WO2010029470A1

WO2010029470A1 - System and method for processing medical images

Info

Publication number: WO2010029470A1
Application number: PCT/IB2009/053837
Authority: WO
Inventors: Sebastian P. M. Dries; Cristian Lorenz
Original assignee: Koninklijke Philips Electronics N.V.; Philips Intellectual Property & Standards Gmbh
Priority date: 2008-09-10
Filing date: 2009-09-03
Publication date: 2010-03-18

Abstract

The invention provides a system, e.g. a Picture Archiving and Communications System (PACS), and a method for identifying the presence of relevant medical images, e.g. medical images ofthe same region of interest or anatomical feature(s) of the same patient, taken at a different time or by a different acquisition modality. The system ofthe invention notifies a user on the presence of any such relevant image, and is arranged for presenting such relevant images on a display. Two or more medical images may be displayed at the same time or one after the other, overlaid, by sliding a space or time axis, etc. The medical images are scaled and orientated so as to maximize the visual correlation between them, i.e. to make them exhibit the anatomy the most possible alike, so as to ease comparison between anatomical features without disturbing diagnostically relevant contrasts on the medical images.

Description

System and method for processing medical images

FIELD OF THE INVENTION

The present invention generally relates to processing and graphical representation of medical images. In particular, the invention relates to a system for registering and processing medical images. The invention moreover relates to a method for registering and processing medical images. Furthermore, the invention relates to a computer program product being adapted to enable a computer system comprising at least one computer having data storage means associated therewith to control a system or method according to the invention.

BACKGROUND OF THE INVENTION

Medical imaging is advantageously done digitally. Digital acquisition techniques or modalities include computerized tomography (CT), magnetic resonance (MR), positron emission tomography (PET), ultrasound (US), X-ray, computed radiography (CR) and digital radiography (DR). The image files may contain three dimensional image data, such as for CT or MR scans, or two-dimensional image data, such as for CR or DR radiography data. An image file will typically also contain other data relevant to the image, such as modality, patient details, details of the image content etc. The non-image data is typically stored as text.

Medical image data files may be archived and accessed in a hospital environment by a Picture Archiving and Communication System (PACS). Picture Archiving and Communications Systems (PACSs) store, process, transmit and display imaging data. Acquired data is stored by means of submission from the place of acquisition, e.g. inside a hospital's radiology department, along with additional information like patient data and image findings. Data is processed by means of indexing and compression, sometimes on demand determined by the connection bandwidth and display capabilities of the viewing client, e.g. a workstation with display and user interaction means. Transmission takes place over any appropriate network connection. Display may be carried out at any of a variety of workstations or clients that may use different means of data access; moreover, the workstations may be thin client computers or fat client computers. The individual workstations on the network may allow a physician to access, interpret, manipulate and analyze previously obtained data in the form of stored image files.

An image file may be quite large and it may be quite cumbersome and time consuming to search and navigate through a plurality of image files to find a particular image file.

A healthcare environment, such as a hospital or a medical clinic, may encompass a large array of professionals, patients and equipment. Personnel in a healthcare facility must manage a plurality of patients, systems and tasks to provide quality service to patients. Thus, an improved system for finding and processing relevant images would be advantageous.

The document US 2007/0127795 describes systems and methods for correlating an acquired image with a historical image. An acquired image and at least one historical image are registered, a metric is applied to the acquired image and each of the historical image(s), and a correlation is identified between the acquired image and one of the historical images based on the metric. The acquired image and the historical image may be displayed based on the correlation.

The document US 2004/0068423 discloses displaying an overview of a patient's history made up of a potentially large number of digital images as a two- dimensional array of thumbnails sorted by date in one direction and anatomical feature in a second direction, in a PACS environment.

Even though the documents US 2007/0127795 and US 2004/0068423 disclose handling of large sets of medical images, the user friendliness of such systems may be improved even further. In particular, a system for registering and processing medical images facilitating the comparison between relevant regions of interest in different images would be advantageous. Moreover, a user friendly system for registering and processing medical images of a particular region of interest that might be the subject of more than one examination with different modalities and at different points in time would be advantageous.

SUMMARY OF THE INVENTION

Accordingly, the invention preferably seeks to provide such an improved system. In particular, it may be seen as an object of the present invention to provide a system that provides user friendliness by finding and processing relevant images for comparison. This object and several other objects are obtained in a first aspect of the invention by providing a system for registering and processing medical images, the system comprising: input means for receiving first image data relating to a first medical image of a patient; means for identifying a second medical image relating to said patient, wherein the second image data relates to said second medical image, said first and second medical image comprising a representation of the same anatomical feature of said patient; and processor means arranged for processing said first image data and/or said second image data so as to align a contour of said anatomical feature in the first and second medical images. Hereby, the system allows for an improved comparison between the first and second medical image in that the system identifies the presence of the second medical image, which typically would be another medical image of a particular region of interest that is also contained in the first medical image from another examination with another modality and/or from a different point in time. This identification is done automatically, viz. without user interaction. A region of interest could e.g. be a particular anatomical feature, such as a body part or an organ of the human body, a tumor, etc. Thus follow-up or cross modality comparison of different images of a specific patient is automated at an extent that significantly improves healthcare workflow e.g. during routine case presentations, where there is only time and need for display of the region currently under diagnosis and/or treatment. It should be noted, that the references to "a contour" does not preclude "a plurality of contours"; instead the term "a contour" should be interpreted as covering "one or more contours".

In an advantageous embodiment of the system, the system moreover comprises means for presenting the first and second medical image. Hereby, the first and second medical image may be displayed to a user, either simultaneously and next to each other, or subsequently, one after the other. Advantageously, the system may further comprise a user interface arranged for indicating to a user that the second medical image has been identified and initiating the presentation of the first and/or second medical image upon user interaction. This notification may be done by icons on screen or by activating an additional control at the client workstation or display client computer. Hereby, the system may provide, at the client workstation, an indication that another data set relating to a second medical image of possible interest exists. By clicking on the icon or using the control a user may request that the other dataset relating to the second medical image is displayed, alone or in combination with the first medical image. The system may display the second medical image in the orientation and scale closest to the orientation and scale of the first medical image. The icon or additional control at the client workstation may advantageously provide the user with information on data modality and/or acquisition time and/or acquisition parameters of the second medical image. Of course, more than one additional medical image may exist, and they may relate to different regions of interest in the first image. In this case the above relates to each such additional medical image. The system might initiate display of the second medical image automatically without waiting for a user to indicate that he/she is interested in viewing it. However, in the case where a plurality of second medical images of possible interest exists, it may be advantageous that the system waits for user interaction before presenting further medical images together with the first medical image, so that the user may choose the one(s) he/she is interested in viewing.

According to an embodiment of the system, the processor means is arranged for changing the orientation of the first and/or second medical image. Moreover, the processor means may be arranged for deforming the first and/or second medical image. The deforming of the first and/or second medical image may comprise rigid, affine, spline based or free form deformation, where the rigid deformation may comprise one or more of the following: rotation, translation, scaling. These embodiments provide approaches for optimizing the correlation between the first and second medical image so as to make it easier for a user viewing the first and second image to compare features, typically anatomical features, shown in the medical images. According to another embodiment, the processor means of the system is furthermore arranged for processing said first image data and/or said second image data so as to change the coloring, brightness and/or contrast of said first and/or said second medical image(s). Hereby, the first and/or second image(s) may be adjusted in order to optimize display so that an anatomical feature on the first and second medical images may be compared readily. Typically, the contrast and brightness of the medical images may be derived from a contoured anatomical feature or structure, and adjusting the contrast and brightness of one or more of the medical images may make it easier for a user to view and compare the anatomical feature occurring on the medical images.

According to yet another embodiment, the system is arranged for segmenting structures contained in medical images in order to identify the anatomical feature(s) to which a medical image relates, said segmentation comprising information on a represented anatomical feature of a patient's body. The term "segment" is meant to cover the process of partitioning the medical image into regions or structures, these structures being useful to identify anatomical feature(s) of the medical image. According to yet another embodiment, the system is arranged for analyzing metadata comprised with medical images in order to identify the anatomical feature(s) to which a medical image relates, said metadata comprising information on a represented anatomical feature of the patient's body. Such metadata may for example be the so-called Digital Imaging and Communications in Medicine (DICOM) information; DICOM is a standard for handling, storing, printing, and transmitting information in medical imaging.

In a second aspect, the invention relates to a corresponding method for registering and processing medical images, the method including similar advantages as the system. In a third aspect, the invention relates to a computer program product being adapted to enable a computer system comprising at least one computer having data storage means associated therewith to control a system according to the first aspect of the invention. This aspect of the invention is particularly, but not exclusively, advantageous in that the present invention may be implemented by a computer program product enabling a computer system to perform the operations of the second aspect of the invention. Thus, it is contemplated that some known system may be changed to operate according to the present invention by installing a computer program product on a computer system controlling the said system. Such a computer program product may be provided on any kind of computer readable medium, e.g. magnetically or optically based medium, or through a computer based network, e.g. the Internet.

The first, second and third aspect of the present invention may each be combined with any of the other aspects. These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

BRIEF DESCRIPTION OF THE FIGURES

The present invention will now be explained, by way of example only, with reference to the accompanying Figures, where

Fig. 1 shows an image file set of medical images;

Figs. 2 and 3 are schematic diagrams showing exemplary networks of systems according to the invention and associated equipment; and

Fig. 4 is a flow-chart of a method according to the invention.

DETAILED DESCRIPTION OF AN EMBODIMENT Figure 1 shows an image file set 4 of medical images 2. Each medical image comprises image data 1 and text information 3. In Figure 1, the text information 3 of each medical image 2 is shown as a header; however, the text information may be shown in any place within the image or alternatively may not be shown within the medical image. The text information 3 may contain metadata relating to the anatomical feature(s) to which the medical image 2 relates, said metadata comprising information on a represented anatomical feature of a patient's body. Such metadata may for example be the DICOM information. The text information may comprise patient information, such as patient name, patient number, accession number, information on the anatomical feature presented, information on the acquisition modality, etc. Of course, an image set relating to one or more patients could comprise a lot more images than the five images shown in Figure 1 ; this small number of images is shown for the sake of clarity only. It should be noted, that even though Figure 1 discloses the images of the image set 4 as physical images, this is only for illustrative purposes in that digital images are stored as digital data sets within a storage means. Figure 2 is a schematic diagram showing an exemplary network 50 of a system

35 according to the invention and associated equipment. The network 50 may be a Picture Archiving and Communication System (PACS) used with an embodiment of the present invention; however, the invention could be used with any alternative appropriate network as well. The network 50 includes an imaging modality 10 arranged for obtaining one or more images of a patient anatomy. The imaging modality 10 may include any device capable of capturing an image of a patient anatomy such as a medical diagnosing imaging device, such as an X-ray imager, ultrasound scanner, magnetic resonance imager, etc. Image data representative of the image(s) is communicated from the imaging modality 10 to an acquisition unit 20. Any communication of data within the network 50 may be preformed electronically over a wired or a wireless connection. Even though only one imaging modality 10 is shown in Figure 2a, it should be noted that any appropriate number of imaging modalities may be used. The acquisition unit 20 is arranged for performing one or more preprocessing functions in order to prepare the image for viewing on a display workstation 40a, 40b, 40c. The acquisition unit 20 may convert raw image data into a DICOM standard format or attach a DICOM header. A preprocessing function is typically a modality specific enhancement applied at the beginning of an imaging and display process. The preprocessing functions differ from processing functions applied to image data in that the processing functions are not modality specific and are instead applied at the end of the imaging and display workflow, for example at a display workstation 40a, 40b, 40c.

Subsequent to preprocessing at the acquisition unit 20, the preprocessed image data are communicated to a server 35. The server 35 may comprise the system according to the invention, including means for selecting an image file set relating to a specific patient, where the image file set comprises a first medical image; means for identifying a second medical image within the image file set relating to the specific patient, the first and second medical image relating to the same anatomical feature of the patient; as well as means for processing the second medical image so as to maximize the visual correlation of the first and second medical image. The selection means, the identification means and the processing means of the server 35 may be part of a processor means of the server 35. The server may be a PACS workstation or any other appropriate server. Even though the acquisition unit 20 is shown as separate from the server 35, it should be noted that it may be integrated within the server 35. The server 35 may communicate with one or more workstations 40a, 40b, 40c, for example PACS workstations. Even though Figure 2 discloses three workstations 40a, 40b, 40c, the number of workstations may be smaller or larger. The workstations 40a, 40b, 40c are arranged for receiving or retrieving image data from the server 35 relating to medical images of anatomical features of patients. The medical images may correspond to different imaging modalities, different patients, different anatomical features, etc. The workstations 40a, 40b, 40c moreover comprises display means for presenting the medical images in order to let a user, such as a physician, examine one or more medical images. The workstations 40a, 40b, 40c may also comprise a user interface for indicating to a user that a second medical image related to a first medical image has been identified and for initiating the presentation of the first and second medical images upon user interaction.

The first and second medical images relate to one or more common regions of interest, i.e. the same anatomical features. The first medical image may be a recently acquired image and the second medical image may be a historical image, typically of the same anatomical feature of a specific patient, so that the first and second medical image that are two images from different instants in time of the same anatomical feature of a specific patient may be compared. The first and second medical images may be acquired by the same acquisition modality or by different acquisition modalities. The server 35 according to the invention is arranged for processing the second medical image so as to maximize the visual correlation between the first and second medical images; however, the server may alternatively be arranged to process the first medical image or both the first and second image in order to maximize the visual correlation between the first and second image. The processing may include changing the orientation of the first and/or second medical image so as to bring the first and second medical image into alignment, and deforming the second medical image and/or changing the coloring of the second medical image. The changing of the coloring may include the change of contrast and/or change of grayscale of an image. The deforming of the second medical image comprises rigid, affine, spline based or free form deformation, wherein rigid deformation may comprise one or more of the following: rotation, translation, scaling. The system 35 is thus arranged for identifying the presence of an image relating to a particular region of interest, for notifying a user of the presence and for displaying the image to the user at a workstation, e.g. a PACS workstation 40a, 40b, 40c.

As client computers 40a, 40b, 40c may not be identically scaled throughout a whole institution, e.g. a hospital, the automated registration part of the medical images may be performed in the backend, viz. the server 35. Hereby user interaction time may be shortened, in that to interaction delay is kept at a minimum. However, the invention is not limited to this, and registration of medical images may also take place at client computers or workstations 40a, 40b, 40c as described in connection with Figure 3.

By means of user assigned location tags or automated registration to a given generic patient model, viz. any method of registration that uses a body atlas or a whole body model as a reference, any new data set in the form of medical image data is classified to indicate whether it spatially intersects with any previous data set of the same or another modality, for a given patient. The model or atlas may be individualized to the patient by any means, e.g. by the image data already available or by other stored patient data like weight, height, gender, age. The registration method may incorporate segmentation of each image data set to derive a representation to register with the model representation, e.g. using model based segmentation with active shape models. However, the registration may also be gray value based incorporating a similarity measure that minimizes a transform distance from the image to the atlas by applying rigid deformation, such as rotation, translation and scaling, or affine, spline based or free form deformation.

In the case of the existence of a spatially intersecting previous image of the same or another modality for a given patient, i.e. the existence of two medical images of the same anatomical feature of a given patient, a transformation field is calculated between both images by image registration. Both aforementioned processes can be done either immediately after acquisition, or in the system 35 's idle time, or on display request through one of the client workstations 40a, 40b, 40c. DICOM tags may be used in the identification of the second medical image as an alternative or in addition to the information on the anatomical part of the patient's body registered with the medical images. On display request the client workstation 40a, 40b, 40c is provided with the image plane to display and additionally with the information on any additional data sets that are or can be registered with the current data set. If a user via the user interface of the client workstation 40a, 40b, 40c requests an additional data set, the registration is calculated or looked up and the client workstation 40a, 40b, 40c is provided with the topographically most similar image of the latter data set. This may also facilitate side-by- side comparison, flicker display, color coded or blunt overlay of both or even more than two medical images registered with each other. Moreover, the first and second medical image may be presented at the same time or one after the other, overlaid, or the difference between them may be calculated and displayed. Furthermore, browsing an image data set may be coupled between the first and second medical image, e.g. sliding a space or time axis may select corresponding views. Of course, other display layouts and configurations may be conceivable.

Figure 3 is a schematic diagram showing an alternative, exemplary network 50 of a system 35' according to the invention and associated equipment. The network 50 may be a Picture Archiving and Communication System (PACS) used with an embodiment of the present invention; however, the invention could be used with any alternative appropriate network as well. The imaging modality 10, acquisition unit 20 and workstations 40a, 40b, 40c are similar to the same units bearing the same reference numerals in Figure 2; therefore they will not be described in more detail here. However, in Figure 3, the system 35' of the invention comprises both a server 30, such as a PACS server, and a workstation 40b, e.g. a PACS workstation. Of course, the server 30 of Figure 2b may form alternative systems with another one of the workstations 40a, 40c, and again the number of workstations 40a, 40b, 40c could be any other appropriate number.

As described above, the processing of medical images, including the automated registration part of the medical images, may be performed in the at the server 35 or at one or more of the workstations 40a, 40b, 40c.

Figure 4 is a flow-chart of a method 100 for registering and processing medical images according to the invention. The method shows similar advantages as described above in relation to the systems in Figures 2 and 3 carrying out the method of the invention. The method starts at step S and continues to step 101, wherein an image file set relating to a specific patient is received or selected. The image set may comprise one or more medical images, acquired by one or more acquisition modalities, at one or more instances in time. The image set includes a first medical image, which may be a recently acquired medical image or a historical medical image. In the subsequent step, step 102, a second medical image within the image file set relating to the specific patient is identified, where the first and second medical image relate to the same anatomical part of the patient. The second medical image may be a historic medical image of an anatomical feature of the patient, whilst the first medical image may be a recently acquired medical image of the same anatomical feature of the patient. In the subsequent step 103, the second medical image is processed so as to maximize the visual correlation of the first and second medical image. This enables presentation of the first and second medical image in a way that makes comparison between the anatomical features easy. For instance, the processing in step 103 may be scaling and orienting the second medical image in a way so as to resemble the first medical image the most. More than one historical medical image may be identified and processed. Moreover, the terminology "first medical image" and "second medical image" is not meant to be limiting; thus, the first medical image may be the historical medical image, and the second medical image may be a recently taken medical image to be processed in order to create resemblance between the first and second medical images. Moreover, both the first and the second medical image may be historical images taken some time ago, or they may be medical images taken at the same time or at very small intervals, possibly by different acquisition modalities.

The processing of the image(s) may be performed upon, e.g. immediately after, acquisition, or in idle time of the system performing the method, or on display request through a client workstations. The method 100 may optionally comprise the step 104 of presenting the first and processed second medical image, e.g. on a display of a workstation, such as a PACS workstation. The method 100 ends in step 105.

The invention can be implemented in any suitable form including hardware, software, firmware or any combination of these. The invention or some features of the invention can be implemented as computer software running on one or more data processors and/or digital signal processors. The elements and components of an embodiment of the invention may be physically, functionally and logically implemented in any suitable way. Indeed, the functionality may be implemented in a single unit, in a plurality of units or as part of other functional units. As such, the invention may be implemented in a single unit, or may be physically and functionally distributed between different units and processors.

In short, the invention provides a system, e.g. a Picture Archiving and Communications System (PACS), and a method for identifying the presence of relevant medical images, e.g. medical images of the same region of interest or anatomical feature(s) of the same patient, taken at a different time or by a different acquisition modality. The system of the invention notifies a user on the presence of any such relevant image, and is arranged for presenting such relevant images on a display. Two or more medical images may be displayed at the same time or one after the other, overlaid, by sliding a space or time axis, etc. The medical images are scaled and orientated so as to maximize the visual correlation between them, i.e. to make them exhibit the anatomy the most possible alike, so as to ease comparison between anatomical features without disturbing diagnostically relevant contrasts on the medical images.

Although the present invention has been described in connection with the specified embodiments, it is not intended to be limited to the specific form set forth herein. Rather, the scope of the present invention is limited only by the accompanying claims. In the claims, the term "comprising" does not exclude the presence of other elements or steps. Additionally, although individual features may be included in different claims, these may possibly be advantageously combined, and the inclusion in different claims does not imply that a combination of features is not feasible and/or advantageous. In addition, singular references do not exclude a plurality. Thus, references to "a", "an", "first", "second" etc. do not preclude a plurality. Furthermore, reference signs in the claims shall not be construed as limiting the scope.

Claims

CLAIMS:

1. A system (35; 35') for registering and processing medical images, the system comprising: input means for receiving first image data relating to a first medical image of a patient, - means for identifying (35; 30) a second medical image relating to said patient, wherein second image data relates to said second medical image, said first and second medical images comprising a representation of the same anatomical feature of said patient, and processor means (35; 30) arranged for processing said first image data and/or said second image data so as to align a contour of said anatomical feature in the first and second medical images.

2. A system according to claim 1, said system moreover comprising means for presenting (40a, 40b, 40c) the first and second medical images.

3. A system according to claim 1, said system further comprising a user interface arranged for indicating to a user that the second medical image has been identified and initiating the presentation of the first and/or second medical image upon user interaction.

4. A system according to any of the claims 1 to 3, wherein the processor means

(35; 30) is arranged for changing the orientation of the first and/or second medical image.

5. A system according to any of the claims 1 to 4, wherein the processor means (35; 30) is arranged for deforming the first and/or second medical image.

6. A system according to any of the claims 1 to 5, wherein the deforming of the first and/or second medical image comprises rigid, affine, spline based or free form deformation.

7. A system according to any of the claims 1 to 6, wherein the processor means (35; 30) is furthermore arranged for processing said first image data and/or said second image data so as to change the coloring, brightness and/or contrast of said first and/or said second medical image(s).

8. A system according to any of the claims 1 to 7, wherein the system (35; 35') is arranged for segmenting structures contained in medical images in order to identify the anatomical feature(s) to which a medical image relates, said segmentation comprising information on a represented anatomical feature of the patient's body.

9. A system according to any of the claims 1 to 8, wherein the system (35; 35') is arranged for analyzing metadata comprised with medical images in order to identify the anatomical feature(s) to which a medical image relates, said metadata comprising information on a represented anatomical feature of the patient's body.

10. A method (100) for registering and processing medical images, the method comprising the steps of: receiving (101) first image data relating to a first medical image of a specific patient; - identifying (102) a second medical image relating to said patient, wherein second image data relates to said second medical image, said first and second medical images comprising a representation of the same anatomical part of said patient; and processing (103) said first image data and/or said second image data so as to align a contour of said anatomical feature in the first and second medical images.

11. A method (100) according to claim 10, further comprising the step of presenting (104) the first and processed second medical image.

12. A computer program product being adapted to enable a computer system comprising at least one computer having data storage means associated therewith to control a system according to claim 1.