensity.

[0218] 77. The computer-implemented method of any of clauses 73-76, wherein said bolus dose is substantially said minimum effective bolus dose.

[0219] 78. The computer-implemented method of any of clauses 73 to 77, wherein injection parameters such as injection frequency are additionally or alternatively used to recognize the vibration pattern.

[0220] 79. The computer-implemented method of any of clauses 73 to 78, wherein the method is used to predict the onset of an ischemic condition.

[0221] 80. The computer-implemented method of any of clauses 73-79, wherein the anatomical region of interest is detected by tracking.

[0222] 81. The computer-implemented method of any of clauses 73 to 80, wherein tracking of the anatomical region of interest is performed in three dimensions.

[0223] 82. The computer-implemented method of any of clauses 73 to 81, wherein the bolus is injected over a longer phase, for example, between 1 and 5 minutes, more preferably between 1 and 4 minutes, even more preferably between 1 and 3 minutes, and most preferably between 1.5 and 2.5 minutes.

[0224] 83. The computer-implemented method of any of clauses 73 to 82, wherein longer breaks, for example 1 to 10 minutes, during which no fluorescent contrast agent is injected into the subject are performed periodically, for example after every 10 to 300 bolus doses, more preferably after every 30 to 300 bolus doses, more preferably after every 90 to 300 bolus doses, and most preferably after every 200 to 300 bolus doses.

[0225] 84. The computer-implemented method of any of clauses 73 to 83, wherein the user is alerted based on a difference between the measured and expected values ​​for the phase, the frequency and / or the amplitude.

[0226] 85. The computer-implemented method of any of clauses 73 to 84, wherein the method is capable of detecting ischemia and / or venous occlusion and / or assessing the perfusion area of ​​an artery.

[0227] 86. The computer-implemented method of any of clauses 73 to 85, wherein the method is configured to compensate measurements for non-continuous intervals, for example when the region of interest drifts in and out of focus, and compare these to the expected pattern.

[0228] 87. A computer-implemented method for detecting changes in perfusion of an anatomical region of interest of a subject by hemodynamic imaging of at least a portion of the region of interest in an image acquired from the subject, the method comprising: - during and / or after multiple boluses containing a fluorescent contrast agent are delivered to said subject; performing an image analysis of at least one acquired video image sequence, wherein the plurality of boluses are delivered according to a predetermined pattern, e.g., in terms of frequency and / or dosage; - calculating subsequent perfusion parameters in one or more regions of interest based on said image analysis; - monitoring said subsequent perfusion parameters to identify changes in perfusion in said region(s) of interest; The method comprising:

[0229] 88. A method according to clause 87, comprising the steps of any of clauses 1 to 46 or clauses 73 to 86.

Claims

1. 1. A fluorescent imaging agent in the form of indocyanine green (ICG) for use in a method for automated perfusion assessment of an anatomical structure of a subject, the method comprising administering a bolus of the fluorescent imaging agent intravenously, the bolus having a predetermined amount of ICG of less than 0.01 mg / kg body weight; acquiring and analyzing time-series fluorescent images of tissue of the anatomical structure after the injection of the bolus; and determining at least one perfusion parameter of the anatomical structure based on the analysis.

2. 2. The fluorescent imaging agent of claim 1, wherein the interval is between 15 and 300 seconds.

3. 3. The fluorescent imaging agent of claim 2, wherein the interval is between 45 and 210 seconds.

4. 4. The fluorescent imaging agent of claim 3, wherein the interval is between 90 and 120 seconds.

5. 5. The fluorescent contrast agent according to claim 1, wherein the predetermined amount of ICG in each bolus corresponds to 0.0001 to 0.01 mg / kg body weight.

6. 6. The fluorescent contrast agent according to claim 1, wherein the predetermined amount of ICG in each bolus corresponds to 0.001 to 0.01 mg / kg body weight.

7. 7. The fluorescent contrast agent according to claim 1, wherein the initial amount of ICG is equivalent to at least 0.001 mg / kg body weight.

8. 8. The fluorescent imaging agent of claim 1, wherein the subsequent boluses increase or decrease from one bolus to the next by an amount corresponding to at least 0.001 mg / kg body weight of ICG.

9. 9. The fluorescent imaging agent of claim 1, wherein the bolus comprises gradually increasing or gradually decreasing amounts of the fluorescent imaging agent.

10. 10. The fluorescent contrast agent of claim 1, wherein the predetermined amount of ICG in a bolus increases in 10% increments or decreases in 10% decrements from one bolus to a subsequent bolus.

11. 11. The fluorescent imaging agent of any one of claims 1 to 10, wherein the minimum bolus that results in a quantifiable fluorescence emission representative of perfusion of the anatomical structure is determined after administering a series of increasing or decreasing boluses.

12. 12. A fluorescent imaging agent according to any preceding claim, wherein the interval between boluses is long enough to allow measurement of the perfusion gradient for each bolus in the anatomical structure.

13. The fluorescent contrast agent of any of claims 1 to 12, wherein a volume of isotonic solution is injected immediately after the administration of the bolus of fluorescent contrast agent.

14. 14. The fluorescent imaging agent of claim 13, wherein the isotonic solution is saline.

15. 15. The fluorescent imaging agent of claim 13 or claim 14, wherein the volume is 1 to 20 mL.

16. 16. The fluorescent imaging agent of claim 15, wherein the volume is between 2.5 and 15 mL.

17. 17. The fluorescent imaging agent of claim 16, wherein the volume is 5 to 10 mL.

18. 18. The fluorescent imaging agent of any of claims 1 to 17, wherein each bolus is a liquid volume of between 0.5 μL and 10 mL.

19. 19. The fluorescent imaging agent of claim 18, wherein the liquid volume is 0.5 to 5 mL.

Images