Training Method, System, Electronic Device and Medium for Evaluating Pancreatic Exocrine Function Model

Abstract

The present application provides a training method, system, electronic device and medium for a pancreatic exocrine function evaluation model. The method includes obtaining magnetic resonance images of the pancreas, extracting radiomics features and image features of the magnetic resonance images; obtaining pancreatic exocrine function parameters, which are obtained through a non-invasive pancreatic exocrine function experiment; obtaining clinical features of the pancreas; using the radiomics features, the image features and the clinical features as samples to train the pancreatic exocrine function evaluation model, and adjusting the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model. This method can evaluate pancreatic exocrine function non-invasively and conveniently.

Description

Technical Field

[0001] The present invention relates to the field of medicine, and particularly to a training method, system, electronic device and medium for a pancreatic exocrine function evaluation model. Background Art

[0002] Insufficiency of pancreatic exocrine function is the main functional change in chronic pancreatitis. When there is insufficiency of pancreatic exocrine function, it will also lead to various complications. Early and accurate diagnosis of pancreatic exocrine function can ensure timely treatment and effectively prevent the occurrence of complications.

[0003] Currently, methods for diagnosing pancreatic exocrine function include direct and indirect pancreatic exocrine function tests. Among them, the direct pancreatic exocrine function test requires injecting secretin intravenously into the patient, and then intubating into the duodenum to collect pancreatic juice, and evaluating the exocrine function of the pancreas according to the secretion volume of pancreatic juice per standard unit time. The disadvantages are invasiveness, time-consuming, laborious and expensive. The accuracy of the indirect pancreatic exocrine function test is easily affected by drugs such as pancreatic enzymes, and the sensitivity is very low or even ineffective in patients with mild or early pancreatitis. Summary of the Invention

[0004] The present application provides a training method, system, electronic device and medium for a pancreatic exocrine function evaluation model, which can non-invasively and conveniently evaluate the pancreatic exocrine function.

[0005] In a first aspect of the present application, a training method for a pancreatic exocrine function evaluation model is disclosed. The method includes obtaining magnetic resonance images of the pancreas, extracting radiomics features and image features of the magnetic resonance images; obtaining pancreatic exocrine function parameters, where the pancreatic exocrine function parameters are obtained through a non-invasive pancreatic exocrine function test; obtaining clinical features of the pancreas; using the radiomics features, the image features and the clinical features as samples to train the pancreatic exocrine function evaluation model, and training the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model.

[0006] In a possible implementation of the above first aspect, the clinical features include one or more of age, gender, body mass index, drinking history, smoking history, hyperlipidemia, history of acute pancreatitis attack, diabetes, abdominal pain.

[0007] In a possible implementation of the above first aspect, the image features include one or more of the diameter of the main pancreatic duct in the pancreatic head, visualization of branch pancreatic ducts, signal ratio of pancreatic parenchyma to left psoas major muscle parenchyma, and signal ratio of pancreatic parenchyma in the arterial phase and portal phase.

[0008] In a possible implementation of the above first aspect, the method further includes feature screening of the radiomics features and dimensionality reduction processing of the radiomics features using the Least Absolute Shrinkage and Selection Operator (LASSO) regression algorithm.

[0009] In a possible implementation of the above first aspect, the method further includes that the variables in the trained pancreatic exocrine function evaluation model include visualization of the branch pancreatic duct, diabetes, abdominal pain, and the group score calculated according to the radiomics features.

[0010] In a possible implementation of the above first aspect, the magnetic resonance image is a T1-weighted image (T1WI).

[0011] A second aspect of the present application discloses an electronic device, which includes a memory storing computer-executable instructions and a processor; when the instructions are executed by the processor, the device implements the training method of the pancreatic exocrine function evaluation model according to the first aspect of the present application.

[0012] A third aspect of the present application discloses a computer-readable medium, 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 training method of the pancreatic exocrine function evaluation model according to the first aspect of the present application.

[0013] A fourth aspect of the present application discloses a training system for a pancreatic exocrine function evaluation model, the system includes a first acquisition module for acquiring magnetic resonance images of the pancreas; a first extraction module for extracting radiomics features of the magnetic resonance images; a second extraction module for extracting image features of the magnetic resonance images; a second acquisition module for acquiring pancreatic exocrine function parameters obtained through a non-invasive pancreatic exocrine function experiment; a third acquisition module for acquiring clinical features of the pancreas; and a training module for using the radiomics features, the image features, and the clinical features as samples to train the pancreatic exocrine function evaluation model and training the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model.

[0014] A fifth aspect of the present application discloses a computer program product, and when the computer program is executed by a processor, it implements the training method of the pancreatic exocrine function evaluation model according to the first aspect of the present application.

[0015] The training method, system, electronic device and medium of the pancreatic exocrine function evaluation model provided by this application train the model by combining magnetic resonance images of the pancreas and the corresponding clinical features, and can provide a non-invasive, convenient, fast and economical evaluation of pancreatic exocrine function. Compared with the direct pancreatic exocrine function experiment, there is no need to inject secretin into the patient, nor to intubate the patient to test the changes in pancreatic juice, that is, a model is established by non-invasive means to evaluate the exocrine function of the pancreas, and it will not cause any pain to the patient. Compared with Secretin-enhanced Magnetic Resonance Cholangiopancreatography (S-MRCP), this examination does not need to rely on expensive human-injected secretin, nor does it require an ultra-long scanning time of about 1 hour per person. In addition, compared with the indirect pancreatic exocrine function experiment, it will not be affected by drugs such as pancreatic enzymes and has higher evaluation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flowchart of the training method of the pancreatic exocrine function evaluation model according to an embodiment of this application;

[0017] Figure 2 It is a schematic diagram of the pancreatic exocrine function evaluation model 201 according to an embodiment of this application;

[0018] Figure 3 It is a schematic diagram of the trained pancreatic exocrine function evaluation model 301 according to an embodiment of this application;

[0019] Figure 4 For an embodiment of this application Figure 3 The nomogram example of visualizing the pancreatic exocrine function evaluation model in;

[0020] Figures 5-6 It shows the comparison results of the prediction using the training method of the pancreatic exocrine function evaluation model of this application and the detection of Secretin-stimulated Magnetic Resonance Cholangiopancreatography (S-MRCP);

[0021] Figure 7 It is a schematic structural diagram of the training system 700 of the pancreatic exocrine function evaluation model according to an embodiment of this application;

[0022] Figure 8 It is a block diagram of an electronic device 800 according to an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The present application will be further described below in conjunction with specific embodiments and the accompanying drawings. It can be understood that the illustrative embodiments of the present disclosure include, but are not limited to, methods, systems, electronic devices, and media for training a pancreatic exocrine function evaluation model. The specific embodiments described herein are merely for explaining the present application and do not limit the present application. In addition, for the sake of convenience of description, only some parts related to the present application rather than all structures or processes are shown in the drawings.

[0024] To make the objectives, technical solutions, and advantages of the present application clearer, the embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings.

[0025] Currently, the incidence of chronic pancreatitis has been increasing year by year in many countries. Chronic pancreatitis is characterized by progressive and irreversible changes in pancreatic parenchyma and function. Pancreatic exocrine insufficiency (PEI) is the most important functional change in chronic pancreatitis. Once pancreatic exocrine insufficiency occurs, it will lead to many complications, such as chronic diarrhea, bone disease, malnutrition, and malabsorption. If pancreatic exocrine insufficiency can be detected early, symptomatic treatment can be applied in time and the occurrence of complications can be effectively prevented.

[0026] In the related art, the methods for diagnosing pancreatic exocrine function are divided into direct and indirect pancreatic exocrine function tests (PFT). The direct pancreatic exocrine function test requires intravenous injection of a standard amount of secretin to induce pancreatic exocrine activity in patients, and then intubation into the duodenum to collect pancreatic juice, and the exocrine function of the pancreas is evaluated according to the secretion volume of pancreatic juice per standard unit time. Although this method is the gold standard, it has disadvantages such as invasiveness, time-consuming, laborious, expensive, and difficult to standardize. At the same time, it will also bring great pain to patients and is difficult to carry out smoothly and routinely in clinical practice.

[0027] Indirect pancreatic exocrine function tests include N-benzoyl-L-tyrosine-p-aminobenzoic acid test, pancreatic lauric acid test, fecal test, etc. Their common drawback is that they are easily affected by drugs such as pancreatic enzymes. Therefore, for patients with mild or early pancreatitis, the sensitivity is very low or even ineffective.

[0028] In addition, secretin-stimulated magnetic resonance cholangiopancreatography (S-MRCP) stimulates pancreatic juice secretion by intravenous injection of secretin, and the exocrine function of the pancreas is evaluated by quantifying the fluid changes of MRCP through multi-phase MRCP scanning. The defect of S-MRCP is that human injectable secretin is difficult to obtain, the scanning time is up to 11 minutes, and the price is expensive.

[0029] To solve the above problems, an embodiment of the present application provides a training method 100 for a pancreatic exocrine function evaluation model. Refer to Figure 1 . The method 100 includes:

[0030] 110. Obtain magnetic resonance images of the pancreas, and extract radiomics features and image features of the magnetic resonance images.

[0031] In the diagnosis of pancreatic diseases, medical imaging is one of the important technical means. The medical imaging techniques for assisting in the diagnosis of pancreatic diseases include: B-ultrasound, CT (Computed Tomography), MRI (Magnetic Resonance Imaging), etc.

[0032] Magnetic resonance plain scan technology can be used to collect magnetic resonance images of the pancreas. Radiomics features and image features are extracted according to the magnetic resonance images of the pancreas.

[0033] In the extraction of radiomics features, radiomics features can be extracted through image intensity normalization, pancreas segmentation, feature extraction, and feature dimensionality reduction. The pancreas region can be delimited in multiple slices through the drawing tool in the editing module of the software 3D Slicer. The open-source Python package Pyradiomics can be used to extract radiomics features. Generally, the omics features of the mass in the portal venous phase can be divided into two categories, the original feature class and the filter class. Among them, the filter class can include: Logarithm, Exponential, gradient, Square, Square Root, lbp-2D, and wavelet (wavelet filtering). The features extracted from the mass in the portal venous phase can be classified into seven categories: (1) first-order statistics; (2) shape features; (3) gray-level co-occurrence matrix features; (4) gray-level dependence matrix features; (5) gray-level run length matrix features; (6) gray-level size zone matrix features; and (7) neighborhood gray-level zone difference matrix features.

[0034] In feature dimensionality reduction, analysis of variance, spearman correlation analysis, and the Least absolute shrinkage and selection operator (LASSO) logistic regression algorithm can be used to screen the extracted features, that is, reduce the dimensionality, to obtain target features.

[0035] In some embodiments, the image features of the magnetic resonance image include one or more of the diameter of the main pancreatic duct (MPD) in the head of the pancreas, the visualization of the branch pancreatic duct (BPD), the signal intensity ratio (SIR) of the pancreatic parenchyma to the left psoas major parenchyma, and the parenchymal signal ratio (A / P) in the arterial and portal phases of the pancreas.

[0036] 120. Obtain pancreatic exocrine function parameters, which are obtained through a non-invasive pancreatic exocrine function test.

[0037] Obtain the corresponding exocrine function parameters of the pancreas, which can indicate whether the pancreatic exocrine function is normal or abnormal. Moreover, the pancreatic exocrine function parameters are obtained through a non-invasive pancreatic exocrine function test.

[0038] The non-invasive pancreatic exocrine function test can be an indirect pancreatic exocrine function test. The indirect pancreatic exocrine function test is to apply a test meal to stimulate the gastrointestinal tract to secrete gastrointestinal hormones and then measure the pancreatic exocrine function, or based on the decrease in pancreatic function, the unabsorbed food (protein, fat) in the feces increases, the enzyme content in the blood and feces decreases, and some synthetic substances (NBT-PABA, fluorescein laurate, isotope-labeled substrates) are decomposed by pancreatic enzymes in the intestinal lumen. The degree of decrease in the concentration of these hydrolyzed substances in the blood, urine, feces, and exhaled breath is measured to evaluate the pancreatic exocrine function.

[0039] 130. Obtain the clinical features of the pancreas.

[0040] Obtain the features manifested on the patient's body corresponding to the pancreas. In some embodiments, the clinical features include one or more of age, gender, body mass index, history of alcohol consumption, history of smoking, hyperlipidemia, history of acute pancreatitis attacks, diabetes, and abdominal pain.

[0041] 140. Use the radiomics features, image features, and clinical features as samples to train the pancreatic exocrine function evaluation model, and adjust the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain the trained pancreatic exocrine function evaluation model.

[0042] Use the radiomics features, image features, and clinical features obtained in 110 and 130 as samples, and train the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters obtained in 120.

[0043] The training method of the pancreatic exocrine function evaluation model in Method 100 trains the model by combining magnetic resonance images of the pancreas and the corresponding clinical features. Method 100 can provide a non-invasive, convenient, fast, and economical evaluation of pancreatic exocrine function. Compared with direct pancreatic exocrine function experiments, it is not necessary to inject secretin into the patient or insert a tube into the patient to test the changes in pancreatic juice. That is, a model is established by non-invasive means to evaluate the exocrine function of the pancreas, which will not cause great pain to the patient. Moreover, it does not need to rely on expensive artificially injected secretin. In addition, compared with indirect pancreatic exocrine function experiments, Method 100 is not affected by drugs such as pancreatic enzymes and has high evaluation accuracy.

[0044] Figure 2 FIG. 4 shows a schematic diagram of the pancreatic exocrine function evaluation model 201 according to an embodiment of the present application. The pancreatic exocrine function evaluation model 201 takes radiomics features 202, image features 203, and clinical features 204 as training samples for input, and trains the pancreatic exocrine function evaluation model 201 according to the pancreatic exocrine function parameter 205.

[0045] In some embodiments, the pancreatic exocrine function parameter 205 can be measured by fecal elastase (FE)-1. The monitoring results of FE-1 can be used to divide the normal exocrine function group and the abnormal group with a cut-off value of 200 μg / L.

[0046] In some embodiments, the magnetic resonance image is a T1-weighted image (T1WI). The radiomics features are extracted by layer-by-layer delineation of the T1WI plain scan pancreatic images of the pancreas. Based on the original feature class and the filtered feature class, a total of 1409 two-dimensional and three-dimensional features are extracted from the T1WI plain scan pancreatic images of the pancreas.

[0047] In the feature screening, first, the features with no significant difference and no significant correlation between the normal group and the abnormal group are removed. A total of 1257 features are removed, and 152 features are obtained. Then, the features are further reduced to 25 omics features through the LASSO logistic regression algorithm.

[0048] Table 1 shows the 25 omics features obtained by dimensionality reduction through the LASSO logistic regression algorithm and the calculation formula of the omics score.

[0049] Table 1 The most relevant 25 features extracted and the LASSO regression equation

[0050]

[0051]

[0052]

[0053] In the expression form of each feature in Table 1, in A_B_C_D, A represents the category to which the feature belongs. Among them, in wavelet filtering, wavelet filtering can be composed of the combination of high-pass (H) and low-pass (L) filtering in each dimension, including Wavelet-LLH, Wavelet-LHL, Wavelet-LHH, Wavelet-HLL, Wavelet-HLH, Wavelet-HHL, Wavelet-LLL, and Wavelet-HHH.

[0054] Using Logistic regression, the values of these 25 omics features and their respective weight β values can be obtained, and finally the omics score is calculated. According to univariate Logistic regression analysis, the omics score is a parameter that significantly affects pancreatic exocrine function.

[0055] In the multivariate Logistic regression analysis, clinical features and image features of magnetic resonance images are considered. Variables such as diabetes and abdominal pain in clinical features and visualization of the branched pancreatic duct (BPD) in the image features of magnetic resonance images also contribute to the pancreatic exocrine function evaluation model.

[0056] Figure 3 A schematic diagram of the trained pancreatic exocrine function evaluation model 301 of an embodiment of the present application is shown. For the sample to be evaluated, the omics score 302 calculated based on the target features obtained by screening the radiomics features of the magnetic resonance image, visualization of the branched pancreatic duct (BPD) 303 in the image features of the magnetic resonance image, diabetes 304, and abdominal pain 305 are used as variables and input into the trained pancreatic exocrine function evaluation model 301, and the probability of whether the pancreatic exocrine function is normal is obtained at the output 306.

[0057] In some embodiments, the trained pancreatic exocrine function evaluation model can be visualized as a nomogram. A nomogram is a calculation chart that visualizes the functional relationship between several variables of a mathematical equation with a calibrated straight line or curve according to certain geometric conditions.

[0058] Figure 4 Shows Figure 3 The nomogram for visualizing the pancreatic exocrine function evaluation model in Figure 4 In

[0059] In some embodiments, a training set and a validation set can be used to verify the accuracy of the constructed model. For example, magnetic resonance images of the pancreas of 159 participants can be collected and divided into two groups, namely a training set (119) and a validation set (40). Among them, the training set is used to train the model, and the validation set is used to verify the model.

[0060] The ROC curves (receiver operating characteristic curves) of the pancreatic exocrine function evaluation model trained by the present application on the training set and the validation set show that the AUC of the trained pancreatic exocrine function evaluation model of the present application on the training set and the validation set are 0.91 (95% CI: 0.94, 0.99) and 0.93 (95% CI: 0.84, 0.99) respectively. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value on the training set are 93.55%, 85.96%, 89.92%, 87.88%, and 92.45% respectively; on the validation set, they are 83.88%, 92.86%, 90.00%, 83.33%, and 92.86% respectively. In addition, the goodness of fit of the model is good, P = 0.77 on the training set, and P = 0.26 on the validation set. All of the above indicate that the prediction effect of the trained pancreatic exocrine function evaluation model of the present application is better.

[0061] In the present application, the trained pancreatic exocrine function evaluation model can effectively predict the exocrine function of the pancreas. For a more detailed description, Figures 5-6 The comparison results between the prediction using the training method of the pancreatic exocrine function evaluation model of the present application and the secretin-stimulated magnetic resonance cholangiopancreatography (S-MRCP) detection are shown.

[0062] Figure 5 a shows the cross-sectional fat-suppressed T1-weighted image of Participant A. Figure 5 b shows that the probability of predicting pancreatic exocrine insufficiency according to the pancreatic exocrine function evaluation model is 0.0702 (indicated by the arrow). Figure 5 c shows that the secretin-stimulated magnetic resonance cholangiopancreatography (S-MRCP) of Patient A shows that the pancreatic branch ducts are not visualized. Figure 5 d shows the S-MRCP of Participant A 11 minutes after secretin stimulation; it can be seen that the fluid filling in the duodenum and jejunum increases. Figure 5 e shows that the pancreatic flow rate (PFR) is 9.6 ml / min. The prediction of the trained pancreatic exocrine function evaluation model in the present application is consistent with the detection results of indirect pancreatic exocrine function, and is also consistent with the diagnostic results of S-MRCP, that is, the pancreatic exocrine function of Participant A is normal.

[0063] Figure 6a shows a cross-sectional fat-suppressed T1-weighted image of Participant B. Figure 6 b shows that the probability of predicting pancreatic exocrine insufficiency according to the pancreatic exocrine function evaluation model is 0.966 (as indicated by the arrow). Figure 6 c shows that the secretin-stimulated magnetic resonance cholangiopancreatography (S-MRCP) of Participant B shows visualization of the pancreatic branch ducts. Figure 6 d shows the S-MRCP of Participant B at 11 minutes after secretin stimulation; an increase in the fluid filling of the duodenum can be seen. Figure 6 e shows that the pancreatic flow rate (PFR) is 4.9 ml / min. The prediction of the trained pancreatic exocrine function evaluation model in this application is consistent with the test results of indirect pancreatic exocrine function, and is also consistent with the diagnostic results of S-MRCP, that is, the pancreatic exocrine function of Participant B is abnormal.

[0064] Now refer to Figure 7 , Figure 7 which shows a schematic structural diagram of a training system 700 for a pancreatic exocrine function evaluation model according to an embodiment of the present application. The training system 700 for a pancreatic exocrine function evaluation model includes a first acquisition module 701 for acquiring magnetic resonance images of the pancreas; a first extraction module 702 for extracting radiomics features of the magnetic resonance images; a second extraction module 703 for extracting image features of the magnetic resonance images; a second acquisition module 704 for acquiring pancreatic exocrine function parameters, where the pancreatic exocrine function parameters are obtained through a non-invasive pancreatic exocrine function experiment; a third acquisition module 705 for acquiring clinical features of the pancreas; and a training module 706 for using the radiomics features, image features, and clinical features as samples to train the pancreatic exocrine function evaluation model, and training the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model.

[0065] The training system 700 for a pancreatic exocrine function evaluation model combines magnetic resonance images of the pancreas and the corresponding clinical features to train the model. The training system 700 for a pancreatic exocrine function evaluation model can provide a non-invasive, convenient, fast, and economical evaluation of pancreatic exocrine function. Compared with a direct pancreatic exocrine function experiment, there is no need to inject secretin into the patient, nor to intubate the patient to test the changes in pancreatic juice, that is, a model is established by non-invasive means to evaluate the exocrine function of the pancreas, which will not cause great pain to the patient. And, there is no need to rely on expensive artificially injected secretin. In addition, compared with an indirect pancreatic exocrine function experiment, Method 100 is not affected by drugs such as pancreatic enzymes and has a high evaluation accuracy.

[0066] Now refer to Figure 8, shown is a block diagram of an electronic device 800 according to an embodiment of the present application. The device 800 may include one or more processors 802, system control logic 808 connected to at least one of the processors 802, system memory 804 connected to the system control logic 808, non-volatile memory (NVM) 806 connected to the system control logic 808, and a network interface 810 connected to the system control logic 808.

[0067] The processor 802 may include one or more single-core or multi-core processors. The processor 802 may include any combination of general-purpose processors and dedicated processors (e.g., graphics processors, application processors, baseband processors, etc.). In the embodiments herein, the processor 802 may be configured to execute one or more embodiments according to various embodiments as Figures 1-7 shown.

[0068] In some embodiments, the system control logic 808 may include any suitable interface controller to provide any suitable interface to at least one of the processors 802 and / or any suitable device or component communicating with the system control logic 808.

[0069] In some embodiments, the system control logic 808 may include one or more memory controllers to provide an interface connected to the system memory 804. The system memory 804 may be used to load and store data and / or instructions. In some embodiments, the memory 804 of the device 800 may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).

[0070] The NVM / memory 806 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, the NVM / memory 806 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a HDD (Hard Disk Drive), a CD (Compact Disc) drive, and a DVD (Digital Versatile Disc) drive.

[0071] The NVM / memory 806 may include a part of the storage resources installed on the device 800, or it may be accessible by the device but not necessarily part of the device. For example, the NVM / storage 806 may be accessed via the network interface 810 through a network.

[0072] In particular, system memory 804 and NVM / memory 806 may respectively include: a temporary copy and a permanent copy of instructions 820. Instructions 820 may include: instructions that, when executed by at least one of processors 802, cause device 800 to implement as Figure 1 shown in the method. In some embodiments, instructions 820, hardware, firmware, and / or their software components may alternatively / additionally be placed in system control logic 808, network interface 810, and / or processor 802.

[0073] Network interface 810 may include a transceiver for providing a radio interface for device 800 to communicate with any other suitable device (such as a front-end module, antenna, etc.) via one or more networks. In some embodiments, network interface 810 may be integrated with other components of device 800. For example, network interface 810 may be integrated with at least one of processor 802, system memory 804, NVM / memory 806, and a firmware device (not shown) having instructions, and when at least one of processors 802 executes the instructions, device 800 implements Figures 1-7 one or more embodiments of the various embodiments shown. Communication module

[0074] Network interface 810 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, network interface 810 may be a network adapter, a wireless network adapter, a telephone modem, and / or a wireless modem.

[0075] In one embodiment, at least one of processors 802 may be logically encapsulated with one or more controllers for system control logic 808 to form a system-in-package (SiP). In one embodiment, at least one of processors 802 may be integrated with the logic of one or more controllers for system control logic 808 on the same die to form a system-on-chip (SoC).

[0076] Device 800 may further include: an input / output (I / O) device 812. I / O device 812 may include a user interface that enables a user to interact with device 800; the design of the peripheral component interface enables peripheral components to also interact with device 800. In some embodiments, device 800 further includes sensors for determining at least one of environmental conditions and location information related to device 800.

[0077] In some embodiments, the user interface may include, but is not limited to, a display (e.g., a liquid crystal display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., a light-emitting diode flash), and a keyboard.

[0078] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.

[0079] In some embodiments, the sensors may include, but are not limited to, a gyroscope sensor, an accelerometer, a proximity sensor, an ambient light sensor, and a positioning unit. The positioning unit may also be part of or interact with the network interface 810 to communicate with components of a positioning network (e.g., Global Positioning System (GPS) satellites).

[0080] It can be understood that the structure illustrated in the embodiments of the present invention does not constitute a specific limitation on the Bluetooth communication device 800. In other embodiments of the present application, the Bluetooth communication device 800 may include more or fewer components than those illustrated, or combine certain components, or split certain components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0081] Program code can be applied to input instructions to perform the various functions described herein and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of the present application, a processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application specific integrated circuit (ASIC), or a microprocessor.

[0082] The program code can be implemented in a high-level procedural language or an object-oriented programming language to communicate with the processing system. When needed, the program code can also be implemented in assembly language or machine language. In fact, the mechanisms described herein are not limited to the scope of any particular programming language. In any case, the language can be a compiled language or an interpreted language.

[0083] One or more aspects of at least one embodiment can be implemented by representative instructions stored on a computer-readable storage medium, the instructions representing various logics in a processor, and the instructions, when read by a machine, cause the machine to fabricate the logics for performing the techniques described herein. These representations, referred to as “IP cores,” can be stored on a tangible computer-readable storage medium and provided to multiple customers or production facilities to be loaded into a manufacturing machine that actually fabricates the logics or processors.

[0084] An embodiment of the present application discloses a computer-readable medium storing one or more programs executable by one or more processors to implement the training method of the pancreatic exocrine function evaluation model of the present application.

[0085] An embodiment of the present application discloses a computer program product including a computer program, which when executed by a processor implements the training method of the pancreatic exocrine function evaluation model of the present application.

[0086] The above describes the implementation manners of the present application by specific embodiments, and those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. Although the description of the present application will be introduced in combination with preferred embodiments, this does not mean that the features of this invention are limited to this implementation manner. On the contrary, the purpose of introducing the invention in combination with the implementation manner is to cover other alternatives or modifications that may be extended based on the claims of the present application. In order to provide a deep understanding of the present application, many specific details will be included in the following description. The present application can also be implemented without using these details. In addition, in order to avoid confusing or obscuring the key points of the present application, some specific details will be omitted in the description. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0087] In addition, various operations will be described as multiple discrete operations in the manner that is most helpful for understanding the illustrative embodiments; however, the described order should not be construed as implying that these operations must be order-dependent. In particular, these operations do not need to be performed in the presented order.

[0088] Unless the context otherwise requires, the terms "comprise", "have" and "include" are synonyms. The phrase "A / B" means "A or B". The phrase "A and / or B" means "(A and B) or (A or B)".

[0089] As used herein, the term "module" or "unit" can refer to, be, or include: an application specific integrated circuit (ASIC), an electronic circuit, a (shared, dedicated, or group) processor and / or memory that executes one or more software or firmware programs, combinational logic circuits, and / or other suitable components that provide the described functionality.

[0090] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored on one or more transitory or non-transitory machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or via other computer-readable media. Accordingly, machine-readable media may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer), but not limited to, floppy disks, optical disks, optical discs, compact disc read-only memory (CD-ROM), magneto-optical discs, read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic or optical cards, flash memory, or tangible machine-readable memories for transmitting information via electrical, optical, acoustic, or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.) over the Internet. Thus, machine-readable media include any type of machine-readable media suitable for storing or transmitting electronic instructions or information in a form readable by a machine (e.g., a computer).

[0091] In the drawings, some structural or method features are shown in a particular arrangement and / or order. However, it should be understood that such a particular arrangement and / or order may not be required. In some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. Additionally, the inclusion of a structural or method feature in a particular figure does not imply that such a feature is required in all embodiments, and in some embodiments, these features may not be included or may be combined with other features.

[0092] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units or data, these units or data should not be limited by these terms. These terms are only used to distinguish one feature from another. For example, without departing from the scope of the exemplary embodiments, a first feature may be referred to as a second feature, and similarly, a second feature may be referred to as a first feature.

[0093] It should be noted that in this specification, like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not require further definition and explanation in subsequent drawings.

[0094] Although the present invention has been illustrated and described with reference to certain preferred embodiments thereof, those of ordinary skill in the art should understand that various changes may be made in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A training method for a pancreatic exocrine function evaluation model, characterized in that, The method includes: Obtaining a magnetic resonance image of the pancreas, and extracting radiomics features and image features of the magnetic resonance image; Obtaining pancreatic exocrine function parameters, which are obtained through a non-invasive pancreatic exocrine function experiment; Obtaining the clinical features of the pancreas; Using the radiomics features, the image features, and the clinical features as samples to train the pancreatic exocrine function evaluation model, and Adjusting the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model, wherein the image features include one or more of the main pancreatic duct diameter of the pancreatic head, visualization of the branch pancreatic duct, signal ratio of the pancreatic parenchyma to the left psoas major muscle parenchyma, and signal ratio of the pancreatic parenchyma in the arterial phase and portal phase; wherein the variables in the trained pancreatic exocrine function evaluation model include visualization of the branch pancreatic duct, diabetes, abdominal pain, and a group score calculated based on the radiomics features.

2. The training method according to claim 1, wherein The clinical features include one or more of age, gender, body mass index, history of alcohol consumption, history of smoking, hyperlipidemia, history of acute pancreatitis attack, diabetes, and abdominal pain.

3. The training method according to claim 2, wherein The method further includes performing feature screening on the radiomics features and performing dimensionality reduction processing on the radiomics features using the least absolute shrinkage and selection operator (LASSO) regression algorithm.

4. The training method according to claim 1, wherein The magnetic resonance image is a T1-weighted image (T1WI).

5. An electronic device, characterized in that, The device includes a memory storing computer-executable instructions and a processor; when the instructions are executed by the processor, the device implements the training method according to any one of claims 1 to 4.

6. A computer-readable medium, characterized in that, The computer-readable medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the training method according to any one of claims 1 to 4.

7. A training system for a pancreatic exocrine function evaluation model, characterized in that, The system includes: A first acquisition module for acquiring a magnetic resonance image of the pancreas; A first extraction module for extracting radiomics features of the magnetic resonance image; A second extraction module for extracting image features of the magnetic resonance image; A second acquisition module for acquiring pancreatic exocrine function parameters, which are obtained through a non-invasive pancreatic exocrine function experiment; A third acquisition module for acquiring the clinical features of the pancreas; A training module for using the radiomics features, the image features, and the clinical features as samples to train the pancreatic exocrine function evaluation model and training the pancreatic exocrine function evaluation model based on the pancreatic exocrine function parameters to obtain a trained pancreatic exocrine function evaluation model, wherein the image features include one or more of the main pancreatic duct diameter of the pancreatic head, visualization of the branch pancreatic duct, signal ratio of the pancreatic parenchyma to the left psoas major muscle parenchyma, and signal ratio of the pancreatic parenchyma in the arterial phase and portal phase; wherein the variables in the trained pancreatic exocrine function evaluation model include visualization of the branch pancreatic duct, diabetes, abdominal pain, and a group score calculated based on the radiomics features.

8. A computer program product, comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the training method according to any one of claims 1 to 4.

Images