METHOD OF PERFORMING TABLESIDE AUTOMATIC VESSEL ANALYSIS IN AN OPERATION ROOM

Field of Invention

The present invention relates to the field of interventional angio procedures. In particular, the present invention relates to a method of performing tableside Automatic Vessel Analysis (AVA) in an Operation Room (OR). Art Background

Interventional angio procedures, such as X-ray angio procedures, are based on the real time 2D minimally invasive image guidance of endovascular material through the vessels of an object, e.g. the human vessels. The well known imaging modality of choice for the interactive tracking of the guide wires and catheters is an X- ray angio machine. As is known, the technique of 3D Rotational Angiography (3D-RA) has significantly improved the standard 2D angio imaging by adding the third dimension and such allows a better understanding of the vessel morphology and mutual relationship of vessel pathology and surrounding branches.

Further, the clinician performs the intervention at an operation tableside in an operation room. To this end, a Table Side Module (TSM) as a touch screen device enables the clinician to perform simple operations on an available volume of 3D-RA data, such as rotate, zoom in and out, store view etc.

The US 2006/0020915 Al describes a system and a method for improved surgical workflow development which enables to create and to edit a modifiable module used to direct a medical procedure through a sequence of images and functions included in the module. The module may be stored on a server or may be loaded onto a touch-

screen display allowing the surgeon to touch the screen with his hand or a stylus to interact with or modify the module.

Further, the CA 2 356 367 Al describes a wireless handheld device screen-capture printing technology. A software on the device may gathers the screen- capture information locally when requested to do so by the device operator and transmits it, with routing information, via a communication network to a wireless server that formats and routes it to the specified destination device.

However, more complex AVA operations still have to be performed in an other room; the control room. This means that during the intervention, the clinician walks to the control room, sits behind a server, e.g. a workstation, and operates the server with a mouse. Summary of the Invention

AVA allows the clinician to select a vessel segment, which is analyzed. The result of this analysis is a cross sectional view of the vessel, at a user-defined location, showing vessel shape related data, such as the minimum and maximum diameter and area of a fitted ellipse at that location. Furthermore, a graph, illustrating the diameter of the vessel along its path, may be shown.

The graph may also indicate the maximum stenosis, percentage of the occlusion, etc. This tool is very valuable to the clinician, and is used frequently, because it allows him to assess a stenosis in the vessel, and helps him to select a stent length and diameter for the further operation.

It may be desirable to enable the clinician to use a simple to use interventional angio method that retains the advantages of AVA and avoids the disadvantage of the laborious repeated covering of the distance between the operation room and the server room. Further, it may be desirable to enable a method that improves the workflow of the physician.

These problems may be solved by the method of claim 1. The method enables the physician to use the Automated Vessel Analysis tool at the table side, in other words, to perform AVA in the Operation Room (OR) via placing probes through a touch-screen computer device.

When the user touches the screen of the computer device (step c, d of claim 1), coordinates are transferred to a storage of the same device or/and a second one, e.g. a server. The server will position a probe with respect to a so called voxel volume. The probe is a geometrical shape, fitted around a vessel at a given position.

The probe may consists of a sphere:

[p - pj = r ² (1) and a plane through the centre of the sphere: n _p - (p - p ₀ ) = 0 (2) with p the position of a point, p ₀ the centre and r the radius of the sphere and n _p the normal of the plane. Preferably, the radius of the sphere should be slightly greater than the radius of the explored vessel. Interactive selection of the correct radius is easy with a proper graphical user interface, when the 3D sphere is rendered together with a 3D triangle surface representation of the boundary of the vessels, created for example by a marching cubes algorithm. The sphere should be positioned so that the vessel intersects the sphere, at least partially. Next, the orientation of the probe's plane, the local co-ordinate system of this plane and the position of the probe's sphere can be adjusted, based by using as origin the coordinates passed by the screen device, and as direction the viewing vector of the camera. The first intersection of the line and the segmented vessels, will serve to position the first probe (Step c) of claim 1). After alignment, the vessel radius is estimated according to the position of the probe. Executing a vessel tracing function

According to the inventive method, the executing of the vessel tracing function is initialized by touching the screen. If the orientation of the plane of the probe is adjusted to the vessel (the probe is aligned), the direction of the normal is reversed if the new plane normal is opposite to the old plane normal. Next, u and v axis of the plane are chosen so that they are orthogonal to the new plane normal and orthogonal to each other and are as close as possible to the old u and v axes (if any). Consistent u and v axes are necessary for a smooth endoview of the vessel during vessel tracing. These u

and v axes are also used for the adjustment of the sphere centre. The vessel tracing function in several variations is closer described in Jan Bruijns "Semi- Automatic Shape Extraction from Tube-like Geometry", Vision Modeling and Visualization 2000, pp. 347 - 356 and in Jan Bruijns "Verification of the Self-adjusting Probe: Shape Extraction from Cerebral Vasculature", Vision Modeling and Visualization 2003, pp. 159 - 166.

The proposed solution to use the method via a touch-screen may be characterised by the fact that it is easy to use. Together with the fact that it improves the workflow of the physician, it can be seen as an added value for 3D-RA. The availability of this function at the tableside may improve the workflow of the intervention considerably.

The proposed method is generally independent of the system architecture.

However, a preferred embodiment is a client-server system. The system is based on the known interaction of a server, e.g. a workstation, which holds the 3D data, and performs the actual rendering of the data to images and a device called client, for example, a remote touch screen device, such as a Tablet PC or handheld device, which displays the images, and handles the user interaction. The client should be able to do some basic processing, but does not need to have the massive processing power of the workstation. It has a touch-screen and preferably can be, but is not necessarily wireless.

Generally, there has to be a data connection between the server and the client, e.g. WLAN. The proposed method may take the limitations of the processing capabilities of the client and the bandwidth of the connection into account.

According to the exemplary embodiment of the invention the method may comprise the following steps: A 3D-RA data set is acquired by the server. An image of vascular data is displayed on the screen of the client. The user selects a position of the first probe, by pressing the screen of the client, either using his hand, or a stylus pen. The probe is placed on the vessel in the voxel volume at the corresponding position. A cross-section of the vessel at the chosen position is displayed in the image. Then, the user selects a position of a second probe, by pressing the screen. Also this probe is placed on the vessel in the voxel volume at the corresponding position. A

cross-section of the vessel at the chosen position may be displayed in the image. The position of the probes can preferably be moved through the vessel by pressing up or down arrow signs, preferably located on the screen next to the cross-section. Finally, the user presses a sign like a "Trace button" that is displayed on the screen. Then, a graph may be shown by the screen, illustrating the diameter along the traced path.

In a further preferred embodiment of the method with client-server architecture the method comprises the steps of: construction of an image of a vessel tree out of a 3D Rotational

Angiography (3D-RA) data set of an object's volume by a server; transferring the image of the vessel tree from the server to a client (touch screen device); displaying the image at a screen of the client; choosing a first vessel section of a vessel of the displayed vessel tree by touching the screen at a corresponding first location; transferring the coordinates of the first vessel section from the client to the server; extracting a first cross section through the vessel tree by the server, wherein the centre of the cross section corresponds to the centre of the first vessel section; transferring an image of the first cross section from the server to the client; displaying the image of the cross section at the screen of the client; choosing a second vessel section of the vessel of the displayed vessel tree by touching the screen at a corresponding second location; transferring the coordinates of the second vessel section from the client to the server; extracting a second cross section through the vessel tree by the server, wherein the centre of the cross section corresponds to the centre of the second vessel section; transferring an image of the second cross section from the server to the client;

displaying the image of the second cross section at the screen of the client; providing a TRACE FUNCTION by the client; activating the TRACE FUNCTION and transferring an appropriate signal from the client to the server; executing a vessel tracing function through a path of the vessel between the first vessel section and the second vessel section, wherein the shape of the vessel between the two vessel sections is analytically explored by the server; transferring of the data corresponding to the shape from the server to the client; displaying an image of the data corresponding to the shape of the vessel at the screen of the client.

These and other aspects of the present invention will become apparent from and elucidated with reference to the embodiments described hereinafter. Brief Description of the Drawings

An exemplary embodiment of the present invention will be described in the following via eight workflow steps, with reference to the following four drawings.

Fig. 1 to 4 show a front view at a computer device with different images on its touch-screen.

Workflow step 1

A rotational angiography run is transferred to a 3D-RA workstation, not shown here, and a voxel volume is reconstructed, as usual. The user can rotate, translate and zoom the volume (using the mouse, the Table Side Module, or follow C-arm), as usual via the workstation. The workstation will also perform a segmentation of vessels in the voxel volume, and build a vessel tree graph. The 3D-RA workstation will act as server in this embodiment of the invention.

Workflow step 2

When the user selects a Tableside AVA application, the server will transfer a present view of the 3D-RA volume to a computer device 1, shown in Fig. 1.

The client will then display the view as an image at a screen 2. This means that only a 2D image (the view on the volume) needs to be transferred. The computer device 1 is now ready to receive the user input. The computer device 1 is a remote touch screen device (e.g. a Tablet PC) and will act as a client.

Workflow step 3

When the user touches the screen 2 of the client, shown in Fig.l, coordinates are transferred to the workstation that acts as a server. The server will position a probe with respect to the voxel volume. This means that a line is followed; using as origin the coordinates passed by the client, and as direction the viewing vector of a camera. The first intersection of the line and segmented vessels 3, will serve to position the first probe 4. The probe's centre 5 is positioned on the appropriate centreline of the segmented vessel tree 6. The probe's normal corresponds to the tangent vector of the vessel centreline at the probe's centre 5.

Workflow step 4

The server creates a cross section 7 corresponding to a first vessel section through the voxel volume, shown in Fig. 2. The centre of the cross section 7 corresponds to the probe's centre 5, and the normal of the cross section corresponds to the probe's normal.

The cross section 7 is transferred as an image 8 to the client, as well as an updated image 9 of the view on the volume. The updated image view on the volume will visualize the selected probe 4 with respect to the vessels 3. When the user presses one of the arrow buttons signs 10, 11 on the screen

2 of the client this is communicated to the server. The server will then move the probe 4 along the centreline of the segmented vessel 3, and send an update of the cross-section 5 and the view on the volume to the client.

Workflow steps 5 and 6

When the user clicks on the view of the volume on the client's screen 2 after the first probe has been placed, the second probe 12 will be placed (Fig. 3). In order to do this, the same actions as for the first probe will be performed. Workflow step 7 When the user presses the "Trace" button sign 13, after both probes 4, 12 have been placed, the client will notify the server to run the tracing algorithm (Fig. 3). Workflow step 8

The server will now perform a trace through the vessel tree 6, which means that the path of centreline is followed from the first probe 4 (first vessel section) to the second probe 12 (a second vessel section). Along this path, the diameter and area of the cross-section of the vessel 3 will be recorded. Using the resulting data, a stenosis can be identified. The data will be represented in a graph 14, which will be sent to the client (shown in Fig. 4)

It should be noted that the term "comprising" does not exclude other elements or steps and the "a" or "an" does not exclude a plurality and that a single processor or system may fulfill the functions of several means or units recited in the claims.

Further, it should be noted, that any reference signs in the claims shall not be construed as limiting the scope of the claims.