WO2010140588A1 - 病理画像診断システム、病理画像診断方法、病理画像診断プログラム - Google Patents
病理画像診断システム、病理画像診断方法、病理画像診断プログラム Download PDFInfo
- Publication number
- WO2010140588A1 WO2010140588A1 PCT/JP2010/059268 JP2010059268W WO2010140588A1 WO 2010140588 A1 WO2010140588 A1 WO 2010140588A1 JP 2010059268 W JP2010059268 W JP 2010059268W WO 2010140588 A1 WO2010140588 A1 WO 2010140588A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- area
- pathological
- image
- importance
- divided
- Prior art date
Links
- 230000001575 pathological effect Effects 0.000 title claims abstract description 129
- 238000002405 diagnostic procedure Methods 0.000 title description 3
- 238000005259 measurement Methods 0.000 claims abstract description 37
- 230000005540 biological transmission Effects 0.000 claims abstract description 34
- 239000000284 extract Substances 0.000 claims abstract description 5
- 238000000605 extraction Methods 0.000 claims description 42
- 238000010827 pathological analysis Methods 0.000 claims description 15
- 230000006835 compression Effects 0.000 claims description 7
- 238000007906 compression Methods 0.000 claims description 7
- 238000003745 diagnosis Methods 0.000 abstract description 23
- 210000001519 tissue Anatomy 0.000 description 96
- 238000004364 calculation method Methods 0.000 description 70
- 230000006870 function Effects 0.000 description 41
- 238000012545 processing Methods 0.000 description 28
- 238000000034 method Methods 0.000 description 12
- 238000004891 communication Methods 0.000 description 11
- 238000003860 storage Methods 0.000 description 9
- 210000003855 cell nucleus Anatomy 0.000 description 7
- 239000006185 dispersion Substances 0.000 description 7
- 238000013507 mapping Methods 0.000 description 7
- 238000010186 staining Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- -1 length average Substances 0.000 description 4
- 206010028980 Neoplasm Diseases 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 3
- 201000011510 cancer Diseases 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- WZUVPPKBWHMQCE-UHFFFAOYSA-N Haematoxylin Chemical compound C12=CC(O)=C(O)C=C2CC2(O)C1C1=CC=C(O)C(O)=C1OC2 WZUVPPKBWHMQCE-UHFFFAOYSA-N 0.000 description 2
- 206010061218 Inflammation Diseases 0.000 description 2
- 210000000805 cytoplasm Anatomy 0.000 description 2
- 230000004054 inflammatory process Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000007170 pathology Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000004663 cell proliferation Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- YQGOJNYOYNNSMM-UHFFFAOYSA-N eosin Chemical compound [Na+].OC(=O)C1=CC=CC=C1C1=C2C=C(Br)C(=O)C(Br)=C2OC2=C(Br)C(O)=C(Br)C=C21 YQGOJNYOYNNSMM-UHFFFAOYSA-N 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000007490 hematoxylin and eosin (H&E) staining Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 238000004171 remote diagnosis Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
Definitions
- the present invention relates to a pathological image diagnostic system, a pathological image diagnostic method, and a pathological image diagnostic program for outputting a pathological tissue image to be a target of pathological diagnosis.
- Pathological diagnosis is generally performed by a pathologist looking at a specimen on a slide glass with a microscope.
- a microscope since the field of view of the microscope is limited, it is necessary to change the place to see in order to see the whole and see every corner. Therefore, instead of a microscope, an image scanner of a type that captures a slide as an electronic image is used.
- the maximum field of view of the microscope corresponds to the monitor resolution of the PC (personal computer), but the entire area of the tissue is the monitor of the PC unless the tissue is small and the magnification during image display is small. Dozens of resolutions, hundreds of times.
- the diagnosis result is not necessarily determined by the whole tissue, and the diagnosis result is often influenced by a partial region of the tissue image. Also, distant parts of the tissue may exhibit different appearances. Therefore, if the tissue is divided into certain divisional areas and processing for diagnosis is performed, this processing can be aborted when the diagnosis result is determined, which is efficient.
- dividing a living body (pathology) tissue image obtained by imaging a living tissue into rectangular blocks and diagnosing each block is disclosed, for example, in a method disclosed in Japanese Patent Application Laid-Open No. 06-003601.
- rectangular division of a pathological tissue image is often considered from the viewpoint of transmission efficiency.
- the relationship between the focused segment area to be diagnosed in the pathological tissue image and the adjacent areas such as upper, lower, left, and right adjacent to the focused segment area may be emphasized as described above.
- the diagnosis efficiency is bad, but also there may be a disadvantage that information obtained from an area across different blocks or the entire image is overlooked.
- image communication apparatuses on the transmission side and the reception side are provided, and the image communication apparatus on the transmission side sets priorities for each of the divided images obtained by dividing the original image, and There has been disclosed a method of performing image communication with high transmission efficiency by transmitting and receiving each divided image by increasing the amount of image data and reducing the amount of image data of divided images with low priority (Japanese Patent Application Laid-Open No. 2008-101118).
- Patent Document 2 a method is disclosed in which a captured fundus image is divided and priority is set to each divided image, and an order of displaying each divided image is set based on the priority to shorten a diagnosis time.
- Patent Document 3 discloses a method of making it selectable.
- the diagnosis can not be performed until transmission of an important part (image area) for diagnosis is completed. It may not be necessary to make a detailed diagnosis, that is, it may be necessary to wait for an image area of low importance to be transmitted, resulting in the deterioration of the overall pathological diagnosis efficiency. There is a disadvantage.
- the present invention provides a pathological image processing system, a pathological image processing method, and a pathological image processing program that improves the disadvantages of the above related art and outputs a pathological tissue image that enables efficient image diagnosis.
- the purpose is to do.
- a pathological image in which a living tissue is imaged is divided into preset division areas, and each division area is preset according to a criterion of pathological diagnosis.
- An importance setting means for setting and a division area in which the area where the biological tissue is imaged in the division area has a fixed ratio is extracted, and the measurement index value and the classification importance are applied to the division area It is characterized in that it comprises: segmented area extracting means for attaching and sending; and segmented area output displaying means for outputting and displaying a pathological image consisting of the sent out segmented areas.
- a pathological image obtained by imaging a living tissue is divided into preset division areas, and a measurement index value according to a criterion of pathological diagnosis is set for each division area. Based on the measurement index value of the divided area and the measurement index value in the divided area adjacent to each divided area, the importance level for pathological measurement is set to each divided area, and the biological tissue is imaged in the divided area It is characterized in that a divided area in which the area shows a certain ratio is extracted, the measurement index value and the divided importance are associated with the divided area, and the pathological image composed of the divided area is output and displayed. There is.
- the community management program divides a pathological image in which a living tissue is imaged into division areas set in advance, and has a function of calculating measurement index values according to the criteria of pathological diagnosis for each division area, A function of setting the importance of the pathological measurement in each of the divided areas based on the measured index value of each divided area and the measured index value in the divided areas adjacent to each of the divided areas; It is characterized in that it has a function of extracting a divided area in which the imaged area has a fixed ratio, and a function of being executed by a computer set in advance.
- diagnostic imaging can be performed efficiently.
- FIG. It is a schematic block diagram which shows one Embodiment of the data processing apparatus in the pathological image diagnostic system disclosed in FIG. It is a schematic block diagram which shows one Embodiment of the image display apparatus in the pathological image diagnostic system disclosed in FIG. It is a schematic block diagram which shows one Embodiment of the image transmission apparatus (area encoding apparatus) in the pathological image diagnostic system disclosed to FIG. It is a flowchart which shows the whole operation processing step in the pathological image diagnostic system disclosed in FIG. It is a flowchart which shows the whole operation processing step in the pathological image diagnostic system disclosed in FIG. It is a flowchart which shows the whole operation processing step in the pathological image diagnostic system disclosed in FIG. It is a flowchart which shows the whole operation processing step in the pathological image diagnostic system disclosed in FIG. It is a flowchart.
- the pathological image diagnostic system 1 of the first embodiment includes a tissue extraction unit 10 for extracting an image area (tissue area) in which a tissue is imaged in a pathological tissue image (S10) sent from the outside.
- An index amount calculation device 20 which receives rectangular blocks S20 which are image areas extracted by the tissue extraction means 10 and calculates an index amount applied to a pathological tissue set in advance for each rectangular block S20; and an index of each rectangular block Importance calculation means 30 for calculating and ordering the importance of quantities, and image display means 40a for outputting and displaying the rectangular block S20 sent from the importance calculation means 30 in association with the index value S40 calculated in advance and the importance.
- the tissue extraction unit 10 extracts a tissue region in a pathological tissue image input from the outside of the pathological image diagnostic apparatus 1.
- the tissue extraction unit 10 detects a tissue region (image region) in which a tissue is imaged as a stained region in a pathological image.
- the tissue extraction unit 10 creates a monochrome image of a specific color (or luminance) from the pathological tissue image, and generates a binarized image creation algorithm (see FIG. 2) using a threshold set in advance for this monochrome image.
- the binarization processing is performed by the discriminant analysis method (Binarization of Otsu).
- the tissue extracting means 10 divides the pathological tissue image S10 into rectangular meshes, and processes only blocks (rectangular blocks) including tissue regions (pixels) having a predetermined number of pixels or more. Extract as a region. Note that it is not necessary to actually draw a mesh in the pathological tissue image, and it may be set to store one vertex of each set rectangular block and the vertical and horizontal sizes of the rectangular block.
- the index amount calculation means 20 calculates the index amount of the rectangular block S20 extracted by the tissue extraction means 10 as described above.
- the index amount calculation means 20 includes index amount calculation units 201, 202, 203,... Which calculate pathological indexes simultaneously and in parallel.
- the index amount calculation unit 20 including the index amount calculation units 201, 202, and 203 is exemplified, but the number of index usages set by the index amount calculation unit 20 is not limited to this.
- imaging diagnosis of pathological tissue by a pathologist looks at the whole tissue first at low magnification, finds cell nucleus, a portion stained in deep cytoplasm, and a portion having a large change in staining intensity, It is a good practice to use a sparse and dense search method to switch and look at each cell nucleus in detail. Therefore, in the present embodiment, as an index amount related to pathological tissue diagnosis, the image of the pathological tissue is divided into block units, and the in-block index amount for examining the global staining density, the global staining change Introduce [b] inter-block index quantity to check, [c] total block index quantity to check if staining concentration or change is an artificial reason.
- the in-block index amount is an index amount which can be calculated from the information of each pixel constituting the rectangular block.
- This indicator amount is an abnormal cell nucleus caused by a disease such as inflammation or cancer, an absolute concentration of a specific color in a region corresponding to the abnormal cell nucleus group, or an amount related to the shape of a pixel block extracted in a specific color, for example, a rectangular block Inner moment, Hu moment, discriminant analysis threshold, pixel number of specific color, number of pixel blocks of specific color, area average, dispersion, length average, dispersion, circularity average, dispersion, etc. are used.
- a processed image is generated by performing signal processing such as Fourier transform or image processing such as edge extraction filter on the original pathological tissue image, and the moment, the Hu moment, and the discrimination analysis threshold value in the processed image.
- the index amount may be calculated based on the number of pixels of a specific color, the number of pixel blocks of a specific color, area average, dispersion, length average, dispersion, circularity average, dispersion, and the like.
- the moments are set as image luminance values I (x, y) in a target image of image processing (here, a pathological tissue image), and m and n as dimensions in the x and y directions. It can be obtained by the calculation formula of 1]. Also, one-dimensional Hu moment sets m 2 , 0 , m 0 , 2 as a normalized central moment, But you are asked. Circularity, S area and L perimeter Determined by
- an HE (Hematoxylin-Eosin) stained image as a pathological tissue image is considered.
- the portion of the cell nucleus in this HE has the property of staining with hematoxylin blue purple and the portion of the cytoplasm with eosin red purple.
- indicators such as Hu moment of color signals of red, blue or its complementary cyan (light blue) and magenta (purple), discriminant analysis threshold, and number of pixel blocks become important indicators for pathological diagnosis. . Therefore, in the present embodiment, the index amount (in-block index amount) is calculated from the value obtained by quantifying, for example, the Hu moment, the discrimination analysis threshold value, and the number of pixels of a specific color.
- the inter-block index amount can be obtained by calculating the degree of change with the neighboring blocks based on the intra-block index amount of (a).
- This index amount reflects changes in the periphery of the specific tissue region corresponding to the proliferation process of the cell nucleus where the pathological diagnosis is emphasized, and its surrounding region, for example, using the block unit laplacian or the dispersion with the peripheral block Be
- the number of extracted blocks adjacent to the index value of the target block is obtained from, for example, the sum of all the index amounts (index values) of the extracted blocks adjacent to the target block (FIG. 5A). It is set as the value which subtracted the value which multiplied by (Equation 3).
- the target block (a) has seven adjacent blocks among the adjacent eight adjacent blocks (upper and lower, left and right diagonal rectangular blocks), so Laplacian is calculated based on [Equation 4] shown below Be done.
- the total block index amount is a value calculated based on the in-block variance value [a], which is a statistic (average, variance, histogram, etc.) calculated from index (quantity) values of rectangular blocks in the entire image ).
- This total block index amount is for verifying whether the above [a] and [b] are caused for pathological reasons or artificial reasons not related to the pathology, and for each rectangular block, Although it is not related to the calculation of the priority, it is used supplementary to issue a warning as an abnormal image due to a failure in slide creation or shooting.
- the one or more index amounts calculated above are sent to the importance degree calculation means 30 as scalar values or vector values.
- the importance degree calculation means 30 orders index amounts which are scalar values or vector values attached to each block.
- the importance can be set as follows based on the pathological diagnostic reason.
- color uniformity within the block for example, in the case of denseness or density of cell nuclei, the higher the color value or density of the pixel, the higher the degree of importance because there is the possibility of diseases such as inflammation and cancer.
- block homogeneity for example, if there is a large amount of large circular pixel blocks (in the case of HE staining, a large blue circular block), there is a possibility of cancer cells and the degree of weight is high.
- color values, density, etc. may be used as they are, or may be divided and ranked in a certain range.
- the order may be a correspondence with unordered groups in addition to the order of magnitude (full order). Assuming that the index quantity is x, the rank or group is y, and the corresponding mapping is f, When the order is full order, f is a function, and the group y and the mapping (function) f have a relation shown in the following [Equation 6].
- the index amount calculation means 20 is configured by a single index amount calculation unit 201, since the index amount is a scalar value, ranking can be performed simply by the magnitude of the value.
- the index amount calculation means 20 is composed of a plurality of index amount calculation units (201, 202,...), It is necessary to give a mapping (function).
- how to give the mapping (function) changes depending on the selection of the index quantity and which index is to be emphasized. If the importance of all the indices is equivalent, the sum of all the indices is calculated.
- the function f may be a map that belongs to a specific order or group regardless of other indices when a certain index falls within a specific numerical value, in addition to the function that can be expressed by a mathematical expression.
- f may be set by machine learning such as a neural network or a support vector machine.
- the importance calculation means 30 associates the ranks (groups) and the index values with the blocks and sends them to the screen display means 40a.
- the image display means 40a associates the order of the blocks input from the importance degree calculation means 30 and the index value S40 with the rectangular block of the original image, and outputs and displays the result on an output device such as a display set in advance.
- FIG. 6 an example of a display screen of the pathological image S10 and an interface are shown in FIG.
- the dark gray block on the left in FIG. 6 and the third row of importance C from the top of the importance column in the table on the right of FIG. 6 are associated.
- the degree of importance may be the same, and in this example, there are two C and F.
- multiple importance levels may be set, in which case, switching is made in the importance level menu.
- the index amount (value) applied to the pathological tissue is set and calculated in each rectangular block (region) set in the pathological tissue image, thereby locally generating the pathological tissue image.
- Can be assigned a quantified value and the original pathological tissue image can be displayed quantitatively according to the importance of the pathological tissue.
- the user can quantitatively understand the state of the entire pathological tissue image, so that the pathological image diagnosis can be performed efficiently and quickly, and furthermore, the diagnosis time can be shortened.
- the importance as a pathological tissue is set by giving an index value to each set rectangular block, so that pathologically important that is grasped across a plurality of blocks including adjacent blocks. It is possible to easily output and display an area of interest.
- Embodiment 1 [Description of Operation of Embodiment 1] Next, an outline of the operation of the pathological image diagnostic system in the first embodiment will be described.
- the pathological image in which the biological tissue is imaged is divided into previously set divided regions, and the index amount calculation unit 20 sets measurement index values according to the criteria of pathological diagnosis for each divided region (index amount setting step), Then, the importance degree calculation unit 30 sets the importance degree for the pathological measurement in each of the divided areas based on the measurement index value of each of the divided areas and the measurement index value in the divided areas adjacent to each of the divided areas (important Degree setting process).
- the importance degree calculation means 30 extracts a divided area in which the area where the biological tissue is imaged shows a constant ratio among the divided areas (tissue area determination and extraction step), and the measurement index is calculated for this divided area.
- the value and the classification importance are associated with each other and sent out, and the image display means 40a outputs and displays the pathological image consisting of the divided regions.
- the execution content of the index amount setting step, the importance degree setting step, and the tissue area determination and extraction step may be programmed to be executed by a computer.
- the program is recorded on a recording medium and is subject to commercial transaction.
- a stained pathological image (original pathological image) is input to the tissue extraction unit 10 (step S1).
- the tissue extraction unit 10 creates a monochrome image of the luminance component of the pathological image from the pathological image (step S2), and generates an image of the tissue image area from the generated monochrome image using an algorithm such as Otsu binarization.
- the extraction is performed (step S3: tissue region extraction function) and transmitted to the index use calculation unit 20.
- the index utilization calculating unit 20 divides the original pathological image and the corresponding binarized image into blocks (rectangular blocks) (step S4), and calculates an index amount in each image block including the tissue image.
- Step S5 Index Amount Calculation Function
- Each block is transmitted to the importance degree calculation means 30.
- the importance calculation means 30 calculates the importance of each block by mapping (function) that relates the index amount and the importance from the single or plural index amounts (step S6: importance calculation function).
- the importance degree calculation means 30 outputs and displays a pathological image on a display device such as a monitor set in advance, based on the association information between each block of the original pathological image and the importance degree information (step S7).
- a display device such as a monitor set in advance
- output display is performed in which each block is associated with importance information and index amount information.
- the execution contents of the tissue area extraction function, index amount calculation function, and importance degree calculation function may be programmed to be executed by a computer provided in advance in the pathological image diagnostic apparatus 1.
- a tissue extraction device 100 and an index amount calculation device 200 each of which is provided as a tissue extraction unit 10, an index amount calculation unit 20, an importance calculation unit 30, and an image display unit 40a provided in the pathological image diagnostic device 1.
- the importance degree calculation device 300 and the image display device 400a may be connected by a communication line or the like.
- a storage device 251 such as a memory or storage disk for storing image data (block) input as input data, index amount information, and importance degree information, and image data
- the data processing apparatus 250 may be configured to include an arithmetic unit 252 such as a CPU that performs extraction of tissue regions in (blocks), index amounts of each block, and calculation of importance.
- the image display device 400a is a memory, disk, etc. for storing input data including image information (block data), importance information, index value information sent from the importance calculation means 30.
- Storage device 351 an input device 352 such as a keyboard or pointing device to which the user instructs to input, a monitor 353 outputting and displaying the processing result by the CPU 352, and calculation of the result to be displayed on the monitor by input data and input instruction
- an arithmetic device 352 such as a CPU may be employed.
- the image transmission means 40b outputs (transmits) the block data S50b to the communication line in the descending order of the importance order based on the importance order of the blocks and the index value S40 sent from the importance calculation means 30 (block output Send function). Further, when the reception of the block data of the importance required by the reception side is completed, the image transmission means 40b receives the stop request signal S60 from the reception side terminal, and transmits the block data based on the stop request signal S60. Stop (data transmission stop function).
- the tissue extracting means 10, the index amount calculating means 20, and the importance degree calculating means 30 in the second embodiment operate in the same manner as in the first embodiment (FIG. 1) described above.
- a pathological image subjected to staining processing is input to the tissue extraction means 10 (step A1).
- the tissue extraction unit 10 creates a monochrome image of the luminance component of the pathological image from the pathological image (step A2), and roughly generates a tissue image region from the generated monochrome image using an algorithm such as Otsu binarization. (Step A3: tissue region extraction function), and transmitted to the index use calculation unit 20.
- a blurring filter may be inserted to suppress the influence of noise outside the tissue region.
- the index usage calculation unit 20 divides the original pathological image and the binarized image corresponding thereto into blocks (rectangular blocks) (step A4), and calculates an index amount in each image block including the tissue image.
- Step A5 Index Amount Calculation Function
- Each block is transmitted to the importance degree calculation means 30.
- the importance calculation means 30 calculates the importance of each block by mapping (function) that relates the index amount and the importance from the single or plural index amounts (step A6: importance calculation function).
- the importance calculation means 30 sorts the blocks in the descending order of the calculated importance (step A7), and transmits the blocks to the image transmission means 40b.
- the image transmission means 40b adds importance information to each block data, and transmits block data to the receiving terminal in descending order of importance (step A8).
- the receiving side transmits a stop request signal requesting stop of transmission of the block data to the image transmitting means 40b which is the transmitting side.
- the image transmission means 40b receives this stop signal (step A9: YES)
- the image transmission means 40b stops transmission of the encoded block data (step A10: transmission end).
- the execution contents of the tissue area extraction function, index amount calculation function, and importance degree calculation function may be programmed to be executed by a computer provided in advance in the pathological image diagnostic apparatus 1.
- a storage device 251 such as a memory or storage disk for storing image data (block), index amount information, and importance level information, and a tissue area in image data (block)
- arithmetic processing unit 250 which consists of operation parts 252, such as CPU which performs extraction, an index quantity of each block, and calculation of importance.
- the image transmitting apparatus 400b reads in and temporarily stores input data including image information (block data), importance information and index value information sent from the importance calculation means 30.
- a storage device disk memory 411) such as a memory or disk and a block to be transmitted are determined from the block data read into the disk memory 411 based on the value of importance information, and the block is cut out from image information
- a configuration comprising a CPU 412 and a network interface 413 connected to a communication line to transmit a clipped image (block) as a bit stream to the network under control of the CPU 412 and to receive a stop request signal sent from a receiving apparatus I assume.
- the CPU 412 and the network interface 413 according to the second embodiment transmit block data (blocks) in descending order of importance as described above.
- Embodiment 3 has almost the same configuration as that of the embodiment 1 (FIG. 1) described above, and as shown in FIG. 3, the image display means 40a of the embodiment 1 (FIG. 1)
- the second embodiment is different from the first embodiment in that an image transmission unit 40c is provided instead of the image transmission unit 40c, and a reception side terminal 40c connected to the image transmission unit 40b via a communication line is further provided.
- the area coding means 40c performs different coding for each block data for each block based on the order of the blocks (data) sent from the importance degree calculation means 30 and the index value S40 (block coding function). For blocks with higher importance of the sent-in block, lossless coding is performed, which can be completely restored to the original image. In addition, as the importance of the block sent in decreases, the area coding means 40c increases the compression rate according to the importance, and performs coding with a higher compression rate (function), Reduce the amount of data of the encoded block (the amount of transmission data).
- the area coding means 40c sends the coded block data S50c to the receiving side, and when all the reception of the coded block data requiring high transmission with high importance is completed on the receiving side (terminal) And stop transmission of the encoded block data (data transmission stop function).
- the pathological image subjected to the staining process is input to the tissue extraction means 10 (step B1).
- the tissue extraction unit 10 creates a monochrome image of the luminance component of the pathological image from the pathological image (step B2), and roughly generates a tissue image region from the generated monochrome image using an algorithm such as Otsu binarization. Extraction (step B3: tissue region extraction function). Transmit to index usage calculation unit 20.
- a blurring filter may be inserted to suppress the influence of noise outside the tissue region.
- the index utilization calculating unit 20 divides the original pathological image and the corresponding binarized image into blocks (rectangular blocks) (step B4), and calculates an index amount in each image block including the tissue image.
- Step B5 Index Amount Calculation Function
- Each block is transmitted to the importance degree calculation means 30.
- the importance calculation means 30 calculates the importance of each block by mapping (function) that relates the index amount and the importance from the single or plural index amounts (step B6: importance calculation function).
- the importance calculation means 30 sorts the blocks in the descending order of the calculated importance (step B7), and transmits the blocks to the area coding means 40c.
- the area coding unit 40c codes each block while changing the compression rate according to the degree of importance of each block in the sorted order (step B8).
- the area coding means 40c adds importance information to the coded block data and transmits it to the receiving side (step B9).
- the receiving side After receiving the block data of the necessary degree of importance, the receiving side transmits a stop request signal requesting stop of transmission of the block data to the area coding means 40c which is the transmitting side.
- the area coding means 40c receives this stop signal (step B10: YES)
- the area coding means 40c stops transmission of the encoded block data (completion of transmission of step B11).
- the execution contents of the tissue area extraction function, index amount calculation function, and importance degree calculation function may be programmed to be executed by a computer provided in advance in the pathological image diagnostic apparatus 1.
- a storage device 251 such as a memory or storage disk for storing image data (block), index amount information, and importance level information, and a tissue area in image data (block) It is good also as composition provided with arithmetic processing unit 250 which consists of operation parts 252, such as CPU which performs extraction, an index quantity of each block, and calculation of importance.
- the area encoding device 400c sends the image information (block data), the importance information, and the index value information sent from the importance calculation means 30 as in the case of the image transmission device 400b.
- a storage device disk memory 411) such as a memory or disk that reads input data including and temporarily storing the data, and determines a block to be transmitted from the block data read into the disk memory 411 based on the value of importance information.
- the CPU 412 is connected to a communication line and the control of the CPU 412 cuts out the block from the image information, transmits the cut image (block) as a bit stream to the network under control of the CPU 412, and receives the stop request signal sent from the receiving device.
- the interface 413 may be provided.
- the arithmetic device 412 such as the CPU according to the third embodiment encodes image data between cutting out of a block and bit stream formation, and the compression ratio at that time changes according to the degree of importance of the block. Do.
- the functions processed by the CPU set in the pathological image diagnostic device 1 can be distributed and allocated to each of the arithmetic processing device 250 and the area coding device 400c, so that the processing load of the CPU in each device Can be suppressed.
- the present invention can be applied to a pathological image diagnosis performed between remote places and a system supporting image diagnosis for telemedicine.
- Pathological imaging system (pathological imaging system) DESCRIPTION OF SYMBOLS 10 Tissue extraction means 20 Index amount calculation part 30 Importance degree calculation means 201, 202, 203, ... Index calculation part 250 Data processing device 251, 351, 411 Disk memory 252, 353, 412 CPU (central processing unit) 354 Monitor 400a Image display apparatus 400b Image transmission apparatus 400c Area coding apparatus 413 Network interface
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Radiology & Medical Imaging (AREA)
- Molecular Biology (AREA)
- Quality & Reliability (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
Description
この場合は、顕微鏡の視野に相当するのは最大でもPC(パーソナル・コンピュータ)のモニタ解像度になるが、組織が小さく画像表示時の倍率が小さくない限り、組織の全体の面積は、PCのモニタ解像度の数十倍、数百倍になる。
また、遠隔診断を行う場合など通信を介して電子画像の送信を行う場合は、伝送効率の関係からも病理組織画像の矩形分割が検討されることが多い。
また、例えば、診断対象である病理組織画像の注目区分領域とこの注目区分領域に隣接する上下左右等の隣接領域との関係が重視されることもあるため、上記特許文献1、2および3に開示された関連技術のように、各分割画像に指標値を割当てる手法では、画像診断効率が悪くなってしまうという不都合がある。
次に、本発明の実施形態について、その基本的構成内容を説明する。
本実施形態1の病理画像診断システム1は、図1に示すように、外部から送り込まれた病理組織画像(S10)における組織が撮像された画像領域(組織領域)を抽出する組織抽出手段10と、組織抽出手段10で抽出された画像領域である矩形ブロックS20を受け、各矩形ブロックS20について予め設定された病理組織にかかる指標量の計算を行う指標量計算装置20と、各矩形ブロックの指標量の重要度を算出し順序付けを行う重要度計算手段30と、重要度計算手段30から送り込まれる矩形ブロックS20を予め算出された指標値S40および重要度と対応づけて出力表示する画像表示手段40aを備えた構成をとっている。
ここで、組織抽出手段10は、組織が撮像された組織領域(画像領域)を、病理画像中における染色された領域として検知する。具体的には、組織抽出手段10は、上記病理組織画像から特定の色(または輝度)のモノクロ画像を作成し、このモノクロ画像に対して、予め設定された閾値による2値化画像作成アルゴリズム(例えば、判別分析法(大津の2値化))により2値化処理を行う。
また、ノイズにより少数の組織領域画素のみのブロックが抽出されるのを回避するために、上記2値化処理を行う前にガウシアンフィルタなどのぼかしフィルタを利用してノイズを抑制する処理を行うことが望ましい。
尚、実際に病理組織画像中にメッシュを描く必要はなく、設定された各矩形ブロックの頂点一つと矩形ブロックの縦横サイズを記憶する設定としてもよい。
尚、指標量計算手段20は、同時並行、分散的に病理指標の計算を行う指標量計算部201、202、203、…を備えている。尚、本実施形態1では、指標量計算部201、202、203からなる指標量計算手段20を例示しているが、指標量計算手段20で設定される指標利用の数はこれに限定されない。
そこで、本実施形態では、病理組織診断にかかる指標量として、病理組織の画像をブロック単位に分け、大域的な染色の濃度を調べるための[a]ブロック内指標量、大域的な染色変化を調べるための[b]ブロック間指標量、染色濃度や変化が人工的な理由であるかを調べるための[c]全ブロック指標量を導入する。
この指標量は、炎症や癌などの病気により生じる異常細胞核、異常細胞核群に対応した領域の特定色の絶対的な濃度や特定色で抽出された画素の塊の形状に関する量、例えば、矩形ブロック内のモーメント、Huモーメント、判別分析閾値、特定色の画素数、特定色の画素数塊の個数、面積平均、分散、長さ平均、分散、円形度平均、分散などが用いられる。
円形度は、Sを面積、Lを周囲長として
この指標量は、病理診断の重視される細胞核の増殖過程に対応した特定の組織領域内の周辺とその周辺領域の変化を反映し、例えば、ブロック単位のラプラシアンや周辺ブロックとの分散などが用いられる。
図5に示すように、対象ブロック(ア)は、隣接する8近傍ブロック(上下左右斜めの矩形ブロック)のうち7つの隣接ブロックをもつので、ラプラシアンは、以下に示す[数4]に基づき算出される。
この全ブロック指標量は、上記の[a],[b]が病理的な理由で生じたのか病理に関係のない人為的な理由で生じたのかを検証するためのもので、各矩形ブロックの優先度の算出には関係しないが、スライド作成時や撮影時の失敗などによる異常な画像として警告を出すために補助的に用いられる。
尚、上記算出された上記単独または複数の指標量は、スカラー値、もしくはベクトル値として重要度計算手段30に送られる。
ブロック内の色の均質性の場合、例えば細胞核の密集や濃さの場合は、画素の色値や密度が大きいほうが炎症やかんなどの病気の可能性があるので重要度は高くなる。
ブロック内の塊の均質性の場合は、例えば大きな円形度の高い画素塊(HE染色の場合は青い大きな円形塊)が多い場合は、癌細胞の可能性があるので重量度は高くなる。
ブロック間のラプラシアンのような周囲との変化の場合、変化量が大きい方が細胞増殖過程の異常として現れる病気の可能性があるので重要度は高くなる。
これらのような具体的な数値で計算される場合は、色値、密度などそのままの数値で使っても良いし、一定レンジで分けてランク化しても良い。
上記順序については、大小順位(全順序)の他、必ずしも順序づけがされていないグループとの対応づけであっても良い。指標量をx、順位もしくはグループをy、対応する写像をfとすると、
指標量計算手段20が、複数の指標量計算部(201、202、…)で構成されている場合は、写像(関数)を与える必要がある。
ここで、写像(関数)の与え方は、指標量の選択とどの指標を重視するかによって変わる。全指標の重要度が等価である場合は、全指標の総和による算出する。
ここでは、図6左図の濃灰色のブロックと図6右図の表の重要度列の上から3番目の重要度Cの行が対応づけられている。
また、複数の重要度を設定してもよく、その場合は、重要度のメニュで切り替えられるようにする。
このため、利用者は病理組織画像全体の様子が定量的に把握できるので、効率的、且つ迅速に病理画像診断を行うことができ、更には、診断時間を短縮することができる。
次に、上記実施形態1における病理画像診断システムの動作の概要を説明する。
生体組織が撮像された病理画像を予め設定された区分領域に分け各区分領域に対して、指標量計算部20が、病理診断の基準にかかる計測指標値を設定し(指標量設定工程)、次いで、重要度計算手段30が、前記各区分領域の計測指標値と前記各区分領域に隣接する区分領域における計測指標値とに基づき病理計測にかかる重要度を前記各区分領域に設定する(重要度設定工程)。
次いで、重要度計算手段30は、前記区分領域のうち前記生体組織が撮像された領域が一定の割合をしめる区分領域を抽出し(組織領域判定抽出工程)、この区分領域に対して前記計測指標値および前記区分重要度を対応付けて送出し、画像表示手段40aが、前記区分領域からなる病理画像の出力表示を行う。
以下、上記実施形態1における病理画像診断システムの動作について、図10のフローチャートに基づいて詳説する。
重要度計算手段30は、単一もしくは複数の指標量から指標量と重要度とを関連づける写像(関数)により各ブロックの重要度を算出する(ステップS6:重要度算出機能)。
ここでは、各ブロックと重要度情報、指標量情報との対応づけが示された出力表示が行われるものとする。
[実施形態2]
次に、本発明にかかる実施形態2の病理画像診断システムについて説明する。ここで、前述した実施形態1と同一の部分については、同一の符号を付するものとする。
この実施形態2は、システムの機器構成部分は前述した実施形態1(図1)とほぼ同一の構成を備えており、図2に示すように、実施形態1(図1)の画像表示手段40aに代えて、画像送信手段40bを備え、更に、この画像送信手段40bと通信回線を介して接続された受信側端末45cを備えた点が実施形態1の場合と相違する。
また、画像送信手段40bは、受信側で必要な重要度のブロックデータの受信が完了したとき、受信側端末から停止要求信号S60を受信すると共に、この停止要求信号S60に基づきブロックデータの送信を停止する(データ送信停止機能)。
次に、上記実施形態3における病理画像診断システムの動作について、図11のフローチャートに基づいて説明する。
重要度計算手段30は、単一もしくは複数の指標量から指標量と重要度とを関連づける写像(関数)により各ブロックの重要度を算出する(ステップA6:重要度算出機能)。
画像送信手段40bは、各ブロックデータに対して重要度情報を付加すると共に、重要度の高い順にブロックデータを受信側端末に送信する(ステップA8)。
画像送信手段40bは、この停止信号を受信した(ステップA9:YES)場合に、符号化済みブロックデータの送信を停止する(ステップA10:送信終了)。
尚、本実施形態2のCPU412、およびネットワークインタフェース413は、上述のように、重要度の高い順にブロックデータ(ブロック)の送信を行う。
次に、本発明にかかる実施形態3の病理画像診断システムについて説明する。ここで、前述した実施形態1と同一の部分については、同一の符号を付するものとする。
この実施形態3は、システムの機器構成部分は前述した実施形態1(図1)とほぼ同一の構成を備えており、図3に示すように、実施形態1(図1)の画像表示手段40aに代えて、画像送信手段40cを備え、更に、この画像送信手段40bと通信回線を介して接続された受信側端末40cを備えた点が実施形態1の場合と相違する。
送り込まれたブロックの重要度がより高いブロックに対しては、元画像に完全に復元可能な、可逆符号化を行う。
また、領域符号化手段40cは、送り込まれたブロックの重要度が低くなるにつれて、その重要度に対応して圧縮率を高くしていき、より高い圧縮率となる符号化を行い(機能)、符号化されたブロックのデータ量(送信データ量)の軽減を行う。
次に、上記実施形態3における病理画像診断システムの動作について、図12のフローチャートに基づいて説明する。
重要度計算手段30は、単一もしくは複数の指標量から指標量と重要度とを関連づける写像(関数)により各ブロックの重要度を算出する(ステップB6:重要度算出機能)。
領域符号化手段40cは、ソートされた順に、各ブロックの重要度に応じて圧縮率を変えながら、各ブロックの符号化を行う(ステップB8)。
受信側は必要な重要度のブロックデータを受信後、ブロックデータの送信停止を要求する停止要求信号を、送信側である領域符号化手段40cに送信する。
領域符号化手段40cは、この停止信号を受信した(ステップB10:YES)場合に、符号化済みブロックデータの送信を停止する(ステップB11送信完了)。
10 組織抽出手段
20 指標量計算部
30 重要度計算手段
201、202、203、… 指標計算部
250 データ処理装置
251、351、411 ディスクメモリ
252、353、412 CPU(中央処理装置)
354 モニタ
400a 画像表示装置
400b 画像送信装置
400c 領域符号化装置
413 ネットワークインタフェース
Claims (8)
- 生体組織が撮像された病理画像を予め設定された区分領域に分け各区分領域に対して病理診断の基準にかかる予め設定された計測指標値を設定する計測指標設定手段と、
前記各区分領域の計測指標値と前記各区分領域に隣接する区分領域における計測指標値とに基づき病理計測にかかる重要度を前記各区分領域に設定する重要度設定手段と、
前記区分領域のうち前記生体組織が撮像された領域が一定の割合をしめる区分領域を抽出すると共に当該区分領域に対して前記計測指標値および前記区分重要度を対応付けて送出する区分領域抽出手段と、
前記送出された区分領域からなる病理画像を出力表示する区分領域出力表示手段とを備えたことを特徴とする病理画像診断システム。 - 前記請求項1に記載の病理画像診断システムにおいて、
前記病理画像中の生体組織が予め設定された特定色により染色処理されている場合に、前記計測指標設定手段は、前記特定色の予め設定された濃度の量に基づき病理診断にかかる計測指標値を設定する色濃度指標値設定機能を備えたことを特徴とする病理画像診断システム。 - 前記請求項1に記載の病理画像診断システムにおいて、
前記組織抽出手段は、前記区分領域間のラプラシアンを計算することにより前記区分領域における画像特徴の抽出を行い、当該抽出結果に基き前記生物組織が含まれたブロックの抽出を行うラプラシアン抽出機能を備えたことを特徴とする病理画像診断システム。 - 前記請求項1に記載の病理画像診断システムにおいて、
前記組織抽出手段は、隣接区分領域のモーメントに基づき画像特徴の抽出を行うことにより生物組織が含まれたブロックを抽出するモーメント抽出機能を備えたことを特徴とする病理画像診断システム。 - 前記請求項1に記載の病理画像診断システムにおいて、
区分領域抽出手段は、
前記区分領域を前記設定された重要度の値が高い順に送信する重要度順送信機能を備えたことを特徴とする病理画像診断システム。 - 前記請求項1、2に記載の病理画像診断システムにおいて、
区分領域抽出手段は、
前記設定された重要度の値がより高い区分領域をより低い圧縮率で符号化すると共に、前記重要度の値がより低い区分領域をより高い圧縮率で符号化する区分領域圧縮機能を備えたことを特徴とする病理画像診断システム。 - 生体組織が撮像された病理画像を予め設定された区分領域に分け各区分領域に対して病理診断の基準にかかる計測指標値を設定し、前記各区分領域の計測指標値と前記各区分領域に隣接する区分領域における計測指標値とに基づき病理計測にかかる重要度を前記各区分領域に設定し、前記区分領域のうち前記生体組織が撮像された領域が一定の割合をしめる区分領域を抽出すると共に当該区分領域に対して前記計測指標値および前記区分重要度を対応付けて送出し、前記区分領域からなる病理画像を出力表示することを特徴とする病理画像診断方法。
- 生体組織が撮像された病理画像を予め設定された区分領域に分け各区分領域に対して病理診断の基準にかかる計測指標値を算出する機能と、前記各区分領域の計測指標値と前記各区分領域に隣接する区分領域における計測指標値とに基づき病理計測にかかる重要度を前記各区分領域に設定する機能と、前記区分領域のうち前記生体組織が撮像された領域が一定の割合をしめる区分領域を抽出する機能と、予め設定されたコンピュータに実行させることを特徴とする病理画像診断プログラム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201080024614.3A CN102667471B (zh) | 2009-06-03 | 2010-06-01 | 病理图像诊断系统、病理图像诊断方法以及病理图像诊断程序 |
US13/376,120 US8594412B2 (en) | 2009-06-03 | 2010-06-01 | Pathologic image diagnostic system, pathologic image diagnostic method, and pathologic image diagnostic program |
EP10783371.7A EP2439528A4 (en) | 2009-06-03 | 2010-06-01 | PATHOLOGICAL IMAGING DIAGNOSTIC SYSTEM, PATHOLOGICAL IMAGING DIAGNOSTIC METHOD, AND PATHOLOGICAL IMAGING DIAGNOSTIC PROGRAM |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009-134219 | 2009-06-03 | ||
JP2009134219A JP5387147B2 (ja) | 2009-06-03 | 2009-06-03 | 病理画像診断システム、病理画像処理方法、病理画像診断プログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010140588A1 true WO2010140588A1 (ja) | 2010-12-09 |
Family
ID=43297722
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/059268 WO2010140588A1 (ja) | 2009-06-03 | 2010-06-01 | 病理画像診断システム、病理画像診断方法、病理画像診断プログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US8594412B2 (ja) |
EP (1) | EP2439528A4 (ja) |
JP (1) | JP5387147B2 (ja) |
CN (1) | CN102667471B (ja) |
WO (1) | WO2010140588A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012102069A1 (ja) * | 2011-01-24 | 2012-08-02 | 日本電気株式会社 | 情報処理システム、情報処理方法、情報処理装置およびその制御方法とその制御プログラムを格納した記憶媒体 |
JPWO2013187148A1 (ja) * | 2012-06-15 | 2016-02-04 | オリンパス株式会社 | 画像処理装置、顕微鏡システム、内視鏡システム及び画像処理方法 |
CN105741266A (zh) * | 2016-01-22 | 2016-07-06 | 北京航空航天大学 | 一种病理图像细胞核快速定位方法 |
Families Citing this family (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5151403B2 (ja) * | 2007-11-06 | 2013-02-27 | 日本電気株式会社 | 低分化癌検出モジュール、これを備えた病理画像診断支援装置、プログラムおよび記録媒体 |
CN103140757A (zh) * | 2010-09-30 | 2013-06-05 | 日本电气株式会社 | 信息处理设备、信息处理系统、信息处理方法、程序和记录介质 |
JP5372068B2 (ja) * | 2011-05-20 | 2013-12-18 | キヤノン株式会社 | 撮像システム、画像処理装置 |
US20140176602A1 (en) * | 2011-08-18 | 2014-06-26 | Nec Corporation | Information processing system, information processing method, information processing apparatus, and control method and control program thereof |
WO2013027399A1 (ja) * | 2011-08-23 | 2013-02-28 | 日本電気株式会社 | 情報処理システム、情報処理方法、情報処理装置およびその制御方法と制御プログラム |
US8755633B2 (en) * | 2012-03-14 | 2014-06-17 | Sony Corporation | Automated synchronized navigation system for digital pathology imaging |
US20150132766A1 (en) * | 2012-03-30 | 2015-05-14 | On-Chip Cellomics Consortium | Imaging cell sorter |
JP6284024B2 (ja) * | 2014-04-28 | 2018-02-28 | 横河電機株式会社 | 細胞生死判定システム、細胞生死判定方法 |
WO2015170183A2 (en) * | 2014-05-05 | 2015-11-12 | Dako Denmark A/S | Method and apparatus for image scoring and analysis |
WO2016120418A1 (en) | 2015-01-31 | 2016-08-04 | Ventana Medical Systems, Inc. | Quality control of automated whole-slide analysis |
CN104732230A (zh) * | 2015-03-27 | 2015-06-24 | 麦克奥迪(厦门)医疗诊断系统有限公司 | 一种基于细胞核统计信息的病理图像局部特征提取方法 |
US10801959B2 (en) | 2015-06-18 | 2020-10-13 | Konica Minolta, Inc. | Objective biological substance analysis device, analysis system, analysis method, and recording medium storing computer readable analysis program |
WO2017150194A1 (ja) * | 2016-03-04 | 2017-09-08 | コニカミノルタ株式会社 | 画像処理装置、画像処理方法及びプログラム |
CN111310893B (zh) * | 2016-08-05 | 2023-11-21 | 中科寒武纪科技股份有限公司 | 一种用于执行神经网络运算的装置及方法 |
JP6670222B2 (ja) * | 2016-11-01 | 2020-03-18 | 株式会社日立ハイテク | 画像診断支援装置及びシステム、画像診断支援方法 |
JP6911995B2 (ja) * | 2018-02-20 | 2021-07-28 | 日本電気株式会社 | 特徴抽出方法、照合システム、およびプログラム |
CN108577803B (zh) * | 2018-04-26 | 2020-09-01 | 上海鹰瞳医疗科技有限公司 | 基于机器学习的眼底图像检测方法、装置及系统 |
CN108898580B (zh) * | 2018-05-31 | 2021-08-27 | 中南大学湘雅医院 | 皮瓣移植术后护理监控系统和方法、管理系统及存储介质 |
CN108903913B (zh) * | 2018-05-31 | 2020-12-01 | 中南大学湘雅医院 | 皮瓣移植术后护理监控设备、系统、方法、产品及服务器 |
CN108694403B (zh) * | 2018-05-31 | 2021-08-31 | 中南大学湘雅医院 | 皮瓣移植术后护理监控系统和方法、数据管理方法和设备 |
CN109785148B (zh) * | 2018-11-30 | 2023-10-24 | 平安科技(深圳)有限公司 | 肝病报销流程的处理方法、装置、设备及可读存储介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH063601A (ja) | 1992-06-22 | 1994-01-14 | Olympus Optical Co Ltd | 顕微鏡静止画像伝送システム |
JP2005057494A (ja) | 2003-08-04 | 2005-03-03 | Matsushita Electric Ind Co Ltd | 画像通信装置及び画像通信方法 |
JP2006102097A (ja) | 2004-10-05 | 2006-04-20 | Canon Inc | 眼科撮影装置及び眼科画像処理システム |
JP2006121578A (ja) | 2004-10-25 | 2006-05-11 | Matsushita Electric Ind Co Ltd | 画像伝送システム、画像送信装置、画像受信装置および画像伝送方法 |
JP2007135065A (ja) | 2005-11-11 | 2007-05-31 | Konica Minolta Photo Imaging Inc | 表示プログラム及び画像データの表示方法 |
JP2009134219A (ja) | 2007-12-03 | 2009-06-18 | Seiko Epson Corp | プロジェクタ |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE19616997A1 (de) * | 1996-04-27 | 1997-10-30 | Boehringer Mannheim Gmbh | Verfahren zur automatisierten mikroskopunterstützten Untersuchung von Gewebeproben oder Körperflüssigkeitsproben |
US6031930A (en) * | 1996-08-23 | 2000-02-29 | Bacus Research Laboratories, Inc. | Method and apparatus for testing a progression of neoplasia including cancer chemoprevention testing |
US6272235B1 (en) * | 1997-03-03 | 2001-08-07 | Bacus Research Laboratories, Inc. | Method and apparatus for creating a virtual microscope slide |
US6535626B1 (en) * | 2000-01-14 | 2003-03-18 | Accumed International, Inc. | Inspection system with specimen preview |
US7627531B2 (en) | 2000-03-07 | 2009-12-01 | American Express Travel Related Services Company, Inc. | System for facilitating a transaction |
US6711283B1 (en) * | 2000-05-03 | 2004-03-23 | Aperio Technologies, Inc. | Fully automatic rapid microscope slide scanner |
US8538770B2 (en) * | 2000-08-01 | 2013-09-17 | Logical Images, Inc. | System and method to aid diagnoses using cross-referenced knowledge and image databases |
US7383134B2 (en) * | 2002-01-15 | 2008-06-03 | Piper James R | Method and/or system for analyzing biological samples using a computer system |
EP1488229B1 (en) * | 2002-03-13 | 2012-10-24 | Q3Dm, Llc | System and method for automatic color segmentation and minimum significant response for measurement of fractional localized intensity of cellular compartments |
US20040202357A1 (en) * | 2003-04-11 | 2004-10-14 | Perz Cynthia B. | Silhouette image acquisition |
CN101002093A (zh) * | 2004-01-15 | 2007-07-18 | 凯玫肯国际有限公司 | 图像分析和试验系统 |
US20060159325A1 (en) * | 2005-01-18 | 2006-07-20 | Trestle Corporation | System and method for review in studies including toxicity and risk assessment studies |
US20070031043A1 (en) * | 2005-08-02 | 2007-02-08 | Perz Cynthia B | System for and method of intelligently directed segmentation analysis for automated microscope systems |
US20070135999A1 (en) * | 2005-12-13 | 2007-06-14 | Applied Spectral Imaging Ltd. | Method, apparatus and system for characterizing pathological specimen |
JP5145672B2 (ja) * | 2006-02-27 | 2013-02-20 | 富士通セミコンダクター株式会社 | 半導体装置の製造方法 |
US20080033302A1 (en) * | 2006-04-21 | 2008-02-07 | Siemens Corporate Research, Inc. | System and method for semi-automatic aortic aneurysm analysis |
US8135189B2 (en) * | 2007-10-03 | 2012-03-13 | Siemens Medical Solutions Usa, Inc. | System and method for organ segmentation using surface patch classification in 2D and 3D images |
-
2009
- 2009-06-03 JP JP2009134219A patent/JP5387147B2/ja not_active Expired - Fee Related
-
2010
- 2010-06-01 US US13/376,120 patent/US8594412B2/en active Active
- 2010-06-01 CN CN201080024614.3A patent/CN102667471B/zh active Active
- 2010-06-01 EP EP10783371.7A patent/EP2439528A4/en not_active Withdrawn
- 2010-06-01 WO PCT/JP2010/059268 patent/WO2010140588A1/ja active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH063601A (ja) | 1992-06-22 | 1994-01-14 | Olympus Optical Co Ltd | 顕微鏡静止画像伝送システム |
JP2005057494A (ja) | 2003-08-04 | 2005-03-03 | Matsushita Electric Ind Co Ltd | 画像通信装置及び画像通信方法 |
JP2006102097A (ja) | 2004-10-05 | 2006-04-20 | Canon Inc | 眼科撮影装置及び眼科画像処理システム |
JP2006121578A (ja) | 2004-10-25 | 2006-05-11 | Matsushita Electric Ind Co Ltd | 画像伝送システム、画像送信装置、画像受信装置および画像伝送方法 |
JP2007135065A (ja) | 2005-11-11 | 2007-05-31 | Konica Minolta Photo Imaging Inc | 表示プログラム及び画像データの表示方法 |
JP2009134219A (ja) | 2007-12-03 | 2009-06-18 | Seiko Epson Corp | プロジェクタ |
Non-Patent Citations (1)
Title |
---|
See also references of EP2439528A4 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012102069A1 (ja) * | 2011-01-24 | 2012-08-02 | 日本電気株式会社 | 情報処理システム、情報処理方法、情報処理装置およびその制御方法とその制御プログラムを格納した記憶媒体 |
US9230154B2 (en) | 2011-01-24 | 2016-01-05 | Nec Corporation | Information processing apparatus, method, and storage medium for assisting with a diagnosis based on a tissue sample |
US9361508B2 (en) | 2011-01-24 | 2016-06-07 | Nec Corporation | Information processing system, information processing method, information processing apparatus, control method therefor, and storage medium storing control program therefor |
JPWO2013187148A1 (ja) * | 2012-06-15 | 2016-02-04 | オリンパス株式会社 | 画像処理装置、顕微鏡システム、内視鏡システム及び画像処理方法 |
US9517013B2 (en) | 2012-06-15 | 2016-12-13 | Olympus Corporation | Image processing apparatus, microscope system, endoscope system, and image processing method |
CN105741266A (zh) * | 2016-01-22 | 2016-07-06 | 北京航空航天大学 | 一种病理图像细胞核快速定位方法 |
CN105741266B (zh) * | 2016-01-22 | 2018-08-21 | 北京航空航天大学 | 一种病理图像细胞核快速定位方法 |
Also Published As
Publication number | Publication date |
---|---|
EP2439528A4 (en) | 2014-09-24 |
CN102667471A (zh) | 2012-09-12 |
JP2010281637A (ja) | 2010-12-16 |
US20120082366A1 (en) | 2012-04-05 |
US8594412B2 (en) | 2013-11-26 |
JP5387147B2 (ja) | 2014-01-15 |
CN102667471B (zh) | 2014-09-10 |
EP2439528A1 (en) | 2012-04-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2010140588A1 (ja) | 病理画像診断システム、病理画像診断方法、病理画像診断プログラム | |
JP7422825B2 (ja) | 顕微鏡スライド画像のための焦点重み付き機械学習分類器誤り予測 | |
JP6091137B2 (ja) | 画像処理装置、画像処理システム、画像処理方法およびプログラム | |
US9014443B2 (en) | Image diagnostic method, image diagnostic apparatus, and image diagnostic program | |
JP5442542B2 (ja) | 病理診断支援装置、病理診断支援方法、病理診断支援のための制御プログラムおよび該制御プログラムを記録した記録媒体 | |
JP5963009B2 (ja) | デジタル標本作製装置、デジタル標本作製方法およびデジタル標本作製サーバ | |
Helin et al. | Optimized JPEG 2000 compression for efficient storage of histopathological whole-slide images | |
Krupinski et al. | Compressing pathology whole-slide images using a human and model observer evaluation | |
Kowalik-Urbaniak et al. | The quest for'diagnostically lossless' medical image compression: a comparative study of objective quality metrics for compressed medical images | |
US20120134556A1 (en) | Image processing device, image processing method, and computer-readable recording device | |
CN115410693B (zh) | 一种数字病理切片的存储系统、浏览系统及方法 | |
CN103561631B (zh) | 图像处理装置和图像处理方法 | |
Sharma et al. | Balancing image quality and compression factor for special stains whole slide images | |
JP5832281B2 (ja) | 画像処理装置、画像処理システム、画像処理方法、およびプログラム | |
RU2295297C2 (ru) | Способ исследования и диагностики состояния биологического объекта или его части | |
JP2011215680A (ja) | 病理診断支援装置、病理診断支援方法、病理診断支援のための制御プログラムおよび該制御プログラムを記録した記録媒体 | |
JPWO2005010206A1 (ja) | 染色体状態の評価方法および評価システム | |
JP2016065821A (ja) | 石炭分析方法、石炭分析装置及びコンピュータプログラム。 | |
US20230334621A1 (en) | Technologies for Improved Whole Slide Imaging | |
US11875893B2 (en) | Method and apparatus for outputting information related to a pathological slide image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080024614.3 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10783371 Country of ref document: EP Kind code of ref document: A1 |
|
REEP | Request for entry into the european phase |
Ref document number: 2010783371 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010783371 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13376120 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |