Operation decision-making method and device based on liver image, equipment and medium

Abstract

The invention provides an operation decision-making method and device based on a liver image, equipment and a medium, relates to the technical field of image processing, and is suitable for the field of medical health. The method comprises the following steps: performing blood vessel branch feature extraction on a liver biopsy slice image to obtain liver blood vessel branch features; performing Shannon entropy calculation on the liver blood vessel branch features to obtain the liver blood vessel Shannon entropy; performing time sequence entropy change fitting according to each liver blood vessel Shannon entropy in the liver image sequence to obtain liver blood vessel entropy change fitting data; amplifying a quadratic function coefficient in the liver blood vessel entropy change fitting data to obtain a microvessel index; performing variation calculation according to any two microvessel indexes to obtain a microvessel entropy change rate; performing fractal detection on the hepatic angiography image to obtain a hepatic blood vessel fractal dimension; and predicting the liver operation according to the capillary entropy change rate and the liver blood vessel fractal dimension to obtain an operation category. The operation decision accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, is applicable to the field of medical health, and particularly relates to a surgical decision-making method and apparatus, equipment, and medium based on liver images. Background Art

[0002] Image processing technology can detect disease states in medical images, thereby predicting whether a patient can undergo surgery and anticipating surgical risks. For example, in the healthcare field, pathological analysis, imaging testing, and clinical decision-making processes operate independently and lack systematic integration. This makes it impossible to accurately determine whether liver surgeries such as transjugular intrahepatic portosystemic shunts can be performed, resulting in poor surgical decision-making accuracy. For example, pathological analysis relies on static observations and cannot capture dynamic changes in the liver. Another example is that clinical decision-making relies primarily on human experience and judgment, which is easily influenced by subjective factors. This fragmented technology makes it impossible to form an effective closed-loop system for the entire diagnosis and treatment process, making it difficult to achieve accurate surgical decisions.

[0003] Therefore, the related technology has the problem of low accuracy of surgical decision-making. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to propose a surgical decision-making method and device, equipment, and medium based on liver images, which can reflect the dynamic changes of the liver by combining the entropy change rate, and also combine the fractal dimension to make surgical decisions together, thereby improving the accuracy of surgical decisions.

[0005] To achieve the above objectives, a first aspect of an embodiment of the present application provides a surgical decision-making method based on liver images, the method comprising:

[0006] Acquiring a liver image sequence of a target object, wherein the liver image sequence includes at least two liver biopsy slice images;

[0007] performing vascular branch feature extraction on each of the liver biopsy slice images to obtain liver vascular branch features;

[0008] Performing Shannon entropy calculation on the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy;

[0009] performing time series entropy change fitting according to the Shannon entropy of each liver blood vessel in the liver image sequence to obtain liver blood vessel entropy change fitting data;

[0010] amplifying the quadratic function coefficient in the liver vascular entropy change fitting data to obtain a microvascular index;

[0011] Calculating the change of any two microvascular indices to obtain the microvascular entropy change rate;

[0012] Acquiring a liver angiography image of the target object, and performing fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels;

[0013] Liver surgery is predicted according to the microvascular entropy change rate and the liver blood vessel fractal dimension to obtain a surgery category; wherein the surgery category represents whether the liver surgery is allowed or prohibited.

[0014] Optionally, the surgery category includes a first surgery category, and the predicting of liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgery category includes:

[0015] Comparing the liver blood vessel fractal dimension with a preset fractal dimension threshold to obtain a fractal dimension comparison result;

[0016] If the fractal dimension comparison result indicates that the liver blood vessel fractal dimension is less than the fractal dimension threshold, then comparing the microvascular entropy change rate with a preset entropy change rate threshold to obtain an entropy change rate comparison result;

[0017] If the entropy change rate comparison result indicates that the liver blood vessel fractal dimension is greater than the entropy change rate threshold, then the first surgery category indicating that the liver surgery is prohibited is generated.

[0018] Optionally, the surgery category further includes a second surgery category, and the predicting of liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgery category further includes:

[0019] If the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is greater than or equal to the fractal dimension threshold, generating the second surgery category indicating that the liver surgery is permitted; or if the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is less than the fractal dimension threshold and if the entropy change rate comparison result indicates that the fractal dimension of the liver blood vessels is less than or equal to the entropy change rate threshold, generating the second surgery category indicating that the liver surgery is permitted;

[0020] The method further comprises:

[0021] acquiring liver ultrasound raw data and serum data of the target subject according to the second surgical category;

[0022] Pressure prediction is performed based on the microvascular entropy change rate, liver ultrasound raw data, and serum data to obtain portal vein pressure prediction data;

[0023] Risk prediction is performed based on the portal vein pressure prediction data to obtain the liver surgery risk.

[0024] Optionally, performing pressure prediction based on the microvascular entropy change rate, liver ultrasound raw data, and serum data to obtain portal vein pressure prediction data includes:

[0025] performing pressure correction on the raw liver ultrasound data to obtain corrected liver pressure data;

[0026] Counting platelets on the serum data to obtain a platelet number fluctuation value;

[0027] Data fusion is performed based on the microvascular entropy change rate, the corrected liver pressure data, and the platelet number fluctuation value to obtain the portal vein pressure prediction data.

[0028] Optionally, the raw liver ultrasound data includes raw liver ultrasound video and raw liver pressure data;

[0029] The performing pressure correction on the raw liver ultrasound data to obtain corrected liver pressure data includes:

[0030] Tracking the diaphragm displacement of the original liver ultrasound video to obtain a diaphragm displacement curve;

[0031] Extracting the displacement value of the diaphragm displacement curve to obtain the current diaphragm displacement value;

[0032] The original liver pressure data is corrected according to the current diaphragm displacement value to obtain the corrected liver pressure data.

[0033] Optionally, extracting blood vessel features from each of the liver biopsy slice images to obtain liver blood vessel branch features includes:

[0034] Extracting a current liver biopsy slice image from the liver image sequence, and performing vascular skeleton extraction on the current liver biopsy slice image to obtain a current liver vascular skeleton feature; wherein the liver image sequence also includes historical liver biopsy slice images before the current liver biopsy slice image;

[0035] performing skeleton registration on the current liver vascular skeleton feature according to the historical liver biopsy slice image to obtain a registered vascular skeleton feature;

[0036] Branch point detection is performed on the registered blood vessel skeleton features to obtain the liver blood vessel branch features.

[0037] Optionally, performing fractal detection on the liver angiography image to obtain the liver blood vessel fractal dimension includes:

[0038] performing blood vessel segmentation on the liver angiography image to obtain a blood vessel region;

[0039] Performing centerline extraction on the blood vessel region to obtain a blood vessel centerline;

[0040] Performing grid coverage on the blood vessel centerline according to each of a plurality of preset grid sizes to obtain the number of coverage boxes corresponding to each of the grid sizes;

[0041] Performing linear fitting according to the grid size and the number of covering boxes to obtain a liver blood vessel fractal line;

[0042] The slope of the liver blood vessel fractal line is extracted to obtain the liver blood vessel fractal dimension.

[0043] To achieve the above-mentioned objectives, a second aspect of an embodiment of the present application provides a surgical decision-making device based on liver images, the device comprising:

[0044] a liver image acquisition module, configured to acquire a liver image sequence of a target object, wherein the liver image sequence includes at least two liver biopsy slice images;

[0045] a branch feature extraction module, configured to extract blood vessel branch features from each of the liver biopsy slice images to obtain liver blood vessel branch features;

[0046] a Shannon entropy calculation module, configured to perform Shannon entropy calculation on the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy;

[0047] a time series entropy change fitting module, configured to perform time series entropy change fitting according to the Shannon entropy of each liver vessel in the liver image sequence to obtain liver vessel entropy change fitting data;

[0048] a coefficient amplification module, configured to amplify the quadratic function coefficient in the liver vascular entropy change fitting data to obtain a microvascular index;

[0049] an entropy change rate calculation module, configured to calculate a change in any two of the microvascular indices to obtain a microvascular entropy change rate;

[0050] a fractal dimension detection module, configured to obtain a liver angiography image of the target object and perform fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels;

[0051] A liver surgery decision module is used to predict liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain a surgery category; wherein the surgery category represents whether the liver surgery is allowed or prohibited.

[0052] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the liver image-based surgical decision-making method described in the first aspect above.

[0053] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, it implements the liver image-based surgical decision-making method described in the first aspect above.

[0054] The surgical decision-making method and device based on liver images, electronic equipment, and storage medium proposed in this application, on the one hand, obtain the microvascular entropy change rate based on the liver image sequence encoding, specifically, first extract the liver vascular branch characteristics, and then perform Shannon entropy calculation, time series entropy change fitting, quadratic function coefficient amplification, and change amount calculation in sequence, to obtain the microvascular entropy change rate with high accuracy. On the other hand, a liver angiography image is obtained, and a fractal detection is performed on the liver angiography image to obtain the fractal dimension of the liver blood vessels. Both the microvascular entropy change rate and the fractal dimension can indicate the functional state of the target object's blood vessels and can be used to assess whether liver surgery (such as transjugular intrahepatic portosystemic shunt) can be supported. Finally, liver surgery is predicted based on the microvascular entropy change rate and the liver blood vessel fractal dimension to obtain the surgical category. In summary, by combining the entropy change rate, the dynamic changes of the liver can be reflected, and the fractal dimension is also combined to make surgical decisions, thereby improving the accuracy of surgical decisions.

[0055] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become obvious from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of a liver image-based surgical decision-making method provided in an embodiment of the present application;

[0057] Figure 2 yes Figure 1 Flowchart of step 102 in FIG.

[0058] Figure 3 yes Figure 1 Flowchart of step 107 in FIG.

[0059] Figure 4 yes Figure 1 Flowchart of step 108 in FIG.

[0060] Figure 5 is a flowchart of a liver image-based surgical decision-making method provided by another embodiment of the present application;

[0061] Figure 6 yes Figure 5 Flowchart of step 502 in FIG.

[0062] Figure 7 This is a block diagram of the module structure of a liver image-based surgical decision-making device provided in an embodiment of the present application;

[0063] Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0065] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0067] First, let’s analyze some of the terms used in this application:

[0068] Artificial intelligence (AI) is a new technical discipline that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. A branch of computer science, AI seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thinking. It also encompasses the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0069] Liver surgery refers to the surgical removal or repair of liver tissue or structures to treat liver disease. Common liver surgeries include partial hepatectomy (lobe or partial lobe resection), total hepatectomy (liver transplantation), resection of liver tumors (such as hepatocellular carcinoma), repair of liver damage, and reconstruction of the liver blood vessels or bile ducts. Transjugular intrahepatic portosystemic shunt is an important interventional treatment for liver disease.

[0070] Transjugular intrahepatic portosystemic stent-shunt (TIPS): The principle is to use a special interventional treatment device, under the guidance of X-ray fluoroscopy, through the jugular vein to establish an artificial shunt channel in the liver between the hepatic vein and the main branches of the portal vein, and maintain its permanent patency with a metal stent, so as to control and prevent rupture of esophageal and gastric varicose veins and promote ascites absorption after reducing portal hypertension.

[0071] CT angiography (CTA) combines CT enhancement technology with thin-slice, wide-area, and rapid scanning techniques. Through appropriate post-processing, it clearly displays vascular details throughout the body. Non-invasive and easy to use, it is invaluable for examining vascular variations, vascular diseases, and the relationship between lesions and blood vessels. In medicine, it is also known as non-invasive vascular imaging (abbreviated as CT angiography, or CTA). Angiography is an interventional testing procedure in which a contrast agent is injected into the blood vessels. Because X-rays cannot penetrate the contrast agent, angiography exploits this property to diagnose vascular lesions by revealing images of the contrast agent under X-rays.

[0072] Ultrasonic testing technology: It is a non-destructive testing method based on echo detection. It mainly controls the generation and propagation of ultrasonic waves, uses a probe to scan the object, records the echo signal, and processes and analyzes it to identify defects and structural features inside the object. Ultrasonic testing technology is the preferred inspection method for various liver diseases. Two-dimensional real-time ultrasound imaging is mainly used for changes in liver morphology, while two-color Doppler blood flow imaging is used for liver vascular lesions and hemodynamic examinations. Ultrasonic examinations show images of liver lesions, which are changes in acoustic physical properties. For the same lesion, the ultrasound image performance is different at different stages of the disease progression; while different lesions have similar acoustic physical properties, and the ultrasound image performance may be the same.

[0073] On the one hand, traditional liver disease diagnostic techniques primarily focus on static density measurements of liver microvessels, failing to effectively capture the dynamic changes in vascular branching structure as the disease progresses. In the diagnosis of early-stage cirrhosis, the lack of quantitative analysis of the evolution of subtle vascular morphology leads to a high rate of missed diagnosis. Furthermore, ultrasound detection technology is susceptible to interference from respiratory motion, making it difficult for existing filtering algorithms to completely eliminate artifacts. This leads to large errors in elasticity measurement, severely limiting diagnostic accuracy. Furthermore, existing diagnostic and treatment methods have failed to establish a quantitative correlation between the fractal dimensions of liver vessels and patients' surgical tolerance. This makes it impossible to accurately predict surgical risks in preoperative assessments for liver procedures such as transjugular intrahepatic portosystemic shunts, resulting in a high incidence of postoperative hepatic encephalopathy. Furthermore, the temporal fusion capabilities of multimodal data are insufficient, resulting in significant lag in portal vein pressure predictions, often significantly exceeding actual pressure changes, making it difficult to meet the needs of timely clinical intervention. Furthermore, in the current field of liver disease diagnosis and treatment, pathological analysis, imaging, and clinical decision-making processes are independent and lack systematic integration. Pathological analysis relies on static observation and lacks dynamic modeling; imaging is subject to artifacts, leading to data inaccuracy; and clinical decision-making relies heavily on human experience and judgment, which is susceptible to subjective influences. The fragmented technology at each stage prevents the entire diagnostic and treatment process from forming an effective closed loop, making it difficult to achieve accurate diagnosis and treatment.

[0074] Based on this, the embodiments of the present application propose a surgical decision-making method based on liver images, a surgical decision-making device based on liver images, an electronic device and a computer-readable storage medium. By combining the entropy change rate, the dynamic changes of the liver can be reflected, and the fractal dimension is also combined to make surgical decisions, thereby improving the accuracy of surgical decisions.

[0075] The liver image-based surgical decision-making method provided in the embodiments of the present application can be applied to terminals and servers, and can also be software running on the server. The server can be configured as an independent physical server, or as a server cluster or distributed system consisting of multiple physical servers, or as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the liver image-based surgical decision-making method, etc., but is not limited to the above forms.

[0076] The present application can be used in many general or special computer system environments or configurations. For example: server computers, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0077] The embodiments of the present application provide a surgical decision-making method based on liver images, a surgical decision-making device based on liver images, an electronic device, and a computer-readable storage medium, which are specifically illustrated by the following embodiments. First, the surgical decision-making method based on liver images in the embodiments of the present application is described.

[0078] It should be noted that in each specific embodiment of the present application, when it comes to performing relevant processing based on the user's image data (such as liver images) and other data related to the user's identity or characteristics, the user's permission or consent will be obtained first, and the collection, use and processing of such data will comply with relevant laws, regulations and standards.

[0079] Reference Figure 1 , Figure 1 This is an optional flowchart of the liver image-based surgical decision-making method provided in an embodiment of the present application, which may include but is not limited to steps 101 to 108.

[0080] Step 101: Acquire a liver image sequence of a target object, where the liver image sequence includes at least two liver biopsy slice images;

[0081] Step 102, extracting blood vessel branch features from each liver biopsy slice image to obtain liver blood vessel branch features;

[0082] Step 103, calculating the Shannon entropy of the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy;

[0083] Step 104, performing time series entropy change fitting based on the Shannon entropy of each liver blood vessel in the liver image sequence to obtain liver blood vessel entropy change fitting data;

[0084] Step 105, amplifying the quadratic function coefficient in the liver vascular entropy change fitting data to obtain a microvascular index;

[0085] Step 106, calculating the change amount based on any two microvascular indices to obtain the microvascular entropy change rate;

[0086] Step 107: Acquire a liver angiography image of the target object, and perform fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels;

[0087] Step 108 : predicting liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain a surgery category; wherein the surgery category represents whether liver surgery is allowed or prohibited.

[0088] Steps 101 to 108 shown in the embodiment of the present application, on the one hand, obtain the microvascular entropy change rate based on the liver image sequence encoding, specifically, first extract the liver vascular branch characteristics, and then perform Shannon entropy calculation, time series entropy change fitting, quadratic function coefficient amplification and change amount calculation in sequence, so as to obtain the microvascular entropy change rate with high accuracy. On the other hand, a liver angiography image is obtained, and a fractal detection is performed on the liver angiography image to obtain the fractal dimension of the liver blood vessels. Both the microvascular entropy change rate and the fractal dimension can indicate the functional state of the target object's blood vessels, and can be used to evaluate whether liver surgery (such as TIPS) can be supported. Finally, liver surgery is predicted based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgical category. In summary, by combining the entropy change rate, the dynamic changes of the liver can be reflected, and the fractal dimension is also combined to make surgical decisions, thereby improving the accuracy of surgical decisions.

[0089] In step 101 of some embodiments, a liver image sequence of a target object is acquired, wherein the liver image sequence includes at least two temporally consecutive liver biopsy slice images.

[0090] In one example, the process of obtaining a liver image sequence may include: a doctor assesses the patient's indications and risks and obtains the patient's consent; performs necessary preoperative examinations (such as blood coagulation function, complete blood cell count, etc.) to ensure safety; uses ultrasound guidance to insert a percutaneous puncture needle into a specific area of ​​the liver to collect liver tissue (a single puncture usually obtains a piece of tissue, but if continuous sectioning is required, multiple collections may be performed or a larger tissue may be obtained in a single puncture); after collection, the tissue is immediately placed in formalin fixative, and after fixation, the tissue block is embedded and sliced ​​(usually 3-5 microns thick); a microtome (microcrystallometer) or a rotary slicer is used to cut the tissue piece by piece to produce a series of continuous slices to obtain a liver image sequence.

[0091] In step 102 of some embodiments, vascular branch feature extraction is performed on each liver biopsy slice image to obtain liver vascular branch features. A pre-trained vascular branch encoder can be used to extract vascular branch features from the liver biopsy slice image.

[0092] In one embodiment, referring to Figure 2 , step 102 may include:

[0093] Step 201: extracting a current liver biopsy slice image from a liver image sequence, and performing vascular skeleton extraction on the current liver biopsy slice image to obtain current liver vascular skeleton features; wherein the liver image sequence also includes historical liver biopsy slice images prior to the current liver biopsy slice image;

[0094] Step 202 , performing skeleton registration on the current liver vascular skeleton features based on the historical liver biopsy slice image to obtain registered vascular skeleton features;

[0095] Step 203: perform branch point detection on the registered blood vessel skeleton features to obtain liver blood vessel branch features.

[0096] In step 201 , the blood vessel skeleton may be extracted using MedialAxis transformation, the main principle of which is to extract the blood vessel centerline through a topological skeletonization algorithm while preserving the branch topology structure.

[0097] In step 202 , a branch tracking feature matrix is ​​constructed, and the vascular skeleton features at consecutive time points t1 and t2 are registered to obtain registered vascular skeleton features.

[0098] In step 203, a branch point dynamic tracking algorithm is used to detect branch points from the registered vascular skeleton features to obtain the liver vascular branch features, which can be expressed as BranchPoint = [x, y, degree, branch_length]. The dynamic tracking formula is ‖BP t1 -BP t2 ‖<δ, BP refers to the pixel position, δ can be 5, which refers to the pixel displacement threshold).

[0099] In one example, a liver image sequence includes A1, A2, and A3, a total of three liver biopsy slice images. If the current liver biopsy slice image is A1, since A1 is the first image in the liver image sequence, there is no historical liver biopsy slice image. A1 can be directly subjected to vascular skeleton extraction and then branch point detection to obtain liver vascular branch features. If the current liver biopsy slice image is A2, the historical liver biopsy slice image is A1. First, the current liver vascular skeleton features of A2 are extracted. Then, the current liver vascular skeleton features of A2 are skeleton-aligned based on the liver vascular skeleton features of A1 to obtain the aligned vascular skeleton features of A2. Finally, branch point detection is performed on the aligned vascular skeleton features of A2 to obtain the liver vascular branch features of A2. If the current liver biopsy slice image is A3, and the historical liver biopsy slice image is A2, the current liver vascular skeleton features of A3 are first extracted, and then the current liver vascular skeleton features of A3 are skeleton-aligned according to the liver vascular skeleton features of A2 to obtain the registered vascular skeleton features of A3. Finally, branch point detection is performed on the registered vascular skeleton features of A3 to obtain the liver vascular branch features of A3.

[0100] The benefit of the embodiment of steps 201 to 203 is that, while being able to extract the blood vessel branch features of each slice image, the correlation of the branch features between the slice images is improved, and the dynamic changes of the blood vessels can be captured.

[0101] In step 103 of some embodiments, the Shannon entropy of the liver blood vessel branch characteristics is calculated to obtain the liver blood vessel Shannon entropy. Specifically, the Shannon entropy formula includes: (1) Branch complexity quantification formula: H(X) = -Σ[p(x i )log2p(x i )], where: p(x i ) = branch angle interval probability (divide 360° into 12 30° intervals); (2) calculate the time differential formula: ΔH / Δt.

[0102] In step 104 of some embodiments, a temporal entropy change fitting is performed based on the Shannon entropy of each liver vessel in the liver image sequence to obtain liver vessel entropy change fitting data. Specifically, a quadratic function expression is used for temporal entropy change fitting, i.e., entropy change acceleration quantification. The dynamic model of temporal entropy change fitting is: H(t) = a*t 2 +b*t+c. Where: t: time or slice sequence position (independent variable), H(t): Shannon entropy observed at time t (dependent variable), a: quadratic term coefficient (also called quadratic function coefficient), which is the mathematical embodiment of the acceleration of entropy change (d2H / dt2). The magnitude and sign of a directly reflect the acceleration direction and intensity of the evolution of vascular branching complexity, b: linear term coefficient (linear rate of change, speed), and c: constant term (initial entropy level).

[0103] In step 105 of some embodiments, the quadratic function coefficients in the liver vascular entropy change fitting data are amplified to obtain a microvascular index. The quadratic function coefficients can be amplified by a preset multiple to obtain the microvascular index. The preset multiple is set according to actual needs, for example, to 100. For example, the fitting process includes: 1. Taking the entropy values ​​of 5 consecutive time points: [2.37, 3.12, 3.85, 4.32, 4.60]; 2. Least squares method fitting of a quadratic curve: a = 0.18 ± 0.02, b = 0.75 ± 0.15, c = 2.15 ± 0.3; 5. Microvascular index: OTI = a × 100 = 18.

[0104] In step 106 of some embodiments, the change in any two microvascular indices is calculated to obtain the microvascular entropy change rate. For example, the microvascular entropy change rate: VGI = the monthly change in the microvascular index (OTI) (usually one month). Calculation example: Last month OTI = 15.2, this month OTI = 18.0, VGI = (18.0 - 15.2) / 1 month = 2.8 / month.

[0105] In step 107 of some embodiments, a liver angiography image of the target subject is obtained, and fractal detection is performed on the liver angiography image to obtain a fractal dimension of the liver blood vessels. Fractal detection can be performed on the liver angiography image using a pretrained fractal detection model. The liver angiography image is obtained using computed tomography (CT) angiography. CT angiography is a CT scanning technique that uses contrast agents to more clearly display internal organs and vascular structures.

[0106] In one embodiment, referring to Figure 3 , step 107 may include:

[0107] Step 301, performing blood vessel segmentation on the liver angiography image to obtain blood vessel regions;

[0108] Step 302: extract the centerline of the blood vessel region to obtain the blood vessel centerline;

[0109] Step 303, performing grid coverage on the blood vessel centerline according to each of a plurality of preset grid sizes, and obtaining the number of coverage boxes corresponding to each grid size;

[0110] Step 304 , performing linear fitting based on the grid size and the number of covering boxes to obtain the liver blood vessel fractal line;

[0111] Step 305 : extract the slope of the liver blood vessel fractal line to obtain the liver blood vessel fractal dimension.

[0112] In one example, the liver vascular fractal dimension calculation process includes: first CT angiography, then 3D reconstruction, then box counting method, and finally liver vascular fractal dimension calculation. The core three steps include: (1) vessel segmentation and centerline extraction; (2) multi-scale grid coverage and counting the number of coverage boxes N(ε); (3) logarithmic coordinate regression and slope calculation (liver vascular fractal dimension = slope value).

[0113] The benefit of the above embodiment is that it can improve the accuracy of calculating the fractal dimension of liver blood vessels.

[0114] In step 108 of some embodiments, a liver surgery is predicted based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain a surgery category. The surgery category indicates whether liver surgery is permitted or prohibited for the target subject. The surgery categories include a first surgery category and a second surgery category. The first surgery category indicates that liver surgery is prohibited. The second surgery category indicates that liver surgery is permitted.

[0115] It should be noted that the liver surgery may specifically be a TIPS surgery.

[0116] In one embodiment, referring to Figure 4 , step 108 may include:

[0117] Step 401: Compare the fractal dimension of the liver blood vessels with a preset fractal dimension threshold to obtain a fractal dimension comparison result;

[0118] Step 402: If the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is less than the fractal dimension threshold, the microvascular entropy change rate is compared with a preset entropy change rate threshold to obtain an entropy change rate comparison result;

[0119] Step 403 : If the entropy change rate comparison result indicates that the liver blood vessel fractal dimension is greater than the entropy change rate threshold, a first surgery category indicating that liver surgery is prohibited is generated.

[0120] In step 401, a fractal dimension threshold is set based on experience. The fractal dimension is generally between 0 and 2, for example, the fractal dimension threshold is set to 1.3. The fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is less than the fractal dimension threshold, or that the fractal dimension of the liver blood vessels is greater than or equal to the fractal dimension threshold.

[0121] In step 402, if the fractal dimension comparison result indicates that the liver vascular fractal dimension is less than the fractal dimension threshold, it indicates that the target subject's microvascular extensibility is poor, and liver surgery is likely unsuitable for the target subject. In this case, further judgment is required based on the microvascular entropy change rate. The entropy change rate comparison result indicates that the liver vascular fractal dimension is greater than the entropy change rate threshold, or that the liver vascular fractal dimension is less than or equal to the entropy change rate threshold.

[0122] In step 403, if the entropy change rate comparison result indicates that the fractal dimension of the liver blood vessels is greater than the entropy change rate threshold, it indicates that the microvascular disorder, complexity, or instability is high, and it is highly likely that liver surgery cannot be performed on the target patient. Therefore, a first surgery category is generated, indicating that liver surgery is prohibited.

[0123] The benefit of the embodiment of steps 401 to 403 is that the accuracy of surgical decision-making can be improved by first making a judgment based on the fractal dimension of the liver blood vessels and then making a second judgment based on the microvascular entropy change rate.

[0124] In one embodiment, step 108 may further include: if the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is greater than or equal to the fractal dimension threshold, generating a second surgical category indicating that liver surgery is permitted; or, if the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is less than the fractal dimension threshold and if the entropy change rate comparison result indicates that the fractal dimension of the liver blood vessels is less than or equal to the entropy change rate threshold, generating a second surgical category indicating that liver surgery is permitted. This embodiment has the advantage of improving the efficiency of surgical decision-making while improving the accuracy of surgical decision-making.

[0125] In one embodiment, referring to Figure 5 After step 108, the liver image-based surgical decision-making method may further include:

[0126] Step 501 , obtaining liver ultrasound raw data and serum data of a target subject according to a second surgical category;

[0127] Step 502 , performing pressure prediction based on the microvascular entropy change rate, liver ultrasound raw data, and serum data to obtain portal vein pressure prediction data;

[0128] Step 503: Perform risk prediction based on the portal vein pressure prediction data to obtain the liver surgery risk.

[0129] In step 501, if the surgical category is the second surgical category, liver surgery is permitted for the target subject, but the surgical risk assessment is still required. This embodiment uses raw liver ultrasound data and serum data, along with the microvascular entropy change rate, to predict the risk of liver surgery. Raw liver ultrasound data is obtained by performing ultrasound imaging on the target subject's liver. Serum data is obtained by testing the target subject's blood using serum testing technology.

[0130] In step 502, portal vein pressure prediction data can represent the pressure condition of the portal vein and is often used to estimate the degree of portal hypertension.

[0131] In step 503, if the portal vein pressure prediction data indicates a higher pressure, the risk of liver surgery is higher. If the portal vein pressure prediction data indicates a lower pressure, the risk of liver surgery is lower. The specific mapping rules can be set as needed and are not specifically limited in this embodiment.

[0132] In one embodiment, referring to Figure 6 , step 502 may include:

[0133] Step 601, performing pressure correction on the raw liver ultrasound data to obtain corrected liver pressure data;

[0134] Step 602: Count the platelets in the serum data to obtain a platelet number fluctuation value;

[0135] Step 603 : performing data fusion based on the microvascular entropy change rate, the corrected liver pressure data, and the platelet count fluctuation value to obtain portal vein pressure prediction data.

[0136] In step 601, the liver ultrasound raw data includes the liver ultrasound raw video and raw liver pressure data. Step 601 may include:

[0137] The diaphragm displacement is tracked on the original liver ultrasound video to obtain the diaphragm displacement curve;

[0138] Extract the displacement value of the diaphragm displacement curve to obtain the current diaphragm displacement value;

[0139] The original liver pressure data is corrected according to the current diaphragm displacement value to obtain corrected liver pressure data.

[0140] In one example, the input is an ultrasound video of 5 consecutive respiratory cycles (30 frames / second), and the output is a diaphragm displacement curve (unit: mm). The calculation formula is: Displacement(t) = [(x_t-x_ref) 2 +(y_t-y_ref) 2 ] 0.5 , where (x_ref, y_ref) are the coordinates of the end-expiratory reference point, and (x_t, y_t) are the coordinates of the real-time tracking point. Correction is performed using a calibration model: kPa_corrected = kPa_raw × [1 + α × (D - D_avg)], where kPa_corrected represents the corrected liver pressure data, kPa_raw represents the unprocessed raw liver pressure data, α = 0.15 (calibration factor, determined through phantom experiments), D represents the current diaphragm displacement, and D_avg represents the average displacement of the target subject (which can be taken as the average of 10 respiratory cycles).

[0141] The benefit of the above embodiment is that the pressure correction accuracy can be improved.

[0142] In step 602, for example, the platelet counting process may include: 1. Taking three consecutive platelet counts (PLT): [85, 78, 70] × 10 9 / L; 2. Calculate the natural logarithmic transformation: LN(PLT) = [ln(85), ln(78), ln(70)] ≈ [4.44, 4.36, 4.25]; 3. Calculate the platelet count fluctuation value: ΔLN(PLT) = |latest value - earliest value| = |4.25-4.44| = 0.19.

[0143] In step 603 , for example, a multimodal fusion equation dominated by the microvascular entropy change rate (weight 67%) is established: portal vein pressure prediction data PP=0.67×VGI+0.23×ΔLN(PLT)+0.1×kPa_corrected.

[0144] The benefit of the above steps 601 to 603 is that, by fusing multi-source data such as ultrasound imaging and serology, non-invasive prediction of portal vein pressure is achieved, and the prediction accuracy is improved.

[0145] In summary, the present application can at least achieve the following beneficial effects: 1. Reduce examination costs: Through accurate microvascular entropy change analysis, the frequency of liver biopsy can be reduced by 50%; 2. Reduce invasive operations: Non-invasive portal vein pressure prediction technology replaces traditional hepatic vein catheterization examination, reducing invasive operations; 3. Improve the accuracy of surgical decision-making and surgical risk prediction.

[0146] See also Figure 7 The embodiment of the present application also provides a surgical decision-making device based on liver images, which can implement the above-mentioned surgical decision-making method based on liver images. Figure 7 This is a block diagram of the module structure of a liver image-based surgical decision-making device provided in an embodiment of the present application. The device includes:

[0147] A liver image acquisition module 701 is configured to acquire a liver image sequence of a target object, wherein the liver image sequence includes at least two liver biopsy slice images;

[0148] A branch feature extraction module 702 is configured to extract blood vessel branch features from each liver biopsy slice image to obtain liver blood vessel branch features;

[0149] A Shannon entropy calculation module 703 is used to calculate the Shannon entropy of the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy;

[0150] A time series entropy change fitting module 704 is configured to perform time series entropy change fitting based on the Shannon entropy of each liver vessel in the liver image sequence to obtain liver vessel entropy change fitting data;

[0151] The coefficient amplification module 705 is used to amplify the quadratic function coefficient in the liver vascular entropy change fitting data to obtain the microvascular index;

[0152] The entropy change rate calculation module 706 is used to calculate the change amount of any two microvascular indices to obtain the microvascular entropy change rate;

[0153] A fractal dimension detection module 707 is used to obtain a liver angiography image of a target object and perform fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels;

[0154] The liver surgery decision module 708 is used to predict liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain a surgery category; wherein the surgery category represents whether liver surgery is allowed or prohibited.

[0155] In one embodiment, the liver image-based surgical decision-making apparatus further includes a surgical risk prediction module for:

[0156] acquiring liver ultrasound raw data and serum data of the target subject according to the second surgical category;

[0157] Pressure prediction is performed based on microvascular entropy change rate, liver ultrasound raw data and serum data to obtain portal vein pressure prediction data;

[0158] Risk prediction is performed based on portal vein pressure prediction data to obtain the risk of liver surgery.

[0159] It should be noted that the specific implementation of the liver image-based surgical decision-making device is basically the same as the specific embodiment of the liver image-based surgical decision-making method described above, and will not be repeated here.

[0160] The present application also provides an electronic device comprising: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for enabling communication between the processor and the memory. When the program is executed by the processor, the aforementioned liver image-based surgical decision-making method is implemented. The electronic device can be any intelligent terminal, including a tablet computer and an in-vehicle computer.

[0161] See also Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0162] The processor 801 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0163] The memory 802 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called by the processor 801 to execute the liver image-based surgical decision-making method of the embodiments of this application.

[0164] Input / output interface 803, used to implement information input and output;

[0165] Communication interface 804, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0166] Bus 805 , which transmits information between various components of the device (e.g., processor 801 , memory 802 , input / output interface 803 , and communication interface 804 );

[0167] The processor 801 , the memory 802 , the input / output interface 803 and the communication interface 804 are connected to each other in communication within the device via a bus 805 .

[0168] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium used for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned surgical decision-making method based on liver images.

[0169] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0170] The surgical decision-making method based on liver images, the surgical decision-making device based on liver images, the electronic device and the storage medium provided by the present application, on the one hand, obtain the microvascular entropy change rate based on the liver image sequence coding, specifically, first extract the liver vascular branch characteristics, and then perform Shannon entropy calculation, time series entropy change fitting, quadratic function coefficient amplification and change amount calculation in sequence, so as to obtain the microvascular entropy change rate with high accuracy. On the other hand, a liver angiography image is obtained, and a fractal detection is performed on the liver angiography image to obtain the fractal dimension of the liver blood vessels. Both the microvascular entropy change rate and the fractal dimension can indicate the functional state of the target object's blood vessels and can be used to evaluate whether liver surgery (such as TIPS) can be supported. Finally, liver surgery is predicted based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgical category. In summary, by combining the entropy change rate, the dynamic changes of the liver can be reflected, and the fractal dimension is also combined to make surgical decisions, thereby improving the accuracy of surgical decisions.

[0171] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0172] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0173] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0174] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0175] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0176] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0177] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0178] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0179] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0180] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes multiple instructions for enabling an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0181] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A surgical decision-making method based on liver images, characterized in that: The method comprises: Acquiring a liver image sequence of a target object, wherein the liver image sequence includes at least two liver biopsy slice images; performing vascular branch feature extraction on each of the liver biopsy slice images to obtain liver vascular branch features; Performing Shannon entropy calculation on the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy; performing time series entropy change fitting according to the Shannon entropy of each liver blood vessel in the liver image sequence to obtain liver blood vessel entropy change fitting data; amplifying the quadratic function coefficient in the liver vascular entropy change fitting data to obtain a microvascular index; Calculating the change of any two microvascular indices to obtain the microvascular entropy change rate; Acquiring a liver angiography image of the target object, and performing fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels; Liver surgery is predicted according to the microvascular entropy change rate and the liver blood vessel fractal dimension to obtain a surgery category; wherein the surgery category represents whether the liver surgery is allowed or prohibited.

2. The method according to claim 1, characterized in that The surgery category includes a first surgery category, and the liver surgery is predicted based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgery category, including: Comparing the liver blood vessel fractal dimension with a preset fractal dimension threshold to obtain a fractal dimension comparison result; If the fractal dimension comparison result indicates that the liver blood vessel fractal dimension is less than the fractal dimension threshold, then comparing the microvascular entropy change rate with a preset entropy change rate threshold to obtain an entropy change rate comparison result; If the entropy change rate comparison result indicates that the liver blood vessel fractal dimension is greater than the entropy change rate threshold, then the first surgery category indicating that the liver surgery is prohibited is generated.

3. The method according to claim 2, characterized in that The surgical category further includes a second surgical category, and the liver surgery is predicted based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain the surgical category, further comprising: If the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is greater than or equal to the fractal dimension threshold, generating the second surgery category indicating that the liver surgery is permitted; or if the fractal dimension comparison result indicates that the fractal dimension of the liver blood vessels is less than the fractal dimension threshold and if the entropy change rate comparison result indicates that the fractal dimension of the liver blood vessels is less than or equal to the entropy change rate threshold, generating the second surgery category indicating that the liver surgery is permitted; The method further comprises: acquiring liver ultrasound raw data and serum data of the target subject according to the second surgical category; Pressure prediction is performed based on the microvascular entropy change rate, liver ultrasound raw data, and serum data to obtain portal vein pressure prediction data; Risk prediction is performed based on the portal vein pressure prediction data to obtain the liver surgery risk.

4. The method according to claim 3, characterized in that The pressure prediction is performed based on the microvascular entropy change rate, liver ultrasound raw data and serum data to obtain portal vein pressure prediction data, including: performing pressure correction on the raw liver ultrasound data to obtain corrected liver pressure data; Counting platelets on the serum data to obtain a platelet number fluctuation value; Data fusion is performed based on the microvascular entropy change rate, the corrected liver pressure data, and the platelet number fluctuation value to obtain the portal vein pressure prediction data.

5. The method according to claim 3 or 4, characterized in that The liver ultrasound raw data includes liver ultrasound raw video and raw liver pressure data; The performing pressure correction on the raw liver ultrasound data to obtain corrected liver pressure data includes: Tracking the diaphragm displacement of the original liver ultrasound video to obtain a diaphragm displacement curve; Extracting the displacement value of the diaphragm displacement curve to obtain the current diaphragm displacement value; The original liver pressure data is corrected according to the current diaphragm displacement value to obtain the corrected liver pressure data.

6. The method according to any one of claims 1 to 4, characterized in that The extracting blood vessel features from each of the liver biopsy slice images to obtain liver blood vessel branch features includes: Extracting a current liver biopsy slice image from the liver image sequence, and performing vascular skeleton extraction on the current liver biopsy slice image to obtain a current liver vascular skeleton feature; wherein the liver image sequence also includes historical liver biopsy slice images before the current liver biopsy slice image; performing skeleton registration on the current liver vascular skeleton feature according to the historical liver biopsy slice image to obtain a registered vascular skeleton feature; Branch point detection is performed on the registered blood vessel skeleton features to obtain the liver blood vessel branch features.

7. The method according to any one of claims 1 to 4, characterized in that The fractal detection is performed on the liver angiography image to obtain the fractal dimension of the liver blood vessels, including: performing blood vessel segmentation on the liver angiography image to obtain a blood vessel region; Performing centerline extraction on the blood vessel region to obtain a blood vessel centerline; Performing grid coverage on the blood vessel centerline according to each of a plurality of preset grid sizes to obtain the number of coverage boxes corresponding to each of the grid sizes; Performing linear fitting according to the grid size and the number of covering boxes to obtain a liver blood vessel fractal line; The slope of the liver blood vessel fractal line is extracted to obtain the liver blood vessel fractal dimension.

8. A surgical decision-making device based on liver images, characterized in that: The device comprises: a liver image acquisition module, configured to acquire a liver image sequence of a target object, wherein the liver image sequence includes at least two liver biopsy slice images; a branch feature extraction module, configured to extract blood vessel branch features from each of the liver biopsy slice images to obtain liver blood vessel branch features; a Shannon entropy calculation module, configured to perform Shannon entropy calculation on the liver blood vessel branch characteristics to obtain the liver blood vessel Shannon entropy; a time series entropy change fitting module, configured to perform time series entropy change fitting according to the Shannon entropy of each liver vessel in the liver image sequence to obtain liver vessel entropy change fitting data; a coefficient amplification module, configured to amplify the quadratic function coefficient in the liver vascular entropy change fitting data to obtain a microvascular index; an entropy change rate calculation module, configured to calculate a change in any two of the microvascular indices to obtain a microvascular entropy change rate; a fractal dimension detection module, configured to obtain a liver angiography image of the target object and perform fractal detection on the liver angiography image to obtain a fractal dimension of the liver blood vessels; A liver surgery decision module is used to predict liver surgery based on the microvascular entropy change rate and the liver vascular fractal dimension to obtain a surgery category; wherein the surgery category represents whether the liver surgery is allowed or prohibited.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the liver image-based surgical decision-making method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the liver image-based surgical decision-making method according to any one of claims 1 to 7 is implemented.

Images