Abstract: A method is disclosed for the automatic determination of the bone density and a method is disclosed for the automatic detection and characterization of spinal column fractures. Both methods enable the fully automatic detection and assessment of damaged vertebrae and reliably enable an analysis of the state of the vertebrae with a high accuracy rate. A computed tomography system to carry out either of the methods is further disclosed.

Abstract: A method for generating a synthetic two-dimensional mammogram with enhanced contrast for structures of interest includes acquiring a three-dimensional digital breast tomosynthesis volume having a plurality of voxels. A three-dimensional relevance map that encodes for the voxels the relevance of the underlying structure for a diagnosis is generated. A synthetic two-dimensional mammogram is calculated based on the three-dimensional digital breast tomosynthesis volume and the three-dimensional relevance map.

Abstract: A method for generating a synthetic two-dimensional mammogram with enhanced contrast for structures of interest includes acquiring a three-dimensional digital breast tomosynthesis volume having a plurality of voxels. A three-dimensional relevance map that encodes for the voxels the relevance of the underlying structure for a diagnosis is generated. A synthetic two-dimensional mammogram is calculated based on the three-dimensional digital breast tomosynthesis volume and the three-dimensional relevance map.

Abstract: An embodiment of the method is disclosed for non-invasive lesion candidate detection in a patient's body includes generating a number of first medical images of the patient's body. The method further includes identifying lesion-like geometrical regions inside the first medical images of the patient's body by applying image processing methods, whereby the identification is at least partly controlled by a number of patient-specific context features which are not directly extractable from the first medical images. In addition, the method includes selecting a number of the identified lesion-like geometrical regions as lesion candidates.

Abstract: A method is disclosed for the automatic determination of the bone density and a method is disclosed for the automatic detection and characterization of spinal column fractures. Both methods enable the fully automatic detection and assessment of damaged vertebrae and reliably enable an analysis of the state of the vertebrae with a high accuracy rate. A computed tomography system to carry out either of the methods is further disclosed.

Abstract: A system and method for regression-based segmentation of the mitral valve in 2D+t cardiac magnetic resonance (CMR) slices is disclosed. The 2D+t CMR slices are acquired according to a mitral valve-specific acquisition protocol introduced herein. A set of mitral valve landmarks is detected in each 2D CMR slice and mitral valve contours are estimated in each 2D CMR slice based on the detected landmarks. A full mitral valve model is reconstructed from the mitral valve contours estimated in the 2D CMR slices using a trained regression model. Each 2D CMR slice may be a cine image acquired over a full cardiac cycle. In this case, the segmentation method reconstructs a patient-specific 4D dynamic mitral valve model from the 2D+t CMR image data.

Abstract: A method and system for providing medical decision support based on virtual organ models and learning based discriminative distance functions is disclosed. A patient-specific virtual organ model is generated from medical image data of a patient. One or more similar organ models to the patient-specific organ model are retrieved from a plurality of previously stored virtual organ models using a learned discriminative distance function. The patient-specific valve model can be classified into a first class or a second class based on the previously stored organ models determined to be similar to the patient-specific organ model.

Abstract: A method in radiological imaging for determining lesions in image data of an examination object is described. In an embodiment, the method includes determining anatomical structures by hierarchical breakdown of the image data of the examination object. The method furthermore includes image data analysis for localizing lesion candidates in the anatomical structures. Moreover, the method also includes determining the lesions by evaluating and filtering the lesion candidates. Moreover, an image processing workstation in radiological imaging for determining lesions in image data of an examination object and an imaging apparatus are described.

Abstract: An embodiment of the method is disclosed for non-invasive lesion candidate detection in a patient's body includes generating a number of first medical images of the patient's body. The method further includes identifying lesion-like geometrical regions inside the first medical images of the patient's body by applying image processing methods, whereby the identification is at least partly controlled by a number of patient-specific context features which are not directly extractable from the first medical images. In addition, the method includes selecting a number of the identified lesion-like geometrical regions as lesion candidates.

Abstract: A method of performing image retrieval includes training a random forest RF classifier based on low-level features of training images and a high-level feature, using similarity values generated by the RF classifier to determine a subset of the training images that are most similar to one another, and classifying input images for the high-level feature using the RF classifier and the determined subset of images.

Abstract: A method and system for providing detecting and classifying coronary stenoses in 3D CT image data is disclosed. Centerlines of coronary vessels are extracted from the CT image data. Non-vessel regions are detected and removed from the coronary vessel centerlines. The cross-section area of the lumen is estimated based on the coronary vessel centerlines using a trained regression function. Stenosis candidates are detected in the coronary vessels based on the estimated lumen cross-section area, and the significant stenosis candidates are automatically classified as calcified, non-calcified, or mixed.

Abstract: A method of performing image retrieval includes training a random forest RF classifier based on low-level features of training images and a high-level feature, using similarity values generated by the RF classifier to determine a subset of the training images that are most similar to one another, and classifying input images for the high-level feature using the RF classifier and the determined subset of images.

Abstract: A system and method for regression-based segmentation of the mitral valve in 2D+t cardiac magnetic resonance (CMR) slices is disclosed. The 2D+t CMR slices are acquired according to a mitral valve-specific acquisition protocol introduced herein. A set of mitral valve landmarks is detected in each 2D CMR slice and mitral valve contours are estimated in each 2D CMR slice based on the detected landmarks. A full mitral valve model is reconstructed from the mitral valve contours estimated in the 2D CMR slices using a trained regression model. Each 2D CMR slice may be a cine image acquired over a full cardiac cycle. In this case, the segmentation method reconstructs a patient-specific 4D dynamic mitral valve model from the 2D+t CMR image data.

Abstract: A method and system for providing detecting and classifying coronary stenoses in 3D CT image data is disclosed. Centerlines of coronary vessels are extracted from the CT image data. Non-vessel regions are detected and removed from the coronary vessel centerlines. The cross-section area of the lumen is estimated based on the coronary vessel centerlines using a trained regression function. Stenosis candidates are detected in the coronary vessels based on the estimated lumen cross-section area, and the significant stenosis candidates are automatically classified as calcified, non-calcified, or mixed.

Abstract: A method and system for providing medical decision support based on virtual organ models and learning based discriminative distance functions is disclosed. A patient-specific virtual organ model is generated from medical image data of a patient. One or more similar organ models to the patient-specific organ model are retrieved from a plurality of previously stored virtual organ models using a learned discriminative distance function. The patient-specific valve model can be classified into a first class or a second class based on the previously stored organ models determined to be similar to the patient-specific organ model.

Abstract: A medical ontology may be used for computer assisted clinical decision support. Multi-level and/or semantically grouped medical ontology is incorporated into a machine learning algorithm. The resulting machine-learnt algorithm outputs information to assist in clinical decisions. For example, a patient record is input to the algorithm. Based on the incorporated medical ontology, similarities are aggregated in different groups. An aggregate similarity of at least one group is a function of an aggregate similarity of at least another group. One or more similar patients and/or outcomes are identified based on similarity. Probability based outputs may be provided.

Abstract: A method and an apparatus retrieve additional information regarding a patient record. Applying the subject-matter of the present invention clinical experts, such as doctors, can be provided with domain specific background knowledge, and can, hence, be supported in making decisions regarding the treatment of patients. Said additional information can be structured and visualised, which makes the retrieved information understandable also for persons without advanced knowledge regarding data processing. The retrieval of additional information is accomplished by a comparison of data being stored in the patient record and data being stored in a predefined textual resource, such as a ontology, and an identification of further terms describing attributes of the patient record. This finds application in supporting the diagnosis of patients and healthcare related information retrieval tasks.

Abstract: A medical ontology may be used for computer assisted clinical decision support. Multi-level and/or semantically grouped medical ontology is incorporated into a machine learning algorithm. The resulting machine-learnt algorithm outputs information to assist in clinical decisions. For example, a patient record is input to the algorithm. Based on the incorporated medical ontology, similarities are aggregated in different groups. An aggregate similarity of at least one group is a function of an aggregate similarity of at least another group. One or more similar patients and/or outcomes are identified based on similarity. Probability based outputs may be provided.