Method and system for recovering and analyzing multi-focus type medical image

Abstract

The invention relates to a recovery and analysis method and system for a multi-focus type medical image, and relates to the technical field of medical image processing. The multi-focus type medical image recovery method comprises the following steps: acquiring a medical image and a focus mask image corresponding to the medical image; wherein the focus mask image is configured to be different mask numerical values corresponding to a plurality of focus types; performing multiplication operation on the medical image and the focus mask image corresponding to the medical image to obtain a focus gray value image; and carrying out gray value recovery on the focus gray value image based on the focus gray value image, and mask numerical values and focus spatial positions corresponding to the focus types in the plurality of focus types in the focus mask image, so as to obtain a focus image with an original gray value. According to the embodiment of the invention, the recovery of the multi-focus type medical image and the analysis of the medical image can be realized.

Description

Technical Field

[0001] The present disclosure relates to the technical field of medical image processing, and in particular to a method and system for restoring and analyzing medical images of multiple lesion types. Background Art

[0002] Medical image segmentation technology based on convolutional neural networks (CNNs) can separate different tissue structures or organs in medical images and has achieved remarkable success in medical image analysis. CNNs segment organs and lesions in medical images through automated and standardized methods, improving diagnostic accuracy and efficiency.

[0003] Medical image segmentation models based on convolutional neural networks can generate multi-lesion mask images corresponding to multiple lesion types. However, in multi-lesion mask images, the different mask values corresponding to multiple lesion types cannot produce lesion images corresponding to the original grayscale values consistent with the medical image. This makes it impossible to measure the lesion grayscale values based on the lesion image in one step, hindering further analysis of the lesion grayscale values based on the lesion image, especially when the lesion overlaps with the organ where the lesion is located. Summary of the Invention

[0004] The present disclosure proposes a method for restoring and analyzing medical images of multiple lesion types and a corresponding technical solution of the system.

[0005] According to one aspect of the present disclosure, a method for restoring a medical image of multiple lesion types is provided, comprising:

[0006] Acquire a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured as different mask values corresponding to multiple lesion types; perform a multiplication operation on the medical image and its corresponding lesion mask image to obtain a lesion grayscale value image; based on the lesion grayscale value image, the mask values corresponding to each lesion type of the multiple lesion types in the lesion mask image, and the spatial position of the lesion, perform grayscale value restoration on the lesion grayscale value image to obtain a lesion image with original grayscale values.

[0007] Preferably, the method of performing grayscale value restoration on the lesion grayscale value image based on the mask values and lesion spatial positions corresponding to each lesion type of the multiple lesion types in the lesion mask image to obtain a lesion image with original grayscale values includes: extracting the mask values and lesion spatial positions corresponding to each lesion type in the lesion mask image; locating the spatial position of the lesion type on the lesion grayscale value image based on the lesion spatial positions corresponding to each lesion type to obtain a spatial position positioning image corresponding to each lesion type; and dividing the spatial position positioning image corresponding to each lesion type by the corresponding mask value to obtain a lesion image with original grayscale values.

[0008] Preferably, before obtaining the medical image and its corresponding lesion mask image, a preset segmentation network, a medical training image corresponding to the preset segmentation network for training, and a multi-mask label fusion image representing organs and lesions corresponding to the medical training image are obtained; the preset segmentation network is trained on the medical training image and its corresponding multi-mask label fusion image to obtain a corresponding organ / lesion segmentation model; based on the organ / lesion segmentation model, the medical image corresponding to the medical training image is subjected to lesion segmentation to obtain a lesion mask image.

[0009] According to one aspect of the present disclosure, a method for analyzing a medical image is provided, comprising:

[0010] Obtaining a lesion map with original grayscale values using the above-mentioned restoration method for multi-lesion type medical images;

[0011] If the dimension corresponding to the medical image is two-dimensional, extracting a lesion image with original grayscale values from the medical image based on the medical image and its corresponding lesion mask image; calculating the average grayscale value corresponding to each lesion type in the lesion mask image, and determining the lesion grayscale value corresponding to each lesion type;

[0012] If the dimension corresponding to the medical image is three-dimensional, then based on the two-dimensional slice lesion mask image in the lesion mask image of each two-dimensional slice medical image and its corresponding slice in the three-dimensional medical image, a three-dimensional lesion image with original grayscale values is extracted from the three-dimensional medical image; the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image is calculated respectively to determine the lesion grayscale value corresponding to each lesion type.

[0013] Preferably, the medical image analysis method further includes:

[0014] If the dimension corresponding to the medical image is two-dimensional, edge detection is performed on each lesion corresponding to each lesion type on the lesion mask image to obtain a two-dimensional lesion edge mask image; based on the two-dimensional lesion edge mask image, one or more of the following information is determined: the lesion shape corresponding to each lesion type, the major diameter and minor diameter of the lesion, and the lesion area;

[0015] If the dimension corresponding to the medical image is three-dimensional, the lesion area corresponding to each lesion type in each two-dimensional slice lesion mask image in the three-dimensional lesion mask image is calculated respectively; multiple lesion areas are obtained; the major axis and minor axis corresponding to the maximum lesion area among the multiple lesion areas corresponding to each lesion type are calculated respectively, and the major axis and minor axis corresponding to each lesion type are obtained; the edge detection is performed on the lesions corresponding to each lesion type on the lesion mask image respectively, and a three-dimensional lesion edge mask line is obtained; the three-dimensional lesion edge mask line corresponding to each lesion type is fitted respectively, and the lesion shape corresponding to each lesion type is obtained; the mask map corresponding to each lesion type in the lesion mask image is three-dimensionally reconstructed respectively, and the lesion volume corresponding to each lesion type is obtained.

[0016] Preferably, the medical image analysis method further includes:

[0017] If the dimension corresponding to the medical image is three-dimensional, the organ / lesion segmentation model is used to perform organ segmentation on the medical image to obtain a three-dimensional organ mask image; according to the three-dimensional organ mask image, the organ mask in the three-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the three-dimensional organ mask image, the lesion position corresponding to each lesion type is determined.

[0018] Preferably, the method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the lateral organ mask and the right organ mask in the three-dimensional organ mask image includes: extracting multiple two-dimensional slice lesion mask images corresponding to each lesion type from the three-dimensional lesion mask image; if the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the left organ mask spatial positions of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the right organ mask spatial positions of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

[0019] Preferably, the medical image analysis method further includes:

[0020] If the dimension corresponding to the medical image is two-dimensional, the organ / lesion segmentation model is used to perform organ segmentation on the medical image to obtain a two-dimensional organ mask image; based on the two-dimensional organ mask image, the organ mask in the two-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the two-dimensional organ mask image, the lesion position corresponding to each lesion type is determined.

[0021] Preferably, the method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the organ mask image includes: if the lesion mask spatial position corresponding to the lesion mask image is within the left organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial position corresponding to the lesion mask image is within the right organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

[0022] According to one aspect of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising:

[0023] An acquisition unit is used to acquire a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured as different mask values corresponding to multiple lesion types; a multiplication unit is used to perform a multiplication operation on the medical image and its corresponding lesion mask image to obtain a lesion grayscale value image; a first restoration unit is used to restore the grayscale value of the lesion grayscale value image based on the lesion grayscale value image, the mask values corresponding to each lesion type of the multiple lesion types in the lesion mask image, and the spatial position of the lesion, to obtain a lesion image with original grayscale value.

[0024] According to one aspect of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: an electronic device configured with a processor and a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned method for restoring medical images of multiple lesion types.

[0025] According to one aspect of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned method for restoring medical images of multiple lesion types.

[0026] According to one aspect of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the above-mentioned method for restoring medical images of multiple lesion types.

[0027] According to one aspect of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: a computer program product, comprising a computer program / instruction, which implements the above-mentioned method for restoring medical images of multiple lesion types when executed by a processor.

[0028] According to one aspect of the present disclosure, a medical image analysis device / system is provided, comprising:

[0029] A second restoration unit is configured to obtain a lesion image with original grayscale values by using the above-mentioned restoration method for medical images of multiple lesion types or the above-mentioned restoration system for medical images of multiple lesion types; a determination unit is configured to extract, if the dimension corresponding to the medical image is two-dimensional, a lesion image with original grayscale values from the medical image based on the medical image and its corresponding lesion mask image; calculate the average grayscale value corresponding to each lesion type in the lesion mask image, and determine the lesion grayscale value corresponding to each lesion type; and / or, if the dimension corresponding to the medical image is three-dimensional, extract a three-dimensional lesion image with original grayscale values from the three-dimensional medical image based on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice; calculate the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image, and determine the lesion grayscale value corresponding to each lesion type.

[0030] According to one aspect of the present disclosure, a medical image analysis device / system is provided, comprising: an electronic device configured with a processor and a memory for storing processor-executable instructions; wherein the processor is configured to call instructions stored in the memory to execute the above-mentioned medical image analysis method.

[0031] According to one aspect of the present disclosure, a medical image analysis device / system is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned medical image analysis method.

[0032] According to one aspect of the present disclosure, a medical image analysis device / system is provided, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions implement the above-mentioned medical image analysis method when executed by a processor.

[0033] According to one aspect of the present disclosure, a medical image analysis device / system is provided, comprising: a computer program product including a computer program / instruction, which implements the above-mentioned medical image analysis method when executed by a processor.

[0034] In the disclosed embodiments, a method and system for restoring and analyzing medical images of multiple lesion types are proposed to solve the technical problem that it is impossible to obtain a lesion image corresponding to the original grayscale value consistent with the medical image, resulting in the inability to perform a one-step lesion grayscale value measurement based on the lesion image, which hinders further analysis of the lesion grayscale value based on the lesion image.

[0035] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0036] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to explain the technical solutions of the present disclosure.

[0038] Figure 1 A flowchart illustrating a method for restoring a multi-lesion type medical image according to an embodiment of the present disclosure is shown;

[0039] Figure 2 is a block diagram of an electronic device 800 according to an exemplary embodiment;

[0040] Figure 3 is a block diagram of an electronic device 1900 according to an exemplary embodiment. DETAILED DESCRIPTION

[0041] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0042] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0043] The term "and / or" herein simply describes an association relationship between associated objects, indicating that three relationships can exist. For example, "A and / or B" can represent the existence of three situations: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" herein refers to any combination of at least two of any one or more of a plurality of items. For example, "at least one of A, B, and C" can represent any one or more elements selected from the set consisting of A, B, and C.

[0044] In addition, numerous specific details are provided in the following detailed description to better illustrate the present disclosure. Those skilled in the art will appreciate that the present disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art are not described in detail in order to highlight the main points of the present disclosure.

[0045] It can be understood that the various method embodiments for restoring and analyzing multi-lesion type medical images mentioned in this disclosure can be combined with each other to form combined embodiments without violating the principle logic. Due to space limitations, this disclosure will not elaborate on them.

[0046] In addition, the present disclosure also provides a recovery and analysis device / system, electronic device, computer-readable storage medium, and program for multi-lesion type medical images. The above can be used to implement any multi-lesion type medical image recovery and analysis method provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding records in the method section and will not be repeated here.

[0047] Figure 1 FIG. 1 is a flow chart showing a method for restoring a multi-lesion type medical image according to an embodiment of the present disclosure. Figure 1As shown, the method for restoring a medical image with multiple lesion types includes: step S101: acquiring a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured with different mask values corresponding to multiple lesion types; step S102: performing a multiplication operation on the medical image and its corresponding lesion mask image to obtain a lesion grayscale value image; and step S103: performing grayscale value restoration on the lesion grayscale value image based on the lesion grayscale value image, the mask values corresponding to each of the multiple lesion types in the lesion mask image, and the spatial positions of the lesions, to obtain a lesion image with original grayscale values. This solves the technical problem of being unable to obtain a lesion image corresponding to the original grayscale values consistent with the medical image, resulting in an inability to perform one-step lesion grayscale value measurement based on the lesion image, hindering further analysis of the lesion grayscale values based on the lesion image.

[0048] Step S101: Acquire a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured with different mask values corresponding to multiple lesion types.

[0049] In the embodiments of the present disclosure and other possible embodiments, the medical image may be configured as any type of medical image corresponding to any part. For example, a specific body part may be scanned or photographed using CT / DR / PET-CT / MRI / ultrasound equipment to obtain a corresponding medical image.

[0050] In the embodiments of the present disclosure and other possible embodiments, the CT device can also be configured as a spectral CT device (dual-energy CT device, spectral CT device), and the spectral CT device (dual-energy CT device, spectral CT device) is used to scan the patient's abdomen to obtain a three-dimensional dual-energy abdominal image (dual-energy abdominal image / dual-energy spectral abdominal image / dual-energy abdominal CT image / dual-energy spectral abdominal CT image / spectral abdominal image / spectral abdominal CT image).

[0051] In the embodiments of the present disclosure and other possible embodiments, the DR device can also be configured as a dual-energy digital X-ray machine, which is used to photograph the patient's abdomen to obtain a two-dimensional dual-energy abdominal image (dual-energy abdominal image / dual-energy abdominal X-ray image).

[0052] In the embodiments of the present disclosure and other possible embodiments, the medical image is configured as a chest medical image; wherein, the corresponding organ is configured as the lungs, and the lesions are configured as one or more of small airway lesions, emphysema, diseased bronchus, bullae, pulmonary nodules, pulmonary embolism, lung cancer, pneumonia, tuberculosis, and lung abscess.

[0053] In the embodiment of the present disclosure and other possible embodiments, the medical image is configured as an abdominal medical image; the corresponding organs are configured as kidneys and / or ureters and / or bladders, and the lesions are configured as stones.

[0054] In the embodiments of the present disclosure and other possible embodiments, the medical image is configured as a brain medical image; the corresponding organ is configured as the cranium; and the lesion is configured as one or more lesions in diseased cerebral blood vessels, traumatic brain areas, brain tumors, meningitis areas, cerebral hemorrhage areas, and cerebral ischemia areas.

[0055] In the embodiments of the present disclosure and other possible embodiments, before obtaining the medical image and its corresponding lesion mask image, a preset segmentation network, a medical training image corresponding to the preset segmentation network for training, and a multi-mask label fusion image representing organs and lesions corresponding to the medical training image are obtained; the preset segmentation network is trained on the medical training image and its corresponding multi-mask label fusion image to obtain a corresponding organ / lesion segmentation model; based on the organ / lesion segmentation model, the medical image corresponding to the medical training image is segmented for lesions to obtain a lesion mask image.

[0056] For example, an organ mask label image corresponding to at least one of the kidneys, ureters, and bladders in the dual-energy abdominal training image and a stone mask label image in the organ are obtained; before obtaining the organ mask label image corresponding to at least one of the kidneys, ureters, and bladders in the dual-energy abdominal training image and a stone mask label image in the organ, the stone mask label image is determined based on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image, and the organ mask label image is determined based on the water-based image / soft tissue image corresponding to the dual-energy abdominal training image.

[0057] In the embodiments of the present disclosure and other possible embodiments, the method for determining a stone mask label image based on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image includes: performing stone delineation on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image to determine the stone mask label image. Simultaneously, the method for determining an organ mask label image based on the water-based image / soft tissue image corresponding to the dual-energy abdominal training image includes: performing organ delineation on the water-based image / soft tissue image corresponding to the dual-energy abdominal training image to determine the organ mask label image.

[0058] In the embodiments of the present disclosure and other possible embodiments, a multi-mask label fusion image corresponding to the medical image is generated based on the organ mask value corresponding to the organ mask label image and the organ spatial position corresponding to the organ mask value, the lesion mask value corresponding to the lesion mask label image and the second spatial position corresponding to the lesion mask value.

[0059] In the embodiments of the present disclosure and other possible embodiments, the method for generating a multi-mask label fusion image corresponding to the medical image based on the organ mask value corresponding to the organ mask label image and the organ spatial position corresponding to the organ mask value, the lesion mask value corresponding to the lesion mask label image and the lesion spatial position corresponding to the lesion mask value, includes: respectively determining the organ spatial position corresponding to the organ mask value in the organ mask label image and the lesion spatial position corresponding to the lesion mask value in the lesion mask image; if the organ spatial position overlaps with the lesion spatial position, updating the mask value corresponding to the overlapping spatial position to the lesion mask value corresponding to the lesion spatial position; otherwise, retaining the organ mask value corresponding to the organ spatial position to generate a multi-mask label fusion image corresponding to the medical image.

[0060] In the embodiments of the present disclosure and other possible embodiments, before respectively determining the organ spatial position corresponding to the organ mask value in the organ mask label image and the lesion spatial position corresponding to the lesion mask value in the lesion mask image, it also includes: if the number corresponding to the lesion mask image is greater than 1, respectively obtaining the lesion spatial positions corresponding to multiple lesion mask images; judging whether there is spatial position overlap between the lesion spatial positions corresponding to the multiple lesion mask images; if there is spatial position overlap, respectively generating multi-mask label fusion images of each lesion and its corresponding organ corresponding to the spatial position overlap and generating multi-mask label fusion images of lesions and their corresponding organs corresponding to non-overlapping spatial positions; otherwise, generating multi-mask label fusion images corresponding to all lesions and their corresponding organs together.

[0061] In the embodiments of the present disclosure and other possible embodiments, in the embodiments of the present disclosure, the method for constructing the multi-mask label fusion image representing the kidney, bladder, ureter, ureter and stone respectively includes: obtaining an organ mask label image corresponding to at least one organ of the kidney, ureter, and bladder in the dual-energy abdominal training image and a stone mask label image in the organ; generating a multi-mask label fusion image corresponding to the dual-energy abdominal training image based on the organ mask label image and the stone mask label image.

[0062] In the embodiments of the present disclosure and other possible embodiments, the method for generating a multi-mask label fusion image corresponding to the dual-energy abdominal training image based on the organ mask label image and the stone mask label image includes: generating a multi-mask label fusion image corresponding to the dual-energy abdominal training image based on the organ mask value corresponding to the organ mask label image and the organ spatial position corresponding to the organ mask value, the stone mask value corresponding to the stone mask label image and the second spatial position corresponding to the stone mask value.

[0063] In the embodiments of the present disclosure and other possible embodiments, before obtaining the organ mask label image corresponding to at least one organ of the kidney, ureter, and bladder in the dual-energy abdominal training image and the stone mask label image in the organ, the stone mask label image is determined based on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image, and the organ mask label image is determined based on the water-based image / soft tissue image corresponding to the dual-energy abdominal training image.

[0064] In the embodiments of the present disclosure and other possible embodiments, the method for determining a stone mask label image based on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image includes: outlining the stone on the iodine-based image / calcium image corresponding to the dual-energy abdominal training image, and determining the stone mask label image.

[0065] In the embodiments of the present disclosure and other possible embodiments, the method for determining the organ mask label image based on the water-based image / soft tissue image corresponding to the dual-energy abdominal training image includes: outlining the organs in the water-based image / soft tissue image corresponding to the dual-energy abdominal training image to determine the organ mask label image.

[0066] In the embodiments of the present disclosure and other possible embodiments, the method for generating a multi-mask label fusion image corresponding to the dual-energy abdominal training image based on the organ mask value corresponding to the organ mask label image and the organ spatial position corresponding to the organ mask value, the stone mask value corresponding to the stone mask label image and the stone spatial position corresponding to the stone mask value, includes: respectively determining the organ spatial position corresponding to the organ mask value in the organ mask label image and the stone spatial position corresponding to the stone mask value in the stone mask image; if the organ spatial position overlaps with the stone spatial position, updating the mask value corresponding to the overlapping spatial position to the stone mask value corresponding to the stone spatial position; otherwise, retaining the organ mask value corresponding to the organ spatial position, and generating a multi-mask label fusion image corresponding to the dual-energy abdominal training image.

[0067] In the embodiments of the present disclosure and other possible embodiments, before respectively determining the organ spatial position corresponding to the organ mask value in the organ mask label image and the stone spatial position corresponding to the stone mask value in the stone mask image, it also includes: if the number corresponding to the stone mask image is greater than 1, then respectively obtaining the stone spatial positions corresponding to multiple stone mask images; judging whether there is spatial position overlap between the stone spatial positions corresponding to multiple stone mask images; if there is spatial position overlap, then respectively generating multi-mask label fusion images of each stone corresponding to the spatial position overlap and its corresponding organ, and generating multi-mask label fusion images of stones corresponding to non-overlapping spatial positions and their corresponding organs; otherwise, generating multi-mask label fusion images corresponding to all stones and their corresponding organs together.

[0068] In the embodiments of the present disclosure and other possible embodiments, the method further includes: obtaining a preset segmentation network, a medical training image corresponding to the preset segmentation network for training, and a multi-mask label fusion image representing organs and lesions corresponding to the medical training image; wherein the multi-mask label fusion image representing organs and lesions is generated using the multi-mask label image fusion image generation method as described above; training the preset segmentation network on the medical training image and its corresponding multi-mask label fusion image to obtain a corresponding organ / lesion segmentation model; performing lesion segmentation on the medical image corresponding to the medical training image based on the organ / lesion segmentation model to obtain a lesion mask image.

[0069] In the embodiments of the present disclosure and other possible embodiments, in the process of training the preset segmentation network for the medical training image and the corresponding multi-mask label fusion image, the method includes: obtaining the organ loss function corresponding to the organ and the corresponding organ loss weight value, the lesion loss function corresponding to the lesion and the corresponding lesion loss weight value which is less than the organ loss weight value; in the process of training the preset segmentation network, if the organ loss value corresponding to the organ loss function is less than the first set loss value and / or the total loss value corresponding to the organ loss function and the lesion loss function is less than the second set loss value, then in the process of training the preset segmentation network, During each training process of continued training, the lesion loss weight value is increased and regulated according to the set ratio to obtain the lesion loss regulation weight value; based on the lesion loss regulation weight value, the organ loss regulation weight value corresponding to the organ loss weight value is determined; based on the lesion loss regulation weight value and the organ loss regulation weight value, the lesion loss value corresponding to the lesion loss function and the organ loss function and the total loss value corresponding to the lesion loss function are calculated; if the lesion loss value is less than the third set loss value and / or the total loss value corresponding to the organ loss function and the lesion loss function is less than the fourth set loss value, the training of the preset segmentation network is stopped.

[0070] In the embodiments of the present disclosure and other possible embodiments, the method for determining the organ loss control weight value corresponding to the organ loss weight value based on the lesion loss control weight value includes: obtaining a set configuration value corresponding to the sum of the organ loss weight value and the lesion loss weight value; subtracting the lesion loss control weight value from the set configuration value to determine the organ loss control weight value corresponding to the organ loss weight value; wherein the set configuration value corresponding to the sum of the organ loss weight value and the lesion loss weight value is configured to be 1.

[0071] In the embodiments of the present disclosure and other possible embodiments, the organ loss function and the lesion loss function are respectively configured as one or more of the cross entropy loss function, Dice loss function, Focal loss function, Tversky loss function, Tversky loss function, and IoU loss function.

[0072] In the embodiments of the present disclosure and other possible embodiments, the preset segmentation network is configured as one or more segmentation networks such as Unet, ResUnet, Unet++, ResUnet++, nnUnet, SegNet, PSPNet, DeepLab, RefineNet, Medformer or improved segmentation networks thereof.

[0073] In the embodiments of the present disclosure and other possible embodiments, a preset segmentation network, a dual-energy abdominal training image corresponding to the preset segmentation network, and a multi-mask label fusion image corresponding to the dual-energy abdominal training image representing the kidneys, ureters, bladder, and stones are obtained; the preset segmentation network is trained using the dual-energy abdominal training image and its corresponding multi-mask label fusion image to obtain a corresponding stone segmentation model; based on the stone segmentation model, the dual-energy abdominal image is segmented to obtain stones in the kidneys, ureters, and bladder to obtain a stone mask image. This solves the current technical problem of insufficient utilization of the anatomical structural information of the kidneys, ureters, bladder, and stones in their spatial locations, and thus lacks the synergistic effect of the anatomical structural information of the kidneys, ureters, bladder, and stones.

[0074] In the embodiments of the present disclosure and other possible embodiments, the method of using the dual-energy abdominal training image and the corresponding multi-mask label fusion image to train the preset segmentation network to obtain the corresponding stone segmentation model includes: obtaining a set loss value corresponding to the stone segmentation model; in the process of training the preset segmentation network using the dual-energy abdominal training image and the corresponding multi-mask label fusion image, if the loss value corresponding to the stone segmentation model is less than or equal to the set loss value, stopping training the preset segmentation network to obtain the corresponding stone segmentation model.

[0075] In the embodiments of the present disclosure and other possible embodiments, the process of training the preset segmentation network using the dual-energy abdominal training image and the corresponding multi-mask label fusion image includes: obtaining the organ loss function corresponding to the organ and the corresponding organ loss weight value, the stone loss function corresponding to the stone and the corresponding stone loss weight value that is less than the organ loss weight value; in the process of training the preset segmentation network, if the organ loss value corresponding to the organ loss function is less than the first set loss value and / or the total loss value corresponding to the organ loss function and the stone loss function is less than the second set loss value, then in the process of training the preset segmentation network, During each training process of the continued training of the network, the stone loss weight value is increased and regulated according to the set ratio to obtain the stone loss regulation weight value; based on the stone loss regulation weight value, the organ loss regulation weight value corresponding to the organ loss weight value is determined; based on the stone loss regulation weight value and the organ loss regulation weight value, the stone loss value corresponding to the stone loss function and the organ loss function and the total loss value corresponding to the stone loss function are calculated; if the stone loss value is less than the third set loss value and / or the total loss value corresponding to the organ loss function and the stone loss function is less than the fourth set loss value, the training of the preset segmentation network is stopped.

[0076] In the embodiments of the present disclosure and other possible embodiments, the method for determining the organ loss control weight value corresponding to the organ loss weight value based on the stone loss control weight value includes: obtaining a set configuration value corresponding to the sum of the organ loss weight value and the stone loss weight value; subtracting the stone loss control weight value from the set configuration value to determine the organ loss control weight value corresponding to the organ loss weight value; wherein the set configuration value corresponding to the sum of the organ loss weight value and the stone loss weight value is configured to be 1.

[0077] In the embodiments of the present disclosure and other possible embodiments, the organ loss function and the stone loss function are respectively configured as one or more of the cross entropy loss function, Dice loss function, Focal loss function, Tversky loss function, Tversky loss function, and IoU loss function.

[0078] In the embodiment of the present disclosure and other possible embodiments, based on the stone segmentation model, the dual-energy abdominal image is segmented for stones in the kidney, ureter and bladder to obtain a stone mask image.

[0079] Step S102: performing a multiplication operation on the medical image and the corresponding lesion mask image to obtain a lesion grayscale value image.

[0080] In the embodiment of the present disclosure and other possible embodiments, if the dimension corresponding to the medical image is two-dimensional, a multiplication operation is directly performed on the medical image and its corresponding lesion mask image to obtain a two-dimensional lesion grayscale value image.

[0081] In the embodiments of the present disclosure and other possible embodiments, if the dimension corresponding to the medical image is three-dimensional, a multiplication operation is performed on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice to obtain a three-dimensional lesion grayscale value image.

[0082] Step S103: Based on the lesion grayscale value image, the mask values corresponding to each of the multiple lesion types in the lesion mask image, and the lesion spatial positions, the grayscale value of the lesion grayscale value image is restored to obtain a lesion image with original grayscale values.

[0083] In an embodiment of the present disclosure, the method of performing grayscale value restoration on the lesion grayscale value image based on the mask values and lesion spatial positions corresponding to each lesion type of the multiple lesion types in the lesion mask image to obtain a lesion image with original grayscale values includes: extracting the mask values and lesion spatial positions corresponding to each lesion type in the lesion mask image; locating the spatial position of the lesion type on the lesion grayscale value image based on the lesion spatial positions corresponding to each lesion type to obtain a spatial position positioning image corresponding to each lesion type; and dividing the spatial position positioning image corresponding to each lesion type by the corresponding mask value to obtain a lesion image with original grayscale values.

[0084] If the dimension corresponding to the medical image is two-dimensional, then based on the medical image and its corresponding lesion mask image, a lesion image with original grayscale values is extracted from the medical image; the average grayscale value corresponding to each lesion type in the lesion mask image is calculated, and the lesion grayscale value corresponding to each lesion type is determined; if the dimension corresponding to the medical image is three-dimensional, then based on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice, a three-dimensional lesion image with original grayscale values is extracted from the three-dimensional medical image; the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image is calculated, and the lesion grayscale value corresponding to each lesion type is determined.

[0085] The disclosed embodiment also proposes a medical image analysis method, comprising: obtaining a lesion image with original grayscale values using the above-mentioned restoration method for medical images of multiple lesion types; if the dimension corresponding to the medical image is two-dimensional, extracting a lesion image with original grayscale values from the medical image based on the medical image and its corresponding lesion mask image; calculating the average grayscale value corresponding to each lesion type in the lesion mask image, and determining the lesion grayscale value corresponding to each lesion type; if the dimension corresponding to the medical image is three-dimensional, extracting a three-dimensional lesion image with original grayscale values from the three-dimensional medical image based on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice; calculating the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image, and determining the lesion grayscale value corresponding to each lesion type.

[0086] In the embodiments of the present disclosure and other possible embodiments, the medical image lesion segmentation method as described above is used to perform lesion segmentation on the medical image to obtain a lesion mask image; based on the medical image and / or the lesion mask image, one or more of the corresponding lesion position, lesion shape, lesion grayscale value, the corresponding major and minor diameters of the lesion, the lesion area or the lesion volume is extracted.

[0087] In the embodiments of the present disclosure and other possible embodiments, the method for extracting one or more of the following information based on the lesion mask image: corresponding lesion location, lesion shape, lesion grayscale value, corresponding major and minor diameters of the lesion, and lesion area or volume, includes: obtaining different configuration colors or different configuration mask values corresponding to different types of lesions and the dimensions corresponding to the medical image; determining the lesion type on the lesion mask image based on the different configuration colors or different configuration mask values corresponding to different types of lesions; and extracting one or more of the following information corresponding to the lesion location, lesion shape, lesion grayscale value, corresponding major and minor diameters of the lesion, and lesion area or volume, respectively.

[0088] In the embodiments of the present disclosure and other possible embodiments, if the dimension corresponding to the medical image is two-dimensional, edge detection is performed on the lesions corresponding to each lesion type on the lesion mask image to obtain a two-dimensional lesion edge mask image; based on the two-dimensional lesion edge mask image, one or more of the following information is determined, including the lesion shape corresponding to each lesion type, the major and minor diameters corresponding to the lesion, and the lesion area.

[0089] In an embodiment of the present disclosure, if the dimension corresponding to the medical image is three-dimensional, the lesion area corresponding to each lesion type in each two-dimensional slice lesion mask image in the three-dimensional lesion mask image is calculated respectively; multiple lesion areas are obtained; the major axis and minor axis corresponding to the maximum lesion area among the multiple lesion areas corresponding to each lesion type are calculated respectively, to obtain the major axis and minor axis corresponding to each lesion type; edge detection is performed on the lesions corresponding to each lesion type on the lesion mask image respectively, to obtain a three-dimensional lesion edge mask line; the three-dimensional lesion edge mask line corresponding to each lesion type is fitted respectively, to obtain the lesion shape corresponding to each lesion type; the mask map corresponding to each lesion type in the lesion mask image is three-dimensionally reconstructed respectively, to obtain the lesion volume corresponding to each lesion type.

[0090] In the embodiments of the present disclosure and other possible embodiments, if the dimension corresponding to the medical image is three-dimensional, the organ segmentation of the medical image is performed using an organ / lesion segmentation model to obtain a three-dimensional organ mask image; based on the three-dimensional organ mask image, the organ mask in the three-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the three-dimensional organ mask image, the lesion position corresponding to each lesion type is determined.

[0091] In the embodiments of the present disclosure and other possible embodiments, the method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the three-dimensional organ mask image includes: extracting multiple two-dimensional slice lesion mask images corresponding to each lesion type from the three-dimensional lesion mask image; if the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the left organ mask spatial positions of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the right organ mask spatial positions of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

[0092] In the embodiments of the present disclosure and other possible embodiments, if the dimension corresponding to the medical image is two-dimensional, the organ segmentation model is used to perform organ segmentation on the medical image to obtain a two-dimensional organ mask image; based on the two-dimensional organ mask image, the organ mask in the two-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the lesion mask in the lesion mask image and the spatial position relationship between the side organ mask and the right organ mask in the two-dimensional organ mask image, the lesion position corresponding to each lesion type is determined.

[0093] In the embodiments of the present disclosure and other possible embodiments, the method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the organ mask image includes: if the lesion mask spatial position corresponding to the lesion mask image is within the left organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial position corresponding to the lesion mask image is within the right organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

[0094] In the embodiments of the present disclosure and other possible embodiments, the medical image is configured as a two-dimensional or three-dimensional chest medical image; wherein the two-dimensional chest medical image can be configured as a chest X-ray image / dual-energy chest X-ray image, and the three-dimensional chest medical image can be configured as a chest CT image / dual-energy chest CT image. Based on the above two-dimensional or three-dimensional chest medical image, the corresponding organs are configured as lungs, including: a left lung and a right lung. In this case, the left lung and the right lung are the left organ and the right organ, respectively; furthermore, the left organ and the right organ correspond to the left organ mask and the right organ mask, respectively. At the same time, each of the lesion types may be one or more lesions selected from small airway lesions, emphysema, diseased bronchus, bullae, pulmonary nodules, pulmonary embolism, lung cancer, pneumonia, tuberculosis, and lung abscess; furthermore, the lesion mask image corresponding to each of the lesion types may be one or more lesion mask images selected from small airway lesion mask image, emphysema mask image, diseased bronchus mask image, bullae mask image, pulmonary nodule mask image, pulmonary embolism mask image, lung cancer mask image, pneumonia mask image, tuberculosis mask image, and lung abscess mask image.

[0095] In the embodiments of the present disclosure and other possible embodiments, the medical image is configured as a two-dimensional or three-dimensional abdominal medical image; wherein the two-dimensional abdominal medical image can be configured as an abdominal X-ray image / dual-energy abdominal X-ray image, and the three-dimensional chest medical image can be configured as an abdominal CT image / dual-energy abdominal CT image. Based on the above two-dimensional or three-dimensional abdominal medical image, the corresponding organs are configured as kidneys and / or ureters and / or bladders, including: the left kidney and / or left ureter and / or left bladder, the right kidney and / or right ureter and / or right bladder. In this case, the left kidney and / or left ureter and / or left bladder, the right kidney and / or right ureter and / or right bladder are the left and right organs, respectively; furthermore, the left and right organs correspond to the left and right organ masks, respectively. At the same time, the lesions are configured as different types of stones, including: one or more of: uric acid stones, calcium oxalate monohydrate stones, calcium oxalate dihydrate stones, carbonate apatite stones, carbonic acid calcium phosphate stones, and struvite stones. Furthermore, the lesion mask image corresponding to the lesion type may be one or several lesion mask images among different types of stone mask images, including: one or several lesion mask images among uric acid stone lesion mask images, calcium oxalate monohydrate stone lesion mask images, calcium oxalate dihydrate stone lesion mask images, carbonated apatite stone lesion mask images, calcium carbonate phosphate stone lesion mask images, and struvite stone lesion mask images.

[0096] In the embodiments of the present disclosure and other possible embodiments, the medical image is configured as a three-dimensional brain medical image; wherein, the two-dimensional abdominal medical image can be configured as a brain X-ray image / dual-energy brain X-ray image, and the three-dimensional brain medical image can be configured as a brain CT image / dual-energy brain CT image. Based on the above two-dimensional or three-dimensional brain medical image, the corresponding organs are configured as the cranium, including: the left brain and the right brain. In this case, the left brain and the right brain are the left organ and the right organ, respectively; furthermore, the left organ and the right organ correspond to the left organ mask and the right organ mask, respectively. At the same time, the lesion is configured as one or more lesions among diseased cerebral blood vessels, traumatic brain areas, brain tumors, meningitis areas, cerebral hemorrhage areas, and cerebral ischemia areas. Furthermore, the lesion mask image corresponding to the lesion type may be one or more of the lesion mask images selected from the group consisting of a diseased cerebral blood vessel lesion mask image, a traumatic brain region lesion mask image, a brain tumor lesion mask image, a meningitis region lesion mask image, a cerebral hemorrhage region lesion mask image, and a cerebral ischemia region lesion mask image.

[0097] In the embodiments of the present disclosure and other possible embodiments, corresponding stone information is extracted based on the dual-energy abdominal image and / or the stone mask image; wherein the stone information is configured as one or more of the following information: stone position, stone shape, stone calcification value, the corresponding long and short diameters of the stone, stone area or stone volume.

[0098] In the embodiment of the present disclosure and other possible embodiments, the stone position corresponding to the stone information is determined based on the kidney and / or ureter and / or bladder mask images corresponding to the dual-energy abdominal image and the stone mask image.

[0099] In the embodiments of the present disclosure and other possible embodiments, the method for determining the stone position corresponding to the stone information based on the kidney and / or ureter and / or bladder mask image and the stone mask image corresponding to the dual-energy abdominal image includes: performing kidney and / or ureter and / or bladder segmentation on the dual-energy abdominal image to obtain a kidney and / or ureter and / or bladder mask image; configuring the kidney and / or ureter and / or bladder mask in the kidney and / or ureter and / or bladder mask image as a left kidney and / or ureter and / or bladder mask and a right kidney and / or ureter and / or bladder mask according to the kidney and / or ureter and / or bladder mask image; and determining the stone position corresponding to each stone type based on the stone mask in the stone mask image and the spatial position relationship between the left kidney and / or ureter and / or bladder mask and the right kidney and / or ureter and / or bladder mask in the kidney and / or ureter and / or bladder mask image.

[0100] In the embodiments of the present disclosure and other possible embodiments, the method for determining the stone position corresponding to each stone type based on the stone mask in the stone mask image and the spatial position relationship between the lateral kidney and / or ureter and / or bladder mask and the right kidney and / or ureter and / or bladder mask in the kidney and / or ureter and / or bladder mask image includes: if the dual-energy abdominal image is configured as a three-dimensional dual-energy abdominal CT image; extracting multiple two-dimensional slice stone mask images corresponding to each stone type from the stone mask image; if the spatial position of the stone mask corresponding to the multiple two-dimensional slice stone mask images is in the kidney and / or ureter and / or bladder mask image is within the left kidney and / or ureter and / or bladder mask spatial position of the two-dimensional slice stone mask image of the slice corresponding to the kidney and / or ureter and / or bladder mask image, then the stone corresponding to the stone mask spatial position is configured as a left kidney and / or ureter and / or bladder stone; if the stone mask spatial positions corresponding to the multiple two-dimensional slice stone mask images are within the right kidney and / or ureter and / or bladder mask spatial position of the two-dimensional slice stone mask image of the slice corresponding to the kidney and / or ureter and / or bladder mask image, then the stone corresponding to the stone mask spatial position is configured as a right kidney and / or ureter and / or bladder stone.

[0101] In the embodiments of the present disclosure and other possible embodiments, the method for determining the stone position corresponding to each stone type based on the stone mask in the stone mask image and the spatial position relationship between the lateral kidney and / or ureter and / or bladder mask and the right kidney and / or ureter and / or bladder mask in the kidney and / or ureter and / or bladder mask image includes: if the dual-energy abdominal image is configured as a two-dimensional dual-energy abdominal X-ray image; performing organ segmentation on the dual-energy abdominal image to obtain a two-dimensional organ mask image; according to the two-dimensional organ mask image, configuring the organ mask in the two-dimensional organ mask image as a left organ mask and a right organ mask; and determining the stone position corresponding to each stone type based on the stone mask in the stone mask image and the spatial position relationship between the lateral organ mask and the right organ mask in the two-dimensional organ mask image.

[0102] In the embodiments of the present disclosure and other possible embodiments, the method for determining the stone position corresponding to each stone type based on the spatial position relationship between the stone mask in the stone mask image and the side organ mask and the right organ mask in the organ mask image includes: if the stone mask spatial position corresponding to the stone mask image is within the left organ mask spatial position corresponding to the organ mask image, then the stone corresponding to the stone mask spatial position is configured as a left organ stone; if the stone mask spatial position corresponding to the stone mask image is within the right organ mask spatial position corresponding to the organ mask image, then the stone corresponding to the stone mask spatial position is configured as a right organ stone.

[0103] In the embodiment of the present disclosure and other possible embodiments, the limescale value corresponding to the stone information is determined based on the dual-energy abdominal image and the corresponding stone mask image.

[0104] In the embodiments of the present disclosure and other possible embodiments, the method for determining the grayscale value corresponding to the stone information based on the dual-energy abdominal image and its corresponding stone mask image includes: if the dual-energy abdominal image is configured as a three-dimensional dual-energy abdominal CT image; based on each two-dimensional slice dual-energy abdominal image and the two-dimensional slice stone mask image of the stone mask image of its corresponding slice in the dual-energy abdominal image, extracting the stone image with the original grayscale value from the dual-energy abdominal image; calculating the average grayscale value corresponding to each stone type in the stone mask image, and determining the stone grayscale value corresponding to each stone type.

[0105] In the embodiments of the present disclosure and other possible embodiments, the method for extracting a stone image with original grayscale values from the dual-energy abdominal image based on the two-dimensional slice stone mask image of each two-dimensional slice dual-energy abdominal image and the stone mask image of its corresponding slice in the dual-energy abdominal image includes: performing a multiplication operation on the two-dimensional slice stone mask image of each two-dimensional slice dual-energy abdominal image and the stone mask image of its corresponding slice in the dual-energy abdominal image to obtain a stone grayscale value image; performing grayscale value restoration on the stone grayscale value image based on the mask values corresponding to each stone type and the spatial position of the stone in the two-dimensional slice stone mask image of the stone grayscale value image and the stone mask image of its corresponding slice to obtain a stone image with original grayscale values.

[0106] In the embodiments of the present disclosure and other possible embodiments, the method of performing grayscale value restoration on the stone grayscale value image based on the mask values corresponding to each stone type and the stone spatial position in the two-dimensional slice stone mask image of the stone mask image of its corresponding slice to obtain a stone image with original grayscale values includes: respectively extracting the mask values and stone spatial positions corresponding to each stone type in the two-dimensional slice stone mask image of the stone mask image; based on the stone spatial positions corresponding to each stone type, respectively performing stone type spatial position positioning on the stone grayscale value image to obtain spatial position positioning images corresponding to each stone type in the two-dimensional slice stone grayscale value image in the stone grayscale value image; dividing the spatial position positioning images corresponding to each stone type in the two-dimensional slice stone grayscale value image in the stone grayscale value image by the mask values corresponding to the stone spatial positions to obtain a stone image with original grayscale values.

[0107] In the embodiments of the present disclosure and other possible embodiments, if the dual-energy abdominal image is configured as a two-dimensional dual-energy abdominal X-ray image, a stone image with original grayscale values is extracted from the dual-energy abdominal image based on the dual-energy abdominal image and its corresponding stone mask image; the average grayscale value corresponding to each stone type in the stone mask image is calculated to determine the stone grayscale value corresponding to each stone type.

[0108] In the embodiment of the present disclosure and other possible embodiments, based on the stone mask image, one or more of the stone shape corresponding to the stone information, the major diameter and minor diameter corresponding to the stone, and the stone volume are determined.

[0109] In the embodiments of the present disclosure and other possible embodiments, the method for determining one or more of the stone shape, the major diameter and minor diameter corresponding to the stone, and the stone volume corresponding to the stone information based on the stone mask image includes: respectively calculating the stone area corresponding to each stone type in each two-dimensional slice stone mask image in the stone mask image corresponding to the dual-energy abdominal image; obtaining multiple stone areas; respectively calculating the major diameter and minor diameter corresponding to the maximum stone area among the multiple stone areas corresponding to each stone type, to obtain the major diameter and minor diameter corresponding to each stone type; and / or, respectively performing edge detection on the stones corresponding to each stone type on the stone mask image corresponding to the dual-energy abdominal image, to obtain stone edge mask lines; respectively fitting the stone edge mask lines corresponding to each stone type, to obtain the stone shape corresponding to each stone type; and / or, respectively reconstructing the mask images corresponding to each stone type in the stone mask image corresponding to the dual-energy abdominal image, to obtain the stone volume corresponding to each stone type.

[0110] In the embodiments of the present disclosure and other possible embodiments, the abdominal medical image processing method further includes: if the dual-energy abdominal image is configured as a three-dimensional dual-energy abdominal CT image, identifying the stone type corresponding to the stone information based on the urinary stone atomic number image corresponding to the dual-energy abdominal image. This addresses the technical problem of difficulty in distinguishing or identifying mixed urinary stones of different components, thereby failing to meet the clinical diagnosis requirements for mixed-component stones, which are more common than single-component stones.

[0111] In the embodiments of the present disclosure and other possible embodiments, a spectral CT device (dual-energy CT device, spectral CT device) is used to scan the patient's abdomen to obtain a three-dimensional dual-energy abdominal image (dual-energy abdominal image / dual-energy spectral abdominal image / dual-energy abdominal CT image / dual-energy spectral abdominal CT image / spectral abdominal image / spectral abdominal CT image).

[0112] In the embodiments of the present disclosure and other possible embodiments, a spectral CT device (dual-energy CT device, spectral CT device), namely a dual-source CT device with fast tube voltage switching, employs instantaneous kVp switching technology to switch between high and low energy within an extremely short time (<0.25ms), achieving the three similarities of dual energy (simultaneous, same-direction, same-source). This overcomes the shortcomings of dual-tube dual-source CT, avoiding subtle angular discrepancies between the two tube scanning planes, improving data acquisition accuracy, and providing superior CT scan images.

[0113] In the embodiments disclosed herein and other possible embodiments, spectral CT scanning and 3D reconstruction can not only clearly display the density and morphology of the renal parenchyma, the orientation and contours of the renal pelvis and calyces, the course and lumen of the ureters, and the bladder wall and lumen, but can also show the morphological location of kidney stones and the renal pelvis and calyces, and further measure the volumes of stones and the renal pelvis and calyces, with imaging quality far superior to that of conventional CT 3D multi-planar reconstruction. In particular, the effective atomic number (Zeff value) of spectral CT can be used to analyze the compositional characteristics of kidney stones, revealing numerical differences between different regions, and based on these differences, clarifying the specific spatial distribution of different components in mixed stones.

[0114] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the stone information based on the urinary stone atomic number image corresponding to the dual-energy abdominal image includes: if the dual-energy abdominal image is configured as a three-dimensional dual-energy abdominal CT image; obtaining the urinary stone atomic number image corresponding to the dual-energy abdominal image; and identifying the stone type corresponding to the dual-energy abdominal image based on the urinary stone atomic number image.

[0115] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the dual-energy abdominal image based on the urinary stone atomic number image includes: extracting multiple component percentages of peak values corresponding to multiple bar graphs corresponding to the effective atomic numbers in the urinary stone atomic number image; identifying the stone type corresponding to the dual-energy abdominal image based on multiple component percentages corresponding to the multiple bar graphs and multiple set stone type component percentage intervals.

[0116] In the embodiments of the present disclosure and other possible embodiments, the method for extracting multiple component percentages of peak values corresponding to multiple bar graphs corresponding to effective atomic numbers in the atomic number image of urinary stones includes: obtaining a set bar graph color and / or a set bar graph width; extracting multiple bar graphs corresponding to effective atomic numbers that meet the set bar graph color and / or the set bar graph width in the atomic number image of urinary stones based on the set bar graph color and / or the set bar graph width; determining multiple component percentages of peak values corresponding to multiple bar graphs based on the multiple bar graphs that meet the set bar graph color and / or the set bar graph width in the atomic number image of urinary stones.

[0117] In the embodiment of the present disclosure and other possible embodiments, those skilled in the art may configure the bar graph color configuration and / or the bar graph width according to actual needs. For example, the bar graph color configuration is set to yellow.

[0118] In the embodiments of the present disclosure and other possible embodiments, before extracting the multiple bar graphs corresponding to the effective atomic numbers in the urinary tract stone atomic number image or extracting the multiple bar graphs corresponding to the effective atomic numbers that meet the set bar graph color and / or set bar graph width in the urinary tract stone atomic number image, the urinary tract stone atomic number image is corrected to obtain a corrected urinary tract stone atomic number image; and multiple component percentages of the peak values corresponding to the multiple bar graphs corresponding to the effective atomic numbers in the corrected urinary tract stone atomic number image are extracted or multiple component percentages of the peak values corresponding to the multiple bar graphs corresponding to the effective atomic numbers that meet the set bar graph color and / or set bar graph width in the corrected urinary tract stone atomic number image are extracted.

[0119] In the embodiments of the present disclosure and other possible embodiments, the method of correcting the urinary tract stone atomic number image to obtain the corrected urinary tract stone atomic number image includes: performing at least one correction processing operation such as angle adjustment, contrast enhancement, and scaling on the urinary tract stone atomic number image to obtain the corrected urinary tract stone atomic number image, so as to facilitate the subsequent extraction of multiple bar graphs corresponding to the effective atomic numbers.

[0120] In the embodiments of the present disclosure and other possible embodiments, optical character recognition technology is used to extract multiple component percentages of peak values corresponding to multiple bar graphs corresponding to effective atomic numbers in the urinary stone atomic number image; or, optical character recognition technology is used to extract multiple component percentages of peak values corresponding to multiple bar graphs corresponding to effective atomic numbers that meet the set bar graph color and / or set bar graph width in the corrected urinary stone atomic number image.

[0121] In the embodiments of the present disclosure and other possible embodiments, optical character recognition (OCR) technology is a computer vision technology that uses image processing and machine learning algorithms to identify and extract text content in images and convert it into a machine-readable and editable text format.

[0122] In an embodiment of the present disclosure, the method for identifying the stone type corresponding to the dual-energy abdominal image based on the multiple component percentages corresponding to the multiple bar graphs and the multiple set stone type component percentage intervals includes: obtaining the set component percentage corresponding to the interval that is less than the minimum value of the set stone type component percentage interval; if the component percentage corresponding to the multiple bar graphs is less than the set component percentage, deleting the component percentage corresponding to the multiple component percentages corresponding to the multiple bar graphs that is less than the set component percentage; otherwise, retaining the component percentage corresponding to the set component percentage that is greater than or equal to the set component percentage; identifying the stone type corresponding to the dual-energy abdominal image based on the retained multiple component percentages and the multiple set stone type component percentage intervals.

[0123] In the embodiments of the present disclosure and other possible embodiments, those skilled in the art may configure the set component percentage corresponding to the interval less than the minimum value in the set stone type component percentage interval according to actual needs. For example, the set component percentage corresponding to the interval less than the minimum value in the set stone type component percentage interval is configured to be 5%.

[0124] In the embodiments of the present disclosure and other possible embodiments, the implementation of OCR requires calling the API of the GOT_OCR2 tool and setting its parameters such as language and text box positioning; atomic number peak recognition and morphological calculation require using Python's NumPy, SciPy or OpenCV library to execute the extraction of multiple component percentages of peaks corresponding to multiple bar graphs corresponding to the effective atomic numbers in the urinary stone atomic number image or multiple component percentages of peaks corresponding to multiple bar graphs corresponding to the effective atomic numbers in the urinary stone atomic number image that meet the set bar graph color and / or set bar graph width.

[0125] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the dual-energy abdominal image based on the multiple component percentages corresponding to the multiple bar graphs and the multiple set stone type component percentage intervals includes: obtaining the set component percentage corresponding to the interval that is less than the minimum value of the interval in the set stone type component percentage interval; if the component percentage corresponding to the multiple bar graphs is less than the set component percentage, deleting the component percentage corresponding to the multiple component percentages corresponding to the multiple bar graphs that is less than the set component percentage; otherwise, retaining the component percentage corresponding to the set component percentage that is greater than or equal to the set component percentage; identifying the stone type corresponding to the dual-energy abdominal image based on the retained multiple component percentages and the multiple set stone type component percentage intervals.

[0126] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the dual-energy abdominal image based on the multiple component percentages corresponding to the multiple bar graphs and the multiple set stone type component percentage intervals also includes: if there is an overlapping interval in the set stone type component percentage intervals corresponding to at least two of the multiple component percentages corresponding to the multiple bar graphs, obtaining an iodine-based image corresponding to the dual-energy abdominal image; and identifying the stone type corresponding to the overlapping interval based on the iodine-based image corresponding to the dual-energy abdominal image.

[0127] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the overlapping interval based on the iodine-based image corresponding to the dual-energy abdominal image includes: determining the stone grayscale value corresponding to the iodine-based image corresponding to the dual-energy abdominal image; respectively calculating the average grayscale value and uniformity corresponding to the stone grayscale value; if the average grayscale value is greater than the set average grayscale value and the uniformity is greater than the set uniformity, configuring the stone type as calcium oxalate dihydrate stone; otherwise, configuring the stone type as struvite stone or carbonate apatite stone.

[0128] In the embodiments of the present disclosure and other possible embodiments, the multiple set stone type composition percentage intervals include: one or more of: uric acid stone composition percentage intervals, calcium oxalate monohydrate stone composition percentage intervals, calcium oxalate dihydrate stone composition percentage intervals, carbonate apatite stone composition percentage intervals, calcium carbonate phosphate stone composition percentage intervals, and struvite stone composition percentage intervals.

[0129] In the embodiments of the present disclosure and other possible embodiments, those skilled in the art may configure the set average grayscale value and the set uniformity according to actual needs.

[0130] In the embodiment of the present disclosure and other possible embodiments, uniformity = (maximum value - minimum value) / (2 * average value) × 100%, where the maximum value represents the maximum value corresponding to the grayscale value of the cementation, the minimum value represents the minimum value corresponding to the grayscale value of the cementation, and the average value represents the average grayscale value corresponding to the grayscale value of the cementation.

[0131] In the embodiments of the present disclosure and other possible embodiments, the percentage interval of the uric acid stone component is configured to be 6.5-10.5, the percentage interval of the calcium oxalate monohydrate stone component is configured to be 13.3-14.0, the percentage interval of the calcium oxalate dihydrate stone component is configured to be 12.0-13.3, the percentage interval of the carbonate apatite stone component is configured to be 14.0-15.0, the percentage interval of the calcium carbonate phosphate stone component is configured to be greater than 12.5, and the percentage interval of the struvite stone component is configured to be less than 12.5.

[0132] In the embodiments of the present disclosure and other possible embodiments, the Zeff peak value (component percentage interval) of uric acid stones is between 6.5-10.5; the Zeff peak value of calcium oxalate monohydrate stones is between 13.3-14.0; the Zeff peak value of calcium oxalate dihydrate stones is between 12.0-13.3; the Zeff peak value of carbonate apatite stones is between 14.0-15.0; the Zeff peak value of calcium carbonate phosphate stones is greater than 12.5 and all have CT value images with uneven density; the Zeff peak value of struvite stones is less than 12.5 and all have CT value images with uneven density.

[0133] In the embodiments of the present disclosure and other possible embodiments, the method for identifying the stone type corresponding to the dual-energy abdominal image based on the multiple component percentages corresponding to the multiple bar graphs and the multiple set stone type component percentage intervals includes: constructing a stone component comparison table according to the multiple set stone type component percentage intervals; based on the stone component comparison table, performing a table lookup and comparison operation on the multiple component percentages corresponding to the multiple bar graphs to identify the stone type corresponding to the dual-energy abdominal image.

[0134] The method for restoring and analyzing medical images of multiple lesion types may be executed by a device / system for restoring and analyzing medical images of multiple lesion types. For example, the method for restoring and analyzing medical images of multiple lesion types may be executed by a terminal device, a server, or other processing device, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, an in-vehicle device, a wearable device, etc. In some possible implementations, the method for restoring and analyzing medical images of multiple lesion types may be implemented by a processor calling computer-readable instructions stored in a memory.

[0135] Those skilled in the art will understand that in the above-mentioned method for restoring and analyzing medical images of multiple lesion types in the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0136] According to one aspect of an embodiment of the present disclosure, a device / system for restoring a medical image of multiple lesion types is provided, comprising: an acquisition unit for acquiring a medical image and a corresponding lesion mask image; wherein the lesion mask image is configured with different mask values corresponding to multiple lesion types; a multiplication unit for performing a multiplication operation on the medical image and the corresponding lesion mask image to obtain a lesion grayscale value image; and a first restoration unit for performing grayscale value restoration on the lesion grayscale value image based on the lesion grayscale value image, the mask values corresponding to each of the multiple lesion types in the lesion mask image, and the spatial location of the lesions, to obtain a lesion image with original grayscale values. This solves the technical problem of being unable to obtain a lesion image corresponding to the original grayscale values consistent with the medical image, resulting in an inability to perform one-step lesion grayscale value measurement based on the lesion image, hindering further analysis of the lesion grayscale values based on the lesion image.

[0137] According to one aspect of an embodiment of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: an electronic device configured with a processor and a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned method for restoring medical images of multiple lesion types.

[0138] According to one aspect of an embodiment of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned method for restoring medical images of multiple lesion types.

[0139] According to one aspect of an embodiment of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the above-mentioned method for restoring medical images of multiple lesion types.

[0140] According to one aspect of an embodiment of the present disclosure, a device / system for restoring medical images of multiple lesion types is provided, including: a computer program product, including a computer program / instruction, which implements the above-mentioned method for restoring medical images of multiple lesion types when executed by a processor.

[0141] According to one aspect of an embodiment of the present disclosure, a medical image analysis device / system is provided, comprising: a second restoration unit, configured to obtain a lesion image with original grayscale values using the above-mentioned method for restoring medical images of multiple lesion types or the above-mentioned system for restoring medical images of multiple lesion types; a determination unit, configured to extract, if the dimension corresponding to the medical image is two-dimensional, a lesion image with original grayscale values from the medical image based on the medical image and its corresponding lesion mask image; calculate the average grayscale value corresponding to each lesion type in the lesion mask image, and determine the lesion grayscale value corresponding to each lesion type; and / or, if the dimension corresponding to the medical image is three-dimensional, extract a three-dimensional lesion image with original grayscale values from the three-dimensional medical image based on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice; calculate the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image, and determine the lesion grayscale value corresponding to each lesion type.

[0142] According to one aspect of an embodiment of the present disclosure, a medical image analysis device / system is provided, comprising: an electronic device configured with a processor and a memory for storing processor-executable instructions; wherein the processor is configured to call instructions stored in the memory to execute the above-mentioned medical image analysis method.

[0143] According to one aspect of an embodiment of the present disclosure, a medical image analysis device / system is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned medical image analysis method.

[0144] According to one aspect of an embodiment of the present disclosure, a medical image analysis device / system is provided, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions implement the above-mentioned medical image analysis method when executed by a processor.

[0145] According to one aspect of an embodiment of the present disclosure, a medical image analysis device / system is provided, including: a computer program product, including a computer program / instruction, which implements the above-mentioned medical image analysis method when executed by a processor.

[0146] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above embodiment of the method for restoring and analyzing medical images of multiple lesion types. Its specific implementation can refer to the description of the above embodiment of the method for restoring medical images of multiple lesion types. For the sake of brevity, it will not be repeated here.

[0147] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the above-mentioned method for restoring and analyzing medical images of multiple lesion types. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0148] The present disclosure also provides an electronic device comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above-mentioned method for restoring and analyzing medical images of multiple lesions. The electronic device can be provided as a terminal, server, or other device.

[0149] Figure 2 8 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

[0150] Reference Figure 2 , the electronic device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0151] The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 802 may include one or more modules to facilitate interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

[0152] The memory 804 is configured to store various types of data to support operations on the electronic device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0153] The power supply component 806 provides power to the various components of the electronic device 800. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 800.

[0154] The multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0155] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), which is configured to receive external audio signals when the electronic device 800 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0156] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0157] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the electronic device 800. For example, the sensor assembly 814 can detect the open / closed state of the electronic device 800, the relative positioning of components, such as the display and keypad of the electronic device 800. The sensor assembly 814 can also detect changes in the position of the electronic device 800 or a component of the electronic device 800, the presence or absence of user contact with the electronic device 800, the orientation or acceleration / deceleration of the electronic device 800, and temperature changes of the electronic device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0158] The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0159] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0160] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions. The computer program instructions can be executed by the processor 820 of the electronic device 800 to perform the above method.

[0161] Figure 3 1 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a server. Figure 3 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute the instructions to perform the above-described method.

[0162] The electronic device 1900 may further include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0163] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by the processing component 1922 of the electronic device 1900 to perform the above method.

[0164] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0165] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0166] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0167] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0168] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0169] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0170] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0171] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the prescribed logical function. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the prescribed function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0172] While various embodiments of the present disclosure have been described above, the foregoing description is intended to be illustrative, non-exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for restoring medical images of multiple lesion types, characterized in that: include: Acquire a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured with different mask values corresponding to multiple lesion types; Performing a multiplication operation on the medical image and the corresponding lesion mask image to obtain a lesion grayscale value image; Based on the lesion grayscale value image, the mask values corresponding to each of the multiple lesion types in the lesion mask image, and the lesion spatial positions, the grayscale value of the lesion grayscale value image is restored to obtain a lesion image with original grayscale values.

2. The method for restoring a multi-lesion type medical image according to claim 1, characterized in that: The method of performing grayscale value restoration on the lesion grayscale value image based on the lesion grayscale value image, the mask values corresponding to the respective lesion types of the multiple lesion types in the lesion mask image, and the lesion spatial positions to obtain a lesion image with original grayscale values includes: Extracting the mask value and lesion spatial position corresponding to each lesion type in the lesion mask image; Based on the spatial position of the lesion corresponding to each lesion type, the lesion grayscale value image is used to locate the spatial position of the lesion type to obtain a spatial position location image corresponding to each lesion type; The spatial position positioning image corresponding to each lesion type is divided by the corresponding mask value to obtain a lesion image with an original grayscale value.

3. The method for restoring a multi-lesion type medical image according to any one of claims 1 or 2, characterized in that: Before obtaining the medical image and its corresponding lesion mask image, a preset segmentation network, a medical training image corresponding to the preset segmentation network for training, and a multi-mask label fusion image representing organs and lesions corresponding to the medical training image are obtained; Training the preset segmentation network on the medical training image and its corresponding multi-mask label fusion image to obtain a corresponding organ / lesion segmentation model; Based on the organ / lesion segmentation model, lesion segmentation is performed on the medical image corresponding to the medical training image to obtain a lesion mask image.

4. A medical image analysis method comprising: A method for restoring a multi-lesion type medical image according to any one of claims 1 to 3 is used to obtain a lesion image with an original grayscale value, characterized in that it further comprises: If the dimension corresponding to the medical image is two-dimensional, then based on the medical image and the corresponding lesion mask image, extracting a lesion image with original grayscale values from the medical image; calculating the average grayscale value corresponding to each lesion type in the lesion mask image, and determining the lesion grayscale value corresponding to each lesion type; and / or, If the dimension corresponding to the medical image is three-dimensional, then based on the two-dimensional slice lesion mask image in the lesion mask image of each two-dimensional slice medical image and its corresponding slice in the three-dimensional medical image, a three-dimensional lesion image with original grayscale values is extracted from the three-dimensional medical image; the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image is calculated respectively to determine the lesion grayscale value corresponding to each lesion type.

5. The medical image analysis method according to claim 4, characterized in that: Also includes: If the dimension corresponding to the medical image is two-dimensional, edge detection is performed on the lesions corresponding to each lesion type on the lesion mask image to obtain a two-dimensional lesion edge mask image; based on the two-dimensional lesion edge mask image, one or more of the following information is determined: the lesion shape corresponding to each lesion type, the major diameter and minor diameter of the lesion, and the lesion area; and / or, If the dimension corresponding to the medical image is three-dimensional, the lesion area corresponding to each lesion type in each two-dimensional slice lesion mask image in the three-dimensional lesion mask image is calculated respectively; multiple lesion areas are obtained; the major axis and minor axis corresponding to the maximum lesion area among the multiple lesion areas corresponding to each lesion type are calculated respectively, and the major axis and minor axis corresponding to each lesion type are obtained; the edge detection is performed on the lesions corresponding to each lesion type on the lesion mask image respectively, and a three-dimensional lesion edge mask line is obtained; the three-dimensional lesion edge mask line corresponding to each lesion type is fitted respectively, and the lesion shape corresponding to each lesion type is obtained; the mask map corresponding to each lesion type in the lesion mask image is three-dimensionally reconstructed respectively, and the lesion volume corresponding to each lesion type is obtained.

6. The medical image analysis method according to any one of claims 4 or 5, characterized in that: Also includes: If the dimension corresponding to the medical image is three-dimensional, the organ / lesion segmentation model is used to perform organ segmentation on the medical image to obtain a three-dimensional organ mask image; according to the three-dimensional organ mask image, the organ mask in the three-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the three-dimensional organ mask image, the lesion position corresponding to each lesion type is determined; and / or, The method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the lateral organ mask and the right organ mask in the three-dimensional organ mask image comprises: extracting a plurality of two-dimensional slice lesion mask images corresponding to each lesion type from the three-dimensional lesion mask image; If the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the left organ mask spatial position of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial positions corresponding to the multiple two-dimensional slice lesion mask images are within the right organ mask spatial position of the two-dimensional slice organ mask image of the slice corresponding to the three-dimensional organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

7. The medical image analysis method according to any one of claims 4 to 6, characterized in that: Also includes: If the dimension corresponding to the medical image is two-dimensional, the organ / lesion segmentation model is used to perform organ segmentation on the medical image to obtain a two-dimensional organ mask image; according to the two-dimensional organ mask image, the organ mask in the two-dimensional organ mask image is configured as a left organ mask and a right organ mask; based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the two-dimensional organ mask image, the lesion position corresponding to each lesion type is determined; and / or, The method for determining the lesion position corresponding to each lesion type based on the spatial position relationship between the lesion mask in the lesion mask image and the side organ mask and the right organ mask in the organ mask image includes: if the lesion mask spatial position corresponding to the lesion mask image is within the left organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a left organ lesion; if the lesion mask spatial position corresponding to the lesion mask image is within the right organ mask spatial position corresponding to the organ mask image, then the lesion corresponding to the lesion mask spatial position is configured as a right organ lesion.

8. A multi-lesion type medical image restoration system, characterized in that: include: An acquisition unit, configured to acquire a medical image and its corresponding lesion mask image; wherein the lesion mask image is configured with different mask values corresponding to multiple lesion types; a multiplication unit, configured to perform a multiplication operation on the medical image and the corresponding lesion mask image to obtain a lesion grayscale value image; a first restoration unit, configured to restore the grayscale value of the lesion grayscale value image based on the lesion grayscale value image, the mask value corresponding to each of the multiple lesion types in the lesion mask image, and the lesion spatial position, to obtain a lesion image with original grayscale value; or The invention comprises: an electronic device, the electronic device being configured with a processor and a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute the method for restoring a medical image of multiple lesions according to any one of claims 1 to 3; or, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the method for restoring a multi-lesion type medical image according to any one of claims 1 to 3; or, Comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein when the computer program instructions are executed by a processor, the method for restoring a multi-lesion type medical image according to any one of claims 1 to 3 is implemented; or The invention comprises: a computer program product, comprising a computer program / instruction, which implements the method for restoring a multi-lesion type medical image according to any one of claims 1 to 3 when executed by a processor.

9. A medical image analysis system, characterized in that: include: a second restoration unit, configured to obtain a lesion image with original grayscale values by using the multi-lesion type medical image restoration method according to any one of claims 1 to 3 or the multi-lesion type medical image restoration system according to claim 8; a determining unit configured to extract, if the dimension corresponding to the medical image is two-dimensional, a lesion image having original grayscale values from the medical image based on the medical image and the lesion mask image corresponding thereto; Calculating the average grayscale value corresponding to each lesion type in the lesion mask image to determine the lesion grayscale value corresponding to each lesion type; and / or, if the dimension corresponding to the medical image is three-dimensional, extracting a three-dimensional lesion image with original grayscale values from the three-dimensional medical image based on each two-dimensional slice medical image in the three-dimensional medical image and the two-dimensional slice lesion mask image in the lesion mask image of its corresponding slice; Calculating the average grayscale value corresponding to each lesion type in the three-dimensional lesion mask image respectively to determine the lesion grayscale value corresponding to each lesion type; or, The device comprises an electronic device, the electronic device being configured with a processor and a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute the medical image analysis method according to any one of claims 4 to 7; or comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the medical image analysis method according to any one of claims 4 to 7; or, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the medical image analysis method according to any one of claims 4 to 7; or The invention comprises: a computer program product, comprising a computer program / instruction, which implements the medical image analysis method according to any one of claims 4 to 7 when the computer program / instruction is executed by a processor.

Images