Koninklijke Philips Corporation N.v.

Other

Philips Gmbh

Christian Buerger

Irina Waechter-Stehle

Frank Michael Weber

Juergen Weese

Steffen Renisch

Michael Grass

Sheng-Wen Huang

Emil George Radulescu

Sabine Mollus

Jochen Peters

Ramon Quido Erkamp

Ivan Salgo

Shougang Wang

Tobias Klinder

Thomas Heiko Stehle

David Prater

Scott Holland Settlemier

Robert Joseph Schneider

Tim Nielsen

Amir Mohammad Tahmasebi Maraghoosh

Heinrich Schulz

Eberhard Sebastian Hansis

Christian Haase

Drazenko Babic

Nicole Schadewaldt

Jens Wiegert

Christian Lorenz

Joerg Sabczynski

Jörg Sabczynski

Harald Sepp Heese

Niels Nijhof

Alexander Schmidt-Richberg

Torbjorn Vik

Matthias Lenga

Irina Wächter-Stehle

Michael Daniel Cardinale

Michael Cardinale

Philippe Paul Mazouer

In a method of segmenting a feature in an image, an image product related to the image is provided () to a model trained using a machine learning process. An indication of a shape descriptor for the feature in the image is received () from the model, based on the image product. The indicated shape descriptor is then used () in a model based segmentation, MBS, to initialize the MBS and segment the feature.

Methods and systems for segmenting images

The present invention relates to matching a field of view for mobile 3D imaging, for example mobile C-arm 3D imaging In order to provide image data that is improved for comparing purposes, for example when using a mobile X-ray imaging system, first location information of a first reconstruction volume based on a first sequence of X-ray images of a region of interest of a subject acquired along a first trajectory in a first position of an X-ray imaging device is received. Further, a planned second trajectory for acquiring a second sequence of X-ray images in a second position of the X-ray imaging device is received and a resulting second reconstruction volume for the second sequence of X-ray images is calculated. Then, second location information for the second reconstruction volume is determined. Further, a degree of comparability for the first reconstruction volume and the second reconstruction volume is determined based on the first location information and the second location information. An adapted second trajectory is calculated that results in an increased degree of comparability of the first reconstruction volume and the second reconstruction volume. The adapted second trajectory is provided for acquiring the second sequence of X-ray images in the second position of the X-ray imaging device.

Field of view matching for mobile 3D imaging

The invention relates to an ultrasound system for imaging a volumetric region comprising a region of interest (ROI). The system comprises a probe; an ultrasound wave controlling unit adapted to control ultrasound wave transmission and provide ultrasound image data of the volumetric region; an image processor; and a ROI identifier, which is adapted to generate identification data indicating the ROI within the volumetric region; wherein the ultrasound wave transmission is configurable by a plurality of use cases in response to respective identifiers of said use cases, each use case being associated with a particular imaging procedure and comprising an anatomical model for said imaging procedure; and wherein the ROI identifier is configurable by the respective anatomical models of said use cases. A method of configuring such an ultrasound system is also disclosed.

Intelligent ultrasound system

A method and system for mapping boundary detecting features of at least one source triangulated mesh of known topology to a target triangulated mesh of arbitrary topology. A region of interest in a volumetric image associated with each triangle of the target triangulated mesh is provided to a feature mapping network. The feature mapping network assigns a feature selection vector to each triangle of the target triangulated mesh. The associated region of interest and assigned feature selection vector for each triangle of the target triangulated mesh are provided to a boundary detection network. A predicted boundary based on features of the associated region of interest selected by the assigned feature selection vector is obtained from the boundary detection network.

Model-based image segmentation

The invention provides for a medical apparatus () comprising a memory () for storing machine executable instructions () and a processor () for controlling the medical apparatus. Execution of the machine executable instructions causes the processor to: receive () a medical image () descriptive of a three-dimensional anatomy of a subject (); and provide () an image segmentation () by segmenting the medical image into multiple tissue regions () using a model-based segmentation. The model-based segmentation assigns a tissue type to each of the multiple regions. The model-based segmentation has a surface mesh (). The segmentation is corrected by using the tissue type assigned to each of the multiple regions to correct for partial volume effects at boundaries formed by the surface mesh between at least some of the multiple tissue regions.

Simultaneous partial volume corection and segmentation refinement

An ultrasound system is disclosed comprising an ultrasound transducer array () comprising a plurality of ultrasound transducer cells (), each of said cell having an independently adjustable position and/or orientation such as to conform an ultrasound transmitting surface of the cell to a region of a body and a controller (). The controller is configured to register the respective ultrasound transducer cells by simultaneously operating at least two ultrasound transducer cells in a transmit mode in which the cells transmit distinguishable ultrasound signals and operating the remaining ultrasound transducer cells in a receive mode. The controller extracts time-of-flight information of the respective ultrasound signals between transmitter and receiver and by systematically selecting different ultrasound transducer cells as transmitters, the controller collects sufficient time-of-flight information from which the respective position and/or relative orientation of the ultrasound transducer cells within the ultrasound transducer array may be derived. A method for operating the ultrasound system in this manner as well as a computer program product is also disclosed.

Ultrasound system, method and computer program product

For delivering an image-guided radiation therapy treatment to a moving structure included in a region of a patient body a series of first images of the region of the patient body in different phases of a motion of the structure is acquired in accordance with a first imaging mode. The series of first images is associated with a series of second images of the patient body in essentially the same phases of the motion of the target structure, the second images being acquired in a second imaging mode. During the treatment, a third image is acquired using the second imaging mode during the radiation therapy treatment and a continuation of the radiation therapy treatment is planned on the basis of data relating to one of the first images selected on the basis of a comparison between the third image and the second images associated with the first images.

Image-guided radiation therapy

An apparatus includes an imaging probe and is configured for dynamically arranging presentation of visual feedback for guiding manual adjustment, via the probe, of a location, and orientation, associated with the probe. The arranging is selectively based on comparisons between fields of view of the probe and respective results of segmenting image data acquired via the probe. In an embodiment, the apparatus includes a sensor which guides a decision that acoustic coupling quality is insufficient, the apparatus issuing a user alert upon the decision.

Anatomically intelligent echochardiography for point-of-care

An imaging steering apparatus includes sensors and an imaging processor configured for: acquiring, via multiple ones of the sensors and from a current position (), and current orientation (), an image of an object of interest; based on a model, segmenting the acquired image; and determining, based on a result of the segmenting, a target position (), and target orientation (), with the target position and/or target orientation differing correspondingly from the current position and/or current orientation. An electronic steering parameter effective toward improving the current field of view may be computed, and a user may be provided instructional feedback () in navigating an imaging probe toward the improving. A robot can be configured for, automatically and without need for user intervention, imparting force () to the probe to move it responsive to the determination.

Imaging view steering using model-based segmentation

An ultrasonic diagnostic imaging system has a user control by which a user positions the user's selection of a heart chamber border in relation to two myocardial boundaries identified by a deformable heart model. The user's border is positioned by a single degree of freedom control which positions the border as a function of a single user-determined value. This overcomes the vagaries of machine-drawn borders and their mixed acceptance by clinicians, who can now create repeatably-drawn borders and exchange the control value for use by others to obtain the same results.

Ultrasonic diagnosis of cardiac performance using heart model chamber segmentation with user control

There is provided a computer-implemented method () and apparatus for determining a transformation for anatomically aligning fragments of a broken bone. An image of the broken bone of the subject is acquired (). The bone is broken into two or more fragments. A model of a corresponding unbroken bone and at least one parameter is acquired. The at least one parameter defines one or more deformations to the model that are permitted when fitting portions of the model of the unbroken bone to corresponding fragments of the broken bone (). Portions of the model of the unbroken bone are fit to corresponding fragments of the broken bone based on the at least one parameter (). A transformation is determined that anatomically aligns the fragments of the broken bone with the corresponding portions of the model ().

Transformation determination for anatomically aligning fragments of a broken bone

The invention relates to a medical image data processing system () for image segmentation. The medical image data processing system () comprises a machine learning framework trained to receive an anatomical position of a voxel and to provide a tissue type classification. An execution of machine executable instructions by a processor () of the medical image data processing system () causes the processor () to control the medical image data processing system () to: &#x2014;receive medical image data () comprising an anatomical structure of interest, &#x2014;fit an anatomical frame of reference () to the medical image data () using model-based segmentation, &#x2014;classify tissue types represented by voxels of the medical image data () using the machine learning framework, wherein anatomical positions of the voxels with respect to the anatomical frame of reference () are used as the input to the machine learning framework.

Tissue classification using image intensities and anatomical positions

The application discloses a computer-implemented method () of providing a model for estimating an anatomical body measurement value from at least one 2-D ultrasound image including a contour of the anatomical body, the method comprising providing () a set of 3-D ultrasound images of the anatomical body; and, for each of said 3-D images, determining () a ground truth value of the anatomical body measurement; generating () a set of 2-D ultrasound image planes each including a contour of the anatomical body, and for each of the 2-D ultrasound image planes, extrapolating () a value of the anatomical body measurement from at least one of an outline contour measurement and a cross-sectional measurement of the anatomical body in the 2-D ultrasound image plane; and generating () said model by training a machine-learning algorithm to generate an estimator function of the anatomical body measurement value from at least one of a determined outline contour measurement and a determined cross-sectional measurement of a contour of the anatomical body within a 2-D ultrasound image using the obtained ground truth values, extrapolated values and at least one of the outline contour measurements and the cross-sectional measurements as inputs of said machine-learning algorithm. A computer-implemented method of deploying such a model, a computer program product, an ultrasound image processing apparatus and an ultrasound imaging system adapted to implement such methods are also disclosed.

Anatomical measurements from ultrasound data

The present invention relates to a medical image processing device (), comprising: &#x2014;a receiving unit () for receiving a first and a second medical image () of an anatomical object of interest (), wherein each of the first and the second medical images () comprises a different field of view of the anatomical object of interest (), and wherein the first medical image and the second medical image () show a same or similar anatomical state of the anatomical object of interest (); &#x2014;a registration unit () that is configured to determine a transformation from an image space of the second medical image () to an image space of the first medical image (); &#x2014;a transformation unit () that is configured to transform the second medical image () into the image space of the first medical image () based on said transformation in order to receive a transformed second medical image ('); and &#x2014;a segmentation unit () that is configured to perform an overall segmentation that makes use of both the first medical image () and the transformed second medical image (&#x2032;) without fusing the first medical image () and the transformed second medical image (&#x2032;), wherein one and the same segmentation model () is simultaneously adapted to both the first medical image () and the transformed second medical image (&#x2032;) by identifying a first set of feature points () of the anatomical object of interest () within the first medical image (), by identifying a second set of feature points () of the anatomical object of interest () within the transformed second medical image (&#x2032;), and by adapting the segmentation model () to both the first and the second set of feature points ().

Medical image processing device and method

The present invention relates to an apparatus for adaptive contouring of a body part. It is described to provide () at least one image; wherein, the at least one image comprises a first image comprising image data of a body part. An initial automatic model based segmentation of image data of the body part in the first image is determined (). Final segmentation data of the body part is determined () in response to a modification of the initial automatic model based segmentation. An updated model based segmentation can be applied () on the basis of the initial automatic model based segmentation and the final segmentation data.

Growing community of inventors

Hamburg, Germany

Christian Buerger