Abstract: For training a machine learning system for representing a patient body a plurality of stored medical imaging data sets each representing at least a part of a respective patient are obtained. A first one of the plurality of stored medical imaging data sets represents a different part of the patient body than a second one of the plurality of stored medical imaging data sets. A plurality of landmarks in the stored medical imaging data sets are estimated, and each of the stored medical imaging data sets are aligned to a predefined pose using the plurality of landmarks. A plurality of points in the aligned medical imaging data sets are sampled, and the machine learning system is trained based on at least the plurality of points. The learned parameters of the machine learning system are then stored and used in a method for inferring a body representation.

Abstract: Object specific in-homogeneities in an MRI system are corrected. Prescan information available at the MR imaging system is determined. The prescan information includes at least object specific information of an object located in the MR imaging system from which an MR image is to be generated. The prescan information does not include a B1 map of the MRI system with the object being present in the MR imaging system. The prescan information is applied to a trained machine learning module provided at the MRI system. The trained machine learning module determines and generates shimming information as output. The shimming information is applied to a shimming module of the MR imaging system, wherein the shimming module uses the shimming information to generate a corrected magnetic field B0.

Abstract: Various examples of the disclosure pertain to using whole-slide images that depict healthy tissue for a training process for at least one machine-learning algorithm for digital pathology. For instance, an autoencoder neural network can be trained based on the healthy tissue.

Abstract: Various disclosed examples pertain to digital pathology, more specifically to training of a segmentation algorithm for segmenting whole-slide images depicting tissue of multiple types. An initial annotation of a whole-slide image is refined to yield a refined annotation based on which parameters of the segmentation algorithm can be set. Techniques of patch-wise weak supervision can be employed for such refinement.

Abstract: Patient biographic data may be estimated by receiving patient image data, applying the patient image data to a machine learned model, the machine learned model trained on second patient data and trained to map the second patient data to associated biographic data using machine learned features, generating the patient biographic data based on the applying and the machine learned features, and outputting the patient biographic data. The patient biographic data may include a patient weight, a patient height, a patient gender, and a patient age.

Abstract: For patient weight estimation in a medical imaging system, a patient model, such as a mesh, is fit to a depth image. One or more feature values are extracted from the fit patient model, reducing the noise and clutter in the values. The weight estimation is regressed from the extracted features.

Abstract: Various example embodiments pertain to processing images that depict tissue samples using a neural network algorithm. The neural network algorithm includes multiple encoder branches that are copies of each other that share the same parameters. The encoder branches can, accordingly, be referred to as Siamese copies of each other.

Abstract: A model dataset is generated based on first image data. The model dataset and second image data map at least a common part of an examination region at a second detail level. The model dataset and the second image data are pre-aligned at a first detail level below the second detail level based on first features that are mapped at the first detail level in the model dataset and the second image data and/or an acquisition geometry of the second image data. The model dataset and the second image data are registered at the second detail level based on second features that are mapped at the second detail level in the model dataset and the second image data. The second class of features is mappable at the second detail level or above. The registered second image data and/or the registered model dataset is provided.

Abstract: For training for and performance of patient modeling from surface data in a medical system, a progressive multi-task model is used. Different tasks for scanning are provided, such as landmark estimation and patient pose estimation. One or more features learned for one task are used as fixed or constant features in the other task. This progressive approach based on shared features increases efficiency while avoiding reductions in accuracy for any given task.

Abstract: A method for controlling a scanner comprises: sensing an outer surface of a body of a subject to collect body surface data, using machine learning to predict a surface of an internal organ of the subject based on the body surface data, and controlling the scanner based on the predicted surface of the internal organ.

Abstract: Object specific in-homogeneities in an MRI system are corrected. Prescan information available at the MR imaging system is determined. The prescan information includes at least object specific information of an object located in the MR imaging system from which an MR image is to be generated. The prescan information does not include a B1 map of the MRI system with the object being present in the MR imaging system. The prescan information is applied to a trained machine learning module provided at the MRI system. The trained machine learning module determines and generates shimming information as output. The shimming information is applied to a shimming module of the MR imaging system, wherein the shimming module uses the shimming information to generate a corrected magnetic field B0.

Abstract: Various example embodiments pertain to processing images that depict tissue samples using a neural network algorithm. The neural network algorithm includes multiple encoder branches that are copies of each other that share the same parameters. The encoder branches can, accordingly, be referred to as Siamese copies of each other.

Abstract: Synthetic CT is estimated for planning or other purposes from surface data (e.g., depth camera information). The estimation uses parameterization, such as landmark and/or segmentation information, in addition to the surface data. In training and/or application, the parameterization may be used to correct the predicted CT volume. The CT volume may be predicted as a sub-part of the patient, such as estimating the CT volume for scanning one system, organ, or type of tissue separately from other system, organ, or type of tissue.

Abstract: A stream of virtual topograms, in particular live virtual topograms, is predicted. Sets of surface data of an outer surface of a subject are continuously received. Based on each received set of surface data a (live) virtual topogram is continuously generated by a trained machine learning algorithm (MLA). Thereto, a representation of body landmarks is updated based on each received set of surface data by a trained body marker detector (BMD), of the trained MLA, and the (live) virtual topogram is predicted based on the updated spatial marker map and on the corresponding set of surface data by a trained topogram generator (TG) of the trained MLA.

Abstract: For topogram predication from surface data, a sensor captures the outside surface of a patient. A generative adversarial network (GAN) generates the topogram representing an interior organ based on the outside surface of the patient. To further adapt to specific patients, internal landmarks are used in the topogram prediction. The topogram generated by one generator of the GAN may be altered based on landmarks generated by another generator.

Abstract: For training a machine learning system for representing a patient body a plurality of stored medical imaging data sets each representing at least a part of a respective patient are obtained. A first one of the plurality of stored medical imaging data sets represents a different part of the patient body than a second one of the plurality of stored medical imaging data sets. A plurality of landmarks in the stored medical imaging data sets are estimated, and each of the stored medical imaging data sets are aligned to a predefined pose using the plurality of landmarks. A plurality of points in the aligned medical imaging data sets are sampled, and the machine learning system is trained based on at least the plurality of points. The learned parameters of the machine learning system are then stored and used in a method for inferring a body representation.

Abstract: Patient biographic data may be estimated by receiving patient image data, applying the patient image data to a machine learned model, the machine learned model trained on second patient data and trained to map the second patient data to associated biographic data using machine learned features, generating the patient biographic data based on the applying and the machine learned features, and outputting the patient biographic data. The patient biographic data may include a patient weight, a patient height, a patient gender, and a patient age.

Abstract: A method is disclosed for adjusting a collimator of an X-ray source. In an embodiment, the method includes detecting an arrangement of an X-ray detector with respect to the X-ray source; automatically determining an adjustment for the collimator based on the detected position of the X-ray detector with respect to the X-ray source; and automatically adjusting the collimator based on the determined adjustment for the collimator. An X-ray device and computer readable medium are also disclosed.

Abstract: Synthetic CT is estimated for planning or other purposes from surface data (e.g., depth camera information). The estimation uses parameterization, such as landmark and/or segmentation information, in addition to the surface data. In training and/or application, the parameterization may be used to correct the predicted CT volume. The CT volume may be predicted as a sub-part of the patient, such as estimating the CT volume for scanning one system, organ, or type of tissue separately from other system, organ, or type of tissue.

Abstract: During the generation of a panoramic x-ray recording, the use of semi-transparent x-ray screens allows the patient's x-ray exposure to be reduced when partial x-ray images are created, in spite of relatively large overlapping areas between the partial x-ray images.