Siemens Aktiengesellschaft

Siemens Healthcare Gmbh

Siemens Medical Solutions Usa, Inc.

Siemens Corporation

Peng Wang

Dorin Comaniciu

Terrence Chen

Martin Ostermeier

Simone Prummer

Shaohua Kevin Zhou

Yefeng Zheng

Matthias John

Ying Zhu

Guillaume Dumont

Jan Boese

Ali Kamen

Shanhui Sun

Marcus Pfister

Wei Zhang

Gareth Funka-Lea

Norbert Strobel

Michael Kelm

Wen Wu

Michael Wels

Andreas Meyer

Peter Durlak

Olivier Ecabert

Sara Good

Ulrich Bill

Jeffrey Stoll

Renbin Peng

Sheng Yi

Wei-shing Liao

Jingfei Jia

Stratis Tzoumas

An overlay of anatomy is created for a fluoroscopy image. Anatomy detected in angiographic images is used to locate anatomy in the fluoroscopy image. For each cardiac phase, the anatomy is detected multiple times from different cardiac cycles using angiographic images. The fluoroscopic overlay for each cardiac phase is formed from a combination of angiographic candidates fit from the different cardiac cycles to the fluoroscopic image. To further enhance the combination, prediction of the anatomy position from one phase to the next phase is used.

Dynamic overlay of anatomy from angiography to fluoroscopy

A method for detecting polyps in endoscopy images includes pruning a plurality of two dimensional digitized images received from an endoscopy apparatus to remove images that are unlikely to depict a polyp, where a plurality of candidate images remains that are likely to depict a polyp, pruning non-polyp pixels that are unlikely to be part of a polyp depiction from the candidate images, detecting polyp candidates in the pruned candidate images, extracting features from the polyp candidates, and performing a regression on the extracted features to determine whether the polyp candidate is likely to be an actual polyp.

Accurate and efficient polyp detection in wireless capsule endoscopy images

A periodic time-varying model of a guide wire, other device, or vessel is created from higher contrast CINE acquisition. A multivariate curve regression is used for periodic, phase dependent motion modeling. This model is then used to locate or track the guide wire, other device or vessel in lower contrast fluoroscopy images during intervention.

Regression for periodic phase-dependent modeling in angiography

A needle is enhanced in a medical diagnostic ultrasound image. The image intensities associated with a needle in an image are adaptively increased and/or enhanced by compounding from a plurality of ultrasound images. Filtering methods and probabilistic methods are used to locate possible needle locations. In one approach, possible needles are found in component frames that are acquired at the same time but at different beam orientations. The possible needles are associated with each other across the component frames and false detections are removed based on the associations. In one embodiment of needle detection in an ultrasound component frame, lines are found first. The lines are then searched to find possible needle segments. In another embodiment, data from different times may be used to find needle motion and differences from a reference, providing the features in additional to features from a single component frame for needle detection.

Needle enhancement in diagnostic ultrasound imaging

A method and system for extracting coronary vessels fluoroscopic image sequences using coronary digital subtraction angiography (DSA) are disclosed. A set of mask images of a coronary region is received, and a sequence of contrast images for the coronary region is received. For each contrast image, a motion estimate is calculated between each of the mask images and a background region of the contrast image and a covariance is calculated for each motion estimate. Multiple background layer predictions are generated by generating a background layer prediction for each mask image based on the calculated motion estimate and covariance. The multiple background layer estimates are combined using statistical fusion to generate a final estimated background layer. The final estimated background layer is subtracted from the contrast image to extract a coronary vessel layer for the contrast image.

System and method for coronary digital subtraction angiography

A method and system for device detection in a 2D medical image. In order to account for shape variation of a 3D object in a 2D imaging plane, a hierarchical tree-structured array of trained classifiers is used to detect a 3D object, such as a pigtail catheter in a 2D medical image, such as a fluoroscopic image. The hierarchical tree-structured array of trained classifiers increases a dimensionality of the search space with each hierarchical level, and as the search space is increased, the classification is split into object sub-classes using trained classifiers independently trained for each sub-class.

Method and system for device detection in 2D medical images

A method and system for detecting a curve structure in a 2D fluoroscopic image is disclosed. A plurality of curve segments are detected in the image. A graph is generated based on the detected curve segments. A curve structure is detected in the image by determining a path between a source node and a destination node in the graph. A hyper-graph can be constructed from the graph in order to impose geometric constraints on segments of the detected the curve structure, and the curve structure can be detected by finding a shortest path in the hyper-graph.

Method and system for graph based interactive detection of curve structures in 2D fluoroscopy

A method and system for adaptive discriminant learning and measurement fusion for image based catheter tracking is disclosed. An adaptive discriminant model is trained online based on a tracked object, such as a pigtail catheter tip, in at least one previous frame of a fluoroscopic image sequence. The object is tracked in the current frame of the fluoroscopic image sequence based at least on the adaptive discriminant model trained online. The object may be tracked in the current frame based on a fusion of three types of measurement models including the adaptive discriminant model trained online, an object detection model trained offline, and an online appearance model.

Method and system for catheter tracking in fluoroscopic images using adaptive discriminant learning and measurement fusion

A method and system for detecting and tracking coronary sinus (CS) catheter electrodes in a fluoroscopic image sequence is disclosed. An electrode model is initialized in a first frame of the fluoroscopic image sequence based on input locations of CS sinus catheter electrodes in the first frame. The electrode model is tracked in subsequent frames of the fluoroscopic image sequence by detecting electrode position candidates in the subsequent frames of the fluoroscopic image sequence using at least one trained electrode detector, generating electrode model candidates in the subsequent frames based on the detected electrode position candidates, calculating a probability score for each of the electrode model candidates, and selecting an electrode model candidate based on the probability score.

Method and system for detection and tracking of coronary sinus catheter electrodes in fluoroscopic images

Multiple candidates are optimized in medical device or feature tracking. Possible locations of medical devices or features for each of a plurality of different times are received. The possible locations of devices are modeled using a probability function. An iterative solution to obtain the maximum of the probability function determines the possible locations to be used as the locations of the medical devices or features for each time. Where two or more medical devices or features are provided with a geometric relationship, such as being connected by a detected guide wire, the probability function may account for the geometric relationship, such as a geodesic distance between the possible locations for the two medical devices.

Optimization of multiple candidates in medical device or feature tracking

Methods and Systems for training a learning based classifier and object detection in medical images is disclosed. In order to train a learning based classifier, positive training samples and negative training samples are generated based on annotated training images. Features for the positive training samples and the negative training samples are extracted. The features include an extended Haar feature set including tip features and corner features. A discriminative classifier is trained based on the extracted features.

Method and system for learning based object detection in medical images

A method and system for tracking a needle in a fluoroscopic image sequence is disclosed. In order to track a needle in a fluoroscopic image sequence, the needle is initialized in a first frame of the fluoroscopic image sequence. Needle segments are detected in each subsequent frame of the fluoroscopic image sequence, and the needle is detected in each frame of the fluoroscopic image by tracking the needle from a previous frame of the fluoroscopic image sequence based on the detected needle segments in the current frame.

Method and system for needle tracking in fluoroscopic image sequences

A method and system for co-registration of angiography data and intra vascular ultrasound (IVUS) data is disclosed. A vessel branch is detected in an angiogram image. A sequence of IVUS images is received from an IVUS transducer while the IVUS transducer is being pulled back through the vessel branch. A fluoroscopic image sequence is received while the IVUS transducer is being pulled back through the vessel branch. The IVUS transducer and a guiding catheter tip are detected in each frame of the fluoroscopic image sequence. The IVUS transducer detected in each frame of the fluoroscopic image sequence is mapped to a respective location in the detected vessel branch of the angiogram image. Each of the IVUS images is registered to a respective location in the detected vessel branch of the angiogram image based on the mapped location of the IVUS transducer detected in a corresponding frame of the fluoroscopic image sequence.

Method and system for image based device tracking for co-registration of angiography and intravascular ultrasound images

A method and system for detecting a guiding catheter in a 2D fluoroscopic image is disclosed. A plurality of guiding catheter centerline segment candidates are detected in the fluoroscopic image. A guiding catheter centerline connecting an input guiding catheter centerline ending point in the fluoroscopic image with an image margin of the fluoroscopic image is detected based on the plurality of guiding catheter centerline segment candidates.

Method and system for guiding catheter detection in fluoroscopic images

A method and system for tracking a guidewire in a fluoroscopic image sequence is disclosed. In order to track a guidewire in a fluoroscopic image sequence, guidewire segments are detected in each frame of the fluoroscopic image sequence. The guidewire in each frame of the fluoroscopic image sequence is then detected by rigidly tracking the guidewire from a previous frame of the fluoroscopic image sequence based on the detected guidewire segments in the current frame. The guidewire is then non-rigidly deformed in each frame based on the guidewire position in the previous frame.

Growing community of inventors

Princeton, NJ, United States of America

Peng Wang