# US 2014/0153804 A1 · Method for evaluating image data records > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0153804 A1 - Document ID: US20140153804A1 - Status: Published application - Kind: A1 - Application number: 14/084,892 - Filed: 2013-11-20 - Published: 2014-06-05 - Earliest claimed: 2012-12-03 - Assignees: Siemens Aktiengesellschaft - Applicants: Siemens Aktiengesellschaft - Inventors: Kirstin JATTKE; Sebastian SCHMIDT; Harald WERTHNER - Web page: https://patentgrep.com/patent/US20140153804A1 ## CPC - G06T 7/0012: {Biomedical image inspection} - G06T 7/11: Region-based segmentation - G06T 7/174: involving the use of two or more images - G06T 2207/10084: Hybrid tomography; Concurrent acquisition with multiple different tomographic modalities - G06T 2207/10088: Magnetic resonance imaging [MRI] - G06T 2207/10104: Positron emission tomography [PET] - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract In an embodiment of a method, a PET image data record, a functional magnetic resonance image data record and a morphological magnetic resonance image data record, the spatial resolution of which is better than that of the functional magnetic resonance image data record, of the target area are recorded with the combination image recording facility, whereby a center of the target structure is localized in the PET image data record. The center is transmitted to the functional magnetic resonance image data record, based on the center the target structure is segmented in the functional magnetic resonance image data record, the segmentation of the target structure in the functional magnetic resonance image data record is transmitted to the morphological magnetic resonance image data record and is improved there within the scope of a fine segmentation. ## Claims ### Claim 1 A method for evaluating image data records recorded using a combination image recording facility configured to record magnetic resonance image data and positron emission tomography image data in a shared coordinate system so as to determine position and extent of a target structure in a target area of a human body, the method comprising: recording a PET image data record, a functional magnetic resonance image data record and a morphological magnetic resonance image data record, the spatial resolution of which is better than that of the functional magnetic resonance image data record, of the target area using the combination image recording facility; localizing a center of the target structure in the PET image data record; transmitting the localized center to the functional magnetic resonance image data record; segmenting, based on localized the center, the target structure in the functional magnetic resonance image data record; and transmitting the segmentation of the target structure in the functional magnetic resonance image data record to the morphological magnetic resonance image data record. ### Claim 2 (depends on 1) The method of claim 1, wherein the functional magnetic resonance image data record is recorded as at least one of a diffusion-weighted magnetic resonance image data record, a Dynamic Contrast Enhancement magnetic resonance image data record, an Arterial Spin Labeling magnetic resonance image data record and a perfusion magnetic resonance image data record. ### Claim 3 (depends on 1) The method of claim 1, wherein the morphological magnetic resonance image data record is recorded in at least one of a proton-density-weighted manner, a T1-weighted manner and a T2-weighted manner. ### Claim 4 (depends on 1) The method of claim 1, wherein the PET image data record and the magnetic resonance image data record are recorded at least partially at the same time and/or with a stationary body. ### Claim 5 (depends on 1) The method of claim 1, wherein, in order to determine the center, at least one of a maximum positron emission tomography image datum of the target structure is selected; and the center area is segmented on a threshold value basis. ### Claim 6 (depends on 1) The method of claim 1, wherein, for segmentation in the functional magnetic resonance image data record, at least one of a region growing algorithm and a random walker algorithm is used. ### Claim 7 (depends on 1) The method of claim 1, wherein an edge is sought in the morphological magnetic resonance image data record for fine segmentation in a search area lying about the edge determined in the functional magnetic resonance image data record. ### Claim 8 (depends on 7) The method of claim 7, wherein the search area at least one of corresponds to a voxel of the functional magnetic resonance image data record in terms of size, and is adjustable by a user. ### Claim 9 (depends on 7) The method of claim 7, wherein a threshold value for detecting an edge in the morphological magnetic resonance image data record is determined as a function of a noise value describing local noise. ### Claim 10 (depends on 7) The method of claim 7, wherein the boundary of the target structure located in the functional magnetic resonance image data record is not retained in any detectable edge in the search area. ### Claim 11 (depends on 1) The method of claim 1, wherein the target area is the lungs and the morphological magnetic resonance image data record is recorded in a proton-weighted manner. ### Claim 12 A combination image recording facility, comprising: a control facility configured to at least record a PET image data record, a functional magnetic resonance image data record and a morphological magnetic resonance image data record, the spatial resolution of which is better than that of the functional magnetic resonance image data record, of the target area using the combination image recording facility; localize a center of the target structure in the PET image data record; transmit the localized center to the functional magnetic resonance image data record; segment, based on localized the center, the target structure in the functional magnetic resonance image data record; and transmit the segmentation of the target structure in the functional magnetic resonance image data record to the morphological magnetic resonance image data record. ### Claim 13 (depends on 1) A computer program, configured to perform the method of claim 1, when run on a computing facility. ### Claim 14 (depends on 1) The method of claim 1, wherein the target structure is a tumor. ### Claim 15 (depends on 1) The method of claim 1, wherein the localizing of the center includes localizing at least one of a center area and a center point of the target structure in the PET image data record. ### Claim 16 (depends on 2) The method of claim 2, wherein the morphological magnetic resonance image data record is recorded in at least one of a proton-density-weighted manner, a T1-weighted manner and a T2-weighted manner. ### Claim 17 (depends on 8) The method of claim 8, wherein a threshold value for detecting an edge in the morphological magnetic resonance image data record is determined as a function of a noise value describing local noise.