JP5179902B2 - Medical image interpretation support device - Google Patents

Medical image interpretation support device Download PDF

Info

Publication number
JP5179902B2
JP5179902B2 JP2008050692A JP2008050692A JP5179902B2 JP 5179902 B2 JP5179902 B2 JP 5179902B2 JP 2008050692 A JP2008050692 A JP 2008050692A JP 2008050692 A JP2008050692 A JP 2008050692A JP 5179902 B2 JP5179902 B2 JP 5179902B2
Authority
JP
Japan
Prior art keywords
feature
plurality
types
value
unit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
JP2008050692A
Other languages
Japanese (ja)
Other versions
JP2009207541A (en
Inventor
重治 大湯
仁 山形
アルトウーロ・カルデロン
恭子 佐藤
敦子 杉山
Original Assignee
株式会社東芝
東芝メディカルシステムズ株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社東芝, 東芝メディカルシステムズ株式会社 filed Critical 株式会社東芝
Priority to JP2008050692A priority Critical patent/JP5179902B2/en
Publication of JP2009207541A publication Critical patent/JP2009207541A/en
Application granted granted Critical
Publication of JP5179902B2 publication Critical patent/JP5179902B2/en
Application status is Active legal-status Critical
Anticipated expiration legal-status Critical

Links

Images

Description

  The present invention relates to a medical image interpretation support apparatus having a function of analyzing an abnormal part depicted in a medical image.

  In recent years, a medical image interpretation support apparatus (hereinafter referred to as CAD) having a computer analysis function has been developed and has been partially used by doctors and the like when interpreting examination images. The CAD analyzes the image and presents the detection result of the abnormal part, the type of abnormality, and the result of the malignancy determination to the doctor. Further, a technique for presenting a feature amount that is intermediate data for CAD determination is known (for example, Patent Document 1 and Patent Document 2).

Doctors and others make their own judgments regarding the presence or absence of abnormalities and the degree of malignancy, and examine treatment policies based on these decisions. At this time, the doctor or the like refers to the detection result or determination result by CAD, so that the diagnosis accuracy of the doctor or the like is improved or the work load is reduced. However, when the CAD determination result and the doctor determination result are different, a great problem arises for the determining doctor. In other words, it is not shown how the feature quantity that is the basis of the judgment result is calculated and how much the feature quantity contributes to the judgment result. Which judgment should doctors etc. adopt? I'm in trouble.
JP 2004-222864 A JP 2006-430007 A

  An object of the present invention is to provide a medical image interpretation support apparatus that realizes improvement in diagnostic accuracy of image interpretation.

A medical image interpretation support apparatus according to a first aspect of the present invention includes a storage unit that stores volume data relating to a subject, an abnormal candidate region extraction unit that extracts an abnormal candidate region from the volume data, and the extracted abnormality For a candidate area, a feature amount calculation unit that calculates a plurality of types of feature amounts, a comprehensive determination value calculation unit that calculates a comprehensive determination result based on the calculated types of feature amounts, the plurality of types of feature amounts, and An image display unit that displays the comprehensive determination result; a medical image interpretation support apparatus that includes the feature amount selection unit that selects a desired one from the plurality of types of feature amounts; and the selected feature A cross-section position / direction determination unit that determines the position and direction of the cross section of the abnormal candidate region whose amount is maximum or minimum, and a cross section that generates a cross-sectional image based on the determined position and direction of the cross section Further comprising an image generating unit, a,
The image display unit displays the generated cross-sectional image.
A medical image interpretation apparatus according to a second aspect of the present invention includes a storage unit that stores volume data relating to a subject, an abnormal candidate region extraction unit that extracts an abnormal candidate region from the volume data, and the extracted abnormal candidates A feature amount calculation unit that calculates a plurality of types of feature amounts, an overall determination value calculation unit that calculates a comprehensive determination result based on the calculated plurality of types of feature amounts, the plurality of types of feature amounts, and the region An image display unit that displays a comprehensive determination result, and a medical image interpretation support apparatus that further includes a feature amount changing unit that changes at least one value of the plurality of types of feature amounts, and the comprehensive determination value calculation The unit recalculates the comprehensive determination result based on the changed feature amount, and the image display unit displays the changed feature amount and the recalculated comprehensive determination value. The features.

A medical image interpretation support apparatus according to a third aspect of the present invention includes a storage unit that stores volume data relating to a subject, an abnormal candidate region extraction unit that extracts an abnormal candidate region from the volume data, and the extracted abnormality A feature amount calculation unit that calculates a plurality of types of feature amounts for a candidate region, and applying the plurality of types of feature amounts to a plurality of malignant / benign discriminant expressions given in advance, thereby determining the malignancy / benignity of the abnormal candidate region Malignant / benign discriminating unit for discriminating, a comprehensive judgment value calculating unit for calculating a total judgment value based on the plurality of types of feature values, and the abnormality candidate by the malignant / benign discriminating unit among the plurality of types of feature values and displayed differently from the determination result different from the result of the determination at feature amount for the abnormality candidate area feature quantity determination result same discrimination result of the region, the said comprehensive judgment value more Comprising an image display unit for displaying the feature quantity of the class list, the.

A medical image interpretation support apparatus according to a fourth aspect of the present invention includes a storage unit that stores volume data relating to a subject, an abnormal candidate region extraction unit that extracts an abnormal candidate region from the volume data, and the extracted abnormality A feature amount calculation unit that calculates a plurality of types of feature amounts for the candidate region; and a comprehensive determination value calculation unit that applies the plurality of types of feature amounts to a predetermined malignancy / benignity determination diagram to calculate a total determination value; A determination value calculation unit that calculates a plurality of types of index values indicating contributions of the plurality of types of feature amounts to the total determination value for each feature amount based on the malignancy / benignity determination diagram; and the total determination value And an image display unit that displays the plurality of types of index values in a list.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the medical image interpretation assistance apparatus which implement | achieves the improvement of the diagnostic accuracy of image interpretation.

  Hereinafter, a medical image interpretation support apparatus according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a configuration of a medical image interpretation support apparatus (hereinafter referred to as CAD) 1 according to the present embodiment. As illustrated in FIG. 1, the CAD 1 includes a storage unit 11, an operation unit 12, an image display unit 13, an image processing unit 15, a feature amount calculation unit 16, and a comprehensive determination value calculation unit 17 with the control unit 10 as a center. .

  The storage unit 11 stores volume data of the subject generated by scanning the subject with an X-ray computed tomography apparatus or the like. The operation unit 12 includes a pointing device such as a mouse, detects the coordinates of the cursor and pointer displayed on the image display unit 13, and outputs the detected coordinates to the control unit 10. Further, the operation unit 12 may include a touch panel provided so as to cover the image display unit 13 constituted by a CRT, an LCD, or the like, and is based on a coordinate reading principle such as an electromagnetic induction type, a magnetostriction type, or a pressure sensitive type. The touched coordinate is detected, and the detected coordinate is output to the control unit 10 as a position signal.

  The image processing unit 15 reads the volume data stored in the storage unit 11 and automatically extracts an abnormal candidate region in the volume data based on the existing description. Further, the image processing unit 15 calculates the position and direction of a cross-section for confirming a feature amount, which will be described later, by a cross-section determination routine, and generates an MPR image of the calculated cross-section (hereinafter referred to as a cross-sectional image). Note that the abnormal candidate region in the present embodiment assumes a nodule candidate region inside the subject. The image processing unit 5 also performs image processing such as processing for attaching color information to the abnormal candidate region of the cross-sectional image, image rotation, enlargement / reduction processing, and the like so that the abnormal candidate region can be easily identified.

  The feature amount calculation unit 16 calculates the feature amount of the extracted abnormality candidate region. The feature amount is a numerical value calculated as intermediate data for determining whether the nodule candidate site is malignant or benign. The types of feature amounts include, for example, CT value density, CT value density deviation in the abnormal candidate area, sphericity, flatness, columnarity, and abnormal candidate area in the abnormal candidate area. Examples include the cavity ratio, the number of cavities, the concentration in the cavity, the number of columnar cavities, the number of protrusions in the abnormal candidate region, the maximum protrusion length, the number of depressions, and the maximum depth of depression. All of these feature quantities are not calculated with the same cross section, and different cross sections are used according to the calculated feature quantities. However, some features can be calculated with the same cross section. For example, a combination of sphericity, flatness, and columnarity, a combination of cavity ratio, number of cavities, and concentration in the cavity. The feature amount calculation unit 16 normalizes the calculated feature amount in a range of 0 to 1.

  The comprehensive determination value calculation unit 17 calculates a comprehensive determination value that is an index value indicating whether the abnormal shadow candidate region is malignant or benign based on the calculated and normalized feature amount. Specifically, the comprehensive determination value is calculated by weighted addition, averaging, or the like of the plurality of types of feature amounts described above. The comprehensive judgment value has a value of 0-1. The integrated judgment value is more likely to be malignant as it is closer to 1, and is more likely to be benign as it is closer to 0.

  The image display unit 13 displays the feature amount and the value of the comprehensive determination result calculated therefrom in a list so as to be selectable / changeable. The image display unit 13 displays various cross-sectional images along with a list of feature amounts and comprehensive determination values. Moreover, you may display on the screen the button for displaying the cross-sectional image used for calculating a feature-value as needed. When an instruction to display a cross-sectional image is given, the image processing unit 15 obtains an image serving as a basis for calculating the feature value, for example, the position and orientation of the cross-section of the abnormal candidate region suitable for the feature value calculation, and the image display unit 13 Then, a cross-sectional image of the cross section is displayed.

  Next, the operation of the CAD 1 under the control of the control unit 10 will be described.

  FIG. 2 is a diagram illustrating a processing flow from extraction of an abnormal candidate region to display of a list of feature amounts and comprehensive determination results. As shown in FIG. 2, the control unit 10 first causes the image processing unit 15 to perform an abnormal candidate region extraction process. In the abnormal candidate region extraction process, the image processing unit 5 extracts an abnormal candidate region from the volume data using an existing technique (step S01).

  When the abnormal candidate region is extracted, the control unit 10 causes the feature amount calculation unit 16 to perform a feature amount calculation process. In the feature amount calculation process, the feature amount calculation unit 16 calculates various feature amounts by applying various feature amount calculation algorithms to the abnormality candidate region (step S02). For example, feature quantities such as sphericity, flatness, columnarity, cavity ratio, number of cavities, concentration in the cavity, and CT value concentration deviation are calculated. Note that a method for determining a cross section suitable for confirmation of a feature amount (cross section determination routine) will be described later. A feature amount calculation algorithm that calculates a feature amount using a cross section obtained by the method described later. It is also good. The calculated feature value is normalized (normalized) to a range of 0 to 1 by the feature value calculation unit 16 (step S03).

  When all the feature values have been calculated and normalized, the control unit 10 causes the comprehensive determination value calculation unit 17 to perform a comprehensive determination value calculation process. In the calculation process of the total determination value, the total determination value calculation unit 17 calculates a plurality of calculated feature values (sphericity, flatness, columnarity, cavity ratio, number of cavities, concentration in the cavity, CT value concentration, deviation, etc.). A comprehensive judgment value is calculated based on at least one (step S04). The comprehensive determination value is calculated by weighted addition values or average values of various feature amounts. The value of the weight is obtained and stored in advance based on the importance in determining the abnormality.

  When the comprehensive determination value is calculated, the control unit 10 causes the image display unit 13 to display the comprehensive determination value and the standardized feature values in a list (step S05). FIG. 3 is a diagram illustrating an example of a screen displayed in step S05. As shown in FIG. 3, a diagnostic image is displayed in the area 20 on the left side of the screen, and a list is displayed in the area 21 on the right side of the screen. The diagnostic image 20 includes an abnormal candidate region for which a feature amount has been calculated. If necessary, the diagnostic image 20 may be subjected to an enlargement process by the image processing unit 15 or a color tone assignment process for an abnormal candidate area. In the list 21, the names of feature amounts are displayed in the left column, and the normalized values of the feature amounts are displayed in circles in the right column. This is a graph in which the value of each feature amount is plotted within the range of 0 to 1, which is the normalization range. In the bottom line, the comprehensive judgment value is shown. Similarly to each feature amount, the comprehensive determination value is indicated by a circle with values of 0 to 1. A user (such as a doctor) sees the comprehensive judgment values in the list 21 and finally determines whether the nodule candidate site of the subject is malignant or benign. At this time, the user (physician or the like) can check all values used for calculating the comprehensive determination value at a glance by referring to the list 21.

  When referring to the list, the user (physician or the like) sometimes wonders whether it is really possible to trust this comprehensive judgment value. In this case, the user (such as a doctor) displays a cross-sectional image suitable for confirming the feature amount by selecting the desired feature amount displayed in the list on the screen via the operation unit 12. Can be made. The cross-sectional image suitable for confirming the feature amount is, for example, a cross-sectional image related to a cross section where the selected feature amount is maximum or minimum.

  Hereinafter, the cross-section determination routine for confirming the feature amount will be described with three specific examples. The first is a routine (cross-section determination routine R1) for determining a cross-section of a cross-sectional image displayed when confirming a feature amount related to a shape such as sphericity, flatness, and columnarity. The second is a routine (cross-section determination routine R2) that determines the cross-section of the cross-sectional image displayed when confirming the feature quantities related to the cavities such as the cavity ratio, the number of cavities, and the concentration in the cavity. The third routine is a routine (cross-section determination routine R3) for determining a cross-section of a cross-sectional image displayed when confirming a feature amount related to density non-uniformity such as a CT value density deviation.

Hereinafter, the process of the section determination routine R1 will be described with reference to FIG. As shown in FIG. 4, first, the control unit 10 waits for the operator to select a feature amount displayed on the screen via the operation unit 12 (step SA1). For example, this is performed by selecting the feature amount name displayed in the feature amount column shown in FIG. When the feature amount related to the shape such as sphericity, flatness, columnarity, etc. displayed in the feature amount item of FIG. 3 is selected via the operation unit 12 (step SA1: YES), the control unit 10 The image processing unit 15 is caused to perform a section determination routine R1. In the section determination routine R1, the image processing unit 15 calculates an average value (vector A) of the vectors Xi based on the coordinate values (vectors Xi) of the voxels of the abnormal candidate region extracted in step S01 (step SA2). . Next, the image processing unit 15 calculates a 3 × 3 matrix R based on the vector Xi and the vector A (step SA3). Here, the matrix R is as the following equation (1). However, the superscript T is a symbol indicating a transposed matrix.

Next, the image processing unit 15 spectrally decomposes the matrix R as shown in the following equation (2) to calculate eigenvalues λi 2 and eigenvectors V1, V2, and V3 (step SA4).

  Then, the image processing unit 15 determines a cross section that passes through the center (or center of gravity) of the abnormal candidate region and is stretched between the eigenvector V1 and the eigenvector V3 (step SA5). The determined cross section is set as a display cross section. The cross section determined here is regarded as a cross section in which the abnormal candidate region is most flat. Above, description of cross-section determination routine R1 is complete | finished.

Next, the processing of the section determination routine R2 will be described with reference to FIGS. FIG. 5 is a diagram showing a flow of processing of the section determination routine R2, and FIG. 6 is a diagram showing a hollow region and a section related to processing of the section determination routine R2. As shown in FIG. 6, first, the control unit 10 waits for the operator to select a feature amount displayed on the screen via the operation unit 12 (step SB1). When the feature amount related to the cavity such as the cavity ratio, the number of cavities, and the concentration in the cavity is selected via the operation unit 12, the control unit 10 sends the image processing unit 15 to the image processing unit 15. The section determination routine R2 is performed. In the section determination routine R2, the image processing unit 15 extracts a hollow region (51 in FIG. 6) existing in the abnormal candidate region extracted in step S01 (step SB2). Next, the image processing unit 15 calculates an average value (vector B) of the vectors Yi based on the coordinate values (vector Yi) of each voxel in the extracted hollow area (51 in FIG. 6) (step SB3). . Next, the image processing unit 15 calculates a 3 × 3 matrix S based on the vector Yi and the vector B (step SB4). Here, the matrix S is as the following equation (3). However, the superscript T is a symbol indicating a transposed matrix.

The image processing unit 15 spectrally decomposes the matrix S as in the following equation (4) to calculate the eigenvalue λi 2 and eigenvectors V1, V2, and V3 (step SB5).

  Next, the image processing unit 15 sets a plurality of cross sections parallel to the cross section formed by the vector V1 and the vector V2 to a hollow area at an interval of 1 mm, for example, and a plane where each cross section and the hollow area intersect ( The area of 52) in FIG. 6 is calculated individually (step SB6). Then, the image processing unit 15 determines a cross section (53 in FIG. 6) that maximizes the area of the intersecting surface (52 in FIG. 6) from the plurality of cross sections (step SB7). The determined cross section is set to the display cross section. In the method of the section determination routine R2, the section having the largest area of the plane intersecting the cavity area is selected from the various sections of the abnormality candidate area. Therefore, the cross-sectional image related to the cross-section determined by the cross-section determination routine R2 is suitable for observing the non-uniformity of the voxel values. This is the end of the description of the processing steps of the section determination routine R2.

  Next, the process of the section determination routine R3 will be described with reference to FIG. As shown in FIG. 7, first, the control unit 10 waits for the operator to select a feature amount displayed on the screen via the operation unit 12 (step SC1). When a feature quantity related to density non-uniformity such as a CT value density deviation displayed in the feature quantity item shown in FIG. 3 is selected, the control unit 10 causes the image processing 15 to execute a section determination routine R3.

  In the section determination routine R3, the image processing unit 15 reduces the abnormal candidate region extracted in step S01 by several voxels (step SC2). Next, the image processing unit 15 performs projection processing on each of the reduced abnormality candidate regions in a plurality of directions indicated by a plurality of preset three-dimensional direction vectors, and a plurality of maximum values related to the plurality of projection directions. A projection image and a minimum value projection image are generated respectively (step SC3). For example, 300 three-dimensional direction vectors are prepared in advance. In this case, 300 maximum value projection images and minimum value projection images are generated.

  Next, the image processing unit 15 adds the maximum value projection image and the minimum value projection image that belong to the same projection direction among the generated maximum value projection image and minimum value projection image, and obtains the plurality of addition images. Occurs (step SC4). Next, the image processing unit 15 calculates a plurality of variances of the generated plurality of added images (step SC5). Next, the image processing unit 15 determines an added image having the maximum variance among the plurality of calculated variances (step SC6). Next, the image processing unit 15 sets a plurality of cross sections perpendicular to the cross section of the determined addition image for the reduced abnormality candidate region (step SC7). Next, the image processing unit 15 calculates a plurality of variances of the set plurality of cross sections, respectively (step SC8). Then, the image processing unit 15 determines a cross section having the maximum variance among the plurality of calculated variances (step SC9). The determined cross section is set as a display cross section. In the method of the cross-section determination routine R3, a cross-section having the largest variance of voxel values is selected from various cross-sectional images of the abnormal candidate region. Therefore, the cross-sectional image based on the cross-section determined by the cross-section determination routine R3 is suitable for observing the non-uniformity of the voxel values. Above, description of the process of cross-section determination routine R3 is complete | finished.

  The cross-sectional image to be displayed is, for example, a planar reformatted image, a maximum value projection image with a thickness of about 1 cm (Maximum Intensity Projection), and a minimum value projection image with a thickness (Minimum Intensity Projection) centered on the determined cross-section. ), An average projection image with thickness (Average Projection), and the like, which can be selected as appropriate according to the cross-sectional image to be displayed.

  Note that the feature quantity selection method directly selects the feature quantity name displayed in the feature quantity item of FIG. 3 as described in steps SA1, SB1, and SC1. However, it is not necessary to limit to this, for example, you may select with a button as shown in FIG. FIG. 8 is a diagram showing an example of a screen on which related feature amounts can be selected together. As shown in FIG. 8, a diagnostic image is displayed in the area 20 on the left side of the screen. In the area 22 on the right side of the screen, a list and a plurality of buttons (221 to 228) for displaying a cross-sectional image used for calculating each feature amount are displayed. A button for displaying the cross-sectional image is referred to as a cross-section display button. The cross-section display button 221 is a button for displaying a cross-sectional image for calculating the CT value density. Similarly, the cross-section display button 222 is a button for displaying CT value density deviation, and the cross-section display button 223 is a button for displaying a cross-sectional image for calculating sphericity, flatness, and columnarity, and cross-section display. A button 224 is a button for displaying a cross-sectional image for calculating the cavity ratio, the number of cavities, and a concentration in the cavity, and a cross-section display button 225 is a button for displaying a cross-sectional image for calculating the number of columnar cavities, The cross-section display button 226 is a button for displaying a cross-sectional image for calculating the number of protrusions, the cross-section display button 227 is a button for displaying a cross-sectional image for calculating the maximum protrusion length, and the cross-section display button 228 is depressed. A button for displaying a cross-sectional image for calculating the number and a cross-sectional display button 229 are buttons for displaying a cross-sectional image for calculating the maximum depression depth. In this way, displaying the cross-section display button also has an effect of clearly indicating the related feature amount.

  When a user (such as a doctor) refers to the list, he or she may be wondering whether it is really possible to trust this feature value or the comprehensive judgment value. In this case, a user (such as a doctor) can change a desired feature amount or a comprehensive determination value displayed in the list via the operation unit 12. In the following, with reference to FIG. 9, the processing from the recalculation of the feature amounts and the comprehensive determination values displayed in the list format to the processing after step S05 and the display of the calculation results will be described. Note that the screen displayed first by the image display unit 13 is a screen that displays a diagnostic image and a list of feature amounts and comprehensive determination values as shown in FIGS. 3 and 8.

  First, the control unit 10 waits for a feature amount displayed in the list to be selected by the user (doctor or the like) via the operation unit 12 (step S11). Here, for the sake of explanation, it is assumed that flatness is selected. When the cross-section display button shown in FIG. 8 is displayed, a desired cross-section display button is selected via the operation unit 12.

  When the feature amount is selected (step S11: YES), the control unit 10 causes the image display unit 13 to display a cross-sectional image for confirming the feature amount according to the selected feature amount between the feature amount and the comprehensive determination value. A list is displayed (step S12). The cross-sectional image for confirming this feature amount is a cross-sectional image related to the cross-section determined by any of the cross-section determination routines R1, R2, and R3 described above. The cross-sectional image may be displayed on the image display unit 13 after being subjected to enlargement processing, color tone assignment processing for the abnormal candidate region, or the like by the image processing unit 15 as necessary.

  FIG. 10 is a diagram illustrating an example of the screen displayed in step S12. In the area 24 on the left side of the screen, a cross-sectional image used to calculate the selected feature amount is displayed. Since the flatness is selected in step S11, this cross-sectional image is a cross-sectional image in which the extracted abnormality candidate region 241 appears flattened. A list is displayed in the area 25 on the right side of the screen. In this list 25, all feature quantities related to the feature quantity selected in step S11 are displayed separately from other unrelated feature quantities. Since the flatness is selected in step S11, the sphericity and the columnarity, which are the feature quantities related to the shape, are displayed separately from the other feature quantities in the same manner as the flatness.

  FIG. 11 is a diagram showing an example different from FIG. 10 of the screen displayed in step S11, and is a diagram showing a screen displaying a cross-sectional image (a cross-sectional image related to the cavity) for calculating the feature quantity related to the cavity. As shown in FIG. 11, a cross-sectional image relating to the cavity is displayed in the area 30 on the left side of the screen, and a list is displayed in the area 31 on the right side of the screen. The area of the cavity depicted in the cross-sectional image relating to the cavity is clearly displayed. For this reason, it is possible to clearly see the hollow region that has not been extracted and the extracted hollow region. The list 31 displays the feature amounts (cavity ratio, the number of cavities, and the density in the cavity) calculated from the cross-sectional images related to the cavities, separately from other feature amounts.

  Next, in step S12, when the cross-sectional images and the list are displayed, the user (physician or the like) observes the screen and determines whether the CAD determination is correct. If it is determined that the determination is correct, the user (such as a doctor) selects the end button 26. When receiving an instruction to end interpretation by the end button 26 (step S13: NO), the control unit 10 ends the process.

  If it is determined that the determination is not correct, the user (doctor or the like) changes the value of the feature value or the value of the comprehensive determination value displayed in the list on the screen via the operation unit 12. For example, when the operation unit 12 includes a mouse or a touch panel, the circle displayed in the list may be moved. When the operation unit 12 includes a keyboard, a feature value is input. May be. In the example of FIG. 10, there is a region 242 that is not extracted as the abnormal candidate region 241 even though it is an abnormal candidate region. Therefore, the user (such as a doctor) can determine that the feature value or the comprehensive determination value of the abnormal candidate region is incorrect because there is an error in the extraction of the abnormal candidate region. Therefore, the user (doctor or the like) has obtained a basis for changing the value of the feature amount or the value of the comprehensive determination value in the list 25. Conversely, if the user (such as a doctor) determines that the extraction of the abnormal candidate region, the feature amount, and the overall determination value are correct, the interpretation ends. In any case, the user (physician or the like) can perform interpretation with certainty in the CAD determination. In addition, the storage unit 11 stores data of change contents. Thereby, the user (physician or the like) can browse the data of the change content later, and the data of the change content becomes useful statistical information for improving the quality of interpretation.

  When the feature amount is changed in step S13 (step S13: feature amount), the control unit 10 causes the comprehensive determination value calculation unit 17 to perform a comprehensive determination value change process. In the comprehensive determination value changing process, the comprehensive determination value calculation unit 17 recalculates the comprehensive determination value in accordance with the change of the feature value (step S14). When the comprehensive determination value is recalculated, the control unit 10 causes the image display unit 13 to display the changed feature amount, the feature amount that has not been changed, and the recalculated comprehensive determination value (step S15).

  FIG. 12 is a diagram illustrating an example of a screen displayed in step S15. In the area 24 on the right side of the screen, the same cross-sectional image as displayed in step S12 is displayed. In the area 27 on the right side of the screen, the changed feature values, the feature values that have not been changed, and the comprehensive judgment values are displayed in a list. FIG. 12 shows a case where the sphericity is changed in step S13. The sphericity shown in the list 27 is updated, and the comprehensive determination value is also updated. The feature quantity or comprehensive judgment value before change is indicated by ◯, and the feature quantity or comprehensive judgment value after change is indicated by □.

  When the comprehensive determination value is changed in step S13 (step S13), the control unit 10 causes the image display unit 13 to display the changed comprehensive determination value (step S16). Also in this case, as in step S15, the comprehensive judgment value before the change is displayed as ◯, and the comprehensive judgment value after the change is displayed as □.

  Interpretation ends when the user (doctor or the like) checks the screen displayed in step S15 or step S16.

  Thus, according to the present embodiment, it is possible to provide a CAD that improves the diagnostic accuracy of image interpretation.

(Modification 1)
In the above embodiment, the user (physician or the like) determines that the abnormal candidate area is malignant when the comprehensive determination value displayed in the list is 0.5 or more, and the abnormal candidate area when the comprehensive determination value is 0.5 or less. Is determined to be benign. However, the method for determining malignancy / benignity of the abnormal candidate region according to the present embodiment is not limited to the above example. The first modification is a method of automatically determining whether the condition is malignant or benign by applying a malignant / benign discrimination conditional expression stored in advance in the storage unit 10 to a plurality of feature amounts. In the following description, components having substantially the same functions as those of the present embodiment are denoted by the same reference numerals, and redundant description will be given only when necessary.

  FIG. 13 is a diagram illustrating a configuration of the CAD 1a according to the first modification of the present embodiment. The CAD 1a includes a storage unit 11, an operation unit 12, an image display unit 13, an image processing unit 15, a feature amount calculation unit 16, a comprehensive determination value calculation unit 17, and a malignant / benign determination unit 18 with the control unit 10 as a center. Have. The malignant / benign determining unit 18 applies the plurality of feature amounts calculated by the feature amount calculating unit 16 to the malignant / benign discriminant conditional expressions and uses them in the malignant / benign discriminant conditional expressions when the conditions are satisfied. Data indicating malignancy (hereinafter referred to as malignant data) is added to the feature amount data. This malignant / benign discrimination conditional expression is called a decision tree. The decision tree is devised based on medical knowledge and is given a comparative discriminant between each feature quantity and a reference value. The image display unit 13 displays the comprehensive determination value and the plurality of feature amounts in a list. The image display unit 13 distinguishes and displays the feature amount to which the malignant data is given by the malignant / benign discrimination unit 18 and the feature amount to which the malignant data is not given.

  FIG. 14 is a diagram illustrating a simple example of a decision tree. An example of a process for adding malignant data will be described taking the decision tree shown in FIG. 14 as an example. As shown in FIG. 14, first, the malignant / benign determination unit 18 determines whether or not the standardized sphericity is 0.5 (reference value) or more. When it is determined that the normalized sphericity is 0.5 (reference value) or more, the malignancy / benignity determination unit 18 adds malignant data (for example, 1) to the feature value data of sphericity. To do. When it is determined that the normalized sphericity is 0.5 (reference value) or less, the malignancy / benignity determination unit 18 determines whether the normalized CT value concentration is 0.5 (reference value) or more. Is determined. When it is determined that the standardized CT value density is 0.5 (reference value) or more, the malignant / benignity determination unit 18 adds malignant data to the feature amount called CT value density. When it is determined that the standardized CT value concentration is 0.5 (reference value) or less, the malignant / benignity determination unit 9 similarly determines whether the standardized number of depressions is 0.5 (reference value) or more. Determine if: When it is determined that the standardized number of depressions is 0.5 (reference value) or more, the malignancy / benignity determination unit 18 adds malignant data to the feature amount data called the number of depressions. If the standardized number of depressions is determined to be 0.5 (reference value) or less, the malignant / benign determination unit 18 determines that the abnormal candidate region is benign, and ends the decision tree.

  FIG. 15 is a diagram showing an example of a list reflecting the result of the decision tree, and is a diagram when the value of sphericity is involved in the determination of malignancy. As illustrated in FIG. 15, the image display unit 13 displays the sphericity item separately from other feature amount items according to the malignant data assigned to the sphericity item. In the example of FIG. 15, the image display unit 13 distinguishes the sphericity item from the other items by displaying the sphericity item as Δ and the other items as ○. There are other ways of distinguishing, for example, the color of an item to be distinguished may be changed or blinked.

  For example, in the second comparison discriminant of the decision tree of FIG. 14 (is the normalized CT value concentration 0.5 or higher?), The malignant / benign discriminating unit 18 assigns malignant data to the feature value of CT value concentration. In this case, it is the CT value concentration that is involved in the determination of malignancy. In this case, the image display unit 13 displays the CT value density item by Δ.

  Also in the first modification, a cross section used when a user (physician or the like) calculates a desired feature amount in the image display unit 13 by selecting a desired feature amount or a section display button via the operation unit 12. It is also possible to display an image. Further, when the user (doctor or the like) changes the feature value values displayed in the list on the image display unit 13 via the operation unit 12, the comprehensive determination value calculation unit 17 calculates the value of the comprehensive determination value. In addition, the malignancy / benignity determination unit 18 may apply the changed feature amount to the malignancy / benignity discriminant.

  Furthermore, the determination of malignancy / benignity by the malignancy / benignity determination unit 18 in the first modification may be performed using a neural network that is an existing technology instead of a decision tree.

  Thus, according to the first modification, it is possible to provide a CAD that improves the diagnostic accuracy of image interpretation.

(Modification 2)
In the second modification of the present embodiment, an example in which an overall determination value is calculated by applying a plurality of feature amounts to a malignancy / benignity determination diagram is shown. In the following description, components having substantially the same functions as those of the present embodiment are denoted by the same reference numerals, and redundant description will be given only when necessary.

  FIG. 16 is a diagram illustrating a configuration of the CAD 1b according to the second modification of the present embodiment. The CAD 1b is centered on the control unit 10, and includes a storage unit 11, an operation unit 12, an image display unit 13, an image processing unit 15, a feature amount calculation unit 16, a comprehensive determination value calculation unit 17, a malignant / benign determination unit 18, and a determination. And an index value calculation unit 19.

  The comprehensive determination value calculation unit 17 calculates a comprehensive determination value by applying a plurality of feature amounts to the malignancy / benignity determination diagram stored in advance in the storage unit 11. The determination index value calculation unit 19 uses an index value (hereinafter referred to as a determination index value) indicating the contribution of each of the plurality of feature amounts to the overall determination value based on the plurality of feature amounts and the malignancy / benignity determination diagram. Calculate for each. The image display unit 13 displays the comprehensive determination value and the plurality of determination index values in a list.

  FIG. 17 is a diagram illustrating an example of a malignant / benign determination diagram. The horizontal axis of the malignant / benign judgment diagram is the normalized feature quantity 1, and the vertical axis is the normalized feature quantity 2. In FIG. 17, the value of the degree of abnormality is indicated by contour lines. In FIG. 17, the lower left is the degree of abnormality a = 0, the upper right of FIG. 17 is the degree of abnormality a = 1, and the degree of abnormality a = 0.5. The line is divided into a benign range and a malignant range. In addition, although this malignant / benign judgment diagram shows a diagram using two feature values for the sake of explanation, it is not necessary to limit to this and three or more feature values may be used. Hereinafter, the process of the modified example 2 will be described using a case where two feature values are used as an example.

  The comprehensive determination value calculation unit 17 applies the two feature amounts to a malignancy / benignity determination diagram as shown in FIG. 17, and specifies the intersection position of the two feature amounts on the malignancy / benignity determination diagram as a comprehensive determination value. When a list showing the values of the comprehensive determination value and the feature amount specified on the image display unit 13 is displayed, the user (doctor or the like) simply sees the feature value and the feature amount and the comprehensive determination value. It is not possible to understand how these are related. Therefore, the determination index value calculation unit 19 calculates a plurality of determination index values indicating how each feature amount contributes to the overall determination value. Then, the image display unit 13 displays the calculated plurality of determination index values and the total determination value as a list.

  In the description of the second modification of the present embodiment, it is assumed that the determination index values are a normal suggestion degree, an abnormality suggestion degree, a minimum abnormality degree, and a maximum abnormality degree. These four determination index values are determined one by one for one feature amount. Hereinafter, four determination index values (normal suggestion degree, abnormality suggestion degree, minimum abnormality degree, and maximum abnormality degree) will be described.

  Now, it is assumed that the intersection point a (p1, p2) of two feature quantities is at x in FIG. A broken line intersecting with x shown in FIG. 17 indicates a range of possible values of the feature amount 1 when the feature amount 2 is fixed to p2. The minimum comprehensive judgment value that the feature quantity 1 can take when the feature quantity 2 takes the value p2 is defined as the minimum abnormality degree amin1p2, and the maximum comprehensive judgment value that the feature quantity 2 takes the value p2 is defined as the maximum abnormality degree amax1p2. . Further, the abnormality suggestion degree b (p1, p2) = a (p1, p2) −amin1p2, and the normal suggestion degree c (p1, p2) = amax1p2−a (p1, p2) are defined.

  FIG. 18 is a diagram showing a graph of the value of feature amount 1 and the overall determination value a at the position of the broken line in FIG. In this graph, the normalized feature quantity 1 is defined on the horizontal axis, and the overall determination value a is defined on the vertical axis. As shown in FIG. 18, when the value of the feature quantity 1 is 0, the comprehensive determination value a has a minimum abnormality degree amin1p2. The comprehensive determination value increases as the value of the feature amount 1 increases. When the value of the feature quantity 1 is 1, the comprehensive determination value a has the maximum abnormality degree amax1p2.

  The degree of abnormality suggestion b (p1, p2) and the degree of normal suggestion c (p1, p2) indicate how much the corresponding feature amount is involved in the comprehensive judgment value. That is, there is a high possibility that the comprehensive determination value a moves to a normal range by changing a feature value having a high degree of abnormality suggestion. For example, since the feature amount 1 has a minimum abnormality degree amin1p2 of less than 0.5, the total judgment value may move to a normal range when the value of the feature amount 1 is lowered. On the other hand, the overall judgment value does not move into the benign range no matter how much the value of the feature amount having the minimum abnormality degree of 0.5 or more is lowered.

  FIG. 19 shows a list indicating the names and values of the overall determination values and the four determination index values (normal suggestion level, abnormal suggestion level, minimum abnormal level, and maximum abnormal level) corresponding to each of the plurality of feature amounts. FIG. The comprehensive judgment value displayed in the feature quantity list and the comprehensive judgment value displayed in the minimum abnormality degree and maximum abnormality degree list indicate the same value.

  FIG. 20 is a diagram illustrating an example of a screen displaying a diagnostic image and a list of determination index values. As shown in FIG. 20, a diagnostic image is displayed in the area 60 on the right side of the screen. In the area 61 on the right side of the screen, a list of determination index values is displayed. When a list of determination index values is presented, a user (such as a doctor) may determine that the comprehensive determination value derivation method itself is incorrect. That is, the feature values are all valid, but the user thinks that the comprehensive judgment value is obviously not valid. In that case, the user (doctor or the like) selects the combination of feature amounts in the malignancy / benignity determination diagram by selecting the “Register Normal Example” button 62 or the “Register Abnormal Example” button 63 via the operation unit 12. It is possible to change.

  In addition to the normal / abnormal determination, the benign / malignant determination of the nodule candidate site is performed separately from the numerical value related to the normal / abnormal determination. , Minimum malignancy, malignancy suggestion, benign suggestion) may be calculated and displayed.

  Also in the second modification of the present embodiment, the user (physician or the like) selects a desired feature value or a section display button via the operation unit 12, thereby obtaining the cross-sectional image used when calculating the desired feature value. It is also possible to display. It is also possible for the user (doctor or the like) to recalculate the value of the comprehensive determination value by changing the value of the determination index value displayed in a list on the image display unit 13 via the operation unit 12. .

  When inference is performed using a neural network or the like, the malignancy / benignity determination diagram can be changed by adding a specified combination of feature amounts to a set of teaching signals.

  Thus, according to the second modification, it is possible to provide a CAD that improves the diagnostic accuracy of image interpretation.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

The figure which shows the structure of the medical image interpretation assistance apparatus (CAD) which concerns on this embodiment. The figure which shows the flow of a process from extraction of an abnormality candidate area | region in this embodiment to displaying the list of a feature-value and a comprehensive determination value. The figure which shows an example of the screen displayed by step S05 of FIG. The figure which shows the process of cross-section determination routine R1 which concerns on this embodiment. The figure which shows the process of the cross-section determination routine R2 which concerns on this embodiment. The figure which shows the area | region and cross section of the cavity in connection with the process of cross section determination routine R2 which concerns on this embodiment. The figure which shows the process of the cross-section determination routine R3 which concerns on this embodiment. The figure which shows an example of the screen which displayed in step S05 of FIG. 2, and displayed the cross-section display button. The figure which shows the flow of a process until it recalculates the feature-value and comprehensive determination value which were displayed by list format regarding the process after step S05 of FIG. 2, and displays the calculation result. The figure which shows an example of the screen displayed by FIG.9 S12 The figure which shows an example different from FIG. 10 of the screen displayed by FIG.9 S12. The figure which shows an example of the screen displayed by FIG.9 S15. The figure which shows the structure of the medical image interpretation assistance apparatus (CAD) in the modification 1 of this embodiment. The figure which shows the very simple example of the decision tree which concerns on the modification 1. FIG. The figure which shows an example of the list which reflected the result of the decision tree of FIG. The figure which shows the structure of the medical image interpretation assistance apparatus (CAD) in the modification 2 of this embodiment. The figure which shows an example of the malignancy / benign judgment figure in the modification 2. The figure which showed the graph of the value of the feature-value 1 (PA) and the comprehensive determination value a in the broken-line position of FIG. The figure which shows the list of four determination parameter | index values displayed on the image display part of FIG. The figure which shows an example of the screen which displayed the diagnostic image displayed on the image display part of FIG. 16, and the list of determination index values.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Medical image interpretation assistance apparatus (CAD), 10 ... Control part, 11 ... Memory | storage part, 12 ... Operation part, 13 ... Image display part, 15 ... Image processing part, 16 ... Feature-value calculation part, 17 ... Comprehensive determination value Calculation unit, 18 ... malignant / benign discrimination unit, 19 ... determination index value calculation unit

Claims (11)

  1. A storage unit for storing volume data relating to the subject;
    An abnormal candidate region extraction unit for extracting an abnormal candidate region from the volume data;
    A feature amount calculation unit that calculates a plurality of types of feature amounts for the extracted abnormality candidate region;
    A comprehensive determination value calculation unit for calculating a comprehensive determination result based on the calculated plural types of feature amounts;
    An image display unit for displaying the plurality of types of feature amounts and the comprehensive determination result;
    A medical image interpretation support apparatus comprising :
    A feature quantity selection unit for selecting a desired one from the plurality of types of feature quantities;
    A cross-sectional position / direction determining unit that determines the cross-sectional position and direction of the abnormal candidate region where the selected feature amount is maximum or minimum;
    A cross-sectional image generator for generating a cross-sectional image based on the determined position and direction of the cross-section;
    Further comprising
    The image display unit displays the generated cross-sectional image;
    A medical image interpretation support apparatus characterized by that.
  2. A storage unit for storing volume data relating to the subject;
    An abnormal candidate region extraction unit for extracting an abnormal candidate region from the volume data;
    A feature amount calculation unit that calculates a plurality of types of feature amounts for the extracted abnormality candidate region;
    A comprehensive determination value calculation unit for calculating a comprehensive determination result based on the calculated plural types of feature amounts;
    An image display unit for displaying the plurality of types of feature amounts and the comprehensive determination result;
    A medical image interpretation support apparatus comprising :
    A feature amount changing unit that changes at least one value of the plurality of types of feature amounts;
    The comprehensive determination value calculation unit recalculates the comprehensive determination result based on the changed feature amount,
    The image display unit displays the changed feature amount and the recalculated comprehensive determination value.
    A medical image interpretation support apparatus characterized by that.
  3. The medical image interpretation support apparatus according to claim 2 , wherein the image display unit displays the changed feature quantity separately from the feature quantity that has not been changed.
  4. The medical image interpretation support apparatus according to claim 1 or 2, wherein the image display unit displays a plurality of selectable buttons for displaying cross-sectional images used for calculating the plurality of types of feature amounts.
  5. The feature amount calculation unit normalizes the values of the plurality of types of feature amounts to a certain range,
    The image display unit displays a list of names and normalized values related to the plurality of types of feature values.
    The medical image interpretation support apparatus according to claim 1 or 2 .
  6. The medical image interpretation support apparatus according to claim 5 , wherein the image display unit displays a normalized value related to the plurality of types of feature amounts as a graph plotted within the certain range.
  7. A storage unit for storing volume data relating to the subject;
    An abnormal candidate region extraction unit for extracting an abnormal candidate region from the volume data;
    A feature amount calculation unit for calculating a plurality of types of feature amounts for the extracted abnormality candidate region;
    Applying the plurality of types of feature quantities to a plurality of malignant / benign discriminant conditional expressions given in advance, a malignant / benign discriminating unit for discriminating the malignant / benign of the abnormal candidate region,
    A comprehensive determination value calculation unit for calculating a comprehensive determination value based on the plurality of types of feature amounts;
    Features of determination results to differ determination result for the abnormality candidate area feature quantity determination result same determination result for the abnormality candidate region by the malignant, benign determination unit among the plurality of types of feature quantities An image display unit that displays the comprehensive judgment value and the plurality of types of feature quantities in a list,
    A medical image interpretation support apparatus comprising:
  8. The medical image interpretation support apparatus according to claim 7 , wherein the plurality of malignant / benign discriminant conditional expressions are decision trees including a plurality of comparison discriminants for individually comparing a feature amount and a predetermined reference value.
  9. A storage unit for storing volume data relating to the subject;
    An abnormal candidate region extraction unit for extracting an abnormal candidate region from the volume data;
    A feature amount calculation unit that calculates a plurality of types of feature amounts for the extracted abnormality candidate region;
    A comprehensive determination value calculating unit that calculates a comprehensive determination value by applying the plurality of types of feature amounts to a predetermined malignancy / benignity determination diagram;
    A determination value calculation unit that calculates a plurality of types of index values indicating the contribution of the plurality of types of feature amounts to the total determination value for each of the feature amounts based on the malignancy / benignity determination diagram;
    An image display unit for displaying the comprehensive judgment value and the plurality of types of index values in a list;
    A medical image interpretation support apparatus comprising:
  10. The medical image interpretation support apparatus according to claim 9 , wherein the image display unit displays a list of names and values of the plurality of types of index values.
  11. The malignant / benign judgment diagram is a diagram showing contour lines of the comprehensive judgment value defined by at least two types of feature amounts;
    The comprehensive determination value calculation unit specifies the value of the contour line at the position of the intersection of the values of the at least two types of feature values as the comprehensive determination value;
    The medical image interpretation support apparatus according to claim 9 .
JP2008050692A 2008-02-29 2008-02-29 Medical image interpretation support device Active JP5179902B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008050692A JP5179902B2 (en) 2008-02-29 2008-02-29 Medical image interpretation support device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2008050692A JP5179902B2 (en) 2008-02-29 2008-02-29 Medical image interpretation support device

Publications (2)

Publication Number Publication Date
JP2009207541A JP2009207541A (en) 2009-09-17
JP5179902B2 true JP5179902B2 (en) 2013-04-10

Family

ID=41181300

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2008050692A Active JP5179902B2 (en) 2008-02-29 2008-02-29 Medical image interpretation support device

Country Status (1)

Country Link
JP (1) JP5179902B2 (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8798227B2 (en) 2010-10-15 2014-08-05 Kabushiki Kaisha Toshiba Medical image processing apparatus and X-ray computed tomography apparatus
WO2012050166A1 (en) * 2010-10-15 2012-04-19 株式会社 東芝 Medical image-processing apparatus and x-ray computer tomography apparatus
US10049445B2 (en) 2011-07-29 2018-08-14 Canon Kabushiki Kaisha Image processing apparatus and image processing method of a three-dimensional medical image
WO2016121811A1 (en) * 2015-01-29 2016-08-04 富士フイルム株式会社 Image processing device, image processing method, and endoscope system
JP6336949B2 (en) * 2015-01-29 2018-06-06 富士フイルム株式会社 Image processing apparatus, image processing method, and endoscope system
JP6147304B2 (en) * 2015-09-01 2017-06-14 キヤノン株式会社 Medical diagnosis support device and medical diagnosis support system
JP6257564B2 (en) * 2015-09-01 2018-01-10 キヤノン株式会社 Diagnosis support apparatus, diagnosis support system, operation method and program for diagnosis support apparatus
WO2017057574A1 (en) * 2015-09-29 2017-04-06 富士フイルム株式会社 Image processing apparatus, endoscope system, and image processing method
EP3357405A4 (en) * 2015-09-29 2018-11-07 FUJI-FILM Corporation Image processing device, endoscopic system, and image processing method
WO2017057572A1 (en) * 2015-09-29 2017-04-06 富士フイルム株式会社 Image processing device, endoscopic system, and image processing method
JP6313741B2 (en) * 2015-12-25 2018-04-18 キヤノン株式会社 Image processing apparatus and method of operating image processing apparatus
JP6334767B1 (en) * 2017-03-21 2018-05-30 株式会社東芝 Information processing apparatus, program, and information processing method
JP6403832B2 (en) * 2017-05-18 2018-10-10 キヤノン株式会社 Diagnosis support apparatus, diagnosis support method and program

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4253055B2 (en) * 1998-04-27 2009-04-08 東芝医用システムエンジニアリング株式会社 Computer-aided diagnosis device
JP2002074325A (en) * 2000-08-31 2002-03-15 Fuji Photo Film Co Ltd Method and detector for detecting abnormal shadow candidate
JP2002112985A (en) * 2000-10-06 2002-04-16 Konica Corp Diagnostic imaging support device
JP4604451B2 (en) * 2003-02-24 2011-01-05 コニカミノルタホールディングス株式会社 Medical image processing apparatus and malignancy determination method
DE10319546A1 (en) * 2003-04-30 2004-11-25 Siemens Ag Automatic vein structure anomaly detection procedure recognises vein structure in tomography data, prepares skeletal path and classifies by comparison with characteristic values
JP2005168962A (en) * 2003-12-15 2005-06-30 Canon Inc Diagnostic support apparatus
JP4513365B2 (en) * 2004-03-12 2010-07-28 コニカミノルタエムジー株式会社 Medical image processing apparatus and medical image processing program
JP4549093B2 (en) * 2004-04-12 2010-09-22 キヤノン株式会社 Image processing apparatus and method, and program
JP4480508B2 (en) * 2004-08-02 2010-06-16 富士通株式会社 Diagnosis support program and diagnosis support apparatus
US8139829B2 (en) * 2005-02-09 2012-03-20 Hitachi Medical Corporation Diagnostic imaging support system and diagnostic imaging support program
JP4599191B2 (en) * 2005-03-01 2010-12-15 国立大学法人神戸大学 Diagnostic imaging processing apparatus and diagnostic imaging processing program
JP2006255065A (en) * 2005-03-16 2006-09-28 Fuji Photo Film Co Ltd Image output method, image output device and program
JP2007105164A (en) * 2005-10-12 2007-04-26 Hitachi Medical Corp Medical diagnostic imaging support apparatus

Also Published As

Publication number Publication date
JP2009207541A (en) 2009-09-17

Similar Documents

Publication Publication Date Title
US7876939B2 (en) Medical imaging system for accurate measurement evaluation of changes in a target lesion
JP4820680B2 (en) Medical image display device
US7633501B2 (en) Graphical user interface for display of anatomical information
EP1764686A1 (en) System and method for dynamic configuration of pacs workstation displays
US7149334B2 (en) User interface for computed tomography (CT) scan analysis
EP1704816B1 (en) Information processing apparatus and method
EP2710957A2 (en) Method and system for intelligent qualitative and quantitative analysis of digital radiography softcopy reading
JP2008541889A (en) System and method for computer-aided detection
US6925200B2 (en) Graphical user interface for display of anatomical information
JP2008521468A (en) Digital medical image analysis
JP4949264B2 (en) System and method for automatically detecting and segmenting tumor boundaries in medical image data
JP2005103055A (en) Medical image processor
US9168007B2 (en) Computer-aided diagnostic apparatus
US6901156B2 (en) Method, system and computer readable medium for an intelligent search workstation for computer assisted interpretation of medical images
US20080240494A1 (en) Case image search apparatus and system
Andriole et al. Optimizing analysis, visualization, and navigation of large image data sets: one 5000-section CT scan can ruin your whole day
JP5312801B2 (en) Medical image viewing protocol
JP2010246776A (en) Diagnosis assisting apparatus, diagnosis assisting program, and diagnosis assisting method
WO2006042191A2 (en) Systems and methods for interactive navigation and visualization of medical images
US8797350B2 (en) Dynamic customizable human-computer interaction behavior
AU2004266022A1 (en) Computer-aided decision support systems and methods
JP3083606B2 (en) Medical diagnosis support system
US8160320B2 (en) Medical image display apparatus, method and program, and recording medium for the program
WO2007139633A2 (en) Digital mammography system with improved workflow
US10332251B2 (en) Methods and systems for automatically mapping biopsy locations to pathology results

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20110228

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20120928

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20121002

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20121122

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20121218

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20130110

R150 Certificate of patent or registration of utility model

Ref document number: 5179902

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150

S111 Request for change of ownership or part of ownership

Free format text: JAPANESE INTERMEDIATE CODE: R313117

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350

S533 Written request for registration of change of name

Free format text: JAPANESE INTERMEDIATE CODE: R313533

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350