Agfa Healthcare Nv

Vrvis Zentrum Fur Virtual Reality Und Visualisierung Forschungs-gmbh

Agfa Healthcare

Vrvis Zentrum Fur Virtual Reality Und Visualisierung

Imp Forschunginstitut Für Molekulare Pathologie Gmbh

Vrvis Forschungs Gmbh

Vrvis Zentrum Für Virtual Reality Und Visualisierung Forschungs-gmb

Imp Forschungsinstitut Fur Molekulare Pathologie Gmbh

Ipm Forschunginstitut Für Molekulare Pathologie Gmbh

David Major

Katja Buehler

Maria Wimmer

Alexey Novikov

Dimitrios Lenis

Jiri Hladuvka

Katja Bühler

Astrid Berg

Gert Sluiter

Florian Schulze

A method for finding image regions that significantly influence classification in a tool for pathology classification in a medical image wherein image regions that influence classification are inpainted, by replacing the pixels in these regions by representations of heathy tissue, resulting in an image with a healthy classification-score in the classification tool.

Method for finding image regions that significantly influence classification in a tool for pathology classification in a medical image

Method of segmenting anatomical structures such as organs in 3D scans in an architecture that combines U-net, time-distributed convolutions and bidirectional convolutional LSTM.

Sequential segmentation of anatomical structures in 3D scans

A method for training an artificial neural network on an additional untrained segmentation task prevents the loss of previously acquired segmentation skills on originally trained segmentation tasks.

Avoiding catastrophic interference while training an artificial neural network on an additional task

A pipe-line method for multi-label segmentation of anatomic structures in a medical image using a convolutional neural network trained with a weighted loss function takes into account under&#x2014;representation of at least one anatomical structure in a ground-truth mask relative to other anatomical structures. Different architectures for the convolutional neural network are described.

Multi-class image segmentation method

A method and system for automatically detecting systemic arteries in arbitrary field-of-view computed tomography angiography (CTA) includes fully-automatically analyzing a medical image represented by a digital image representation.

Method for automatically detecting systemic arteries in arbitrary field-of-view computed tomography angiography (CTA)

A method, and an according apparatus and system, for analyzing medical images of blood vessels includes the steps of a) classifying a surrounding of at least one vessel represented in at least one medical image by applying a first classifier to the medical image such that the surrounding of the vessel is assigned to one of at least two surrounding classes, and b) segmentation of the at least one vessel dependent on the surrounding class to which the surrounding of the vessel has been assigned. The invention allows for a reliable segmentation and/or shape detection, in particular bifurcation detection, of blood vessels represented in medical images.

Method, apparatus and system for analyzing medical images of blood vessels

A method, apparatus, and system for reliably identifying a specific part of a spine in an image of a human or animal body, includes the steps of determining one or more parts of the spine in the image, determining one or more discriminative parameters for each of the one or more parts of the spine in the image, the discriminative parameters relating to at least one anatomical property of each of the one or more parts of the spine, classifying the discriminative parameters of the one or more parts of the spine in the image, and identifying a specific part of the spine based on the classification of the discriminative parameters of the one or more parts of the spine in the image. An identified vertebra, in particular the T12 vertebra and/or its associated intervertebral discs, can be used advantageously as a starting point of powerful automatic spine labeling algorithms.

Method, apparatus and system for identifying a specific part of a spine in an image

A method, an apparatus, and a system label one or more parts of a spine in an image, in particular a computed tomography (CT) image, of a human or animal body, and in order to achieve a reliable spine labeling and a high throughput of images, match a model of a spine segment with segments of the spine in the image by starting matching the model of a spine segment with an initial segment of the spine in the image, wherein the initial segment of the spine in the image is located at an initial position along the spine in the image, and continue to match the model of a spine segment with one or more further segments of the spine in the image, wherein the further segments of the spine in the image are located at positions farther along the spine in the image.

Method, apparatus and system for automated spine labeling

A method and a corresponding apparatus and system localizes a spine in an image, in particular a computed tomography (CT) image, of a human or animal body, allowing for a reduced need for computational power and/or memory on the one hand and assuring a reliable localization of the spine on the other hand. The method includes a) acquiring a plurality of slice images of at least a part of a human or animal body, and b) automatically selecting slice images and/or parts of slice images from the acquired plurality of slice images by considering at least one parameter (&#x3bc;, &#x3bd;, &#x3c3;, &#x39b;) characterizing a distribution of bones in the acquired slice images, wherein the selected slice images and/or parts of the slice images includes image information about the spine.

Growing community of inventors

Mortsel, Belgium

David Major