Siemens Healthcare Gmbh

Siemens Aktiengesellschaft

Siemens Healthineers Ag

Other

Siemens Corporation

Friedrich-Alexander-Universitat Erlangen-Nürnberg

Andreas Wimmer

Vivek Kumar Singh

Terrence Chen

Yao-Jen Chang

Ankur Kapoor

Grzegorz Soza

Kai Ma

Birgi Tamersoy

Ruhan Sa

Shaohua Kevin Zhou

Michael Suehling

Rui Liao

Joachim Hornegger

Norbert Strobel

Michael Wels

Andreas Krauss

Alexander Brost

Klaus J Kirchberg

Rainer Raupach

Ziyan Wu

David D Liu

Klaus Engel

Christian Hofmann

Wen Wu

Thomas J O'Donnell

Carsten Thierfelder

Noha Youssry El-Zehiry

Thilo Hannemann

Christian Tietjen

Martin Kramer

Sven Kohle

Michael Sühling

Brian Teixeira

Jiangping Wang

Philipp Hölzer

Ferdinand Distler

Eva Eibenberger

Andreas Prokein

Christian Petry

Amitkumar Bhupendrakumar Shah

Martin Willibald Koch

Tobias Betz

Murali Tharan Gnanamani

Florian Maier

Santosh Pai

Jianping Wang

Stefanie Steding

Zhuokai Zhao

Jumana Jaber

Klaus Kirchberg

Machine learning is used to train a network to estimate a three-dimensional (3D) body surface and body regions of a patient from surface images of the patient. The estimated 3D body surface of the patient is used to determine an isocenter of the patient. The estimated body regions are used to generate heatmaps representing visible body region boundaries and unseen body region boundaries of the patient. The estimation of 3D body surfaces, the determined patient isocenter, and the estimated body region boundaries may assist in planning a medical scan, including automatic patient positioning.

Dense body marker estimation from camera data for patient positioning in medical imaging

A computer-implemented method is for operating a medical imaging system. The imaging system includes an imaging unit and a positioning apparatus for positioning an examination object. In an embodiment, the method includes ascertainment of object information relating to the examination object; determination of a region of interest for the examination object, the region of interest being defined as a region of the examination object to be examined by the imaging unit in the context of an examination to be performed; arrival at a determination concerning a positional state of the examination object relative to the positioning apparatus and, to be precise, based on the region of interest and the object information; and provision of user output for a user of the imaging system based on the determination.

Optimizing the positioning of a patient on a patient couch for medical imaging

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.

Estimating patient biographic data parameters

The method comprises receiving an image sequence of the subject from the camera during the medical imaging scan; receiving at least one of the current position or velocity of the patient table during the medical imaging scan; performing a motion tracking analysis of the image sequence to extract a motion model, wherein at least one of the motion tracking analysis or the motion model is tailored to the body region of interest and takes into account the at least one of the current patient table position or velocity; and analysing the motion model to detect subject motion and, if the detected motion is above a threshold, at least one of adapting the medical imaging examination or issuing an alert.

Method for controlling a medical imaging examination of a subject, medical imaging system and computer-readable data storage medium

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.

Multi-task progressive networks for patient modeling for medical scans

For patient model estimation from surface data in a medical system, a stream or sequence of depth camera captures are performed. The fitting of the patient model is divided between different times or parts of the sequence, using the streaming capture to distribute processing and account for patient movement. Less manual involvement may be needed due to the regular availability of image captures. Subsequent fitting may benefit from previous fitting.

Patient model estimation from camera stream in medicine

A framework for sensor-based patient treatment support. In accordance with one aspect, one or more sensors are used to acquire sensor data of one or more objects of interest. The sensor data is then automatically interpreted to generate processing results. One or more actions may be triggered based on the processing results to support treatment of a patient, including supporting medical scanning of the patient.

Sensor-based patient treatment support

A method is for determining a patient weight and/or a body mass index of a patient. The method includes acquiring image data containing depth information of the patient; generating a surface model of the patient based upon the image data acquired; determining density information or X-ray attenuation information of at least part of the patient; and determining at least one of the total patient weight and the body mass index of the patient using the surface model generated and the density information or X-ray attenuation information determined.

Method and apparatus for determining a patient weight and/or a body mass index

In an embodiment, a method includes providing a three-dimensional structure image including surface data relating to external contours and structure data relating to internal structures of a body region; recording the body region via a camera system, to produce a recording while a patient is situated in or on the medical imaging system; registering, at least locally and as a structure image, the three-dimensional surface image in relation to the three-dimensional structure image; producing an overlay image including at least part of the structure image registered including structure data of the three-dimensional structure image in a form of an overlay with at least one of the three-dimensional surface image and a real view of a corresponding part of the body region of the patient; and controlling the medical imaging system to record body regions of the patient based upon an arrangement of the structure data registered in the overlay image.

Controlling a medical imaging system

A method of obtaining a medical image includes obtaining, via a camera, at least one surface image of a patient. A pose of the patient is determined from the at least one surface image of the patient using at least one spatial information module. The patient is positioned, via a moveable bed, to an imaging start position and a medical image of the patient is obtained using a medical imaging modality.

System and method for assisted patient positioning

Machine learning is used to train a network to predict the location of an internal body marker from surface data. A depth image or other image of the surface of the patient is used to determine the locations of anatomical landmarks. The training may use a loss function that includes a term to limit failure to predict a landmark and/or off-centering of the landmark. The landmarks may then be used to configure medical scanning and/or for diagnosis.

Internal body marker prediction from surface data in medical imaging

A method and a device are disclosed for the perspective representation, via an output image, of at least one virtual scene component arranged within a real scene. In the method, depth image data of the real scene is captured from a first perspective via a depth image sensor, and 2D image data of the real scene is captured from a second perspective via a 2D camera. Further, a virtual three-dimensional scene model of the real scene is created with reference to depth information from the depth image data, and at least one virtual scene component is inserted into the three-dimensional virtual model. Finally, an output image is generated by way of perspective projection of the 2D image data corresponding to the second perspective onto the virtual three-dimensional scene model comprising the virtual scene component.

Perspective representation of a virtual scene component

The present embodiments relate to cinematic volume renderings and/or volumetric Monte-Carlo path tracing. By way of introduction, the present embodiments described below include apparatuses and methods for cinematic rendering of unfolded three-dimensional volumes. An image analysis algorithm is performed on an input volume to extract one or more structures of interest, such as a rib centerline, a liver surface or another three-dimensional volume. Based on the extracted three-dimensional structure(s), a geometric transformation is computed to generate an unfolded three-dimensional volume of the structure(s). Cinematic volume rendering techniques are used to generate a rendered image from the unfolded three-dimensional volume.

Cinematic rendering of unfolded 3D volumes

For human body representation, bone length or other size characteristic that varies within the population is incorporated into the geometric model of the skeleton. The geometric model may be normalized for shape or tissue modeling, allowing modeling of the shape without dedicating aspects of the data-driven shape model to the length or other size characteristic. Given the same number or extent of components of the data-driven shape model, greater or finer details of the shape may be modeled since components are not committed to the size characteristic.

Human body representation with non-rigid parts in an imaging system

Growing community of inventors

Forchheim, Germany

Andreas Wimmer