Medical image processing method, electronic device and storage medium

Abstract

The present invention discloses a medical image processing method, electronic device and storage medium, the method comprising: obtaining a target case image of the same patient; obtaining a reference image in the target case image; registering other images in the target case image to the reference image to obtain a registration result; and displaying the reference image and the registration result. The present invention realizes registering all case images to the same posture, and then sequentially displays the reference image and the registration result in a set order, thereby improving the image registration efficiency and the robustness of the registration effect, while facilitating doctors to timely check and overall analyze the patient's condition, and being able to quickly understand and grasp the current patient's condition change trend, greatly improving the doctor's film reading efficiency, and optimizing the diagnosis process, effectively improving the doctor's film reading experience.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a medical image processing method, electronic equipment and storage medium. Background Art

[0002] In recent years, deep learning technology has made rapid progress, especially in the fields of computer vision and medical imaging. Computer-aided diagnosis technology based on artificial intelligence provides a reading system with accurate and detailed quantitative information of lesions, helping radiologists and clinicians improve their daily work efficiency and diagnostic accuracy. However, in the doctor's work scenario, patients often have multiple cases, and radiologists need to mention the follow-up comparison and follow-up suggestions of the patient in the patient's imaging report. Radiologists need to manually compare cases and flip through images layer by layer for comparison to initially obtain the change trend of lesions, related tissues and areas of interest. After repeated manual measurements, a rough quantitative information can be obtained, and then the language is organized to write into the patient's imaging report. This operation is undoubtedly cumbersome and time-consuming, and it is impossible to directly obtain accurate and detailed quantitative information. The overall follow-up comparison process is cumbersome, time-consuming and inconvenient to interact. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the fact that the mainstream film reading system in the prior art can only support some rough, one-to-one follow-up image displays, can only perform a rough analysis of the follow-up images, and the display form is not intuitive, which affects the doctor's judgment and seriously affects the user's interactive experience and work efficiency; during the follow-up process, the doctor's focus will be adjusted due to the condition or the doctor's personal habits, such as some mobile lesions, and the interactive mode of aligning the lesions and linking them will greatly speed up the doctor's diagnosis process. However, these film reading systems are all aligned based on images, which is difficult to meet the needs of doctors. In addition, after the doctor completes the image diagnosis, he needs to summarize the condition and organize the language to edit it into a diagnosis report. This process undoubtedly increases the doctor's workload and time consumption, that is, the existing medical image processing method cannot meet the doctor's actual film reading needs. Provide a medical image processing method, electronic device and storage medium.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] The present invention provides a method for processing a medical image, the method comprising:

[0006] Obtain target case images of the same patient;

[0007] Wherein, the target case image includes a current case image and / or several historical case images;

[0008] Acquire a reference image in the target case image;

[0009] Registering other images in the target case image with the reference image to obtain a registration result;

[0010] The reference image and the registration result are displayed.

[0011] Preferably, the obtaining of the reference image in the target case image comprises:

[0012] Calculate and obtain a reference image corresponding to the target case image;

[0013] Traversing and calculating the target similarity between each image in the target case image and the reference image;

[0014] An image whose target similarity among the target case images meets the set conditions is selected as the reference image.

[0015] Preferably, the calculation to obtain the reference image corresponding to the target case image includes:

[0016] Performing binarization processing on each of the target case images to obtain a first case image after binarization;

[0017] Performing computational processing on the first case image after binarization to obtain a corresponding target average image;

[0018] The target average image is used as the reference image.

[0019] Preferably, the calculation to obtain the reference image corresponding to the target case image includes:

[0020] Using a set of set window widths and set window positions to perform binarization processing on each image in the target case image to obtain a first processing result;

[0021] Calculating and obtaining a first average image corresponding to all the first processing results;

[0022] The first average image is used as the reference image.

[0023] Preferably, the calculation to obtain the reference image corresponding to the target case image includes:

[0024] Using multiple sets of set window widths and set window positions to perform binarization processing on each of the target case images to obtain multiple sets of second processing results;

[0025] Calculating and obtaining a second average image corresponding to each group of the second processing results;

[0026] Each of the second average images is used as the reference image, and a plurality of the reference images are acquired.

[0027] Preferably, the calculation to obtain the reference image corresponding to the target case image includes:

[0028] Performing averaging processing on the target case image to obtain the reference image;

[0029] The traversal calculation of the target similarity between each image in the target case image and the reference image includes:

[0030] Traversing and calculating the absolute error rate between each image in the target case image and the reference image;

[0031] The step of selecting the case image whose similarity is less than a set threshold from among the target case images as the reference image comprises:

[0032] The target case image corresponding to the minimum absolute error rate is selected as the reference image.

[0033] Preferably, the displaying of the reference image and the registration result comprises:

[0034] Obtaining time information corresponding to each image in the target case image;

[0035] According to the time information, the registration result corresponding to each image in the reference image and the target case image is displayed in turn;

[0036] and / or,

[0037] The method further comprises:

[0038] Obtain external operation instructions;

[0039] Obtaining a corresponding target lesion according to the external operation instruction matching;

[0040] The registration result and a lesion image corresponding to the target lesion in the reference image are obtained, and the lesion image is displayed.

[0041] Preferably, the displaying of the reference image and the registration result further comprises:

[0042] Obtaining analysis results corresponding to the current case image and / or the historical case image;

[0043] Wherein, the analysis result includes at least one of the following: overall evaluation information of the condition, follow-up time information, and quantitative information of lesions at each follow-up;

[0044] Obtaining the user's habit data of viewing imaging reports;

[0045] Based on the habit data and the analysis results, generating a corresponding target image report;

[0046] and / or,

[0047] When the target case image includes a current case image, the method further includes:

[0048] Processing the current case image to generate a lesion image and corresponding lesion quantitative information;

[0049] comparing the lesion quantification information with the symptom preset value, and generating prompt information when the lesion quantification information is greater than or equal to the symptom preset value;

[0050] The prompt information includes case information representing the severity of the patient's condition and / or triage information corresponding to the case information.

[0051] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned medical image processing method when executing the computer program.

[0052] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned medical image processing method is implemented.

[0053] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present invention.

[0054] The positive and progressive effects of the present invention are:

[0055] In the present invention, a reference image is obtained from the current case image and the historical case images, and other case images in all case images are aligned to the reference image to achieve alignment of all case images to the same posture, and then the reference image and the alignment results are displayed in sequence according to the set order, thereby improving the image alignment efficiency and the robustness of the alignment effect. At the same time, it is convenient for doctors to check and analyze the patient's condition in a timely manner, and can quickly understand and grasp the current patient's condition change trend, which greatly improves the doctor's reading efficiency and optimizes the diagnosis process, effectively improving the doctor's reading experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flowchart of a medical image processing method according to Embodiment 1 of the present invention.

[0057] Figure 2 This is a first flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0058] Figure 3 This is a second flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0059] Figure 4 This is the third flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0060] Figure 5 This is the fourth flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0061] Figure 6 This is the fifth flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0062] Figure 7 This is the sixth flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0063] Figure 8 This is the seventh flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0064] Fig. 9 This is the eighth flow chart of the medical image processing method according to embodiment 2 of the present invention.

[0065] Fig.10 This is a schematic diagram of the display interface after medical image registration in Example 2 of the present invention.

[0066] Fig.11 This is a schematic diagram of a target image report in Example 2 of the present invention.

[0067] Fig.12 This is a module diagram of a medical image processing system according to Embodiment 3 of the present invention.

[0068] Fig.13 This is a schematic diagram of the structure of an electronic device for implementing a medical image processing method according to Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0069] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0070] Example 1

[0071] like Figure 1 As shown, the medical image processing method of this embodiment includes:

[0072] S101, obtaining a target case image of the same patient;

[0073] The target case images include current case images and / or several historical case images, which include but are not limited to the brain CT (computer tomography) images corresponding to the follow-up reading of the patient's intracranial hemorrhage.

[0074] S102, obtaining a reference image in the target case image;

[0075] S103, registering other images in the target case image with the reference image to obtain a registration result;

[0076] S104, displaying the reference image and the registration result;

[0077] Among them, the comparison between the current case image and the historical case image after registration processing is displayed in the same floating window, which can be displayed in multiple ways such as but not limited to secondary windows, sidebars, etc.

[0078] In addition, the comparison of analysis results between the current case image and the historical case image is in the form of a table and a line graph, and also includes but is not limited to a bar graph, a pie chart, a text description and other forms of expression.

[0079] In this embodiment, a reference image is obtained from the current case image and historical case images, and other case images in all case images are aligned with the reference image to achieve alignment of all case images to the same posture. Then, the reference image and the alignment results are displayed in sequence according to the set order, thereby improving the image alignment efficiency and the robustness of the alignment effect. At the same time, it is convenient for doctors to check and analyze the patient's condition in a timely manner, and can quickly understand and grasp the changing trend of the patient's condition, which greatly improves the doctor's reading efficiency and optimizes the diagnosis process, effectively improving the doctor's reading experience.

[0080] Example 2

[0081] The medical image processing method of this embodiment is a further improvement on the embodiment 1. Specifically:

[0082] In one possible embodiment, Figure 2 As shown, step S102 includes:

[0083] S1021, calculating and obtaining a reference image corresponding to the target case image;

[0084] S1022, traversing and calculating the target similarity between each image in the target case image and the reference image;

[0085] S1023, selecting an image whose target similarity meets the set conditions among the target case images as a reference image.

[0086] By first calculating the benchmark image corresponding to all case images, and then calculating the similarity between each case image and the benchmark image (characterizing the difference or distance between any two compared case images), the image with a similarity greater than a set value is selected as the reference point for registration (for example, an image with a similarity greater than 98% is selected, and of course the set value can be determined and adjusted according to actual needs). Other case images are registered to it, and this registration method is used to register all case images to the same posture, ensuring the comparability of all case images. This can help doctors accurately understand changes in the patient's condition, clarify the direction of disease progression, and provide further treatment plans in a timely manner.

[0087] In one possible embodiment, Figure 3 As shown, step S1021 includes:

[0088] S102111, performing binarization processing on each of the target case images to obtain a first case image after binarization;

[0089] S102112, performing calculation processing on the binarized first case image to obtain a corresponding target average image;

[0090] S102113. Use the target average image as a reference image.

[0091] By binarizing the case images and then taking the sum and average or weighted average of all the binarized images, the corresponding target average image is calculated, and then the target average image is used as the reference image to quickly obtain a reference image that can characterize all case images, thereby ensuring the efficiency and accuracy of the registration.

[0092] In one possible embodiment, Figure 4 As shown, step S1021 includes:

[0093] S102121, using a set of set window widths and set window positions to perform binarization processing on each image in the target case image to obtain a first processing result;

[0094] S102122. Calculate and obtain a first average image corresponding to all first processing results;

[0095] S102123, using the first average image as a reference image;

[0096] Step S1022 includes:

[0097] S102211, traversing and calculating the target similarity between the first case image after binarization in the target case image and the reference image;

[0098] Among them, target similarity corresponds to mean absolute difference, intersection-over-union ratio, Euclidean distance or absolute error rate, etc.

[0099] By adopting a set of set window widths and set window positions (such as window position 800HU, window width 800HU is actually equivalent to using a threshold of 400 for binarization, HU is Hausfeld unit), all case images are binarized, and then all binarized images are summed and averaged or weighted averaged to calculate the corresponding target average image, and then the target average image is used as the reference image, which corresponds to a reference image, and then the similarity between all case images and this reference image is traversed and calculated, and all case images are aligned to the same posture to ensure the alignment effect and efficiency.

[0100] In one possible embodiment, Figure 5 As shown, step S1021 includes:

[0101] S102131, using multiple sets of set window widths and set window positions to perform binarization processing on each image in the target case image, to obtain multiple sets of second processing results;

[0102] S102132. Calculate and obtain a second average image corresponding to each group of second processing results;

[0103] S102133, taking each second average image as a reference image, and acquiring a plurality of reference images;

[0104] Step S1022 includes:

[0105] S102221, calculating first similarities between the same first case image after binarization and a plurality of reference images;

[0106] S102222. Calculate and obtain a third average image corresponding to all first similarities, and use the third average image as a target similarity corresponding to the current first case image;

[0107] Among them, target similarity corresponds to mean absolute difference, intersection over union ratio, Euclidean distance or absolute error rate.

[0108] By adopting multiple groups of set window widths and set window positions (such as using window positions of 700HU, 800HU, and 900HU respectively), all case images are binarized, and then all the binarized images are summed and averaged or weighted averaged to calculate the average image corresponding to each group, and then each average image is used as a reference image, which corresponds to multiple reference images; then for the same case image in all case images, the similarity between the same case image and each reference image is calculated, and the average value corresponding to these similarities is calculated as the target similarity corresponding to the current case image, and so on, each case image in all case images is traversed and calculated to calculate the target similarity corresponding to each case image.

[0109] For example: for all case images (100 images), three sets of set window widths and set window positions (700HU, 800HU, and 900HU, respectively) are used to binarize these 100 case images, and the binarized images are summed and averaged to obtain three sets of average images, that is, three reference images (image A, image B, image C) are obtained at this time; for any case image among the 100 case images, it is necessary to calculate the three similarities (a, b, c) between the same case image and the three reference images (image A, image B, image C), and then the three similarities are summed and averaged to obtain the corresponding average value avg=(a+b+c) / 3 as the target similarity corresponding to the current case image; and so on, the target similarity corresponding to each case image in the 100 images is calculated.

[0110] That is, by acquiring multiple reference images, the calculation accuracy of the similarity corresponding to each case image is improved, thereby ensuring the accuracy of the final reference image, thereby improving the overall registration accuracy and the robustness of the registration effect.

[0111] In one possible embodiment, Figure 6 As shown, step S1021 includes:

[0112] S102141. Perform average processing on the target case image to obtain a reference image;

[0113] Step S1022 includes:

[0114] S102231, traversing and calculating the absolute error rate between each image in the target case image and the reference image;

[0115] Step S1023 includes:

[0116] S10231. Select the case image corresponding to the minimum absolute error rate as the reference image.

[0117] All case images are directly averaged to obtain a benchmark image, and then the absolute error rate between each image and the benchmark image is traversed and calculated, and the image with the minimum absolute error rate is selected as the reference image. This ensures that the acquisition efficiency of the benchmark image is improved while ensuring the accuracy of the benchmark image, thereby improving the overall registration efficiency.

[0118] In one possible embodiment, Figure 7 As shown, step S104 includes:

[0119] S1041, obtaining time information corresponding to each image in the target case image;

[0120] S1042. Display the registration results corresponding to each image in the reference image and the target case image in turn according to the time information.

[0121] According to the follow-up time corresponding to all case images, the case images that have been registered are displayed in chronological order, so that doctors can view them one by one in sequence, and communicate with the current patients based on the corresponding conditions at different follow-up time points. They can also quickly understand and grasp the changing trends of the current patients' conditions, which greatly improves the doctor's reading efficiency and optimizes the diagnosis process.

[0122] In one possible embodiment, Figure 8 As shown, after step S104, the following steps are further included:

[0123] S105, obtaining external operation instructions;

[0124] Among them, external operation instructions include clicking, text input, voice input and other methods.

[0125] S106, obtaining a corresponding target lesion according to the external operation instruction;

[0126] S107, obtaining the registration result and the lesion image corresponding to the target lesion in the reference image, and displaying the lesion image.

[0127] Taking into account that the doctor's focus may be adjusted during the follow-up process due to the condition or the doctor's personal habits, such as some mobile lesions, the doctor can enter the corresponding operation instructions at this time, such as clicking on the target lesion position in any case image in the display interface or clicking on the access link associated with the target lesion, and then automatically determine the target lesion that the doctor currently needs to view, and then the current case image will be partially enlarged and displayed according to the center of the lesion. At the same time, the remaining case images will be controlled to display the target lesion, which is convenient for the doctor to diagnose and analyze the changes in the condition of a specific lesion. It not only meets the overall diagnosis and analysis of the entire diagnostic area, but also meets the specific diagnosis of any local lesion. It effectively meets the different diagnostic needs and habits of different doctors, and greatly improves the doctor's film reading experience.

[0128] In one possible embodiment, Fig. 9 As shown, after step S104, the following steps are further included:

[0129] S108, obtaining analysis results corresponding to the current case image and / or historical case images;

[0130] Among them, the analysis results include overall disease evaluation information, follow-up time information, and lesion quantitative information at each follow-up;

[0131] S109, obtaining the user's habit data of viewing imaging reports;

[0132] S1010: Generate a corresponding target image report based on the habit data and analysis results.

[0133] All case images have corresponding analysis results. These analysis results and the doctor's habit of viewing reports are combined to automatically generate a structured imaging report that is compatible with the current doctor. There is no need for the doctor to fill in the report manually, which greatly simplifies the report processing process and makes it easier for the doctor to copy and use it later or directly modify the generated imaging report to form an imaging report that better meets the current doctor's requirements. Of course, when the analysis results change, the image can be automatically updated based on the changed analysis results, which improves the degree of automation in obtaining imaging reports, reduces labor costs and ensures the accuracy of imaging reports.

[0134] By adding image acquisition devices (such as cameras) or fingerprint acquisition devices, the user information of the current doctor can be collected in a timely manner, and then the expression habit data of the image report that is compatible with the user information can be queried and matched in the database. When the doctor's habit data for viewing image reports changes, the corresponding record information can be updated in time to ensure that the subsequently generated image reports are more appropriate for the current doctor, and more effectively ensure the doctor's user experience throughout the entire reading process.

[0135] Take the brain CT image corresponding to the follow-up reading of the patient's intracranial hemorrhage as an example. Fig.10 As shown, the reference image and the registration result can be displayed in conjunction in the display window, and the follow-up time, analysis results, case quantitative data and other information corresponding to each case image can be displayed in the corresponding image position to guide doctors to view and analyze, and can be easily identified without causing confusion.

[0136] like Fig.11 As shown, in the current imaging diagnosis page, a structured imaging report for imaging-assisted diagnosis of intracranial hemorrhage is provided according to the doctor's expression habits.

[0137] In addition, the structured imaging report includes but is not limited to the analysis results of the current case image, the analysis results corresponding to the historical case images at different historical follow-up time points, or a combination of the analysis results of the current case image and the analysis results corresponding to the historical case images, which can be determined or adjusted according to actual needs.

[0138] In a possible embodiment, when the target case image includes the current case image, the medical image processing method of this embodiment further includes:

[0139] Processing the current case image to generate a lesion image and corresponding lesion quantitative information;

[0140] Among them, the lesion quantitative information is used to characterize the parameter information of the patient's specific lesion condition. The lesion quantitative information includes but is not limited to lesion type, lesion level, lesion location, lesion volume and lesion CT value.

[0141] The lesion images are typeset and displayed; wherein the typeset and display methods can be adjusted and set according to actual needs.

[0142] The film reading system intuitively displays the lesion images corresponding to different patients in a certain format, allowing doctors to directly view the patient's condition without other tedious operations, thereby simplifying the operation process and improving the overall film reading efficiency.

[0143] Comparing the lesion quantitative information with the symptom preset value, and generating a prompt message when the lesion quantitative information is greater than or equal to the symptom preset value;

[0144] The symptom preset values ​​include, but are not limited to, values ​​used to determine whether the patient's condition has reached a critical state (critical value), or values ​​used to determine whether the patient has a certain condition (positive) (lesion characterization value), etc. The symptom preset values ​​can be preset by the hospital based on actual experience. The prompt information includes case information characterizing the severity of the patient's condition and / or triage information corresponding to the case information.

[0145] By timely comparing the quantitative lesion information of each patient with the preset symptom value, once it is greater than or equal to the preset symptom value, it indicates that the current patient is in critical condition or has a hidden danger, and the doctor needs to intervene in time. At this time, the hospital's reporting requirements are met, specifically by timely triggering the information prompt module to push prompt messages to the doctor's terminal, and timely reporting the case information and triage information of the patient to be processed in the doctor's terminal, so as to notify the doctor to deal with the emergency patient as soon as possible and let the patient enter the diagnosis and treatment process as soon as possible, thereby effectively improving the processing efficiency of the critical process in the hospital, avoiding the uncertainty and randomness of the existing manual processing, and better meeting the hospital's needs for critical processes.

[0146] While prompt information is pushed to the doctor's terminal, case information and triage information are simultaneously and intuitively displayed in the processing system to facilitate further review and processing by doctors and other relevant personnel.

[0147] The prompt interface may correspond to a window widget, and the interactive form may be a floating component or other form. The prompt part may be composed of numbers, colors, texts and other elements that are easy to distinguish and warn. For example, numbers can be used to display the number of patients that the doctor needs to treat, and colors can be used to display whether the patient's condition is critical.

[0148] In addition, the corresponding case information and triage information can also be popped up on the doctor's terminal through this prompt method.

[0149] The medical image processing method of this embodiment also includes:

[0150] When the lesion quantitative information is greater than or equal to the symptom preset value, the patient list information is generated and displayed, and the detailed information corresponding to the patient list information is displayed.

[0151] That is, the patient information that needs to be processed within a period of time is listed in the form of a list, which includes the patient's basic information, disease condition and other specific information.

[0152] The triage processing system generates a list of all patients who need to be treated, so that doctors can further review the case information corresponding to each patient based on the list information, and then diagnose and treat the patient based on the viewed content.

[0153] Specifically, the doctor clicks on the patient list information displayed on the display screen of the triage processing system, and then directly jumps to the film reading system to view the case results of different patients in detail for subsequent diagnosis.

[0154] Among them, the patient list information includes but is not limited to the disease information, location information, type information, criticality information, time information, processing status information and department information corresponding to each patient. This information can be filtered through the filter box to display the content that needs to be displayed in a targeted manner.

[0155] In order to avoid the huge number of patient cases and the difficulty for doctors to search, general operation functions such as disease type and timed clearing can also be set to further optimize the processing flow of medical images in the hospital and improve the processing efficiency of medical images.

[0156] In this embodiment, a reference image is obtained from the current case image and historical case images, and other case images in all case images are aligned to the reference image to achieve alignment of all case images to the same posture, and then the reference image and the alignment results are displayed in turn according to the set order, thereby improving the efficiency of image alignment and the robustness of the alignment effect. At the same time, it is convenient for doctors to timely view and comprehensively analyze the patient's condition, and can quickly understand and grasp the changing trend of the current patient's condition, greatly improving the doctor's film reading efficiency, and optimizing the diagnosis process, effectively improving the doctor's film reading experience; at the same time, all case images have corresponding analysis results combined with the doctor's habit of viewing reports to automatically generate a structured image report that is compatible with the current doctor, without the need for doctors to manually fill in and process, greatly simplifying the report processing process, facilitating doctors to copy and use it later, etc., improving the degree of automation in obtaining image reports, reducing labor input costs and ensuring the accuracy of image reports.

[0157] Example 3

[0158] like Fig.12 As shown, the medical image processing system of this embodiment includes:

[0159] The target case image acquisition module 1 is used to acquire the target case image of the same patient;

[0160] Wherein, the target case image includes a current case image and / or several historical case images;

[0161] A reference image acquisition module 2 is used to acquire a reference image in a target case image;

[0162] A registration result acquisition module 3 is used to register other images in the target case image with the reference image to obtain a registration result;

[0163] The display module 4 is used to display the reference image and the registration result.

[0164] In this embodiment, a reference image is obtained from the current case image and historical case images, and other case images in all case images are aligned with the reference image to achieve alignment of all case images to the same posture. Then, the reference image and the alignment results are displayed in sequence according to the set order, thereby improving the image alignment efficiency and the robustness of the alignment effect. At the same time, it is convenient for doctors to timely check and comprehensively analyze the patient's condition, effectively improving the doctor's experience of reading films.

[0165] Example 4

[0166] The medical image processing system of this embodiment is a further improvement of Embodiment 3. Specifically:

[0167] In a possible embodiment, the reference image acquisition module 2 includes:

[0168] A reference image acquisition unit, used for calculating and obtaining a reference image corresponding to a target case image;

[0169] A target similarity calculation unit, used for traversing and calculating the target similarity between each image in the target case image and the reference image;

[0170] The reference image acquisition unit is used to select an image whose target similarity meets the set conditions in the target case image as a reference image.

[0171] In one embodiment, the reference image acquisition unit includes:

[0172] A first case image acquisition subunit is used to perform binarization processing on each image in the target case image to obtain a binarized first case image;

[0173] A target average image acquisition subunit is used to perform calculation processing on the binarized first case image to obtain a corresponding target average image;

[0174] The reference image acquisition subunit is used to use the target average image as the reference image.

[0175] In one embodiment, the reference image acquisition unit includes:

[0176] A first processing result acquisition subunit is used to perform binarization processing on each image in the target case image by using a set of set window widths and set window positions to obtain a first processing result;

[0177] A first average image acquisition subunit, used for calculating and obtaining a first average image corresponding to all first processing results;

[0178] A reference image acquisition subunit, configured to use the first average image as a reference image;

[0179] The target similarity calculation unit is used to traverse and calculate the target similarity between the first case image after binarization in the target case image and the reference image;

[0180] Among them, target similarity corresponds to mean absolute difference, intersection over union ratio, Euclidean distance or absolute error rate.

[0181] In one embodiment, the reference image acquisition unit includes:

[0182] A second processing result acquisition subunit is used to perform binarization processing on each image in the target case image by using multiple sets of set window widths and set window positions to obtain multiple sets of second processing results;

[0183] A second average image acquisition subunit, used for calculating and obtaining a second average image corresponding to each group of second processing results;

[0184] A reference image acquisition subunit, used to use each second average image as a reference image to acquire multiple reference images;

[0185] The target similarity calculation unit includes:

[0186] A first similarity calculation subunit is used to calculate first similarities between the same first case image after binarization and the plurality of reference images;

[0187] A target similarity acquisition subunit is used to calculate and obtain a third average image corresponding to all first similarities, and use the third average image as the target similarity corresponding to the current first case image;

[0188] Among them, target similarity corresponds to mean absolute difference, intersection-over-union ratio, Euclidean distance or absolute error rate, etc.

[0189] In a possible embodiment, the reference image acquisition unit is used to perform average processing on the target case image to acquire the reference image;

[0190] The target similarity calculation unit is used to traverse and calculate the absolute error rate between each image in the target case image and the reference image;

[0191] The reference image acquisition unit is used to select the case image corresponding to the minimum absolute error rate as the reference image.

[0192] In a possible embodiment, the display module 4 includes:

[0193] A time information acquisition unit, used to acquire time information corresponding to each image in the target case image;

[0194] The display unit is used to display the registration results corresponding to each image in the reference image and the target case image in turn according to the time information.

[0195] In a possible embodiment, the medical image processing system of this embodiment further includes:

[0196] An external operation instruction acquisition block is used to acquire external operation instructions;

[0197] A target lesion determination module is used to obtain the corresponding target lesion according to the external operation instruction;

[0198] The display module 4 is also used to obtain the registration result and the lesion image corresponding to the target lesion in the reference image, and to display the lesion image.

[0199] In a possible embodiment, the medical image processing system of this embodiment further includes:

[0200] An analysis result acquisition module is used to acquire analysis results corresponding to the current case image and / or historical case images;

[0201] The analysis results include at least one of the following: overall evaluation information of the condition, follow-up time information, and quantitative information of lesions at each follow-up;

[0202] A habit data acquisition module is used to acquire the user's habit data of viewing image reports;

[0203] A target image report generation module is used to generate a corresponding target image report based on the habit data and analysis results;

[0204] In a possible embodiment, when the target case image includes the current case image, the medical image processing system of this embodiment further includes:

[0205] A lesion quantitative information generation module is used to process the current case image to generate a lesion image and corresponding lesion quantitative information;

[0206] An information comparison module is used to compare the lesion quantitative information with the symptom preset value, and when the lesion quantitative information is greater than or equal to the symptom preset value, the prompt information generation module is called to generate prompt information;

[0207] The prompt information includes case information representing the severity of the patient's condition and / or triage information corresponding to the case information.

[0208] It should be noted that the working principle of the medical image processing system of this embodiment is similar to the implementation principle corresponding to the medical image processing method of Embodiment 1 or 2, and therefore will not be described in detail in this embodiment.

[0209] In this embodiment, a reference image is obtained from the current case image and historical case images, and other case images in all case images are aligned with the reference image to achieve alignment of all case images to the same posture. Then, the reference image and the alignment results are displayed in sequence according to the set order, thereby improving the image alignment efficiency and the robustness of the alignment effect. At the same time, it is convenient for doctors to check and analyze the patient's condition in a timely manner, and can quickly understand and grasp the changing trend of the patient's condition, which greatly improves the doctor's reading efficiency and optimizes the diagnosis process, effectively improving the doctor's reading experience.

[0210] Example 5

[0211] Fig.13 This is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present invention. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the medical image processing method in Embodiment 1 or 2 when executing the program. Fig.13 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0212] like Fig.13 As shown, the electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).

[0213] The bus 33 includes a data bus, an address bus, and a control bus.

[0214] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .

[0215] The memory 32 may also include a program / utility 325 having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0216] The processor 31 executes various functional applications and data processing by running the computer programs stored in the memory 32, such as the medical image processing method in Embodiment 1 or 2 of the present invention.

[0217] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the model generating device 30 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 36. Fig.13 As shown, the network adapter 36 communicates with other modules of the model-generated device 30 via the bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the model-generated device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0218] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0219] Example 6

[0220] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the medical image processing method in Embodiment 1 or 2 are implemented.

[0221] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0222] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps in the medical image processing method in Example 1 or 2.

[0223] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0224] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for processing medical images, characterized in that: The method comprises: Obtain target case images of the same patient; Wherein, the target case image includes a current case image and / or several historical case images; Acquire a reference image in the target case image; Registering other images in the target case image with the reference image to obtain a registration result; Displaying the reference image and the registration result; The step of acquiring a reference image in the target case image comprises: Calculate and obtain a reference image corresponding to the target case image; Traversing and calculating the target similarity between each image in the target case image and the reference image; Selecting an image whose target similarity meets a set condition among the target case images as the reference image; The calculating and obtaining a reference image corresponding to the target case image includes: Performing binarization processing on each of the target case images to obtain a first case image after binarization; Performing computational processing on the first case image after binarization to obtain a corresponding target average image; The target average image is used as the reference image.

2. The method for processing medical images according to claim 1, characterized in that: The calculating and obtaining a reference image corresponding to the target case image includes: Using a set of set window widths and set window positions to perform binarization processing on each image in the target case image to obtain a first processing result; Calculating and obtaining a first average image corresponding to all the first processing results; The first average image is used as the reference image.

3. The medical image processing method according to claim 1, characterized in that: The calculating and obtaining a reference image corresponding to the target case image includes: Using multiple sets of set window widths and set window positions to perform binarization processing on each of the target case images to obtain multiple sets of second processing results; Calculating and obtaining a second average image corresponding to each group of the second processing results; Each of the second average images is used as the reference image, and a plurality of the reference images are acquired.

4. The method for processing medical images according to claim 1, characterized in that: The calculating and obtaining a reference image corresponding to the target case image includes: Performing averaging processing on the target case image to obtain the reference image; The traversal calculation of the target similarity between each image in the target case image and the reference image includes: Traversing and calculating the absolute error rate between each image in the target case image and the reference image; The step of selecting the case image whose similarity is less than a set threshold from among the target case images as the reference image comprises: The target case image corresponding to the minimum absolute error rate is selected as the reference image.

5. The method for processing medical images according to any one of claims 1 to 4, characterized in that: The displaying of the reference image and the registration result includes: Obtaining time information corresponding to each image in the target case image; According to the time information, the registration results corresponding to each image in the reference image and the target case image are displayed in turn.

6. The method for processing medical images according to any one of claims 1 to 4, characterized in that: The method further comprises: Obtain external operation instructions; Obtaining a corresponding target lesion according to the external operation instruction matching; The registration result and a lesion image corresponding to the target lesion in the reference image are obtained, and the lesion image is displayed.

7. The method for processing medical images according to any one of claims 1 to 4, characterized in that: The display of the reference image and the registration result further includes: Obtaining analysis results corresponding to the current case image and / or the historical case image; Wherein, the analysis result includes at least one of the following: overall evaluation information of the condition, follow-up time information, and quantitative information of lesions at each follow-up; Obtaining the user's habit data of viewing imaging reports; Based on the habit data and the analysis result, a corresponding target image report is generated.

8. The method for processing medical images according to any one of claims 1 to 4, characterized in that: When the target case image includes a current case image, the method further includes: Processing the current case image to generate a lesion image and corresponding lesion quantitative information; comparing the lesion quantification information with the symptom preset value, and generating prompt information when the lesion quantification information is greater than or equal to the symptom preset value; The prompt information includes case information representing the severity of the patient's condition and / or triage information corresponding to the case information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the medical image processing method described in any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the medical image processing method according to any one of claims 1 to 8 is implemented.

Images