Atlas-Based Determination of Tumor Growth Direction

Abstract

The invention relates to a method for determining the spatial development of tumor tissue, by acquiring patient medical image data describing sequences of patient medical images of tumors in parts of patient bodies, wherein the patient medical images of each sequence have been taken at subsequent points in time and each sequence has been taken tier a different patient; determining, by additively fusing subsequent patient medical images of each sequence to one another, patient spatial development data describing the spatial development of a tumor in each patient body; acquiring atlas data describing an atlas representation of the parts of patient bodies; determining, based on the atlas data and the patient development data, development probability data describing a probability for a spatial development of a tumor.

Claims

1-15. (canceled)

16. A method for determining the spatial development of tumor tissue, executed. by one or more processors, comprising: acquiring, by one or more of the processors, patient medical image data describing sequences of patient medical images of tumors in parts of patient bodies, wherein the patient medical images of each sequence have been taken at subsequent points in time and each sequence has been taken for a different patient; determining, by one or more of the processors, by additively fusing subsequent patient medical images of each sequence to one another, patient spatial development data describing the spatial development of a tumor in each patient body; acquiring, by one or more of the processors, atlas data describing an atlas representation of the parts of patient bodies; determining, by one or more of the processors, based on the atlas data and the patient development data, development probability data describing a probability for a spatial development of a tumor.

17. The method according to claim 16, wherein the development probability data is determined, by one or more of the processors, based on transforming the patient spatial development data into an atlas reference system in which spatial relationships in the atlas representation are defined.

18. The method according to claim 17, wherein transforming the patient spatial development data into the atlas reference system includes fusing, by one or more of the processors and to the atlas data, the result of additively fusing the patient medical images of each sequence.

19. The method according to claim 16, wherein the position of the tumor in the first patient medical image of each sequence is used as a starting condition for determining the development probability data.

20. The method according to claim 19, wherein determining the development probability data includes determining a growth cone of the tumor for each starting condition, the growth cone describing a probability of a spatial development of the tumor relative to a specific main development direction

21. The method according to claim 16, comprising: acquiring, by one or more of the processors, patient-specific medical image data describing a tumor in a patient to be treated; determining, by one or more of the processors, based on the patient-specific medical image data and the development probability data, patient-specific development probability data describing a probability for a spatial development of the tumor in the patient to be treated.

22. The method according to claim 21, wherein the patient-specific development probability data is determined by one or more of the processors by registering the patient-specific medical image data with the development probability data.

23. The method according to claim 21, comprising: determining, by one or more of the processors and based on the patient-specific medical image data and the patient-specific development probability data, patient-specific probability indication data describing an indication signal to be output to a user using the information content of the patient-specific development probability data.

24. The method according to claim 23, comprising: outputting, to a user and using an indication device for indicating digital information, the indication signal.

25. The method according to claim 24, wherein the indication device includes a graphical output device and wherein the indication to be output includes a visualization of the information content of the patient-specific probability indication data.

26. The method according to claim 21, Wherein the patient medical image data has been taken by application of a computed x-ray tomography imaging method or a magnetic resonance imaging method to the patients' bodies, or by imaging of radiation emitted from a substance emitting ionizing radiation introduced into the patients' bodies.

27. The method according to claim 21, wherein a colored map is determined based on the development probability data, wherein the colored map indicates areas of potential tumor growth and is overlaid as texture on a three-dimensional tumor object.

28. The method according to claim 27, wherein the colored map is displayed on at least one of a display device of a surgical microscope and a medical head-up display.

29. At least one non-transitory storage medium storing instructions for determining the spatial development of tumor tissue, the instructions comprising: a plurality of instructions which, when executed by one or more processors, causes the one or more processors to: acquire, by one or more of the processors, patient medical image data describing sequences of patient medical images of tumors in parts of patient bodies, wherein the patient medical images of each sequence have been taken at subsequent points in time and each sequence has been taken for a different patient; determine, by one or more of the processors, by additively fusing subsequent patient medical images of each sequence to one another, patient spatial development data describing the spatial development of a tumor in each patient body; acquire, by one or more of the processors, atlas data describing an atlas representation of the parts of patient bodies; determine, by one or more of the processors, based on the atlas data and the patient development data, development probability data describing a probability for a spatial development of a tumor.

30. A system for determining the spatial development of tumor tissue, comprising: memory storing instructions to cause one or more processors to: acquire, by one or more of the processors, patient medical image data describing sequences of patient medical images of tumors in parts of patient bodies, wherein the patient medical images of each sequence have been taken at subsequent points in time and each sequence has been taken for a different patient; determine, by one or more of the processors, by additively fusing subsequent patient medical images of each sequence to one another, patient spatial development data describing the spatial development of a tumor in each patient body; acquire, by one or more of the processors, atlas data describing an atlas representation of the parts of patient bodies; determine, by one or more of the processors, based on the atlas data and the patient development data, development probability data describing a probability for a spatial development of a tumor; a display device operatively coupled to one or more of the processors for displaying information content of data determined by one or more of the processors.

Description

DESCRIPTION OF THE FIGURES

[0047] In the following, the invention is described with reference to the enclosed figures which represent a specific embodiment of the invention. The scope of the invention is not however limited to the specific features disclosed in the context of the figures, wherein

[0048] FIG. 1 illustrates a flow diagram of steps of the method in accordance with an aspect of the invention; and

[0049] FIG. 2 illustrates the properties of a growth cone.

[0050] According to FIG. 1, the method starts ins step S1 with acquisition of the patient medical image data which encompasses acquiring image data describing series of scans from multiple patients, that have been taken for the purpose of tumour monitoring/tracking (the tumour has been segmented). Subsequent step S2 is directed to determining the patient spatial development data which encompasses performing image fusions for all scans in the series of one patient. In the following step S3, the atlas data comprising a description of a universal atlas is acquired. Then, for example one registration of a four-dimensional tumour monitoring series per patient to the universal atlas is performed. In subsequent step S4, the four-dimensional series of tumour objects per patient is transformed into the reference system (coordinate space) in which spatial relationships of the universal atlas are defined. Step S4 also encompasses determination of the development probability data, specifically calculation of all probable growth cones for all obtained starting configurations of tumour positions and sizes.

[0051] The resulting four-dimensional growth cone map may be saved on a non-transitory electronic (digital) computer-readable storage medium. When a new patient is treated, that patient's scan can be registered with the saved four-dimensional growth cone map. Onto that patient's anatomy, the information obtained on the basis of the four-dimensional growth cone visualization for that patient can be overlaid. The resulting information can be used during image-guided surgery, e.g. with the Navigated Brush offered by Brainlab AG or another application (e.g. in microscope head-up display) visualizing (statistical) information on the tumour in question on the patients anatomy or in the surgical situation.

[0052] Using the universal atlas as input data allows to bring information from multiple patients with differing tumour locations into one common frame of reference. Through this image, time series with segmented tumours are brought into overlay and clusters of spatially similar tumours can be identified. Such an averaging map focuses on one disease indication (such as e.g. Low-grade gliomas) and clusters the data into spatially similar tumour location groups. Once one group is identified (e.g. frontal left LGGs) the typical growth pattern can be averaged by calculating the 4D tumour growth cones and averaging them.

[0053] The universal atlas-based technology offers a method to bring four-dimensional information on tumour growth patterns into a format, from which it can be applied to individual patients and be utilized during image guided surgery. This allows real-time utilization of statistical information and clinical decision support for the surgeon in a systematic and unprecedented manner.

[0054] By now visualizing this information during surgery via Navigated Brush, the surgeon can be precisely informed on what part of the tumour on which he/she should focus on most for to be removed during surgery. Real-time removal tracking of the tumour is allowed for by the Navigated Brush by offering a possibility of intraoperative modification of a pre-planned tumour object after tumour resection. This modification is calculated by an algorithm that interpolates points that have been acquired on the surgical margin of the residual tumour using a navigated instrument. Hence, the tumour growth cone might change based on information on the growth/recurrence patterns of such residual tumours.

[0055] Advanced visualization via Navigated Brush may provide: [0056] a coloured map indicating areas of potential tumour growth as texture on a three-dimensional tumour object; [0057] three-dimensional simulation of potential tumour growth (rendering of a three-dimensional model); [0058] “guided mode” visualization/workflow as real-time clinical decision support (e.g. surgeon decides/wants to be guided to only focus on areas with high potential for tumour growth); [0059] utilization of a stereoscopic head-up display or three-dimensional video-overlay in a next-generation surgical microscope; and [0060] volumetric report of different tumour growth areas (high, mid, low).

[0061] FIG. 2 shows determination of a growth cone for a specific case of tumour growth, wherein the following is illustrated in the sub-figures of FIG. 2: [0062] Sub-figure 1: The tumour is segmented in an image. [0063] Sub-figure 2: Follow-up scans show the growth from time point 1 to 2. [0064] Sub-figure 3: Follow-up scans show the growth from time point 1 to 2, 2 to 3. [0065] Sub-figure 4: Dominant growth directions can be derived from the preceding three steps. [0066] Sub-figure 5: By using the universal atlas, multiple patient tumour growth information can be included, thereby averaging dominant growth directions from multiple patient image time series to be aggregated in one common frame of reference. [0067] Sub-figure 6: From this averaging, information zones of the tumour at time point 1 can be highlighted which will most likely grow out most aggressively. [0068] Sub-figure 7: During image guided surgery, the information made available in sub-figures 5 and 6 can be brought into the space of the individual patient image through registration of the patient image to the common frame of reference (sub-figure 5), now the tumour at hand can be enriched by highlighting zones with growth prediction information—this information can be visualized both on a medical navigation system (which is suitable for us e.g. in surgery) via e.g. Navigated Brush as well as in a surgical microscope with a head-up display or video-overlay configuration.