WO2009009145A1 - Procédé de détection de tissu permettant une visualisation et un diagnostic assistés par ordinateur en présence d'un marquage - Google Patents

Procédé de détection de tissu permettant une visualisation et un diagnostic assistés par ordinateur en présence d'un marquage Download PDF

Info

Publication number
WO2009009145A1
WO2009009145A1 PCT/US2008/008564 US2008008564W WO2009009145A1 WO 2009009145 A1 WO2009009145 A1 WO 2009009145A1 US 2008008564 W US2008008564 W US 2008008564W WO 2009009145 A1 WO2009009145 A1 WO 2009009145A1
Authority
WO
WIPO (PCT)
Prior art keywords
tissue
voxels
volume
probability
updating
Prior art date
Application number
PCT/US2008/008564
Other languages
English (en)
Other versions
WO2009009145A9 (fr
Inventor
Sarang Lakare
Original Assignee
Siemens Medical Solutions Usa, Inc.
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
Priority claimed from US12/170,596 external-priority patent/US8213696B2/en
Application filed by Siemens Medical Solutions Usa, Inc. filed Critical Siemens Medical Solutions Usa, Inc.
Publication of WO2009009145A1 publication Critical patent/WO2009009145A1/fr
Publication of WO2009009145A9 publication Critical patent/WO2009009145A9/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/143—Segmentation; Edge detection involving probabilistic approaches, e.g. Markov random field [MRF] modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based 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
    • G06T2207/10072—Tomographic images
    • 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

Definitions

  • the present disclosure relates to a tissue detection method for computer aided diagnosis, more specifically, to a tissue detection method for computer aided diagnosis and visualization in the presence of tagging.
  • the diagnostically superior information available from data acquired from current imaging systems enables the detection of potential problems at earlier and more treatable stages.
  • various methods must be developed to efficiently and accurately process image data.
  • advances in image processing are generally performed on digital or digitized images.
  • Digital images are created from an array of numerical values representing a property (such as a grey scale value or magnetic field strength) associable with a set of anatomical location points referenced by a particular array location.
  • the set of anatomical location points comprises the domain of the image.
  • 2 -D digital images, or slice sections the discrete array locations are termed pixels.
  • Three-dimensional digital images can be constructed from stacked slice sections through various construction techniques known in the art.
  • the 3-D images are made up of discrete volume elements, also referred to as voxels, composed of pixels from the 2 -D images.
  • the pixel or voxel properties can be processed to ascertain various properties about the anatomy of a patient associated with such pixels or voxels.
  • Computer-aided diagnosis (“CAD”) systems play a critical role in the analysis and visualization of digital imaging data.
  • volumetric datasets are important for many applications, including medical imaging, finite element analysis, mechanical simulations, etc.
  • 3- dimemsional datasets obtained from scanning modalities such as computed tomography (CT) , magnetic resonance imaging (MRI) , positron emission tomography (PET), ultrasound (US), etc., are usually quite complex, and contain many different objects and structures .
  • CT computed tomography
  • MRI magnetic resonance imaging
  • PET positron emission tomography
  • US ultrasound
  • the region of interest to the user is surrounded either partially or completely by other objects and structures. There is often a need to either remove an obstructing surrounding object, or to keep the region of interest and remove all other objects.
  • Visualization of an image can be accomplished by volume rendering the image, a set of techniques for displaying, three-dimensional volumetric data onto a two-dimensional display image.
  • resulting intensity values or ranges of values can be correlated with specific types of tissue, enabling one to discriminate, for example, bone, muscle, flesh, and fat tissue, nerve fibers, blood vessels, organ walls, etc., based on the intensity ranges of the voxels within the image.
  • the raw intensity values in the image can serve as input to a transfer function whose output is a transparency or opacity value that can characterize the type of tissue.
  • a user can then generate a synthetic image from a viewing point by propagating rays from the viewing point to a point in the 2 -D image to be generated and integrating the transparency or opacity values along the path until a threshold opacity is reached, at which point the propagation is terminated.
  • the use of opacity values to classify tissue also enables a user to select which tissue is to be displayed and only integrate opacity values corresponding to the selected tissue. In this way, a user can generate synthetic images showing, for example, only blood vessels, only muscle, only bone, etc.
  • Three-dimensional volume editing is performed in medical imaging applications to provide for an unobstructed view of an object of interest, such as a fetus face.
  • an object of interest such as a fetus face.
  • the view of the fetus face may be obstructed by the presence of the umbilical cord in front of the fetal head. Accordingly, the obstructing cord should be removed via editing techniques to provide an unobstructed image of the face.
  • Existing commercial software packages perform the clipping either from one of three orthogonal two-dimensional (2D) image slices or directly from the rendered 3D image.
  • Tagging using a contrast agent is a commonly used technique for highlighting a particular object in imaged data. Tagging is often used to highlight an object of interest, and at times, is also used to highlight an object that is not desirable, but whose physical removal is either impossible or difficult and impractical. For example, tagging is often used in virtual colonoscopy to highlight residual material inside the colon. Physical removal of the residual material is impractical as that can cause significant discomfort for the patient being examined.
  • tagging is commonly used in medical imaging to distinguish structures of interest from the rest of the data.
  • the structures of interest are tissues, tagging is used to highlight non-tissue structures that are not of interest .
  • Three dimensional (3D) datasets obtained from scanning modalities such as computed tomography (CT) , magnetic resonance imaging (MRI), etc.
  • CT computed tomography
  • MRI magnetic resonance imaging
  • a method for obtaining a tissue volume includes the steps of inputting a dataset including a plurality of voxels; initializing a tissue probability volume for the plurality of voxels to a pre-determined value; updating, by one of increasing or decreasing the tissue probability volume of each of the plurality of voxels, based on corresponding intensity values of each of the plurality of voxels; and generating the tissue volume by combining the updated tissue probability volume and the inputted dataset.
  • the step of updating may further comprise updating the tissue probability volume of each of the plurality of voxels based on corresponding entropy values .
  • the step of updating may further comprise updating the tissue probability volume of each of the plurality of voxels based on whether the voxels belong to partial volume regions.
  • the step of updating may further comprise updating the tissue probability volume of each of the plurality of voxels based on whether the voxels are connected to each other.
  • the step of updating may further comprise updating the tissue probability volume of each of the plurality of voxels based on a probability of the voxels being disconnected from each other.
  • the dataset may be an output from a computed tomography (CT) scanner.
  • the tissue volume may be used for computer aided detection (CAD) of tissues of interest.
  • CAD computer aided detection
  • the tissues of interest may be colonic polyps.
  • the method may further comprise the step of tagging non- tissue structures.
  • a method for obtaining a tissue volume includes the steps of scanning an anatomical structure to produce a dataset including a plurality of voxels; updating a tissue probability volume of each of the plurality of voxels based on a calculated probability of the voxels belonging to a tissue; and generating the tissue volume by combining the updated tissue probability volume and the dataset.
  • the calculated probability of the voxels belonging to a tissue may further be based on intensity values of the voxels.
  • the calculated probability of the voxels belonging to a tissue may further be based on entropy values of the voxels.
  • the calculated probability of the voxels belonging to a tissue may further be based on whether the voxels belong to partial volume regions.
  • the calculated probability of the voxels belonging to a tissue may further be based on whether the voxels are connected to each other.
  • the calculated probability of the voxels belonging to a tissue may further be based on a probability of the voxels being disconnected from each other.
  • the step of scanning may be performed by a computed tomography (CT) scanner.
  • CT computed tomography
  • the anatomical structure may be a colon.
  • the tissue volume may be used for computer aided detection (CAD) of colonic polyps of the colon.
  • the method may further comprise the step of tagging non- tissue structures.
  • a computer system includes a processor; and a program storage device readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for obtaining a tissue volume, the method including scanning an anatomical structure to produce a dataset including a plurality of voxels; updating a tissue probability volume of each of the plurality of voxels based on a calculated probability of the voxels belonging to a tissue; and generating the tissue volume by combining the updated tissue probability volume and the dataset.
  • FIG. 1 is a flowchart of a method for detecting tissue regions, according to an exemplary embodiment of the present invention
  • FIG. 2 is a diagram showing a step of updating a tissue probability based on voxel intensity, according to an exemplary embodiment of the present invention
  • FIG. 3 is a flowchart of a method for finding connected tissue regions which appear to be disconnected, according to an exemplary embodiment of the present invention
  • FIG. 4 is a flowchart of a method for generating tissue volume, according to an exemplary embodiment of the present invention.
  • FIG. 5 is a diagram showing a step of assigning intensities to voxels based on their tissue probability when generating the tissue volume, according to an exemplary embodiment of the present invention
  • FIG. 6 is a flowchart of a method for computer aided detection (CAD) , visualization, and measurement of tissue, according to an exemplary embodiment of the present invention.
  • CAD computer aided detection
  • FIG. 7 shows an example of a computer system capable of implementing the tissue detection methods according to exemplary embodiments of the present invention.
  • Exemplary embodiments of the present invention seek to provide methods for detecting tissue regions while isolating non-tissue regions.
  • FIG. 1 is a flowchart of a method for detecting tissue regions, according to an exemplary embodiment of the present invention.
  • a tissue probability volume is used to represent the presence of tissue in any given voxel of the input dataset (scan data) .
  • a dataset is input (step 110) and after initializing the probabilities of each voxel in the initialization stage (step 120) , the tissue probability of each voxel is updated in each of the steps as described below.
  • a probability update is performed by either adding or subtracting a small delta from the probability associated with the voxel at that time.
  • the delta amount is defined for each step.
  • the end result is a tissue probability- volume that has higher probabilities for voxels which represent tissue versus those that belong to other, non-tissue materials in the scan.
  • the probability volume is initialized to a value of 0.5 at every voxel, indicating equal probability of the voxel belonging to a tissue material or to a non-tissue material.
  • step 130 an intensity based probability update is performed.
  • the tissue probability at each voxel is updated based on the intensity value (scalar value from the scan) at that voxel.
  • the voxel intensity match the tissue intensity, the probability of being tissue for the voxel is increased, otherwise it is decreased.
  • the expected intensity range for tissue material is based on a priori knowledge of tissue intensities.
  • the tissue voxels usually have a value between -200 and 200 Hounsfield units (HU) .
  • HU Hounsfield units
  • the probability of a voxel belonging to tissue is increased if its intensity is between -300 and 300 HU and decreased if the voxel intensities are outside of that range.
  • FIG. 2 is an example graph showing the mapping of voxel intensity values to the delta change in tissue probability, according to an exemplary embodiment of the present invention.
  • the tissue probability at each voxel is updated by applying the delta change as per the mapping.
  • step 130 intensity values are mapped into a bounded change in the voxel's probability.
  • step 140 an entropy based probability update is performed.
  • Tissue regions usually have homogeneous intensity voxels without much noise. However, non-tissue regions may have noise which can result in voxels with intensities in the tissue range. In order to differentiate these voxels from the real tissue voxels, noise is measured using an entropy computation.
  • Entropy is defined as a quantity specifying the amount of disorder or randomness. We compute the entropy at a given voxel as :
  • tissue probability of voxels is updated in this step based on the entropy value, by lowering the tissue probability of voxels that have higher entropy values .
  • step 150 a partial volume detection is performed.
  • Partial volume may result in incorrect classification of some voxels as tissue voxels.
  • a PV detector is used to determine voxels that belong to PV regions.
  • voxels that belong to PV between non-tissue regions are separated from those that might be next to tissue regions. For voxels in the PV region between non-tissue regions, the tissue probability is reduced.
  • the 3D neighborhood of the voxel is analized and the number of voxels with intensities above and below the tissue range is counted. If both these counts are above a certain threshold t, the voxel is marked as a PV voxel and its tissue probability is reduced.
  • step 160 a connected tissue detection is performed.
  • Connected tissues are determined by first finding a seed point to perform a region grow in a region that already has a high tissue probability.
  • a high tissue probability is defined by a threshold T h ⁇ g h . Any voxel with probability value greater than T h i gh is considered a high tissue probability voxel.
  • the size of the region for finding the seed point may be configured to make sure small, noisy structures are ignored.
  • a connected component region grow is performed to determine all those high tissue probability voxels that are connected to each other to form a connected region.
  • the tissue probability of all these connected voxels is increased, whereas for the disconnected voxels, it is decreased.
  • FIG. 3 is a flowchart of a method for finding connected tissue regions which appear to be disconnected, according to an exemplary embodiment of the present invention.
  • tissue regions are physically connected in the body, a separation may be shown in the scanned data because of a low sampling frequency or the presence of partial volume. In this step, such regions are identified which otherwise might be considered as disconnected.
  • the method of FIG. 3 is used after a first connected region (Rl) is determined in the connected tissue detection step 160 of FIG. 1. From each voxel on the surface of the first connected region (step 360) , lower probability voxels are searched in the neighborhood that have not been considered as part of the first connected region (step 310) .
  • a lower tissue probability is defined by thresholds T h ig h and Ti o w Any voxel with probability value between T h i gh and Ti owt is considered a lower tissue probability voxel.
  • a region grow (step 320) is performed on this lower probability region (LPR) .
  • this region touches a higher probability region (HPR) that has not been part of the first region (step 330) , then a low probability connection is found. Then, a region grow is performed in the higher probability region (step 340) , and the probability of the connection as well as the newly discovered regions is increased and merged with Rl (step 350) .
  • HPR higher probability region
  • the steps are repeated until no new lower probability voxels are found next to the first connected region (Rl) .
  • FIG. 4 is a flowchart of a method for generating the tissue volume, according to an exemplary embodiment of the present invention.
  • the input data (410) and a tissue probability volume (430) are combined to generate and output the tissue volume (460) .
  • the tissue volume is a recreation of the input volume such that it only has voxels that belong to the tissue. The rest of the voxels are assigned air intensities so that they can be easily identified and ignored.
  • tissue volume (for example bones in CT scans) can also be preserved in this tissue volume by pre-segmenting them (440) , using techniques known in the art, and then copying the voxels from this pre- segmentation into the tissue volume.
  • the tissue volume is created and output (460) using the tissue probabilities associated with the voxels at the end of the previous tissue probability detecting steps (420) .
  • the goal is to create a volume that resembles the input scanned volume if it were to only have the tissue regions.
  • FIG. 5 is a diagram showing a step of assigning intensities to voxels based on their tissue probability when generating the tissue volume, according to an exemplary embodiment of the present invention.
  • an additional step may be performed. If a voxel is assigned intensity in the tissue range after the mapping, its original intensity in the input data is checked. If the original intensity was also in the tissue range, the original intensity is copied over to the tissue volume for that voxel. This preserves the original intensities of voxels that belong to the tissues.
  • FIG. 6 is a flowchart of a method for computer aided detection (CAD) , visualization, and measurement of tissue, according to an exemplary embodiment of the present invention.
  • CAD computer aided detection
  • the input data (610) , the tissue probability volume (630), and the tissue volume (620) are used for Computer Aided Detection (CAD) .
  • CAD Computer Aided Detection
  • One such application is Colon CAD that aims to detect colonic polyps. These target polyps are always present on the colon wall, which is a tissue. Another example is the detection of lung nodules.
  • the tissue volume (620) is analyzed to detect polyp shaped objects. Since all the tissue regions have been identified, the presence of other unwanted material such as tagged stool in the input data does not affect the detection.
  • Visualization is a very important tool in medical imaging and is extensively used by physicians and radiologists to perform diagnosis.
  • the tissue detection method may be used in applications that have tissue as the region of interest; for example, colonic polyp detection, lung nodule detection, etc.
  • the tissue volume may be used as input to multi-planar reconstruction algorithms and volume rendering algorithms.
  • a combination of input volume (610) , tissue probability volume (630) , and tissue volume (620) is also used for providing additional information during visualization. This additional information may include, for example, showing the user the presence of non-tissue regions in the data.
  • Measurement (660) of structures of interest is often an integral part of a diagnosis workflow.
  • the tissue volume may be used as input for measurement tools and algorithms where the structures of interests are made of tissue.
  • the use of tissue volume prevents non-tissue voxels in the input data from affecting the measurement.
  • the exemplary methods described above are general and well suited for a broad range of applications where tagging or contrast-enhancement is used and the structures of interest are tissues.
  • the aforementioned methods may be used to detect other regions besides tissues, e.g., a bone region.
  • the "bone probability" of each voxel is updated and a bone probability volume is generated that has higher probabilities for voxels which represent bone versus those that belong to tissue in the scan.
  • the colonic polyps are tissues that are connected to the colon wall, which itself is a tissue.
  • the residual stool and liquid in a patient's colon is tagged and it appears as high intensity in the imaged data .
  • FIG. 7 shows an example of a computer system which may implement the methods of the present disclosure.
  • the system and method of the present disclosure may be implemented in the form of a software application running on a computer system, for example, a mainframe, personal computer (PC) , handheld computer, server, etc.
  • the software application may be stored on a recording media locally accessible by the computer system and accessible via a hard wired or wireless connection to a network, for example, a local area network, or the Internet.
  • the computer system referred to generally as system 1000 may include, for example, a central processing unit (CPU) 1001, random access memory (RAM) 1004, a printer interface 1010, a display unit 1011, a local area network (LAN) data transmission controller 1005, a LAN interface 1006, a network controller 1003, an internal bus 1002, and one or more input devices 1009, for example, a keyboard, mouse etc.
  • the system 1000 may be connected to a data storage device, for example, a hard disk, 1008 via a link 1007.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Software Systems (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

Un procédé d'obtention d'un volume de tissu comprend les étapes consistant à entrer un ensemble de données comprenant une pluralité de voxels (110) ; initialiser un volume probable de tissu pour la pluralité de voxels à une valeur prédéterminée (120) ; procéder à une actualisation en augmentant ou diminuant le volume probable de tissu de chacun de la pluralité de voxels sur la base des valeurs d'intensité correspondantes de chacun de la pluralité de voxels (130) ; puis obtenir le volume de tissu en combinant le volume probable de tissu actualisé (180) et l'ensemble de données entré.
PCT/US2008/008564 2007-07-12 2008-07-11 Procédé de détection de tissu permettant une visualisation et un diagnostic assistés par ordinateur en présence d'un marquage WO2009009145A1 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US94935907P 2007-07-12 2007-07-12
US60/949,359 2007-07-12
US12/170,596 2008-07-10
US12/170,596 US8213696B2 (en) 2007-07-12 2008-07-10 Tissue detection method for computer aided diagnosis and visualization in the presence of tagging

Publications (2)

Publication Number Publication Date
WO2009009145A1 true WO2009009145A1 (fr) 2009-01-15
WO2009009145A9 WO2009009145A9 (fr) 2009-04-30

Family

ID=39822242

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2008/008564 WO2009009145A1 (fr) 2007-07-12 2008-07-11 Procédé de détection de tissu permettant une visualisation et un diagnostic assistés par ordinateur en présence d'un marquage

Country Status (1)

Country Link
WO (1) WO2009009145A1 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006086467A1 (fr) * 2005-02-10 2006-08-17 Siemens Corporate Research, Inc. Systeme et procede d'utilisation de modeles d'apprentissage discriminatif pour la segmentation de donnees d'image du colon en 3d

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006086467A1 (fr) * 2005-02-10 2006-08-17 Siemens Corporate Research, Inc. Systeme et procede d'utilisation de modeles d'apprentissage discriminatif pour la segmentation de donnees d'image du colon en 3d

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
BEZDEK J C ET AL: "Review of MR image segmentation techniques using pattern recognition", MEDICAL PHYSICS, vol. 20, no. 4, July 1993 (1993-07-01), AIP, MELVILLE, NY, USA, pages 1033 - 1048, XP007906032, ISSN: 0094-2405 *
MARROQUIN J L ET AL: "An Accurate and Efficient Bayesian Method for Automatic Segmentation of Brain MRI", IEEE TRANSACTIONS ON MEDICAL IMAGING, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 21, no. 8, 1 August 2002 (2002-08-01), XP011076342, ISSN: 0278-0062 *
PATRICK C TEO ET AL: "Creating Connected Representations of Cortical Gray Matter for Functional MRI Visualization", IEEE TRANSACTIONS ON MEDICAL IMAGING, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 16, no. 6, 1 December 1997 (1997-12-01), XP011035695, ISSN: 0278-0062 *
PHAM D L ET AL: "CURRENT METHODS IN MEDICAL IMAGE SEGMENTATION", ANNUAL REVIEW OF BIOMEDICAL ENGINEERING, ANNUAL REVIEW INCO., PALO ALTO, CA, US, vol. 2, 1 August 2000 (2000-08-01), pages 315 - 337, XP009062827 *
RUCHANEEWAN SUSOMBOON ET AL: "Automatic Single-Organ Segmentation in Computed Tomography Images", DATA MINING, 2006. ICDM '06. SIXTH INTERNATIONAL CONFERENCE ON, IEEE, PI, 1 December 2006 (2006-12-01), pages 1081 - 1086, XP031003140, ISBN: 978-0-7695-2701-7 *
SAHA P K ET AL: "Fuzzy connectedness and image segmentation", PROCEEDINGS OF THE IEEE, IEEE. NEW YORK, US, vol. 91, no. 10, 1 October 2003 (2003-10-01), pages 1649 - 1669, XP011100850, ISSN: 0018-9219 *

Also Published As

Publication number Publication date
WO2009009145A9 (fr) 2009-04-30

Similar Documents

Publication Publication Date Title
Pu et al. Shape “break-and-repair” strategy and its application to automated medical image segmentation
US10339648B2 (en) Quantitative predictors of tumor severity
JP7139479B2 (ja) 画像処理装置、医用画像診断装置、及びプログラム
EP1315125B1 (fr) Procédé de traitement d'image et système pour détecter des maladies
US7840051B2 (en) Medical image segmentation
US7627173B2 (en) GGN segmentation in pulmonary images for accuracy and consistency
Aggarwal et al. Role of segmentation in medical imaging: A comparative study
Bhadoria et al. Comparison of segmentation tools for multiple modalities in medical imaging
Yan et al. Marker‐controlled watershed for lymphoma segmentation in sequential CT images
KR20120086084A (ko) 3차원 의료 영상으로부터 최적의 2차원 의료 영상을 자동으로 생성하는 방법 및 장치
US8165376B2 (en) System and method for automatic detection of rib metastasis in computed tomography volume
US20090016589A1 (en) Computer-Assisted Detection of Colonic Polyps Using Convex Hull
CN107077731B (zh) 成像不确定性的可视化
JP7179521B2 (ja) 医用画像処理装置、画像生成方法、及び画像生成プログラム
EP1941449B1 (fr) Dispositif et procede de rendu
US20060187221A1 (en) System and method for identifying and removing virtual objects for visualization and computer aided detection
CN110570508A (zh) 骨质疏松状况三维可视化渲染方法
US8761470B2 (en) Analyzing an at least three-dimensional medical image
Frimmel et al. Centerline‐based colon segmentation for CT colonography
JP2008510565A (ja) 標識された糞便または虚脱結腸領域の存在下で結腸壁を抽出するためのシステムおよび方法
US8213696B2 (en) Tissue detection method for computer aided diagnosis and visualization in the presence of tagging
US9082193B2 (en) Shape-based image segmentation
Kale et al. Automatic segmentation of human facial tissue by MRI–CT fusion: A feasibility study
Wyatt et al. Segmentation in virtual colonoscopy using a geometric deformable model
WO2009009145A9 (fr) Procédé de détection de tissu permettant une visualisation et un diagnostic assistés par ordinateur en présence d'un marquage

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08780151

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08780151

Country of ref document: EP

Kind code of ref document: A1