Computer-aided evaluation of an image dataset
A method and system for the diagnosis of 3D images are disclosed, which significantly cuts the time required for the diagnosis. The 3D images are for example an image volume dataset of a magnetic resonance tomography system which is saved in an RIS or PACS system. In at least one embodiment, the diagnostic finding are partially automatically generated, and details of the position, size and change in pathological structures are compared to previous diagnostic findings are generated automatically. As a result of this automation the diagnostic work of radiologists is significantly reduced.
1. A method for the computer-aided evaluation of an image dataset, comprising:
generating an unannotated image dataset with the aid of a radiological examination device and saving the image dataset in an image database;
selecting a pathological structure in the image data set for evaluation;
generating a report including a description of appearance and position of a pathological structure and changes in the pathological structure in the image dataset and saving the report in an annotation database without user input; and
automatically determining, via a positioning module, a position of the evaluated pathological structure in relation to one or more anatomical structures and saving the position in the annotation database in an updated report without user input, wherein
a previous evaluation for the pathological structure is saved in the annotation database,
automatically determining a change in the pathological structure in comparison of at least one of size and position with the previous evaluation and saving the change in the annotation database without user input,
at least one of automatically measuring or calculating, via a measurement module, an attribute of the pathological structure in the image dataset based on at least one of a position and a type of the pathological structure and saving the measurement or calculation in the annotation database, and
automatically combining the determined position of the pathological structure, the determined change in the pathological structure and the determined attribute of the pathological structure and therefrom generating and outputting a computer analyzable report in a structured form without user input, the report including at least one of diagnostic findings based on the combination of determined position of the pathological structure, the determined change in the pathological structure and the determined attribute of the pathological structure and a graph of changes of pathological structures; and
wherein a request for the report is received from a Radiology Information System (RIS) or Hospital Information System (HIS) system and contains request parameters,
wherein by way of at least one request parameter an anatomical region in a saved image data set is defined for the evaluation, and
wherein an image parser identifies only anatomical structures which lie in the anatomical region and the report is generated for the requested parameter.
2. The method as claimed in claim 1 ,
wherein the image parser automatically identifies the anatomical structures in the image dataset on the basis of a computerized learning procedure, and
wherein, for each of the anatomical structures, an identifier of the anatomical structure, a position of the anatomical structure in the image dataset and a segmentation of the anatomical structure are automatically saved in the annotation database.
3. The method as claimed in claim 1 ,
wherein each module is a software module or hardware module.
4. The method as claimed in claim 3 ,
wherein the hardware module is an application-specific integrated circuit or circuit board.
5. The method as claimed in claim 1 ,
wherein the image dataset is an image volume dataset or an image series dataset, and
wherein the radiological examination device is a computed tomography system, a magnetic resonance tomography system, a positron emission tomography system, an x-ray device or an ultrasound device.
6. The method as claimed in claim 1 ,
wherein the report is a medical diagnostic finding.
7. A non-transitory computer-readable data carrier, including a computer program, saved thereon, to execute the method as claimed in claim 1 when run in a computer.
8. A non-transitory computer readable medium including program segments for, when executed on a computer device, causing the computer device to implement the method of claim 1 .
9. The method as claimed in claim 1 ,
wherein a measurement module automatically performs a measurement or calculation of an attribute of the pathological structure in the image dataset and saves the measurement or calculation in the annotation database.
10. The method as claimed in claim 1 , further comprising generating a medical diagnosis based on the structured form output from the evaluation generator.
11. A system for the computer-aided evaluation of an image dataset, comprising:
a processor configured to control:
an image database in which an unannotated image dataset is saved;
a diagnostic station including navigation tools configured to allow a user to navigate to and select a pathological structure from the image dataset;
an annotation database in which a report is savable for the pathological structure selected from the image dataset;
a positioning module configured to receive coordinates of the pathological structure from the diagnostic station and automatically determine a position of the pathological structure in relation to one or more anatomical structures and create a report at least partially automatically without user input and save the position information to the annotation database without user input;
a measurement module configured to automatically measure or calculate an attribute of the pathological structure in the image dataset based on at least one of a position and a type of the pathological structure without user input and save the attribute to the annotation database without user input;
a change determination module configured to automatically determine a change in the pathological structure in comparison of at least one of size and position with the saved report without user input and which change is saved in the annotation database as an updated report; and
a report generator configured to automatically combine the determined position of the pathological structure without user input, the determined change in the pathological structure and the determined attribute of the pathological structure and automatically generate a computer analyzable report in a structured form from the combined data without user input, the report including at least one of diagnostic findings based on the combination of determined position of the pathological structure, the determined change in the pathological structure and the determined attribute of the pathological structure and a graph of changes of pathological structures, wherein a request for the report is received from a Radiology Information System (RIS) or Hospital Information System (HIS) system and contains request parameters,
wherein by way of at least one request parameter an anatomical region in a saved image data set is defined for the evaluation, and
wherein an image parser identifies only anatomical structures which lie in the anatomical region and the report is generated for the requested parameter.
12. The system as claimed in claim 11 ,
wherein the image parser is configured to automatically identify the anatomical structures in the image dataset on the basis of a computerized learning procedure.
13. The system as claimed in claim 11 , wherein each module is a software module or a hardware module.
14. The system as claimed in claim 13 , wherein the hardware module is an application-specific integrated circuit or circuit board.