Medical image diagnosis support system, medical image processing device, and medical image processing method
The medical image diagnosis support system efficiently processes multiple tomographic images to generate an extracted image group for quick and accurate lesion confirmation, addressing the time-consuming manual comparison in existing systems.
Patent Information
- Application Number
- JP2020172095
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-10-12
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-10-12
AI Technical Summary
Existing medical image diagnosis systems require doctors to manually compare and process multiple images to confirm lesion candidates, which is time-consuming.
A medical image diagnosis support system that captures and processes multiple types of tomographic image series, including lesion candidate detection and extraction, to generate an extracted image group for efficient comparison.
Enables rapid confirmation of lesion candidates across different types of tomographic images, improving diagnostic efficiency and accuracy.
Smart Images

Figure 0007577502000001 
Figure 0007577502000002 
Figure 0007577502000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a medical image diagnosis support system that supports diagnosis by processing medical images after they are taken. [Background technology]
[0002] As medical imaging devices, such as MRI (Magnetic Resonance Imaging) devices, become more sophisticated, a large number of medical images are being acquired, which increases the burden on doctors who interpret the images to diagnose.
[0003] Patent Document 1 discloses an image diagnosis support device that uses a computer to detect lesion candidate regions from medical images and displays the medical images superimposed on post-processed images, thereby displaying the lesion candidate regions and their surrounding areas in a distinguishable form in order to reduce the burden on doctors who interpret images. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] JP 2012-166047 A Summary of the Invention [Problem to be solved by the invention]
[0005] The technology of Patent Document 1 allows lesion candidates detected by CAD (Computer Assisted Diagnosis: a computer-assisted diagnostic system) to be displayed on a medical image. However, when an image-reading doctor actually makes a diagnosis, he or she does not judge whether or not a lesion candidate is a lesion based only on the image in which the lesion candidate is displayed, but rather compares the lesion candidate image with the image on which the lesion candidate was detected or other contrast images, or performs image processing again to obtain other information. As a result, he or she focuses on an area corresponding to the lesion candidate in other contrast images, etc., and performs detailed observation. For this reason, it may take a long time for the image-reading doctor to make a definitive diagnosis.
[0006] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to enable radiologists to quickly check lesion candidate regions using a plurality of types of tomographic images. [Means for solving the problem]
[0007] In order to achieve the above object, the medical image diagnosis support system of the present invention includes a medical image capturing device that captures or generates multiple types of tomographic image series consisting of multiple slices of tomographic images for the same imaging target part of a subject, and a medical image processing device that processes the multiple types of tomographic image series. The medical image processing device includes a lesion candidate detection unit and an extracted image group generation unit. The lesion candidate detection unit detects a possible lesion area in one type of tomographic image series among the multiple types of tomographic image series. The extracted image group generation unit extracts tomographic images including the possible lesion area detected by the lesion candidate detection unit from one type of tomographic image series, generates an extracted image group, and also extracts tomographic images of slices corresponding to the slices of the extracted tomographic images from other types of tomographic image series and adds them to the extracted image group. Effect of the Invention
[0008] According to the present invention, the extracted image group includes not only cross-sectional images containing possible lesion areas, but also other types of cross-sectional images of the corresponding slices, allowing the radiologist to quickly confirm the areas of suspected lesions using multiple types of cross-sectional images. [Brief description of the drawings]
[0009] [Figure 1] 1 is a block diagram illustrating the overall configuration of a medical image diagnosis support system according to a first embodiment of the present invention. [Diagram 2] 1 is a functional block diagram of a medical image processing apparatus 100 of a medical image diagnosis support system according to a first embodiment. [Diagram 3] 4 is a flowchart for explaining the flow of processing by the medical image-processing apparatus 100 of the first embodiment. [Figure 4]4 is a flowchart for explaining the flow of processing by the medical image-processing apparatus 100 of the first embodiment. [Diagram 5] 2 is a diagram for explaining a series of tomographic images received by the medical image processing apparatus 100 of the first embodiment. FIG. [Figure 6] 2 is a diagram for explaining a lesion candidate 203 detected from a tomographic image by the medical image processing apparatus 100 of the first embodiment. FIG. [Figure 7] 2 is a diagram for explaining the configuration of an extracted image series 204 generated by the medical image-processing apparatus 100 of the first embodiment. FIG. [Figure 8] FIG. 11 is a functional block diagram of a medical image processing apparatus 100 of a medical image diagnosis support system according to a second embodiment. [Figure 9] 10 is a flowchart for explaining the flow of processing by the medical image-processing apparatus 100 of the second embodiment. [Figure 10] FIG. 11 is a functional block diagram of a medical image processing apparatus 100 of a medical image diagnosis support system according to a third embodiment. [Figure 11] 11 is a flowchart for explaining the flow of processing by the medical image-processing apparatus 100 of the third embodiment. [Figure 12] 11 is a flowchart for explaining the flow of processing by the medical image-processing apparatus 100 of the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0010] Hereinafter, preferred embodiments of the medical image diagnosis support system, medical image processing device, and medical image processing method according to the present invention will be described with reference to the attached drawings. In the following description and the attached drawings, components having the same functional configuration are designated by the same reference numerals, and duplicated explanations will be omitted.
[0011] <<Embodiment 1>> A medical image diagnosis support system 1 according to the first embodiment will be described.
[0012] FIG. 1 is a diagram showing a hardware configuration of a medical image diagnosis support system 1, and FIG. 2 is a functional block diagram of a medical image processing apparatus.
[0013] As shown in FIG. 1, a medical image diagnosis support system 1 includes a medical image processing device 100, a medical image pickup device 109, a medical image database 110, and an image reading device 111, which are connected via a network 108 so as to be capable of transmitting and receiving signals.
[0014] The medical imaging device 109 is a device that captures or generates multiple types of sectional image series consisting of multiple slices of sectional images for the same imaging target area of a subject, such as an MRI (Magnetic Resonance Imaging) device, an X-ray CT (computed tomography) device, or an ultrasound CT device.
[0015] The tomographic image series referred to here is a collection of tomographic images of multiple slices, and the multiple slices are preferably consecutive (adjacent) slices. The multiple types of tomographic image series are tomographic image series captured or generated so as to look differently even though they are the imaging target part of the same subject. For example, they include a tomographic image series in which a series of tomographic images are captured using different imaging methods, and a tomographic image series generated so that desired characteristics of the tissue are emphasized by applying image processing to the captured tomographic images. Specifically, they are tomographic image series in which a specified parameter of the subject's tissue is emphasized, and a tomographic image series in which a specified quantitative value of the tissue is used as a pixel value.
[0016] More specifically, when the medical imaging device 109 is an MRI device, examples of the multiple types of tomographic image series include a series of T1 (longitudinal relaxation) weighted images, T2 (transverse relaxation) weighted images, TOF (Time-of-Flight) images, PD (proton density) images, and FLAIR (fluid-attenuated inversion-recovery) images, each obtained for multiple slices. In the case of an X-ray CT device, examples of the series include dual energy imaging images captured by changing the energy of the irradiated X-rays. In the case of an ultrasonic CT device, examples of the series include images captured by changing the frequency of the irradiated ultrasonic waves and images reconstructed by changing the image reconstruction method.
[0017] It is preferable that the slice positions of the multiple types of tomographic image series are consistent with each other, but they do not necessarily have to be consistent. If the slice positions do not match, as will be described later, a calculation such as an interpolation calculation is performed to generate tomographic images in which the slice positions match.
[0018] Furthermore, the multiple types of sectional image series are not limited to images obtained by the same type of imaging device, but may be a combination of a sectional image series obtained by an MRI device and a sectional image series obtained by an X-ray CT device.
[0019] The medical image database 110 is a database system that stores medical images such as tomographic images acquired by the medical image capturing device 109. The medical images are stored in the medical image database 110 together with Digital Imaging and Communications in Medicine (DICOM) tags that include the imaging conditions of the medical images, etc.
[0020] On the other hand, the medical image processing apparatus 100 includes a lesion candidate detection unit 31 and an imaged image group series generation unit 32, as shown in FIG.
[0021] For one type of tomographic image series, when the tomographic image contains an area that is highly likely to be a lesion, lesion candidate detection unit 31 detects the possible lesion area (hereinafter referred to as a lesion candidate).
[0022] The image group generating unit 32 extracts a tomographic image including a lesion candidate from one type of tomographic image series to generate an extracted image group, and extracts a tomographic image of a slice corresponding to the slice of the extracted tomographic image from another type of tomographic image series to add it to the extracted image group. This makes it possible to generate an extracted image group that is a set of a tomographic image including a lesion candidate and other types of tomographic images, such as a tomographic image captured by a different imaging method for the same slice as the tomographic image, a tomographic image with a specified tissue parameter emphasized, or a tomographic image with a specified quantitative value as a pixel value. This allows the radiologist to compare the tomographic image of the lesion candidate with other types of tomographic images for the same slice, thereby providing support in determining whether the lesion candidate is a lesion candidate.
[0023] To realize the functions of the lesion candidate detection unit 31 and the representation image group series generation unit 32, the medical image processing device 100 includes a CPU (Central Processing Unit) 101, a memory 102, a storage unit 103, a network adapter 104, and a display unit 107, and these elements are connected to each other via a bus 105 so as to be capable of transmitting and receiving signals. Here, "capable of transmitting and receiving signals" means a state in which signals can be passed between each other or from one side to the other electrically or optically, regardless of whether it is wired or wireless.
[0024] CPU 101 loads programs stored in storage unit 103 and data required for executing the programs into memory 102 and executes them, thereby realizing the functions of lesion candidate detection unit 31 and extracted image group generation unit 32 by software.
[0025] Lesion candidate detection unit 31 is realized by, for example, AI (Artificial Intelligence). Specifically, a learning model such as a neural network is machine-trained in advance using learning data in which medical images are input data and lesion images are correct answer data, and is used as lesion candidate detection unit 31. The function of detecting lesion candidates is a known technology, and detailed description thereof will be omitted.
[0026] The process performed by the extracted image group generating unit 32 will be described later with reference to a flowchart.
[0027] The medical image processing apparatus 100 may further include a function for detecting the malignancy of a detected lesion candidate and a function for identifying an organ included in a medical image.
[0028] The storage unit 103 stores in advance programs to be executed by the CPU 101 and data necessary for executing the programs. For example, the storage unit 103 may be a recording device such as a hard disk drive (HDD) or a solid state drive (SSD), or a device that reads and writes data from and to a recording medium such as an IC card, an SD card, or a DVD.
[0029] The memory 102 stores programs executed by the CPU 101, intermediate results of arithmetic processing, and the like.
[0030] The CPU 101 receives a series of multiple types of tomographic images to be processed and part of the data required to execute the program from a medical imaging device 109, a medical image database 110, and an image interpretation device 111 via a network .
[0031] The display unit 107 is a device on which the results of program execution and the like are displayed, and specifically is a liquid crystal display or the like.
[0032] The input unit 106 is an operation device with which an operator issues operation instructions to the medical image processing device 100, and is specifically a keyboard, a mouse, or the like. The mouse may be another pointing device such as a track pad or a track ball. In addition, when the display unit 107 is a touch panel, the touch panel also functions as the input unit 106.
[0033] The network adapter 104 is for connecting the medical image processing apparatus 100 to a network 108 such as a LAN, a telephone line, or the Internet.
[0034] The image reading device 111 is a device equipped with a viewer function for reading medical images and a function as an input means for receiving instructions from an image reading doctor, and is equipped with a CPU, a storage unit, a display unit, and an input unit. The CPU of the image reading device 111 reads and executes a program in the storage unit to receive the extracted images generated by the medical image processing device 100, display the images on the display unit, and store the received images in the storage unit as necessary.
[0035] The image reading device 111 is also capable of receiving and displaying images from the medical image capturing device 109 and the medical image database 110 .
[0036] The image interpretation device 111 can also receive designation of data of a plurality of types of tomographic image series to be processed by the medical image processing device 100 from an image interpretation doctor via the input unit.
[0037] The image reading device 111 may be integrated with the medical image processing device 100 .
[0038] Next, the processing of the image processing device 100 will be described using the flowcharts of Fig. 3 and Fig. 4 and examples of tomographic images and the like of Fig. 5 to Fig. 6. Here, as an example, a case will be described in which the imaging target part of the tomographic image is the head, but the imaging target part is not limited to the head.
[0039] First, in step S1 of FIG. 3, the medical image processing apparatus 100 acquires, via the network 108, a plurality of types of tomographic image series (hereinafter also referred to as studies) 201, 202-A, and 202-B, which are obtained by capturing a plurality of slices of the same imaging target part of a subject, from the medical imaging apparatus 109 or the medical image database 110 (see FIG. 5). The study 200 includes a tomographic image series 201 for detecting lesion candidates and other tomographic image series 202-A and 202-B. These may be tomographic images of a type selected by an operator or a radiologist, or may be tomographic images of a type determined in advance. All of them are tomographic images of a plurality of slices of the same imaging target part of the same subject, which are captured by the medical imaging apparatus 109, or tomographic images generated from the captured tomographic images. Here, as an example, the tomographic image series 201 for which lesion candidate detection is performed is a T1 weighted image series, and the other tomographic image series 202-A and 202-B are a T2 weighted image series and a TOF image series.
[0040] Next, in step S2, lesion candidate detection unit 31 of medical image-processing device 100 detects lesion candidates (possible lesion regions) 203 for each tomographic image in tomographic image series 201 for lesion candidate detection in study 200. Fig. 6 shows an example of an image in which lesion candidates 203 are indicated in a tomographic image.
[0041] Next, in step S3, extracted image group generator 32 acquires the center coordinates (x, y, z) of lesion candidate 203. Furthermore, it calculates the size of lesion candidate 203 in the in-plane direction of the tomographic image, and, if lesion candidate 203 extends over multiple adjacent slices, the size of lesion candidate 203 in the slice thickness direction from the slice thickness.
[0042] Next, in step S4, the extracted image group generation unit 32 extracts cross-sectional images including each lesion candidate 203 obtained in step S3 and cross-sectional images of slices corresponding to the slices of these cross-sectional images from other types of cross-sectional image series 202-A, 202-B to generate an extracted image group.
[0043] Step S4 will be described in detail with reference to the flow chart of FIG.
[0044] First, in step S41, the extracted image group generation unit 32 assigns numbers (D1, D2, D3, etc.) to one or more lesion candidates 203 obtained in step S2, extracts the tomographic image 201-SD1 containing the first lesion candidate 203-D1 from the tomographic image series 201, and stores it as an extracted image group series 204, for example, in a specified memory area of the memory unit 103.
[0045] Next, in step S42, the extracted image group generation unit 32 determines whether or not there exists a tomographic image 202-A of slice SD1 that includes the center coordinates (x, y, z) of the lesion candidate 203-D1 obtained in step S3 for the first series 202-A of the other tomographic image series 202-A, 202-B.
[0046] If there is a tomographic image 202-A-SD1 that includes the coordinates (x, y, z) of the lesion candidate 203-D1, in step S43, the extracted image group generator 32 extracts this tomographic image 202-A-SD1.
[0047] In step S42 above, if there is no sectional image 202-A of slice SD1 containing the center coordinates (x, y, z) of lesion candidate 203-D1, the process proceeds to step S44, and the extracted image group generation unit 32 performs known processing such as MPR (Multi Planar Reconstruction) processing on the sectional image series 202-A, and creates an image corresponding to the same slice from the sectional images of other slices.
[0048] Then, in step S45, the tomographic image 202-A-SD1 extracted or created in step S43 or step S44 is added to the extracted image group series 204 (see FIG. 7).
[0049] In step S46, the above steps S42 to S45 for lesion candidate 203-D1 are repeated for the next tomographic image series 202-B. As a result, tomographic image 202-B-SD1 of a slice corresponding to the slice including lesion candidate 203-D1 from tomographic image series 202-B is added to extracted image group series 204 (see FIG. 7).
[0050] When the slices including the lesion candidate 203-D1 and the corresponding slices have been extracted and added to the extracted image group series 204 for all the cross-sectional image series 202-A and 202-B, the extracted image group generator 32 proceeds to step S47.
[0051] In step S47, extracted image group generator 32 repeats steps S41 to S46 for the next lesion candidate 203-D2, thereby extracting sectional image 201-D2 of a slice including lesion candidate 203-D2 and sectional images 202-A-SD2 and 202-B-SD2 of corresponding slices, and adding them to extracted image group series 204 (see FIG. 7).
[0052] The extracted image group generating unit 32 repeats steps S41 to S46 for all of the lesion candidates 203, and can ultimately generate an extracted image group series 204 for the study 200 (see FIG. 7).
[0053] Next, in step S5, the medical image processing apparatus 100 transmits the generated extracted image group series 204 to the image interpretation apparatus 111 via the network .
[0054] The image interpretation device 111 displays the received extracted image group series 201 side by side on the screen of the display device, for example, as shown in FIG.
[0055] By viewing the extracted image group series 201 displayed as in FIG. 7, the radiologist can compare the tomographic images (T1 weighted images) 201-SD1 and 201-SD2 including the lesion candidates 203-D1 and 203-D2 with other types of tomographic images (T2 weighted images) 202-A-SD1 and 202-A-SD2 and tomographic images (TOF images) 202-B-SD1 and 202-B-SD2 for the same slice on the screen. Therefore, the radiologist can easily check the tomographic image including the lesion candidate 203 and other types of tomographic images without preparing a special display device or performing additional operations such as reading the image of the corresponding slice from the medical image database 110, and the radiologist's interpretation accuracy can be improved. Therefore, the radiologist can be assisted in determining whether the lesion candidate is a lesion candidate.
[0056] In this embodiment, the extracted image group generation unit 32 extracts only the tomographic images of slices that include coordinates that match the central coordinates of the lesion candidate 203, but it may also extract multiple tomographic images centered on the coordinates of the lesion candidate.
[0057] Furthermore, the extracted image group generating unit 32 may set an FOV of any size based on the size of the lesion candidate in the slice thickness direction, and extract tomographic images included in the FOV.
[0058] <<Embodiment 2>> Next, the medical image diagnosis support system according to the second embodiment will be described with reference to the functional block diagram of FIG. 8 and the flowchart of FIG.
[0059] 8, the medical image processing device 100 of the medical image diagnosis support system of the second embodiment further includes a quantitative / enhanced image generating unit 33 in addition to a lesion candidate detecting unit 31 and an extracted image group generating unit 32. The quantitative / enhanced image generating unit 33 performs calculations using a plurality of types of tomographic images to generate quantitative images whose pixels have a predetermined quantitative value or enhanced images in which a predetermined parameter is enhanced for a plurality of slices.
[0060] 9, the processing of the medical image diagnosis support system of the second embodiment performs steps S11 and S12 instead of step S1 of acquiring a plurality of types of tomographic image series (study) 200 in the first embodiment. Also, steps S13 and S14 are performed instead of step S4 of generating an extracted image group series 204. Steps S2, S3, and S5 are the same as those of the first embodiment.
[0061] Specifically, in step S11, the medical image processing device 100 acquires, as a study, a number of tomographic images acquired by the medical image imaging device 109 executing a predetermined 3D gradient echo imaging sequence while changing parameters such as echo time (TE), repetition time (TR), flip angle (FA), and RF phase increment value (θ).
[0062] Next, in step S12, the quantitative / weighted image generating unit 33 performs a known Quantitative Parametric Mapping (QPM) analysis to calculate a tomographic image (quantitative value map) having pixels of any one of the quantitative values T1, T2*, PD, and B1 for multiple slices. This is set as a tomographic image series 201 for detecting lesion candidates. Alternatively, two or more types of quantitative value maps may be calculated and processed to generate a predetermined weighted image (e.g., a T2 weighted image). The type of the calculated quantitative value map may be predetermined as in the first embodiment, or may be specified by an operator or a radiologist.
[0063] In step S 2 , lesion candidate detection unit 31 detects lesion candidates 203 in the tomographic images of tomographic image series 201 .
[0064] In step S3, the extracted image group generating unit 32 acquires the center coordinates (x, y, z) of the lesion candidate 203.
[0065] In step S13, the extracted image group generation unit 32 extracts tomographic images including the lesion candidate 203 from the tomographic image series 201 generated in step S12, and stores them as an extracted image group series 204, for example, in a specified memory area of the memory unit 103.
[0066] In step S14, the quantitative / enhanced image generating unit 33 creates another type of quantitative value map or another type of enhanced image from the study acquired in step S11 by QPM analysis only for the slices corresponding to the tomographic images extracted in step S13. The created another type of quantitative value map or another type of enhanced image is added to the extracted image group series 204. This is repeated until all predetermined types of quantitative value maps or enhanced images are generated and added to the visualized image group series 204.
[0067] Next, in step S5, the medical image processing apparatus 100 transmits the generated extracted image group series 204 to the image interpretation apparatus 111 via the network .
[0068] As described above, according to the second embodiment, since the desired quantitative value map and enhanced image can be generated by the QPM analysis, the operator or the radiologist can specify the desired type of image as the tomographic image series 201 for detecting lesion candidates and another type of tomographic image series 202. Therefore, an extracted image group series consisting of the type of tomographic images desired by the operator or the radiologist can be generated, which is expected to improve the efficiency of the examination.
[0069] In addition, by generating a different type of quantitative value map or a different type of enhanced image by QPM analysis only for slices corresponding to the tomographic image from which the lesion candidate was extracted, the time required to generate the tomographic image by QPM analysis can be shortened.
[0070] In the above description, a gradient echo imaging sequence is used as the imaging sequence for QPM. However, other imaging sequences using a spin echo or a pre-pulse may also be used.
[0071] The hardware configuration of the medical image diagnosis support system of the second embodiment is the same as that of the first embodiment, so a description thereof will be omitted.
[0072] <<Embodiment 3>> Next, the medical image diagnosis support system according to the third embodiment will be described with reference to the functional block diagram of FIG. 10 and the flowcharts of FIG. 11 and FIG.
[0073] 10, the medical image processing device 100 of the medical image diagnosis support system of the third embodiment further includes a three-dimensional image generating unit 34 in addition to the lesion candidate detecting unit 31 and the extracted image group generating unit 32 described in the first embodiment. The three-dimensional image generating unit 34 generates a three-dimensional image for one or more of the tomographic image series 201, 202-A, and 202-B, and sets the rotation axis at the center of the lesion candidate 203 or at a position a predetermined distance away from the center of the lesion candidate 203 so that the lesion candidate 203 can be easily observed. The generated three-dimensional image is added to the extracted image group series 204.
[0074] The processing of the medical image processing apparatus 100 will be described using the flowcharts of Figures 11 and 12. In the third embodiment, the same processing as in the flow of Figure 3 of the first embodiment is performed, but the details of the processing of step S4 are different, and in the third embodiment, step S4 is processed as in the flow of Figure 11. In the flow of Figure 11, step S50 is performed between steps S45 and S46 of the flow of Figure 4 of the first embodiment (Figure 11). In this step S50, three-dimensional images are created for each of the multiple types of tomographic image series 201, 202-A, and 202-B.
[0075] Specifically, steps S41 to S45 are performed in the same manner as in the first embodiment, and after the process of step S45, a three-dimensional image is created in step S50.
[0076] Details of step S50 will be described with reference to the flow chart of FIG.
[0077] In step S441, the three-dimensional image generating unit 34 calculates the rotation center coordinates for rotating the three-dimensional image. This may be the same as the center coordinates of the lesion candidate 203, or may be set at a position shifted by a predetermined distance in consideration of ease of observation of the lesion candidate.
[0078] Next, in step S442, the three-dimensional image generating unit 34 calculates the direction of the rotation axis, which is set, for example, to a direction perpendicular to the slice (tomographic image).
[0079] Next, in step S443, the three-dimensional image generator 34 calculates the Field Of View (FOV) of the three-dimensional image. The FOV may be the same as the size of the lesion candidate 203 calculated in step S3 of the first embodiment, or may be a range twice the size of the lesion candidate in consideration of ease of observation.
[0080] Then, in step S444, the three-dimensional image generating unit 34 generates volume data by combining each slice tomographic image data of the tomographic image series 201 while taking into account the slice position information. Similarly, volume data is generated for the tomographic image series 202-A and 202-B.
[0081] In step S445, the rotation center coordinates, rotation center axis, and FOV calculated in steps S441 to S443 are set for each volume data, and a three-dimensional processed image is created. As the three-dimensional processed image, a predetermined projection image such as maximum intensity projection or a volume rendering image is generated. The generated image is added to the extracted image group series 204.
[0082] According to the third embodiment, the image interpreting doctor is able to observe lesion candidates from multiple angles, which leads to improved image interpretation accuracy.
[0083] Note that here, in step S445, three-dimensional processed images such as projection images are added to the extracted image group series 204, but it is also possible to add volume data in which the rotation center coordinates, rotation center axis, and FOV are set to the extracted image group series 204, and have the radiologist perform the rotation and projection operations.
[0084] In the third embodiment, three-dimensional processed images are created for all the tomographic image series 201, 202-A, and 202-B, but three-dimensional processed images may be created for only one or two or more tomographic image series.
[0085] Although the embodiments of the present invention have been described above, the present invention is not limited to these.
[0086] In each of the above-mentioned embodiments, the configuration in which each of the functional units 31 to 34 of the medical image processing device 100 is realized by software has been described, but it is also possible to realize a part or all of each of the functional units 31 to 34 of the medical image processing device 100 by hardware. For example, the medical image processing device 100 may be configured using a custom IC such as an ASIC (Application Specific Integrated Circuit) or a programmable IC such as an FPGA (Field-Programmable Gate Array), and a circuit design may be performed to realize each of the functional units 31 to 34. [Explanation of symbols]
[0087] 1... medical image diagnosis support system, 31... lesion candidate detection unit, 32... visualization image group generation unit, 33... quantitative / enhanced image generation unit, 34... three-dimensional image generation unit, 100... medical image processing device, 101... CPU, 102... memory, 103... storage unit, 104... network adapter, 105... bus, 106... input unit, 107... display unit, 108... network, 109... medical image capturing device, 110... medical image database, 111... image reading device, 200 ...study, 201... series of tomographic images for detecting lesion candidates, 201-SD1, 201-SD2... tomographic images including lesion candidates, 202-A, 202-B... series of tomographic images, 202-A-SD1, 202-A-SD2... different types of tomographic images of slices corresponding to the tomographic images including the lesion candidates, 202-B-SD1, 202-B-SD2... different types of tomographic images of slices corresponding to the tomographic images including the lesion candidates, 204... series of extracted images
Claims
1. A medical imaging apparatus for capturing a plurality of types of tomographic images of a plurality of slices of a same imaging target part of a subject by a predetermined imaging method, and a medical image processing apparatus for processing the plurality of types of the tomographic images, The medical image processing device includes a quantitative / enhanced image generating unit, a lesion candidate detecting unit, and an extracted image group generating unit; the quantitative / enhanced image generating unit performs a calculation using the plurality of types of the tomographic images to generate quantitative images having pixels with a predetermined quantitative value or enhanced images in which a predetermined parameter is enhanced for the plurality of slices; the lesion candidate detection unit detects a possible lesion area in the quantitative image or the enhanced image of the plurality of slices generated by the quantitative / enhanced image generation unit; The medical image diagnosis support system is characterized in that the extracted image group generation unit extracts the quantitative image or the enhanced image containing the possible lesion area detected by the lesion candidate detection unit to generate an extracted image group, and causes the quantitative / enhanced image generation unit to generate a different type of quantitative image or enhanced image only for slices corresponding to the extracted quantitative image or enhanced image, and adds the generated different type of quantitative image or enhanced image to the extracted image group.
2. A medical image processing apparatus including a quantitative / enhanced image generating unit, a lesion candidate detecting unit, and an extracted image group generating unit, The quantitative / enhanced image generating unit receives from a connected external device a plurality of types of tomographic images, each of which is captured by a predetermined imaging method for a plurality of slices of the same imaging target part of the subject, and performs a calculation using the plurality of types of tomographic images to generate a quantitative image having a pixel with a predetermined quantitative value or an enhanced image in which a predetermined parameter is enhanced for the plurality of slices; the lesion candidate detection unit detects a possible lesion area in the quantitative image or the enhanced image of the plurality of slices generated by the quantitative / enhanced image generation unit; The medical image processing device is characterized in that the extracted image group generation unit extracts the quantitative image or the enhanced image containing the possible lesion area detected by the lesion candidate detection unit to generate an extracted image group, and causes the quantitative / enhanced image generation unit to generate a different type of quantitative image or enhanced image only for slices corresponding to the extracted quantitative image or enhanced image, and adds the generated different type of quantitative image or enhanced image to the extracted image group.
3. receiving from an external device a plurality of types of tomographic images, each of which is captured by a predetermined imaging method, for a plurality of slices of the same imaging target part of the subject; performing a calculation using the plurality of types of tomographic images to generate quantitative images having pixels with a predetermined quantitative value or enhanced images in which a predetermined parameter is enhanced for the plurality of slices; Detecting potential lesion regions in the quantitative or enhanced images of a plurality of slices; Extracting the quantitative image or the highlighted image including the possible lesion area to generate an extracted image group; generating another type of quantitative image or enhanced image only for slices corresponding to the extracted slices of the quantitative image or enhanced image; The generated quantitative image or enhanced image of another type is added to the extracted image group. A medical image processing method comprising:
Citation Information
Patent Citations
Magnetic resonance imaging instrument
JP2005028017A
Medical diagnostic imaging supporting instrument and image processing program
JP2007007256A
Program, information recording medium, and freehand drawing similarity judging device
JP2007072569A
Comparative diagnostic reading supporting apparatus
JP2007159934A
Magnetic resonance imaging apparatus and image processing method therefor
JP2011024722A
