Siemens Healthcare Gmbh

Siemens Aktiengesellschaft

The United States of America, As Represented by the Secretary, Department of Health and Human Services

Nathan S Lay

Shaohua Kevin Zhou

David D Liu

Bernhard Geiger

Jan Kretschmer

Li Zhang

Daguang Xu

Noha Youssry El-Zehiry

Chris Schwemmer

Vincent Ordy

Philipp Wolber

Dorin Comaniciu

Bogdan Georgescu

Yefeng Zheng

Vivek Kumar Singh

Daphne Yu

Klaus Engel

Jingdan Zhang

Ronald M Summers

Peter Choyke

Jens Guehring

Neil Birkbeck

Hien Nguyen

Dong Yang

Baris Turkbey

Matthew J McAuliffe

Sonia Gaur

Francesca Mertan

Yohannes Tsehay

Matthew Greer

Ruida Cheng

Holger Roth

Disclosed prostate computer aided diagnosis (CAD) systems employ a Random Forest classifier to detect prostate cancer. System classify individual pixels inside the prostate as potential sites of cancer using a combination of spatial, intensity and texture features extracted from three sequences. The Random Forest training considers instance-level weighting for equal treatment of small and large cancerous lesions and small and large prostate backgrounds. Two other approaches are based on an AutoContext pipeline intended to make better use of sequence-specific patterns. Also disclosed are methods and systems for accurate automatic segmentation of the prostate in MRI. Methods can include both patch-based and holistic (image-to-image) deep learning methods for segmentation of the prostate. A patch-based convolutional network aims to refine the prostate contour given an initialization. A method for end- to-end prostate segmentation integrates holistically nested edge detection with fully convolutional networks. HNNs automatically learn a hierarchical representation that improve prostate boundary detection.

Detection of prostate cancer in multi-parametric MRI using random forest with instance weighting and MR prostate segmentation by deep learning with holistically-nested networks

A method and apparatus for whole body bone removal and vasculature visualization in medical image data, such as computed tomography angiography (CTA) scans, is disclosed. Bone structures are segmented in the a 3D medical image, resulting in a bone mask of the 3D medical image. Vessel structures are segmented in the 3D medical image, resulting in a vessel mask of the 3D medical image. The bone mask and the vessel mask are refined by fusing information from the bone mask and the vessel mask. Bone voxels are removed from the 3D medical image using the refined bone mask, in order to generate a visualization of the vessel structures in the 3D medical image.

Method and system for whole body bone removal and vascular visualization in medical image data

The present embodiments relate to cinematic volume renderings and volumetric Monte-Carlo path tracing. The present embodiments include systems and methods for integrating semantic information into cinematic volume renderings. Scan data of a volume is captured by a scanner and transmitted to a server or workstation for rendering. The scan data is received by a server or workstation. The server or workstation extracts semantic information and/or applies semantic processing to the scan data. A cinematic volume rendering is generated from the scan data and the extracted semantic information.

Semantic cinematic volume rendering

A method and system for approximating a deep neural network for anatomical object detection is discloses. A deep neural network is trained to detect an anatomical object in medical images. An approximation of the trained deep neural network is calculated that reduces the computational complexity of the trained deep neural network. The anatomical object is detected in an input medical image of a patient using the approximation of the trained deep neural network.

Method and system for approximating deep neural networks for anatomical object detection

Systems and methods for segmenting a structure of interest in medical imaging data include generating a binary mask highlighting structures in medical imaging data, the highlighted structures comprising a connected component including a structure of interest. A probability map is computed by classifying voxels in the highlighted structures using a trained classifier. A plurality of detaching operations is performed on the highlighted structures to split the connected component into a plurality of detached connected components. An optimal detaching parameter is determined representing a number of the detaching operations. A detached connected component resulting from performing the number of detaching operations corresponding to the optimal detaching parameter is classified as the structure of interest based on the probability map and the trained classifier.

Learning-based aorta segmentation using an adaptive detach and merge algorithm

Systems and methods for training a region clustering forest include receiving a set of medical training images for a population of patients. A set of image patches is extracted from each image in the set of medical training images. A plurality of region clustering trees are generated each minimizing a loss function based on respective randomly selected subsets of the set of image patches to train the region clustering forest. Each of the plurality of region clustering trees cluster image patches at a plurality of leaf nodes and the loss function measures a compactness of the cluster of image patches at each leaf node in each of the plurality of region clustering trees.

Region clustering forest for analyzing medical imaging data

A visualization method is provided that allows for the unfolding of a human skeleton from a medical image scan and providing increased efficiency for interacting with the image scan and whole body bone reading from such scans. That is, a full head-to-toe unfolded skeleton view (e.g., a 2D unfolded view) is realized for improved visualization and diagnostic capabilities.

Visualization method for a human skeleton from a medical scan

A method and system for segmenting multiple organs in medical image data is disclosed. A plurality of landmarks of a plurality of organs are detected in a medical image using an integrated local and global context detector. A global posterior integrates evidence of a plurality of image patches to generate location predictions for the landmarks. For each landmark, a trained discriminative classifier for that landmark evaluates the location predictions for that landmark based on local context. A segmentation of each of the plurality of organs is then generated based on the detected landmarks.

Method and system for joint multi-organ segmentation in medical image data using local and global context

Growing community of inventors

Bethesda, MD, United States of America

Nathan S Lay