Measuring system for intraoperative bleeding volume

Abstract

The present application discloses a measuring system for an intraoperative bleeding amount. An intraoperative bleeding amount measuring system includes: a body fluid collection device used to collect fluid suctioned by various suction devices during an operation and measure a volume of the fluid; a gauze collection device used to collect gauzes inserted into a surgical area of ​​a patient during the operation and measure the weight of the gauzes; a first camera used to acquire first image information of the fluid in the body fluid collection device; a second camera used to acquire an image of each gauze entering the gauze collection device to obtain second image information; and a measuring and calculating device for determining the volume, weight, color, etc.a first bleeding amount and a second bleeding amount are calculated, and a total intraoperative bleeding amount is calculated. Compared to conventional technology that measures or calculates the bleeding amount using a fluid volume, the technical solution of the present application also fuses the first image information based on the fluid volume.

Description

Area of registration

[0001] The present application relates to the technical field of medical devices, in particular to a measuring system for an intraoperative bleeding amount. State of the art

[0002] Bleeding is more or less inevitable during surgical operations. In most cases, patients can tolerate a certain amount of blood loss without risking their lives, which depends on the individual condition of each patient (e.g., height, weight, whether they were anemic before surgery, etc.) and the rate of blood loss. Therefore, an individual and dynamic assessment of the patient's specific condition is required during each operation. If necessary, emergency blood transfusion (mainly of suspended red blood cells) is required to maintain stable vital functions and ensure patient safety. However, blood transfusions carry many risks, such as allergies, bloodborne diseases, hemolysis, etc. Therefore, voluntary blood transfusions are generally not given to the patient during surgery.This means that the following situation will not occur: The patient will be transfused as much blood as he loses during the operation.

[0003] However, there is currently a lack of tools or devices for accurately calculating intraoperative blood loss. Intraoperative blood loss is mainly estimated comprehensively based on an empirical estimation of the patient's bleeding volume by surgeons and anesthesiologists and dynamic monitoring of the patient's hemoglobin level; and on this basis, it is determined whether the patient requires a blood transfusion and how much blood transfusion is required. However, existing assessment measures can be affected by several factors, such as:Different leakage conditions in the surgical field due to different disease states and different concentrations of body fluids in the patient's body can be affected, which makes it difficult to accurately assess the specific amount of blood loss and can even lead to major errors that can result in unnecessary blood transfusions for patients or, if blood transfusion is not administered in a timely manner, life-threatening situations.

[0004] Blood is a red, opaque, viscous fluid that flows through the blood vessels and heart of humans. It is composed of blood cells and plasma. Plasma contains water, proteins, electrolytes, organic compounds, and other components. Blood cells include red blood cells, white blood cells, and platelets. Red blood cells are red in color, and their primary function is to transport oxygen and expel carbon dioxide. Plasma is generally light yellow in color, and its main functions are nutrition, lipid transport, buffering, osmotic pressure, and participation in immunity, coagulation, and other functions. Red blood cells are most commonly transfused during surgery, followed by plasma. As mentioned above, the functions of the two are different.The surgeon and anesthesiologist must determine whether a blood transfusion is necessary, which blood component is required, and how much of this blood component needs to be transfused based on the blood loss status during surgery. Therefore, the assessment of the blood loss status should include not only an assessment of the volume but also an assessment of the color of the lost fluid (the more red blood cells this fluid contains, the redder it is; the fewer red blood cells it contains, the lighter it is). If a patient loses too many red blood cells and has too few red blood cells in the body, the patient must be given a red blood cell transfusion. During surgery, the patient will experience blood loss, exudation, intravenous infusion, etc. The concentration of red blood cells in the lost fluid (including lost blood) and the body's blood is constantly changing.Since existing assessment measures are based solely on estimating the volume of fluid lost and the recorded hemoglobin concentration (percentage of red blood cells) in the patient, they may sometimes not accurately reflect the patient's actual bleeding status and transfusion needs.

[0005] For example, patients with advanced ovarian cancer often experience large amounts of fluid leakage during surgery. Even if a large number of red blood cells are lost due to bleeding, the patient's hemoglobin concentration may be within the normal range or only slightly below normal due to the concentration of blood in the body. This does not reflect a real condition and may therefore delay the timing of blood transfusion treatment.

[0006] Therefore, there is currently no monitoring system that can accurately measure a patient's actual bleeding volume. This can easily lead to risks associated with surgery.

[0007] US 2014 / 0 128 838 A1 discloses a measuring system for an intraoperative bleeding amount, essentially with the features of the preamble of the main claim, wherein, however, the gauze collecting device is not used to measure the weight of the gauze, the gauze collecting device is not signal-connected to the measuring and calculating device, and the measuring and calculating device does not analyze the second bleeding amount depending on the weight of the gauze in the gauze collecting device.

[0008] US 2021 / 0 052 342 A1 discloses a collection device for increasing the accuracy of the amount of bleeding obtained from image information by means of weight measurement. Task of revelation

[0009] The "Disclosure of the Invention" section is used to present concepts in a simplified and concise form. These concepts are described in detail in the following "Detailed Embodiments" section. The "Disclosure of the Invention" section is neither intended to identify key features or essential features of the claimed technical solutions, nor should it be used to limit the scope of the claimed technical solution.

[0010] To solve the technical problems mentioned in the above-mentioned "Prior Art" section, a measuring system for an intraoperative bleeding amount is provided, which comprises the features of the main claim. Advantageous embodiments are the subject of the dependent claims.

[0011] Compared with conventional technology that measures or calculates the amount of bleeding using a volume of liquid, the technical solution of the present application also fuses the first image information based on the volume of liquid, and analyzes a concentration of red blood cells in the liquid depending on the first image information, so that the information on the concentration of red blood cells is fused in the obtained information on the first bleeding amount, whereby the amount of bleeding of the patient can be measured more accurately; moreover, the weight of the gauze collected during the operation is measured and an image of the gauze is taken, so that the volume of liquid soaked in the gauze is analyzed and the amount of bleeding on the gauze is analyzed, whereby the first bleeding amount and the second bleeding amount respectively detected by the body fluid collecting device and the body fluid collecting device are calculated.the gauze collection device are used to calculate a total amount of bleeding and thus the patient's bleeding status can be accurately assessed during the operation.

[0012] It is further provided that the first bleeding amount comprises a fluid volume and a red blood cell concentration; wherein the second bleeding amount comprises a net weight gain of each gauze and a red blood cell content in each gauze; and wherein the total bleeding amount comprises a fluid loss amount and a red blood cell loss amount.

[0013] In the technical solution of the present application, by calculating the amount of fluid loss and the amount of red blood cell loss, the bleeding condition of the patient during the operation can be well assessed.

[0014] It is further provided that the body fluid collection device comprises a liquid storage tank and a liquid sensor arranged on an inner wall of the liquid storage tank; wherein the liquid sensor is arranged along a height direction of the inner wall of the liquid storage tank.

[0015] In the technical solution of the present application, by disposing a liquid sensor on the inner wall of the liquid storage tank, the height of the liquid level of the liquid in the liquid storage tank can be detected in real time, thereby calculating the volume of the liquid in the liquid storage tank.

[0016] The fluid stored in the fluid storage tank tends to stratify during the operation, and the distribution of red blood cells in the fluid is uneven. Therefore, using the color depth of the fluid in the fluid storage tank directly to determine the red blood cell concentration is a major error. To this end, the present application provides the following technical solution: It is further provided that the liquid storage tank consists of a transparent material, wherein the first camera is aligned with a side surface of the liquid storage tank.

[0017] In the technical solution of the present application, since the liquid storage tank is made of a transparent material and the first camera is directed toward a side surface of the liquid storage tank, image information about the stratification of the liquid in the liquid storage tank can be accurately obtained. After analyzing this image information, the concentration distribution of red blood cells in the liquid storage tank can be more accurately calculated, thereby roughly determining the red blood cell concentration data.

[0018] When using gauze, a physician does not replace each gauze with a new one after soaking it with liquid. Therefore, the usage status of each gauze is not consistent, so the red blood cell concentration must be determined independently for each gauze. If the gauze is placed directly in a container and then image information of all gauze is obtained, subsequent calculations become more difficult because it becomes difficult to detect gauze at different positions in the same image, and after stacking the gauze, obtaining image information of the most recent gauze may be impossible. In light of this problem, the present application provides the following steps: It is further provided that the gauze collecting device comprises a collecting pipe, an electric door at an inlet of the collecting pipe, a pressure sensor on the electric door and a control device which is connected to the electric door and the pressure sensor, respectively; wherein the second camera is arranged directly above the entrance of the collecting tube, wherein upon placing a gauze on the collecting tube, the control unit receives data from the pressure sensor and controls a delayed opening of the electric door so that the second camera captures an image of each gauze placed on the electric door.

[0019] In the technical solution of the present application, a pressure sensor is provided on the electric door, so that the electric door opens with a delay after being subjected to pressure. Therefore, the second camera can capture a single image of each gauze. At the same time, the measuring and calculating device can also determine the gravity of each gauze. Therefore, when the measuring and calculating device measures or calculates the second bleeding amount, it only needs to perform independent calculations for each gauze to obtain an accurate second bleeding amount.

[0020] The gauze collection device separately calculates the weight of each gauze and the corresponding red blood cell concentration to avoid repeated calculation of the blood weight remaining on the electric door, which can lead to inaccurate calculation of the intraoperative bleeding amount. The present application provides the following technical solution: It is further provided that the gauze collecting device further comprises a weight measuring module, wherein the weight measuring module is used to measure the gravity of the collecting tube.

[0021] This solution uses the gauze collection device to measure the gravity of the entire collection tube. Therefore, when a new gauze is placed on the electric door, the increased weight of the collection tube corresponds to the weight of the new gauze.

[0022] Determining red blood cell concentration using only the first image information and the second image information is, in practice, easily influenced by a light source. Different lighting environments could lead to different measurement results. To this end, the present application provides the following technical solution: The intraoperative bleeding amount measuring system further comprises a first laser and a second laser; wherein the liquid storage tank is arranged between the first laser and the first camera, wherein the first laser performs a transmission on the liquid storage tank; wherein the second laser is arranged on the electric door, wherein the second laser performs a transmission on the gauze on the electric door.

[0023] In the technical solution of the present application, a corresponding laser is provided for both the first camera and the second camera, whereby by filling the light with the lasers the difference in the ambient light and its influence on the measurement results can be minimized and the measurement accuracy is increased.

[0024] The analysis of the first image information and the analysis of the second image information essentially involves color differentiation of corresponding features in an image, that is, the corresponding red blood cell concentration is determined based on the color depth. Regardless of whether it is blood-soaked gauze or blood in the liquid storage tank, the color difference is very small. Using RGB information alone for color classification may not necessarily produce the expected effects. In view of this problem, the present application provides the following technical solution: It is further provided that both the first laser and the second laser are multi-wavelength lasers; wherein the first image information includes image information about the first laser performing transmission at the liquid storage tank with light beams of multiple wavelengths; and wherein the second image information includes image information about the fact that the second laser transmits light beams of multiple wavelengths to the gauze.

[0025] In the technical solution of the present application, after using multiple wavelengths of light rays to transmit the gauze or the liquid storage tank, the color difference of the gauze or blood can be further analyzed, whereby the concentration of red blood cells can be accurately distinguished.

[0026] The influence of blood concentration on color is relatively complex. Relying solely on the single color red makes it difficult to analyze blood concentration. This means that the relationship between the single color red and blood concentration is ambiguous. Using only this single color as a feature for subsequent model building can easily lead to overfitting of the model, ultimately resulting in the detected results being inconsistent with the expected results.

[0027] It is further provided that the measuring and calculation device processes the first image information and / or the second image information in order to obtain a corresponding color feature F: F=(w1,1(B / G)1,w2,1(B / G)2 wk,1(B / G)kw1,2(G / R)1,w2,2(G / R)2…wk,2(B / G)kw1,3(R / B)1,w2,3(R / B)2 wk,3(B / G)k); where R, G and B represent a value of three channels, namely a red channel, a green channel and a blue channel, respectively, in image information; where (B / G)1, (G / R)1 and (R / B)1 each represent a ratio of the respective channel of image information at a first transmissive wavelength; where w 1,1 , w 1,2 , w 1,3 each represents a weight coefficient at the first transmissive wavelength, which is preset, where w 1,1 + w 1,2 + w 1,3 = 1 applies; where w 2,1 , w 2,2 , w 2,3 each represent a weight coefficient at a second transmissive wavelength, which is preset, where w 2,1 + w 2,2 + w 2,3 = 1 applies; where w k,1 , w k,2 , w k,3 each represents a weight coefficient at a k-th transmissive wavelength, which is preset, where w k,1 + wk,2 + w k,3 = 1 applies; where k represents the type of wavelength of the transmissive laser light, with k≥4, and where k is an integer.

[0028] In the technical solution of the present application, an appropriate weight coefficient is preset, which can be adaptively adjusted when a target object is transmitted by different transmissive light beams, whereby the information difference in the color of blood of different concentrations can be better represented, so that a closer relationship between color characteristics and blood concentration is established and the accuracy of concentration determination is increased.

[0029] In summary, compared to conventional technology that uses a volume of fluid to measure the amount of bleeding orcalculated, the technical solution of the present application also fuses the first image information based on the liquid volume, and analyzes a red blood cell concentration in the liquid depending on the first image information, so that the red blood cell concentration information is fused in the obtained first bleeding amount information, thereby more accurately measuring the patient's bleeding amount; furthermore, the weight of the gauze collected during the operation is measured and an image of the gauze is taken, so that the volume of liquid soaked in the gauze is analyzed and the bleeding amount on the gauze is analyzed, thereby calculating the first bleeding amount and the second bleeding amount respectively detected by the body fluid collecting device and the blood vessel.the gauze collection device are used to calculate a total amount of bleeding and thus the patient's bleeding status can be accurately assessed during the operation. Brief description of the drawings

[0030] The drawings, which are part of the present application, are intended to provide a further understanding of the present application so that further features, objects, and advantages of the present application will become more apparent. The drawings of the schematic embodiments of the present application and their descriptions are intended to illustrate the present application and do not constitute an undue limitation of the present application.

[0031] Furthermore, the same or similar reference numerals designate the same or similar elements throughout the drawings. It should be understood that the drawings are schematic, and components or elements are not necessarily drawn to scale. Fig. 1 shows a logic diagram of an intraoperative bleeding amount measuring system for measuring bleeding amount. Fig. 2 shows a schematic view of positions of a body fluid collection device, a first camera and a first laser. Fig. Figure 3 shows a schematic view of a gauze collecting device and a second camera. List of reference symbols:

[0032] 11-first camera; 12-fluid storage tank; 13-first laser; 21-manifold; 22-electric door; 221-door plate; 222-cylinder; 23-second camera. Description of the preferred embodiments

[0033] In the following, embodiments of the present application are described in more detail with reference to the drawings. Although certain embodiments of the present application are illustrated in the drawings, it is to be understood that the present application may be implemented in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the present application. It is to be understood that the drawings and embodiments of the present application are exemplary only and are not to be used to limit the scope of the present application.

[0034] It should also be noted that, to facilitate the description, only the parts relevant to the present invention are shown in the drawings. The embodiments of the present application and the features in these embodiments may be combined with one another without conflict.

[0035] The present application will now be described in detail with reference to the drawings in conjunction with the embodiments.

[0036] With reference to Fig. 1, a measurement system for an intraoperative bleeding amount comprises a body fluid collection device, a gauze collection device, a first camera, a second camera, and a measuring and calculating device. The body fluid collection device is used to collect fluid suctioned by various suction devices during an operation and to measure a volume of the fluid. The gauze collection device is used to collect gauzes used on a patient during the operation and to measure the weight of the gauzes. The first camera is used to acquire first image information of the fluid in the body fluid collection device. The second camera is used to acquire an image of each gauze that enters the gauze collection device to obtain second image information.The measuring and calculating device analyzes an intraoperative first bleeding amount depending on the volume of the fluid in the body fluid collection device and the first image information of the fluid, wherein the measuring and calculating device analyzes an intraoperative second bleeding amount depending on the weight of the gauze in the gauze collection device and the second image information of the gauze, and wherein the measuring and calculating device calculates an intraoperative total bleeding amount depending on the first bleeding amount and the second bleeding amount.

[0037] It is further provided that the first bleeding amount comprises a fluid volume and a red blood cell concentration; wherein the second bleeding amount comprises a net weight gain of each gauze and a red blood cell content in each gauze; and wherein the total bleeding amount comprises a fluid loss amount and a red blood cell loss amount. It is prior art how, given a known corresponding red blood cell concentration and a known red blood cell content, volume and weight are converted to determine the total bleeding amount based on the weight of blood collected in the gauze collection device and the blood volume collected in the body fluid collection device. This part will therefore not be discussed further in the present application.

[0038] With reference to Fig. 2, the body fluid collection device specifically includes a fluid storage tank 12 and a fluid sensor. The fluid sensor is arranged along a height direction of the inner wall of the fluid storage tank 12. The fluid sensor is essentially a fluid level sensor that determines the fluid volume in the fluid storage tank 12 based on the immersion depth of the blood. The blood in the fluid storage tank 12 here is a blood-containing fluid that is passed through a flow guide device during a surgery. Furthermore, the fluid storage tank 12 is made of a transparent material, and the first camera is directed toward a side surface of the fluid storage tank 12.To facilitate the first camera 11 to collect more color information, the measurement system in practice further includes a first laser 13, with the liquid storage tank 12 located between the first laser 13 and the first camera 11. Thus, after the laser light generated by the first laser 13 is irradiated onto the liquid storage tank 12 and then collected by the first camera 11, the image information of the liquid storage tank 12 can be obtained.

[0039] The main structure of the body fluid collection device is described above. The key to the body fluid collection device is that the fluid storage tank must be made of a transparent material, and the edge of the fluid storage tank must be a flat structure that does not have any bends and does not scatter light. Typically, the fluid storage tank is constructed in a cube shape. The arrangement of the first camera and the first laser will not be discussed further here.

[0040] The body fluid collection device is used to collect the body fluid (blood) removed from the patient during surgery. However, a lot of blood is lost from the gauze when it is soaked in it. For this purpose, a gauze collection device is required to capture the amount of blood lost on the gauze.

[0041] With reference to Fig.3, the gauze collecting device specifically includes a collection tube 21, an electric door 22, a pressure sensor, a weight measuring module, and a controller. The collection tube 21 is arranged on the weight measuring module. The weight measuring module is a module with a weighing function and can weigh the objects in the collection tube. In practice, after the gauze is placed into the collection tube, the weight measuring module can determine the weight of the newly introduced gauze into the collection tube based on the weight change, and then subtract the initial weight of the gauze to obtain the weight (mass) of the blood water in the gauze. However, there are a plurality of gauzes in the collection tube; and when the gauze falls into the collection tube, blood is easily thrown out of the gauze, causing the color of the gauze to become lighter. This affects the accuracy of determining the amount of bleeding. For this purpose, an electric door 22 is provided above the collection tube.The electric door 22 is not directly connected to the manifold 21, but is fixed to the floor via a set of support devices (omitted from the drawings). Specifically, the electric door 22 includes a door plate 221 and a cylinder 222. Two door plates 221 are provided, and the two door plates 222 are arranged symmetrically on the left and right sides of the manifold 21. In this way, under the action of the cylinder 222, the door plate can be pressed to open or close the manifold 21. Furthermore, the pressure sensor is attached to the door plate, and the control device is signal-connected to the pressure sensor and the cylinder, respectively. After detecting that the pressure sensor on the door plate is transmitting a signal, the control device decelerates for a preset period and then starts the cylinder, causing the door plate 221 to move toward each other, thereby allowing the gauze to fall into the manifold.The second camera 23 is located directly above the electric door 22. By opening the electric door 22 after a preset delay time, the second camera 23 can capture the required image information.

[0042] More specifically, the measurement system further includes a second laser arranged on the door panel 221. The position where the gauze is placed on the door panel is the position where the second laser emits the laser light. The second camera is arranged directly above the gauze. Therefore, after the gauze is placed on the door panel, the second laser emits laser light of different colors in a preset order, allowing the second camera to obtain the corresponding image information.

[0043] Please note: To prevent the gauze from sticking to the electric door and thus falling into the collection tube, the acceleration of the door plate movement can be increased or a corresponding tilting piece can be provided so that the gauze can fall normally into the collection tube. This specific adjustment method is a common technical means in this field, so no further examples are given here.

[0044] Furthermore, both the first laser and the second laser are multi-wavelength lasers; wherein the first image information includes image information that the first laser performs transmission at the liquid storage tank with light beams of multiple wavelengths; and wherein the second image information includes image information that the second laser performs transmission at the gauze with light beams of multiple wavelengths.

[0045] The wavelength type and number of wavelengths of the laser light emitted by the first laser and the second laser are preset. The exact number of types and corresponding wavelengths are not specified here.

[0046] When the measuring and computing device processes the first image information and the second image information, it actually extracts the color features of a target area. Differences in the color features indicate differences in the red blood cell concentration. Although the mapping between the concentration and color of red blood cells in the liquid differs from the mapping between the concentration and color of red blood cells on the gauze, the logic of both is the same. Thus, only the same model trained on a corresponding dataset is required. For this purpose, only one of the ways in which the measuring and computing device processes the first image information and the second image information is given as an example. The following is the expression of a color feature F extracted from the image information by the measuring and computing device: F=(w1,1(B / G)1,w2,1(B / G)2 wk,1(B / G)kw1,2(G / R)1,w2,2(G / R)2…wk,2(B / G)kw1,3(R / B)1,w2,3(R / B)2 wk,3(B / G)k); where R, G and B represent a value of three channels, namely a red channel, a green channel and a blue channel, respectively, in image information; where (B / G)1, (G / R)1 and (R / B)1 each represent a ratio of the respective channel of image information at a first transmissive wavelength; where w 1,1 , w 1,2 , w 1,3 each represents a weight coefficient at the first transmissive wavelength, which is preset, where w 1,1 + w 1,2 + w 1,3 = 1 applies; where w 2,1 , w 2,2 , w 2,3 each represent a weight coefficient at a second transmissive wavelength, which is preset, where w 2,1 + w 2,2 + w 2,3 = 1 applies; where w k,1 , w k,2 , w k,3each represents a weight coefficient at a k-th transmissive wavelength, which is preset, where w k,1 + w k,2 + w k,3 = 1 applies; where k represents the type of wavelength of the transmissive laser light, with k≥4, and where k is an integer.

[0047] Since color irregularities rarely occur on a gauze in practice, for the second image information, after defining the edge contour of the gauze, an average value of the three channels is calculated for all pixels within the edge contour, thus obtaining the color feature F.

[0048] The edge contour of the gauze can be extracted using common algorithms such as edge detection algorithms. In practice, the surface of the door panel can be set to green to reduce the difficulty of contour detection. The specific edge contour extraction method is an existing technology and will not be further demonstrated here.

[0049] The blood is present in the liquid storage tank in the form of a liquid and may be subject to a stratification phenomenon. In some embodiments, edge detection may be performed for the stratification phenomenon to determine the edge of each liquid layer and the corresponding liquid volume of each layer; then, the red blood cell concentration in each layer is calculated separately.

[0050] When a measuring and calculating device calculates red blood cell concentration, the calculation of red blood cell concentration is the most important aspect. To this end, the present application provides a method for calculating red blood cell concentration according to the following solution: S1: m data sets A i , A i ={F i , C q} are prepared in advance; where F i stands for a color feature in the i-th data set; where C q stands for a corresponding red blood cell concentration in the dataset; with q∈Z; where Z stands for a preset gradient set for the red blood cell concentration; where i stands for an index of the dataset, with i∈m.

[0051] Specifically, in this solution, the corresponding concentration gradients are preset according to a required accuracy. When the concentration gradients are set to 5, 1 ≤ q ≤ 5. If q = 1, this means that the current color feature corresponds to a first concentration gradient. The above illustrates how dataset A is set in the present application. It is predictable that after completing subsequent training of a classification model, the corresponding concentration gradient can be obtained by inputting the color features into the classification model.

[0052] S2: For each red blood cell concentration C, its prior probability P(C) is calculated, which is then subjected to Laplace smoothing: P(C)=∑i=1mδ(Ci, C)+1m+NC; where m is the number of training sets, where c istands for a category label of an i-th training set, where N c stands for a number of possible values of the category label, where δ (c i , c) stands for an indicator function which at c i = c takes a value of 1, otherwise 0;

[0053] S3: For each feature F i and each category C will have a conditional probability P(F i | C) which is then subjected to Laplace smoothing P(Fi|C)=∑i=1mδ(Fin,Fn)δ(Ci,c)+1∑i=1mδ(ci,c)+Nn; where Fin represents a value of an n-th element of the color feature Fi in the training set i, with n∈3k; where Fn represents an n-th feature in the color feature, where Nn represents a number of possible values of the n-th feature;

[0054] S3: For a new training set, its feature value F is extracted, using the previously calculated prior probability and conditional probability to calculate a posterior probability of this training set associated with each category, with the category with the largest posterior probability being selected as the prediction category; P(c|x)=P(c)∏nP(Fn|C)P(x); where x represents the new training set, where P(C | X) represents a posterior probability of the training set X associated with a concentration C of red blood cells; where P(C) is the prior probability of the red blood cell concentration C; where P(Fn | C) represents a probability of occurrence of the characteristic Fn under the condition of concentration C of red blood cells; and where P(x) represents a probability of occurrence of the training set X.

[0055] The above three steps are specific implementations for the measurement and calculation device for calculating red blood cell concentration. In subsequent steps, the model must also be evaluated, optimized, and modified.

[0056] In the solution of the present application, the color feature F is used to calculate the concentration, and the color feature does not consist of a single color. To increase the accuracy of the model, this solution acquires images under different laser light conditions to obtain corresponding color features F. This color feature F has the following problems in practice: Because the color feature F is too refined, the color feature data is concentrated in the middle region of the concentration gradient; at the edge of the minimum concentration gradient and the maximum concentration gradient, the number of color features F is very small; thus, the model is easily overfitted due to uneven data distribution. To this end, Laplacian smoothing based on the Bayesian algorithm is introduced in this solution. Laplacian smoothing can reduce data sparsity and overfitting of training data.

[0057] Furthermore, in the solutions of the present application, the measurement and calculation device applies the above-mentioned concept to calculate the blood concentration in the liquid storage tank and the blood concentration in the gauze. In practice, two different classification models must be set up in the measurement and calculation device, respectively, and the structures of these classification models are the same. However, different data are required for training and optimization.

[0058] The solutions of the present application combine the Naive Bayes algorithm and the Laplace algorithm, which can improve the generalization ability of the model and reduce the need for training data sets.

[0059] The embodiments of the present application and the technical principles employed are described above. Those skilled in the art should understand that the scope of the invention pertaining to the embodiments of the present application is not limited to technical solutions consisting of specific combinations of the above-mentioned technical features, but is intended to cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the inventive concept described above. These other technical solutions include, for example, technical solutions obtained by replacing the above-mentioned features with technical features having similar functions to those disclosed in the embodiments of the present application (but are not limited to them).

Claims

[1] Measuring system for an intraoperative bleeding volume, characterized by that it includes: a body fluid collection device used for collecting fluid suctioned by various suction devices during a surgery and for measuring a volume of the fluid; a gauze collection device used to collect gauzes used in a surgical area of a patient during surgery and to measure the weight of the gauzes; a first camera used to obtain first image information of the fluid in the body fluid collection device; a second camera used to acquire an image of each gauze entering the gauze collecting device to obtain second image information; a measuring and calculating device signal-connected to each of the body fluid collecting device, the gauze collecting device, the first camera, and the second camera; wherein the measuring and calculating device analyzes an intraoperative first amount of bleeding depending on the volume of fluid in the body fluid collection device and the first image information of the fluid; wherein the measuring and calculating device analyses an intraoperative second bleeding amount depending on the weight of the gauze in the gauze collecting device and the second image information of the gauze; and wherein the measuring and calculating device calculates an intraoperative total bleeding amount depending on the first bleeding amount and the second bleeding amount characterized by , that the gauze collecting device comprises a collecting pipe, an electric door at an inlet of the collecting pipe, a pressure sensor on the electric door, and a control device connected to the electric door and the pressure sensor, respectively; wherein the second camera is arranged directly above the entrance of the collecting tube, wherein upon placing a gauze on the collecting tube, the control unit receives data from the pressure sensor and controls a delayed opening of the electric door such that the second camera captures an image of each gauze placed on the electric door. [2] Intraoperative bleeding quantity measuring system according to claim 1, characterized by that the gauze collecting device further comprises a weight measuring module, wherein the weight measuring module is used to measure the gravity of the collecting tube. [3] Intraoperative bleeding quantity measuring system according to claim 1, characterized by , that The body fluid collection device comprises a liquid storage tank and a liquid sensor arranged on an inner wall of the liquid storage tank; wherein the liquid sensor is arranged along a height direction of the inner wall of the liquid storage tank; the intraoperative bleeding amount measuring system further comprises a first laser and a second laser; wherein the liquid storage tank is arranged between the first laser and the first camera, wherein the first laser performs a transmission on the liquid storage tank; wherein the second laser is arranged on the electric door, wherein the second laser performs a transmission on the gauze on the electric door. [4] Intraoperative bleeding quantity measuring system according to claim 3, characterized by that both the first laser and the second laser are multi-wavelength lasers; wherein the first image information includes image information about the first laser performing transmission at the liquid storage tank with light beams of multiple wavelengths; and wherein the second image information includes image information about the fact that the second laser transmits light beams of multiple wavelengths to the gauze. [5] Intraoperative bleeding quantity measuring system according to claim 4, characterized by that the measuring and calculation device processes the first image information and / or the second image information to obtain a corresponding color feature F: F=(w1,1(B / G)1,w2,1(B / G)2 wk,1(B / G)kw1,2(G / R)1,w2,2(G / R)2…wk,2(B / G)kw1,3(R / B)1,w2,3(R / B)2 wk,3(B / G)k); where R, G and B represent a value of three channels, namely a red channel, a green channel and a blue channel, respectively, in image information; where (B / G)1, (G / R)1 and (R / B)1 each represent a ratio of the respective channel of image information at a first transmissive wavelength; where w 1,1 , w 1,2 , w 1,3 each represents a weight coefficient at the first transmissive wavelength, which is preset, where w 1,1 + w 1,2 + w 1,3 = 1 applies; where w 2,1 , w 2,2 , w 2,3 each represents a weight coefficient at a second transmissive wavelength that is preset, where w 2,1 + w 2,2 + w 2,3 = 1 applies; where w k,1 , w k,2 , w k,3 each represents a weight coefficient at a k-th transmissive wavelength, which is preset, where w k,1 + w k,2 + w k,3 = 1 applies; where k represents the type of wavelength of the transmissive laser light, with k≥4, and where k is an integer.

Images