Method and system for compressing and reconstructing three-dimensional medical images

Abstract

A method of compressing a three-dimensional (3D) medical image. The method includes acquiring image data of a three-dimensional medical image, performing a data conversion operation to convert the image data of the three-dimensional medical image into video data of frame sequences, each frame sequence corresponding to a respective two-dimensional image, and performing a video encoding operation to encode the video data of the frame sequences, thereby acquiring encoded content data. The encoded content data may be used to reconstruct a three-dimensional medical image. The scheme in some embodiments of the present invention can adapt to different bit rate requirements, and has relatively good performance in high bit rate and low bit rate coding scenarios.

Description

Technical Field

[0001] The present invention generally relates to the compression and / or reconstruction of medical images, in particular three-dimensional (3D) medical images. Background Art

[0002] Medical imaging techniques such as ultrasound, magnetic resonance imaging (MRI), and computed tomography (CT) can be used to generate three-dimensional medical images. Such three-dimensional medical images may contain a relatively large amount of data, and therefore have a relatively large file size. Compressing three-dimensional medical images can reduce the amount of data and thus the file size. However, compression needs to be done in an optimal way to avoid excessive loss of useful data. Summary of the invention

[0003] In a first aspect, the present invention provides a method for compressing a three-dimensional medical image. The method includes acquiring image data of the three-dimensional medical image and performing a data conversion operation; converting the image data of the three-dimensional medical image into video data of a frame sequence. Each frame corresponds to a corresponding two-dimensional image (such as a two-dimensional medical image). The method also includes performing a video encoding operation to encode the video data of the frame sequence to obtain encoded content data. The encoded content data can be used to reconstruct the three-dimensional medical image.

[0004] Optionally, the step of performing a data conversion operation includes: processing image data of a three-dimensional medical image to obtain image data of a plurality of two-dimensional images; and performing a bit-depth conversion operation on the image data of the plurality of two-dimensional images to obtain video data of a frame sequence.

[0005] Optionally, processing the image data of the three-dimensional medical image to obtain image data of a plurality of two-dimensional images includes: segmenting the three-dimensional medical image along an axis perpendicular to the segmentation plane to obtain a plurality of slices that can be represented as a plurality of two-dimensional images, and for each pixel of each two-dimensional image in the plurality of two-dimensional images, mapping the pixel to an n-bit integer to obtain an n-bit pixel. In some examples, n may be 8. In some examples, n may be 10. In some examples, n may be 16. In some examples, n may be another integer.

[0006] In some examples, the segmentation plane is a transverse plane associated with the three-dimensional medical image. In some examples, the segmentation plane is a coronal plane associated with the three-dimensional medical image. In some examples, the segmentation plane is a sagittal plane associated with the three-dimensional medical image.

[0007] Optionally, the mapping is performed according to a data range or a data boundary of the three-dimensional medical image. In some embodiments, the method further comprises: before mapping, determining a data range or a data boundary of the three-dimensional medical image.

[0008] Optionally, mapping is performed according to a specified or defined (eg, user specified or user defined) data range or data boundary. In some embodiments, the method further comprises: prior to mapping, receiving user input associated with the user specified data range or data boundary.

[0009] In some embodiments, the bit-depth conversion operation is based on bit-interlacing.

[0010] Optionally, the bit depth conversion operation based on bit interleaving segmentation includes: converting each n-bit pixel into two respective m-bit pixels, and forming two m-bit sequences based on the m-bit pixels of multiple two-dimensional images. The video data of the frame sequence includes two m-bit sequences. In some examples, the video data of the frame sequence may include other data content. In some examples, n=16, m=8. Optionally, for each n-bit pixel, the above conversion includes: applying a first bit mask to the n-bit pixel to extract the first m bits, applying a second bit mask to the n-bit pixel to extract the second m bits, connecting the first m bits to the m-bit pixel, and connecting the second m bits to another m-bit pixel. Optionally, applying a first bit mask to the n-bit pixel to extract the first m bits includes applying a first bit mask to the n-bit pixel to extract m even position bits. Optionally, applying a second bit mask to the n-bit pixel to extract the second m bits includes applying a second bit mask to the n-bit pixel to extract m odd position bits.

[0011] In some embodiments, the bit depth conversion operation is based on bit-shifting.

[0012] Optionally, the bit depth conversion operation based on the displacement includes: converting each n-bit pixel into a respective m-bit pixel, and forming an m-bit sequence based on the m-bit pixels of the multiple two-dimensional images. The video data of the frame sequence includes the m-bit sequence. In some examples, the video data of the frame sequence may include other data content. In some examples, n=16, m=8 or 10. In some examples, n and m may be other integers. Optionally, for each n-bit pixel, the conversion includes: performing a bit right shift on the higher m bits of the n-bit pixel, and discarding or abandoning the lower nm bits of the n-bit pixel, thereby forming an m-bit pixel.

[0013] Optionally, performing the data conversion operation further comprises: performing a color conversion operation to convert the color space of the video data of the frame sequence. Optionally, performing the data conversion operation further comprises: performing a color conversion operation to convert the color space of the image data of the three-dimensional medical image. Optionally, performing the data conversion operation further comprises: performing a color conversion operation to convert the color space of the image data of the plurality of two-dimensional images.

[0014] Optionally, performing the data conversion operation further includes: generating a data conversion flag (e.g., a flag) representing one or more attributes related to processing image data of the three-dimensional medical image and / or performing a bit depth conversion operation, and performing an encoding operation to encode the data conversion flag to obtain encoded data conversion flag data. The encoded data conversion flag data can be used to facilitate reconstruction of the three-dimensional medical image.

[0015] Optionally, the data conversion flag indicates one or more of: a data range or a data boundary of the three-dimensional medical image, or one or more attributes of a bit depth conversion operation performed. Optionally, the one or more attributes of the bit depth conversion operation performed include: a method based on which the bit depth conversion operation has been performed. Optionally, the method based on which the bit depth conversion operation performed is bit interleaving or bit shifting. Optionally, if the bit depth conversion operation is based on bit shifting, the one or more attributes of the bit depth conversion operation performed further include: a bit depth associated with a sequence generated by the bit shifting.

[0016] The video encoding operation may be arranged to reduce inter-frame redundancy of video data of a frame sequence and / or reduce intra-frame redundancy of video data of a frame sequence. The encoded content data may be in a bitstream format.

[0017] Optionally, the video encoding operation is based on Versatile Video Coding (VVC). Optionally, the video encoding operation is based on High Efficiency Video Coding (HEVC). Optionally, the video encoding operation is based on Advanced Video Coding (AVC). Optionally, the video encoding operation is based on Audio Video Coding (AVS), such as AVS3.

[0018] The video encoding operation is performed by a video encoder, which can be implemented by hardware and / or software. Optionally, the video encoder includes an encoder based on Versatile Video Coding (VVC). Optionally, the video encoder includes an encoder based on High Efficiency Video Coding (HEVC). Optionally, the video encoder includes an encoder based on Advanced Video Coding (AVC). Optionally, the video encoder includes an encoder based on Audio Video Coding (AVS). Optionally, the encoder based on Audio Video Coding (AVS) includes an encoder based on AVS3.

[0019] Optionally, the method further comprises storing the encoded content data as one or more data files. Optionally, the method further comprises storing the encoded data conversion flag data as one or more data files. Optionally, the method further comprises storing the encoded content data and the encoded data conversion flag data as one or more data files.

[0020] Optionally, the three-dimensional medical image is a magnetic resonance imaging (MRI) image. Optionally, the three-dimensional medical image is a computed tomography (CT) image. In some other examples, the three-dimensional medical image may be a medical image of other imaging modalities (except CT and MRI). The three-dimensional medical image may be an image of a human or animal body part, a phantom image, etc.

[0021] In a second aspect of the present invention, a system for compressing a three-dimensional medical image is provided, the system comprising: one or more processors, and a memory storing one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs comprising instructions for performing or facilitating the performance of the method of the first aspect. In some embodiments, the system may further comprise a display for displaying an image (a two-dimensional or three-dimensional image, for example, generated from image data or video data).

[0022] In a third aspect of the present invention, a non-transitory computer-readable storage medium is provided, in which one or more programs are stored, which are configured to be executed by one or more processors, and the one or more programs include instructions for executing or facilitating the execution of the method of the first aspect.

[0023] In a fourth aspect of the present invention, a method for reconstructing a three-dimensional medical image is provided. The method comprises: performing a video decoding operation to decode the encoded content data to obtain video data of a frame sequence, each frame corresponding to a respective two-dimensional image (e.g., a two-dimensional medical image); and performing an inverse data conversion operation to convert the video data of the frame sequence into image data of the three-dimensional medical image. The image data of the three-dimensional medical image can be used to reconstruct the three-dimensional medical image.

[0024] Optionally, the encoded content data may be encoded content data obtained using the method of the first aspect. The encoded content data may be in a bitstream format.

[0025] Optionally, the video decoding operation is based on Versatile Video Coding (VVC). Optionally, the video decoding operation is based on High Efficiency Video Coding (HEVC). Optionally, the video decoding operation is based on Advanced Video Coding (AVC). Optionally, the video decoding operation is based on Audio Video Coding (AVS), such as AVS3.

[0026] The video decoding operation is performed by a video decoder, which can be implemented by hardware and / or software. Optionally, the video decoder includes a decoder based on Versatile Video Coding (VVC). Optionally, the video decoder includes a decoder based on High Efficiency Video Coding (HEVC). Optionally, the video decoder includes a decoder based on Advanced Video Coding Standard (AVC). Optionally, the video decoder includes a decoder based on Audio Video Coding Standard (AVS). Optionally, the decoder based on Audio Video Coding Standard (AVS) includes an AVS3 decoder.

[0027] Optionally, performing the inverse data conversion operation includes: performing an inverse bit depth conversion operation on the video data of the frame sequence to obtain image data of a plurality of two-dimensional images, and processing the image data of the plurality of two-dimensional images to obtain image data of a three-dimensional medical image.

[0028] In some embodiments, the inverse bit depth conversion operation is based on inverse bit interleaving segmentation. Optionally, the video data of the frame sequence includes two m-bit sequences (the video data of the frame sequence may include other data content), and the inverse bit depth conversion operation based on inverse bit interleaving segmentation includes: determining multiple m-bit pixels of multiple two-dimensional images according to the two m-bit sequences, and converting two respective m-bit pixels of each pixel in the multiple two-dimensional images into one respective n-bit pixel, so that each image in the multiple two-dimensional images includes multiple n-bit pixels. In some examples, n=16, m=8.

[0029] In some embodiments, the inverse bit depth conversion operation is based on inverse displacement. Optionally, the video data of the frame sequence includes an m-bit sequence (the video data of the frame sequence may include other data content), and the inverse bit depth conversion operation based on inverse displacement includes: based on the m-bit sequence, determining multiple m-bit pixels of multiple two-dimensional images, and converting each m-bit pixel of each pixel in the multiple two-dimensional images into a corresponding n-bit pixel, so that each two-dimensional image in the multiple two-dimensional images includes multiple n-bit pixels. In some examples, n=16, m=8 or 10. In some examples, n and m can be other integers. Optionally, for each m-bit pixel, the above conversion includes: performing a bit shift to the left on the m bits of the m-bit pixel, and after the bit shift to the left, filling the lower nm bits with zeros to form an n-bit pixel.

[0030] Optionally, the plurality of two-dimensional images include a plurality of n-bit pixels, and processing the image data of the plurality of two-dimensional images to obtain the image data of the three-dimensional medical image includes: for each n-bit pixel of each image in the plurality of two-dimensional images, respectively inversely mapping the n-bit pixel into a pixel according to a data range or a data boundary, and combining (e.g., splicing) a plurality of slices represented by the plurality of two-dimensional images with the inversely mapped pixels along an axis perpendicular to the segmentation plane to obtain the three-dimensional medical image. In some examples, n is 8. In some examples, n is 16. In some examples, n may be another integer.

[0031] Optionally, the data range or data boundary is a data range or data boundary of a three-dimensional medical image. Optionally, the data range or data boundary is a specified or defined (eg, user-specified or user-defined) data range or data boundary.

[0032] Optionally, the segmentation plane is a transverse plane associated with the three-dimensional medical image. Optionally, the segmentation plane is a coronal plane associated with the three-dimensional medical image. Optionally, the segmentation plane is a sagittal plane associated with the three-dimensional medical image.

[0033] Optionally, performing the inverse data conversion operation further includes: performing a decoding operation to decode the encoded data conversion flag data to obtain a data conversion flag (e.g., a flag), wherein the data conversion flag indicates one or more attributes associated with processing image data of a three-dimensional medical image and / or performing a bit depth conversion operation. The data conversion flag can be used to facilitate reconstruction of the three-dimensional medical image.

[0034] Optionally, the data conversion flag indicates one or more of: a data range or a data boundary of the three-dimensional medical image, or one or more properties of a bit depth conversion operation that has been performed. Optionally, the one or more properties of the bit depth conversion operation that has been performed include: a method based on which the bit depth conversion operation has been performed. Optionally, the method based on which the bit depth conversion operation that has been performed is bit interleaving or bit shifting. Optionally, if the bit depth conversion operation is based on bit shifting, the one or more properties of the bit depth conversion operation that has been performed further include a bit depth associated with a sequence generated by the bit shifting.

[0035] Optionally, performing the inverse data conversion operation further includes: performing an inverse color conversion operation to convert the color space of the video data of the frame sequence or the color space of the image data of the three-dimensional medical image or the color space of the image data of the plurality of two-dimensional images.

[0036] Optionally, the three-dimensional medical image is a magnetic resonance imaging (MRI) image. Optionally, the three-dimensional medical image is a computed tomography (CT) image. In some other examples, the three-dimensional medical image may be a medical image of another imaging modality (other than CT and MRI). The three-dimensional medical image may be an image of a human or animal body part, a phantom image, etc.

[0037] Optionally, the method further comprises reconstructing the three-dimensional medical image according to the image data of the three-dimensional medical image. Optionally, the method further comprises reconstructing the three-dimensional medical image according to the image data of the three-dimensional medical image and a data conversion flag.

[0038] Optionally, the method further comprises outputting the reconstructed three-dimensional medical image for display.

[0039] In a fifth aspect of the present invention, a system for compressing a three-dimensional medical image is provided, comprising: one or more processors, and a memory storing one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs comprising instructions for performing or facilitating the performance of the method of the fourth aspect. In some embodiments, the system may further include a display for displaying an image (a two-dimensional or three-dimensional image, for example, generated from image data or video data).

[0040] In a sixth aspect of the present invention, a non-transitory computer-readable storage medium is provided, in which one or more programs are stored, which are configured to be executed by one or more processors, and the one or more programs include instructions for executing or facilitating the execution of the method of the fourth aspect.

[0041] In a seventh aspect of the present invention, a method is provided which includes the method of the first aspect and the method of the fourth aspect.

[0042] In an eighth aspect of the present invention, a system is provided, the system comprising: one or more processors, and a memory storing one or more programs, the programs being configured to be executed by the one or more processors, the one or more programs comprising instructions for executing or facilitating the execution of the method of the seventh aspect. In some embodiments, the system may further comprise a display for displaying an image (a two-dimensional or three-dimensional image, for example, generated by image data or video data).

[0043] In a ninth aspect, the present invention provides a non-transitory computer-readable storage medium storing one or more programs configured to be executed by one or more processors, the one or more programs comprising instructions for executing or facilitating the execution of the method of the seventh aspect.

[0044] Other features and aspects of the invention will become apparent from consideration of the detailed description and accompanying drawings.Any feature described herein in relation to one aspect or embodiment may be combined with any other feature described herein in relation to any other aspect or embodiment, as appropriate.

[0045] Terms of degree, such as "generally," "approximately," "substantially," or similar terms, are used to explain one or more of the following, depending on the context: manufacturing tolerances, degradation, trends, tendencies, imperfect physical conditions, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Embodiments of the present invention will now be described by way of example with reference to the accompanying drawings:

[0047] Figure 1 is a schematic diagram of a medical image compression operation in some embodiments of the present invention.

[0048] Figure 2 is a schematic diagram of a medical image reconstruction operation in some embodiments of the present invention.

[0049] Figure 3 is a flow chart of a method for compressing a three-dimensional medical image in some embodiments of the present invention.

[0050] Figure 4 is a flowchart of a method for three-dimensional medical image reconstruction in some embodiments of the present invention.

[0051] Figure 5 Schematic diagram of encoding operation for compressing a three-dimensional medical image in one embodiment of the present invention.

[0052] Fig. 6A An example of a slice image generated from a 3D CT image.

[0053] Figure 6B Here is an example of another slice generated from a 3D CT image.

[0054] Figure 6C is an image of another slice generated from a three-dimensional CT image in one instance.

[0055] Fig. 7A An example of a slice image generated by 3D MRI.

[0056] Figure 7B An example of another slice generated by 3D MRI is shown here.

[0057] Figure 7C An example of another slice generated by 3D MRI is shown here.

[0058] Figure 8It is a schematic diagram of a bit depth conversion operation based on bit interleaving segmentation in one embodiment of the present invention.

[0059] Fig.9A An example 16-bit image of a slice generated from a 3D CT image.

[0060] Fig. 9B In an example Fig.9A The 16-bit image is implemented Figure 8 The 8-bit image is obtained after the bit depth conversion operation based on bit interleaving segmentation.

[0061] Fig. 9C In an example Fig.9A The 16-bit image is implemented Figure 8 Another 8-bit image is obtained after the bit depth conversion operation based on bit interleaving segmentation.

[0062] Fig.10 FIG. 4 is a schematic diagram of a bit depth conversion operation based on displacement in one embodiment of the present invention.

[0063] Fig.11A An example 16-bit image of a slice generated from a 3D CT image.

[0064] Fig. 11B In an example Fig.11A The 16-bit image is implemented Fig.10 The 8-bit image obtained after the bit-shift-based bit depth conversion operation in .

[0065] Fig. 11C In an example Fig.11A The 16-bit image is implemented Fig.10 The resulting 8-bit image after a bit-shift-based bit-depth conversion operation.

[0066] Fig.12 It is a schematic diagram of the decoding operation of reconstructing a three-dimensional medical image in one embodiment of the present invention.

[0067] Fig.13 is a schematic diagram showing an inverse bit depth conversion operation based on inverse bit interleaving partitioning in one embodiment of the present invention; and

[0068] Fig.14 is a schematic diagram of a data processing system in some embodiments of the present invention. DETAILED DESCRIPTION

[0069] Figure 1 A system 100 is shown which, in some embodiments, is arranged to process image data of a three-dimensional medical image to provide encoded or compressed video data. The system 100 may be implemented using hardware and / or software.

[0070] Figure 2 A system 200 is shown, which in some embodiments is used to process encoded or compressed video data to provide image data of a three-dimensional medical image. The encoded or compressed video data processed by the system 200 may be the encoded or compressed video data generated by the system 100. The system 200 may be implemented using hardware and / or software.

[0071] In some embodiments, systems 100, 200 may be implemented as separate systems. In some embodiments, systems 100, 200 may be implemented together as a single system (eg, combined into a single system).

[0072] Figure 3 A method 300 for compressing a three-dimensional medical image in some embodiments of the present invention is shown. The method 300 may be used Figure 1 The system 100 in is executed.

[0073] The method 300 includes acquiring image data of a three-dimensional medical image in step 302. The three-dimensional medical image may be a magnetic resonance imaging (MRI) image, a computed tomography (CT) image, or a medical image of other imaging modalities (except CT and MRI). The three-dimensional medical image may be an image of a human or animal body part, a phantom image, etc. In some embodiments, the image data of the three-dimensional medical image is acquired from an imaging device (such as a nuclear magnetic resonance imaging machine, a CT machine, etc.).

[0074] Method 300 also includes performing a data conversion operation in step 304 to convert the image data of the three-dimensional medical image into video data of a frame sequence. Each frame corresponds to a corresponding two-dimensional image (e.g., a two-dimensional medical image if the image includes relevant medical content). Although the three-dimensional medical image is not a video, after the image data of the three-dimensional medical image is converted into video data, the video data can be processed using video data processing techniques (e.g., video encoding techniques) to process the image data of the three-dimensional medical image. The three-dimensional medical image can be appropriately processed in this manner because the three-dimensional medical image may generally include at least some slices or images with relevant content (e.g., different images of adjacent slices of the same body part).

[0075] In some embodiments, performing the data conversion operation in step 304 may include processing image data of a three-dimensional medical image to obtain image data of a two-dimensional image, and performing a bit depth conversion operation on the image data of the two-dimensional image to obtain video data of a frame sequence (the frames correspond to two-dimensional images, and due to the bit depth conversion operation, these frames are different from the two-dimensional images of the image data of the two-dimensional image).

[0076] Processing image data of a three-dimensional medical image to obtain image data of a two-dimensional image may include, for example, segmenting the three-dimensional medical image along an axis perpendicular to a segmentation plane to obtain slices that can be represented as two-dimensional images, and mapping each pixel of each two-dimensional image to an n-bit integer to obtain an n-bit pixel. The segmentation plane may be a transverse plane associated with the three-dimensional medical image, a coronal plane associated with the three-dimensional medical image, or a sagittal plane associated with the three-dimensional medical image. The mapping may be based on: a data range or data boundary of the three-dimensional medical image, or a specified or defined (e.g., user-specified or user-defined) data range or data boundary. In some examples, method 300 also includes, before mapping: determining a data range or data boundary of the three-dimensional medical image, or receiving user input related to a specified data range or data boundary.

[0077] In some embodiments, the bit depth conversion operation may be based on bit interleaving. In some examples, the bit depth conversion operation based on bit interleaving may include: converting each n-bit pixel into two respective m-bit pixels, and forming two m-bit sequences based on the m-bit pixels of the two-dimensional image. The video data of the frame sequence may include at least two m-bit sequences. In some examples, n=16, m=8. In some examples, for each n-bit pixel, the conversion includes: applying a first bit mask to the n-bit pixel to extract m bits, applying a second bit mask to the n-bit pixel to extract another m bits, connecting the above m bits to an m-bit pixel, and connecting the above other m bits to another m-bit pixel. In some examples, applying the first bit mask to the n-bit pixel to extract m bits includes applying the first bit mask to the n-bit pixel to extract m even position bits. In some examples, applying the second bit mask to the n-bit pixel to extract another m bits includes applying the second bit mask to the n-bit pixel to extract m odd position bits.

[0078] In some embodiments, the bit depth conversion operation may be based on displacement. In some examples, the bit depth conversion operation based on displacement may include: converting each n-bit pixel into a respective m-bit pixel, and forming an m-bit sequence based on the m-bit pixels of the two-dimensional image. The video data of the frame sequence includes at least an m-bit sequence. In some examples, n=16, m=8 or 10. In some examples, for each pixel in the n-bit pixel, the conversion includes: performing a bit right shift on the higher m bits of the n-bit pixel, discarding or abandoning the lower nm bits of the n-bit pixel, thereby forming an m-bit pixel.

[0079] In some embodiments, performing the data conversion operation in step 304 may include generating a data conversion flag (e.g., a flag) indicating one or more attributes related to image data processing of the three-dimensional medical image and / or performing a bit depth conversion operation. The data conversion flag may indicate one or more of the following attributes: a data range or a data boundary of the three-dimensional medical image, or one or more attributes of the bit depth conversion operation that has been performed. Examples of the one or more attributes include: a method based on which the bit depth conversion operation performed is based (e.g., bit interleaving or shifting), a bit depth associated with a sequence of shift generation (if the bit depth conversion operation is based on shifting). In these embodiments, performing the data conversion operation in step 304 may further include performing an encoding operation to encode the data conversion flag to obtain encoded data conversion flag data. The encoded data conversion flag data may be used to facilitate reconstruction of the three-dimensional medical image.

[0080] In some embodiments, performing the data conversion operation in step 304 may include performing a color conversion operation to convert the color space of video data of a frame sequence, or performing a color conversion operation to convert the color space of image data of a three-dimensional medical image, and / or performing a color conversion operation to convert the color space of image data of a two-dimensional image.

[0081] The method 300 also includes, in step 306, performing a video encoding operation to encode the video data of the frame sequence to obtain encoded content data. The encoded content data can be used for reconstruction of a three-dimensional medical image. The video encoding operation can be arranged to reduce inter-frame redundancy of the video data of the frame sequence and / or reduce intra-frame redundancy of the video data of the frame sequence. The encoded content data can be in a bitstream format. The video encoding operation can be based on a video encoding method / standard, such as but not limited to the following methods / standards: Versatile Video Coding (VVC), High Efficiency Video Coding Standard (HEVC), Advanced Video Coding Standard (AVC), Audio Video Coding Standard (AVS), such as AVS3, etc.

[0082] Although not specifically described, in some embodiments, method 300 also includes storing the encoded content data and / or the encoded data conversion flag data as one or more data files. In some examples, the encoded content data and the encoded data conversion flag data are merged and then stored as a single data file.

[0083] Figure 4 A method 400 for reconstructing a three-dimensional medical image in some embodiments of the present invention is shown. The method 400 may be used Figure 2 The system 200 in is executed.

[0084] Method 400 includes performing a video decoding operation to decode the encoded content data in step 402 to obtain video data of a sequence of frames. Each frame corresponds to a respective two-dimensional image (e.g., a two-dimensional medical image, if the image includes relevant medical content). The encoded content data may be encoded using Figure 3 The method 300 obtains the encoded content data. The encoded content data may be in a bitstream format. In step 402, the video decoding operation may be based on a video decoding method / standard, such as but not limited to the following methods / standards: Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), Audio Video Coding (AVS), such as AVS3, etc.

[0085] The method 400 also includes performing an inverse data conversion operation in step 404 to convert the video data of the frame sequence into image data of the three-dimensional medical image. The image data of the three-dimensional medical image can be used to reconstruct the three-dimensional medical image. The three-dimensional medical image can be a magnetic resonance imaging (MRI) image, a computed tomography (CT) image, or a medical image of other imaging modes (except CT and MRI). The three-dimensional medical image can be an image of a human or animal body part, a phantom image, etc. In some embodiments, the image data of the three-dimensional medical image is obtained from an imaging device (such as a nuclear magnetic resonance imaging machine, a CT machine, etc.).

[0086] In some embodiments, performing the inverse data conversion operation in step 404 may include performing an inverse bit depth conversion operation on the video data of the frame sequence to obtain image data of a two-dimensional image, and processing the image data of the two-dimensional image to obtain image data of a three-dimensional medical image.

[0087] In some embodiments, the inverse bit depth conversion operation can be based on inverse bit interleaving segmentation. In some examples, the video data of the frame sequence includes at least two m-bit sequences, and the inverse bit depth conversion operation based on the inverse bit interleaving segmentation includes: determining m-bit pixels of the two-dimensional image according to the two m-bit sequences, and converting the two respective m-bit pixels into respective n-bit pixels of each pixel of each two-dimensional image, so that each two-dimensional image includes n-bit pixels. In some examples, n=16, m=8.

[0088] In some embodiments, the inverse bit depth conversion operation is based on an inverse bit shift. In some examples, the video data of the frame sequence includes at least an m-bit sequence, and the inverse bit depth conversion operation based on the inverse shift includes: determining the m-bit pixels of the two-dimensional image according to the m-bit sequence, and converting the respective m-bit pixels of each pixel of each two-dimensional image into respective n-bit pixels, so that each two-dimensional image includes n-bit pixels. In some examples, n=16, m=8 or 10. In some examples, for each pixel in the m-bit pixel, the conversion includes: performing a bit left shift on the m bits of the m-bit pixel, and filling the lower nm bits with zeros after the bit left shift, thereby forming an n-bit pixel.

[0089] In some embodiments, the two-dimensional image includes n-bit pixels, and processing the image data of the two-dimensional image to obtain the image data of the three-dimensional medical image includes: for each n-bit pixel of each two-dimensional image, inversely mapping the n-bit pixel into pixels according to a data range or a data boundary, and combining (e.g., splicing) the slice represented by the two-dimensional image with the inversely mapped pixels along an axis perpendicular to the segmentation plane to obtain the three-dimensional medical image. The data range or data boundary can be a data range or data boundary of the three-dimensional medical image, or a specified data range or data boundary. The segmentation plane can be a transverse plane associated with the three-dimensional medical image, a coronal plane associated with the three-dimensional medical image, or a sagittal plane associated with the three-dimensional medical image.

[0090] In some embodiments, performing the inverse data conversion operation in step 404 may include performing a decoding operation to decode the encoded data conversion flag data to obtain a data conversion flag (eg, a flag). The data conversion flag may be Figure 3 Data conversion flag in method 300. The data conversion flag may be used to facilitate reconstruction of a three-dimensional medical image.

[0091] In some embodiments, performing an inverse data conversion operation in step 404 may include performing an inverse color conversion operation to convert the color space of video data of a frame sequence, or performing an inverse color conversion operation to convert the color space of image data of a three-dimensional medical image, or performing an inverse color conversion operation to convert the color space of image data of a two-dimensional image.

[0092] Although not specifically described, in some embodiments, method 400 further includes reconstructing a three-dimensional medical image based on the image data of the three-dimensional medical image. In some embodiments, method 400 further includes reconstructing a three-dimensional medical image based on the image data of the three-dimensional medical image and a data conversion flag. In some embodiments, method 400 further includes outputting the reconstructed three-dimensional medical image for display.

[0093] Figure 5An encoding operation 500 for compressing a three-dimensional medical image in accordance with one embodiment of the present invention is shown. Figure 5 The encoding operation 500 in can be viewed as Figure 3 A specific example of method 300, because the encoding operation 500 as a whole also includes acquiring image data of a three-dimensional medical image, performing a data conversion operation to convert the image data of the three-dimensional medical image into video data of a frame sequence, and performing a video encoding operation to encode the video data of the frame sequence, thereby obtaining encoded content data. Figure 5 The various modules and encoders shown in the encoding operation 500 may be implemented in hardware and / or software.

[0094] In this embodiment, in the encoding operation 500, the original three-dimensional medical image is processed and converted into a two-dimensional medical image sequence by the data conversion module 502. Specifically, the three-dimensional medical image is first segmented to generate 16-bit two-dimensional slices, and then the 16-bit two-dimensional slices are further converted into an 8-bit or 10-bit two-dimensional medical image sequence that can be compressed by a video encoder. The video encoder 504 compresses the 8-bit or 10-bit two-dimensional medical image sequence into a video bitstream. In this embodiment, the flag is used to indicate one or more attributes related to the processing performed by the data conversion module 502. The flag can be compressed by the flag encoder 502C. The flag encoder 502C compresses the flag into a flag bitstream. In this embodiment, the video bitstream and the flag bitstream are merged into a single bitstream and then written to the data file.

[0095] In this embodiment, the data conversion module 502 may convert the 3D medical image into a 2D medical image sequence, which is then compressed by the video encoder 504. In this embodiment, the data conversion module 502 includes a slice segmentation module 502A, a bit depth conversion module 502B, and a flag encoder 502C.

[0096] In this embodiment, the slice segmentation module 502A is used to transform the input three-dimensional medical image X∈R H×W×D Converted into a set of slices that can be represented as a set of 16-bit two-dimensional images Y.

[0097] In the slice segmentation module 502A, the data boundary x of the 3D medical image x is first calculated. min and x max ,Right now:

[0098]

[0099] Subsequently, the three-dimensional medical image x is segmented into a set of slices (representable as a set of two-dimensional images Y) along a specified axis, wherein each pixel of each two-dimensional image is mapped to a 16-bit integer.

[0100] In this embodiment, the mapping process can be described as follows:

[0101]

[0102] Where x is the pixel of x, y is the two-dimensional image (the pixel of the two-dimensional image Y), and round(·) indicates a rounding operation.

[0103] In one example, the segmentation axis is specified as an axis perpendicular to the transverse plane of a CT medical image. In one example, for a MRI medical image, the segmentation axis is specified as an axis perpendicular to the sagittal plane.

[0104] FIG. 6A to FIG. 6C A two-dimensional slice (image) generated from a three-dimensional CT image using the slice segmentation module 502A is shown.

[0105] Figures 7A to 7C A two-dimensional slice (image) generated from a three-dimensional MRI image using the slice segmentation module 502A is shown.

[0106] In one embodiment, the bit depth conversion module 502B is used to process each 16-bit pixel of the two-dimensional image Y to generate an 8-bit or 10-bit two-dimensional medical image sequence Z that can be compressed by the video encoder 504 .

[0107] In one example, the bit depth conversion module 502B is arranged to perform bit depth conversion based on bit interleaving partitioning. Figure 8 An example of bit depth conversion based on bit interleaving is shown. In this example, two sets of bit masks are used to extract the odd and even bits of the 16-bit pixel y respectively. The extraction process can be described as follows (Note: & is a bitwise logical AND operator):

[0108]

[0109] Then, the extracted bits are concatenated into 8-bit pixels. In this way, the 16-bit sequence of the two-dimensional image Y can be divided into two 8-bit sequences.

[0110] Figures 9A to 9C shows that the bit depth conversion module 502B uses Figure 8 Before the bit depth conversion process based on bit interleaving ( Fig.9A ) and after ( Fig. 9B and 9C ) images.

[0111] In another example, the bit depth conversion module 502B is arranged to perform bit depth conversion based on bit shifting. For example, the bit depth conversion based on bit shifting can generate a 10-bit or 8-bit sequence through a bit shifting operation. Fig.10 An example of bit depth conversion based on displacement is given below. Fig.10As shown, by right shifting the high 8 bits of the 16-bit pixel y and discarding the low 8 bits, an 8-bit pixel can be obtained. Similarly, by right shifting the high 10 bits of the 16-bit pixel y and discarding the low 6 bits, a 10-bit pixel can be obtained. This process can be expressed as follows (Note: >> is a right shift operator):

[0112]

[0113] FIG. 11A to FIG. 11C shows that the bit depth conversion module 502B uses Fig.10 Before processing based on bit depth conversion in the displacement ( Fig.11A ) and after processing ( Fig. 11B and Fig. 11C ) images.

[0114] exist Figure 5 In an embodiment of the encoding operation 500, a flag is used and written to the bitstream to indicate one or more properties of the processing performed by the data conversion module 502 (e.g., the technology / method used in the data conversion module 502). In one example, the data range {x min , x max} is indicated by two 32-bit flags, and the bit depth conversion method is indicated by a 1-bit flag. For example, if bit-interleaving-based bit depth conversion is used, the 1-bit flag can be set to 0; if displacement-based bit depth conversion is used, the 1-bit flag can be set to 0, and vice versa. In one example, if displacement-based bit depth conversion is used, an additional flag can be used to indicate whether the bit depth of the conversion sequence is equal to 8. In some examples, the bit consumption of the flag bit is at most 63 bits and can be ignored.

[0115] exist Figure 5 In an embodiment of the encoding operation 500, the video encoder 504 uses an advanced video encoding method to compress the two-dimensional medical image sequence to obtain improved encoding efficiency of the two-dimensional medical image sequence. In one example, a hierarchical B coding structure and a bidirectional prediction technique are used to reduce inter-frame redundancy in the two-dimensional medical image sequence. For example, the frames of the two-dimensional medical image sequence can be divided into four different time layers and predicted by forward and backward reference frames to eliminate inter-frame redundancy. In one example, intra-frame prediction technology is used to reduce spatial redundancy in the two-dimensional medical image sequence. The video bitstream generated by the video encoder 504 is written to a file for storage.

[0116] Fig.12 A decoding operation 1200 for reconstructing a three-dimensional medical image in one embodiment of the present invention is shown. Fig.12 The decoding operation 1200 in can be viewed as Figure 4A specific example of method 400, because the decoding operation 1200 as a whole also includes performing a video decoding operation to decode the encoded content data to obtain video data of a frame sequence, and performing an inverse data conversion operation to convert the video data of the frame sequence into image data of a three-dimensional medical image. Fig.12 The various modules and decoders shown in the decoding operation 1200 may be implemented in hardware and / or software.

[0117] In this embodiment, in the decoding operation 1200, the video bitstream is loaded from the file and processed by the video decoder 1204 to reconstruct the two-dimensional medical image sequence. In addition, the flag bitstream corresponding to the flag of the data conversion module 502 is decoded by the flag decoder 1202C and used to prepare or initialize the inverse data conversion module 1202. The inverse data conversion module 1202 is used to convert the reconstructed two-dimensional medical image sequence into a reconstructed three-dimensional medical image, wherein the 16-bit two-dimensional slices are retrieved by the inverse bit depth conversion module 1202A and then spliced ​​into the reconstructed three-dimensional medical image by the slice splicing module 1202B. In this embodiment, in the decoding operation 1200, the video decoder 1204 is used to read the video bitstream corresponding to the two-dimensional medical image sequence and decompress the video bitstream to obtain the reconstructed two-dimensional medical image sequence.

[0118] In this embodiment, the inverse data conversion module 1202 helps reconstruct a three-dimensional medical image from a decoded two-dimensional medical image sequence. In this embodiment, the inverse data conversion module 1202 includes a flag decoder 1202C, an inverse bit depth conversion module 1202A, and a slice splicing module 1202B. In an example operation, the flag of the data conversion module 502 is first decoded by the flag decoder 1202C and used to prepare or initialize the inverse bit depth conversion module 1202A and the slice splicing module 1202B. Then, the inverse bit depth conversion module 1202A converts the decoded two-dimensional medical image sequence into two-dimensional slices.

[0119] In one embodiment, the inverse bit depth conversion module 1202A is arranged to perform an inverse bit depth conversion based on inverse bit interleaving partitioning, which generally corresponds to the inverse operation of bit interleaving partitioning (e.g., Figure 8 In one embodiment, the inverse bit depth conversion module 1202A is arranged to perform an inverse bit depth conversion based on an inverse bit shift, which generally corresponds to the inverse operation of the bit shift (e.g., Fig.10 In one embodiment, a selection is made between an inverse bit depth conversion based on inverse bit shifting and an inverse bit depth conversion based on inverse bit interleaving according to a decoding flag.

[0120] Fig.13An example of inverse bit depth conversion based on inverse bit interleaving is shown. In this example, two 8-bit decoded 2D medical image sequences are concatenated to form 16-bit slices, where the pixel bits of the two 8-bit sequences are interleaved to generate 16-bit pixels. In the example of inverse bit depth conversion based on shifting, 16-bit pixels are generated by bit-shifting the decoded 8-bit or 10-bit pixels to the left, and the low bits vacated after the shift are filled with 0.

[0121] After the inverse bit depth conversion, the reconstructed 16-bit slice will be decoded from the flag bitstream according to {x min , x max}Map to the data range of the three-dimensional medical image, and stack the slices along the specified axis to reconstruct the three-dimensional medical image. The mapping process can be described as follows:

[0122]

[0123] in, Represents a decoded 16-bit pixel, Represents the reconstructed three-dimensional medical image data.

[0124] Fig.14 An example of a data processing system 1400 is shown for certain embodiments of the present invention. The data processing system 1400 may be used to perform one or more method embodiments of the present invention. The data processing system 1400 may be used to implement, for example, Figure 1 System 100, Figure 2 System 200, Figure 5 The system or system component (e.g., encoder, module, etc.) in the encoding operation 500, Fig.12The system or system component (e.g., decoder, module, etc.) in the decoding operation 1200 of the data processing system 1400. The data processing system 1400 generally includes appropriate components required to receive, store and execute appropriate computer instructions, commands and / or codes. The main components of the data processing system 1400 are the processor 1402 and the memory (storage) 1404. The processor 1402 may include one or more: CPU, MCU, GPU, logic circuit, Raspberry Pi chip, digital signal processor (DSP), application specific integrated circuit (ASIC), field programmable gate array (FPGA) or any other digital or analog circuit / circuit configured to interpret and / or execute program instructions and / or process signals and / or information and / or data. The memory 1404 may include one or more volatile memories (such as RAM, DRAM, SRAM, etc.), one or more non-volatile memories (such as ROM, PROM, EPROM, EEPROM, FRAM, MRAM, FLASH, SSD, NAND, NVDIMM, etc.) or any combination thereof. Appropriate computer instructions, commands, codes, information and / or data may be stored in the memory 1404. Computer instructions for executing or facilitating the execution of embodiments of the methods of the present invention may be stored in memory 1404. Processor 1402 and memory (storage) 1404 may be integrated or separate (and operably connected). Optionally, data processing system 1400 further includes one or more input devices 1406. Examples of such input devices 1406 include: keyboard, mouse, stylus, image scanner, microphone, tactile / touch input device (such as touch sensitive screen), image / video input device (such as camera), etc. The input device may be used to receive user input related to data range or data boundary. Optionally, data processing system 1400 further includes one or more output devices 1408. Examples of such output devices 1408 include: display (such as monitor, screen, projector, etc.), speaker, earphone, headset, printer, additive manufacturing machine (such as 3D printer), etc. The display screen may include LCD display screen, LED / OLED display screen or other suitable display screen, which may be touch sensitive or non-touch sensitive. The display may be used to display an image (two-dimensional or three-dimensional image) provided by processor 1402 (for example, may be generated by processor 1402). Data processing system 1400 may further include one or more disk drives 1412, which may include one or more of a solid-state drive, a hard disk drive, an optical drive, a flash drive, a tape drive, etc. A suitable operating system may be installed in data processing system 1400, for example, on disk drive 1412 or in memory 1404. Memory 1404 and disk drive 1412 may be operated by processor 1402.Optionally, the data processing system 1400 further includes a communication device 1410 for establishing one or more communication links (not shown) with one or more other computing devices (e.g., a server, a personal computer, a terminal, a tablet computer, a mobile phone, a watch, an IoT device, or other wireless computing devices). The communication device 1410 may include one or more of the following devices: a modem, a network interface card (NIC), an integrated network interface, an NFC transceiver, a ZigBee transceiver, a Wi-Fi transceiver, etc. Transceiver, radio frequency transceiver, cellular (2G, 3G, 4G, 5G, 5G or above or similar) transceiver, optical port, infrared port, USB connection or other wired or wireless communication interface. The transceiver can be implemented by one or more devices (integrated transmitter and receiver, independent transmitter and receiver, etc.). The communication link can be wired or wireless and is used to communicate commands, instructions, information and / or data. In one example, the processor 1402, memory 1404 (optionally input device 1406, output device 1408, communication device 1410 and disk drive 1412 (if any)) are directly or indirectly connected to each other via a bus, peripheral component interconnect (PCI) (such as PCI Express), universal serial bus (USB), optical bus or other similar bus structure. In one embodiment, at least some of these components can be connected wirelessly, such as through a network, such as the Internet or a cloud computing network. It will be appreciated by those skilled in the art that the data processing system 1400 is only an example, and in other embodiments, the data processing system 1400 may have a different configuration (e.g., including additional components, having fewer components, etc.).

[0125] Although not required, one or more of the embodiments described with reference to the accompanying drawings may be implemented as an application programming interface (API) or a series of libraries for use by developers, or may be included in another software application, such as a terminal or computer operating system or a portable computing device operating system. In one or more embodiments, since program modules include routines, programs, objects, components, and data files that assist in performing specific functions, it can be understood by a skilled person that the functionality of a software application may be distributed among multiple routines, objects, and / or components to achieve the same functionality required herein.

[0126] In addition, it is also understood that if the methods and systems of the present invention are implemented entirely or in part by a computing system, any suitable computing system architecture may be used. This includes stand-alone computers, network computers, dedicated or non-dedicated hardware devices. When the terms "computing system" and "computing device" are used, these terms are intended to include (but not limited to) any suitable arrangement of computers or information processing hardware capable of implementing the described functions.

[0127] Some embodiments of the present invention provide a three-dimensional medical image compression paradigm based on a video codec. In some embodiments, the video codec may include a data conversion module, an encoder, a decoder, and an inverse data conversion module. In some embodiments, the data format conversion module is intended to convert a three-dimensional medical image into a two-dimensional medical image sequence, and the data format conversion module includes a slice segmentation module for segmenting the three-dimensional medical image into two-dimensional slices and a bit depth conversion module for converting the segmented slices into a two-dimensional medical image sequence that can be compressed by the video codec. In some embodiments, an advanced video encoder is used to effectively compress the two-dimensional medical image sequence into a compact bitstream. In some embodiments, a video decoder is used to decompress the bitstream into a reconstructed two-dimensional medical image sequence. In some embodiments, the inverse data conversion module includes an inverse bit depth conversion module and a slice splicing module for converting the reconstructed two-dimensional medical image sequence into a reconstructed three-dimensional medical image. Through one or more of the above functions, some embodiments of the present invention can provide good performance in three-dimensional medical image compression.

[0128] Some embodiments of the present invention provide a three-dimensional medical image compression method based on video coding, which achieves competitive coding performance by converting the three-dimensional medical image into a sequence and compressing it using a video coding method. Some embodiments of the present invention provide a pipeline for converting a three-dimensional medical image into a two-dimensional sequence, in which the three-dimensional medical image is divided into slices, and then the slice set is converted into a sequence that can be processed by a video encoder using a bit depth conversion module. Some embodiments of the present invention provide bit depth conversion techniques, such as bit depth conversion based on bit interleaving segmentation and bit depth conversion based on displacement, for efficiently converting a slice set into a sequence. Some embodiments of the present invention provide a pipeline for inversely converting a decoded two-dimensional sequence into a three-dimensional medical image, wherein the decoded two-dimensional sequence is converted into a slice set using an inverse bit depth conversion, and then the slice set is spliced ​​into a reconstructed three-dimensional medical image. Some embodiments of the present invention can effectively convert a three-dimensional medical image into a sequence that can be compressed by an existing video codec through a data conversion module. Some embodiments of the present invention use advanced video coding methods for the converted sequence to achieve significant compression efficiency. Some embodiments of the present invention use an inverse data conversion module to effectively convert a decoded sequence into a reconstructed three-dimensional medical image. Through one or more of these functions, some embodiments of the present invention can provide competitive performance in three-dimensional medical image compression.Some embodiments of the present invention can be deployed on existing encoding platforms.

[0129] The scheme in some embodiments of the present invention can adapt to different bit rate requirements, and the performance is relatively good in high bit rate and low bit rate encoding scenarios. The method in some embodiments of the present invention can be configured or reconfigured. For example, the video encoder and video decoder can be configured by the user, and the target bit rate can also be specified or adjusted according to the application.

[0130] Tests were conducted using CT medical image and MRI medical image datasets to verify the performance of the embodiments of operations 500 and 1200. In one test, operations 500 and 1200 were tested on a CT medical image dataset using four target bit rates (2.0, 3.0, 4.0, and 5.0). It was found that operations 500 and 1200 can provide high-quality reconstructed medical images with a peak-to-noise ratio (PSNR) of up to 58.61 dB. In another test, operations 500 and 1200 were tested on a nuclear magnetic resonance medical image dataset using four target bit rates (0.8, 1.2, 1.6, and 2.0). It was found that operations 500 and 1200 can provide high-fidelity, high-quality reconstructed medical images with a peak-to-noise ratio (PSNR) of up to 63.2 dB.

[0131] Those skilled in the art will appreciate that the description and / or illustrated embodiments of the present invention may be varied and / or modified to provide other embodiments of the present invention. Therefore, the embodiments described and / or illustrated in the present invention should be considered in all respects to be illustrative and not restrictive. Examples of optional features of certain embodiments of the present invention are provided in the summary and description. Some embodiments of the present invention may include one or more of these optional features (some of which are not specifically described in the drawings). Some embodiments of the present invention may lack one or more of these optional features (some of which are not specifically described in the drawings). The systems and / or methods disclosed herein can be used to process other three-dimensional images (e.g., non-medical three-dimensional images).

Claims

1. A method for compressing a three-dimensional medical image, comprising: acquiring image data of a three-dimensional medical image; performing a data conversion operation to convert the image data of the three-dimensional medical image into video data of a sequence of frames, each frame corresponding to a respective two-dimensional image; as well as Performing a video encoding operation to encode the video data of the frame sequence to obtain encoded content data; The encoded content data can be used to reconstruct the three-dimensional medical image.

2. The method according to claim 1, wherein the step of performing a data conversion operation comprises: processing the image data of the three-dimensional medical image to obtain image data of a plurality of two-dimensional images; as well as A bit depth conversion operation is performed on the image data of the plurality of two-dimensional images to obtain video data of the frame sequence.

3. The method according to claim 2, wherein the step of processing the image data of the three-dimensional medical image to obtain image data of a plurality of two-dimensional images comprises: segmenting the three-dimensional medical image along an axis perpendicular to a segmentation plane to obtain a plurality of slices representable as the plurality of two-dimensional images; as well as For each pixel of each two-dimensional image of the plurality of two-dimensional images, the pixel is respectively mapped to an n-bit integer to obtain an n-bit pixel. The method according to claim 3 , wherein n is 8, 10 or 16. The method according to claim 3 , wherein the mapping is performed based on a data range or a data boundary of the three-dimensional medical image. The method of claim 3 , wherein the bit depth conversion operation is based on bit interleaving.

7. The method of claim 6, wherein the bit depth conversion operation based on bit interleaving comprises: Converting each of the n-bit pixels into corresponding two m-bit pixels; as well as forming two m-bit sequences based on the m-bit pixels of the plurality of two-dimensional images; The video data of the frame sequence includes the two m-bit sequences. The method according to claim 7 , wherein n=16 and m=8.

9. The method according to claim 7, wherein for each of the n-bit pixels, the conversion comprises: Applying a first bit mask to the n-bit pixel to extract the first m bits; applying a second bit mask to the n-bit pixel to extract a second m bits; Connecting the first m bits to the m-bit pixel; as well as The second m bits are connected to another m-bit pixel.

10. The method according to claim 9, wherein: The step of applying a first bit mask to the n-bit pixel to extract the first m bits comprises applying the first bit mask to the n-bit pixel to extract m even-numbered position bits; and The step of applying a second bit mask to the n-bit pixels to extract the second m bits comprises applying the second bit mask to the n-bit pixels to extract m odd-numbered position bits. The method of claim 3 , wherein the bit depth conversion operation is based on bit shifting.

12. The method of claim 11, wherein the bit depth conversion operation based on bit shifting comprises: Converting each of the n-bit pixels into a respective m-bit pixel; as well as forming an m-bit sequence based on the m-bit pixels of the plurality of two-dimensional images; The video data of the frame sequence includes the m-bit sequence. The method according to claim 12 , wherein n=16, m=8 or 10.

14. The method according to claim 12, wherein for each of the n-bit pixels, the converting comprises: performing a bit right shift on the upper m bits of the n-bit pixel, and The lower nm bits of the n-bit pixel are discarded or abandoned to form the m-bit pixel.

15. The method according to claim 3, wherein the data conversion operation further comprises: generating a data conversion flag indicating one or more attributes related to processing of the image data of the three-dimensional medical image and / or performing the bit depth conversion operation; as well as Performing an encoding operation to encode the data conversion flag bit to obtain encoded data conversion flag bit data; The encoded data conversion flag data may be used to facilitate reconstruction of the three-dimensional medical image.

16. The method of claim 1, wherein the video encoding operation is configured to: reducing inter-frame redundancy of the video data of the frame sequence; and / or Intra-frame redundancy of the video data of the sequence of frames is reduced.

17. The method of claim 1, wherein the video encoding operation is based on the Versatile Video Coding (VVC) standard.

18. The method of claim 1, wherein the video encoding operation is based on the High Efficiency Video Coding (HEVC) standard.

19. The method of claim 1, wherein the video encoding operation is based on the Advanced Video Coding (AVC) standard.

20. The method of claim 1, wherein the video encoding operation is based on the Audio Video Coding Standard (AVS).

21. The method of claim 1, wherein the three-dimensional medical image is a Magnetic Resonance Imaging (MRI) image.

22. The method of claim 1, wherein the three-dimensional medical image is a computed tomography (CT) image.

23. A system for compressing a three-dimensional medical image, comprising: one or more processors; as well as A memory storing one or more programs configured to be executed by the one or more processors, the one or more programs comprising instructions for performing or facilitating the performance of the method according to claim 1.

24. A method for reconstructing a three-dimensional medical image, comprising: performing a video decoding operation to decode the encoded content data to obtain video data of a sequence of frames, each frame corresponding to a respective two-dimensional image; as well as performing an inverse data conversion operation to convert the video data of the frame sequence into image data of a three-dimensional medical image; The image data of the three-dimensional medical image may be used to reconstruct the three-dimensional medical image.

25. The method of claim 24, wherein the video decoding operation is based on the Versatile Video Coding standard (VVC).

26. The method of claim 24, wherein the video decoding operation is based on the High Efficiency Video Coding (HEVC) standard.

27. The method of claim 24, wherein the video decoding operation is based on the Advanced Video Coding (AVC) standard.

28. The method of claim 24, wherein the video decoding operation is based on the Audio Video Coding Standard (AVS).

29. The method of claim 24, wherein performing the inverse data conversion operation comprises: performing an inverse bit depth conversion operation on the video data of the frame sequence to obtain image data of a plurality of two-dimensional images; as well as The image data of the plurality of two-dimensional images are processed to obtain the image data of the three-dimensional medical image.

30. The method of claim 29, wherein the inverse bit-depth conversion operation is based on inverse bit-interleave partitioning.

31. The method of claim 29, wherein the inverse bit-depth conversion operation is based on inverse bit shifting.

32. The method of claim 29, wherein: wherein the plurality of two-dimensional images comprises a plurality of n-bit pixels; as well as Wherein, the step of processing the image data of the plurality of two-dimensional images to obtain the image data of the three-dimensional medical image comprises: For each n-bit pixel of each of the plurality of two-dimensional images, inversely map the n-bit pixel to a pixel according to a data range or a data boundary; and The plurality of slices represented by the plurality of two-dimensional images are combined with pixels inversely mapped along an axis perpendicular to the segmentation plane to obtain the three-dimensional medical image.

33. The method of claim 29, wherein the inverse data conversion operation further comprises: performing a decoding operation to decode the encoded data conversion flag data, thereby obtaining a data conversion flag; the data conversion flag indicates one or more attributes related to processing the image data of the three-dimensional medical image and / or performing a bit depth conversion operation; The data conversion flag may be used to facilitate reconstruction of the three-dimensional medical image.

34. A system for reconstructing a three-dimensional medical image, comprising: one or more processors; as well as A memory storing one or more programs configured to be executed by one or more processors, the one or more programs comprising instructions for performing or facilitating the performance of the method according to claim 24.

Images