General Electric Company

Sheshadri Thiruvenkadam

Dattesh Dayanand Shanbhag

Rakesh Mullick

Florian Wiesinger

Rahul Venkataramani

Vivek Prabhakar Vaidya

Sandeep Suryanarayana Kaushik

Prasad Sudhakar

Ravindra Mohan Manjeshwar

Krishna Seetharam Shriram

Hariharan Ravishankar

Kris Filip Johan Jules Thielemans

Albert Henry Roger Lonn

Srikrishnan Viswanathan

Christopher Judson Hardy

Thomas Kwok-Fah Foo

Scott David Wollenweber

Kevin Matthew Koch

Dirk Ryan Padfield

Chandan Kumar Mallappa Aladahalli

Sudhanya Chatterjee

Sohan Rashmi Ranjan

Mithun Das Gupta

Hua Qian

Navneeth Subramanian

Sumit Nath

Navneeth Subamanian

Anand Magadi Narasimhamurthy

Priyavrat Jhunjhunwala

Harihan Ravishankar

A method for image segmentation includes receiving an input image. The method further includes obtaining a deep learning model having a triad of predictors. Furthermore, the method includes processing the input image by a shape model in the triad of predictors to generate a segmented shape image. Moreover, the method includes presenting the segmented shape image via a display unit.

System and method for image segmentation using a joint deep learning model

A method for determining optimized deep learning architecture includes receiving a plurality of training images and a plurality of real time images corresponding to a subject. The method further includes receiving, by a medical practitioner, a plurality of learning parameters comprising a plurality of filter classes and a plurality of architecture parameters. The method also includes determining a deep learning model based on the plurality of learning parameters and the plurality of training images, wherein the deep learning model comprises a plurality of reusable filters. The method further includes determining a health condition of the subject based on the plurality of real time images and the deep learning model. The method also includes providing the health condition of the subject to the medical practitioner.

System and method for optimization of deep learning architecture

A method for image segmentation includes receiving an input image (). The method further includes obtaining a deep learning model () having a triad of predictors (). Furthermore, the method includes processing the input image by a shape model in the triad of predictors () to generate a segmented shape image (). Moreover, the method includes presenting the segmented shape image via a display unit ().

An imaging system and method are disclosed. An MR image and measured B0 field map of a target volume in a subject are reconstructed, where the MR image includes one or more bright and/or dark regions. One or more distinctive constituent materials corresponding to the bright regions are identified. Each dark region is iteratively labeled as one or more ambiguous constituent materials. Susceptibility values corresponding to each distinctive and iteratively labeled ambiguous constituent material is assigned. A simulated B0 field map is iteratively generated based on the assigned susceptibility values. A similarity metric is determined between the measured and simulated B0 field maps. Constituent materials are identified in the dark regions based on the similarity metric to ascertain corresponding susceptibility values. The MRI data is corrected based on the assigned and ascertained susceptibility values. A diagnostic assessment of the target volume is determined based on the corrected MRI data.

Method and system for improved classification of constituent materials

A method for synchronization of a longitudinal data set from a subject includes receiving a first ensemble registration estimate having a first reference image corresponding to a first image ensemble and receiving a second image ensemble different from the first image ensemble. The method includes determining a second reference image based on the second image ensemble and the first reference image. Further, the method includes determining a second ensemble registration estimate based on the first ensemble registration estimate, the second reference image, the first image ensemble and the second image ensemble using an optimization technique. The method further includes generating a synchronized image ensemble corresponding to the first image ensemble and the second image ensemble based on the second ensemble registration estimate. The method also includes determining a medical condition of the subject by a medical practitioner based on the synchronized image ensemble.

System and method for longitudinal data processing

A system and method for estimating image intensity bias and segmentation tissues is presented. The system and method includes obtaining a first image data set and at least a second image data set, wherein the first and second image data sets are representative of an anatomical region in a subject of interest. Furthermore, the system and method includes generating a baseline bias map by processing the first image data set. The system and method also includes determining a baseline body mask by processing the second image data set. In addition, the system and method includes estimating a bias map corresponding to a sub-region in the anatomical region based on the baseline body mask. Moreover, the system and method includes segmenting one or more tissues in the anatomical region based on the bias map.

System and method for image intensity bias estimation and tissue segmentation

A method implemented using at least one processor includes receiving time-varying image dataset generated by a medical imaging modality. The image dataset corresponds to a bed position and is affected by quasi-periodic motion data. The method also includes applying a signal decomposition technique to the time-varying image dataset to generate a plurality of dataset components and a plurality of motion signals. The method also includes determining reference data based on the time-varying image dataset, wherein the reference data is representative of a direction of the quasi-periodic motion. The method further includes deriving polarity of each of the plurality of motion signals based on the reference data to generate a plurality of sign corrected motion signals. The method also includes determining a gating signal corresponding to the bed position based on at least one of the plurality of sign corrected motion signals.

System and method for data driven gating of multiple bed positions

In one embodiment, a method includes performing a magnetic resonance (MR) imaging sequence to acquire MR image slices or volumes of a first station representative of a portion of a patient; applying a first phase field algorithm to the first station to determine a body contour of the patient in the first station; identifying a contour of a first anatomy of interest within the body contour of the first station using the first phase field algorithm or a second phase field algorithm; segmenting the first anatomy of interest based on the identified contour of the first anatomy of interest; correlating first attenuation information to the segmented first anatomy of interest; and modifying a positron emission tomography (PET) image based at least on the first correlated attenuation information.

Attenuation correction in positron emission tomography using magnetic resonance imaging

Methods and systems for segmentation in echocardiography are provided. One method includes obtaining echocardiographic images and defining a search space within the echocardiographic images using a pair of one-dimensional (1D) profiles. The method also includes using an energy based function constrained by non-local temporal priors within the defined search space to automatically segment a contour of a cardiac structure with the 1D profiles.

Methods and systems for segmentation in echocardiography

Exemplary embodiments of the present disclosure are directed to correcting lung density variations in positron emission tomography (PET) images of a subject using a magnetic resonance (MR) image. A pulmonary vasculature and an outer extent of a lung cavity can be identified in a MR image corresponding to a thoracic region of the subject in response to an intensity associated with pixels in the MR image. The pixels within the outer extent of the lung cavity are classified as corresponding to the pulmonary vasculature or the lung tissue. Exemplary embodiments of the present disclosure can apply attenuation coefficients to a reconstruction of the PET image based on the classification of the pixels within the outer extent of the lung cavity.

Method and system for correction of lung density variation in positron emission tomography using magnetic resonance imaging

A method for generating a positron emission tomography (PET) attenuation correction map. The method includes obtaining a magnetic resonance (MR) image dataset of a subject of interest, obtaining a positron emission tomography (PET) emission dataset of the subject of interest, segmenting the MR image dataset to identify at least one object of interest, determining a volume of the object of interest, and generating a PET attenuation correction map using the determined volume. A medical imaging system and a non-transitory computer readable medium are also described herein.

Methods and systems for generating a positron emission tomography attenuation correction map

A method for automatically measuring interventricular septum thickness is provided. The method in one embodiment comprises acquiring a series of ultrasound images of a heart, acquiring a septum mask by applying a septum segmentation algorithm on the series of ultrasound images, localizing a mitral valve tip using a valve tip localization algorithm and calculating the thickness of the interventricular septum using an interventricular septum thickness algorithm. The interventricular septum thickness algorithm uses the septum mask and the localized mitral valve tip as inputs.

Methods and apparatus for automated measuring of the interventricular septum thickness

A method for generating an image is provided. The method comprises: acquiring a first set of image data using a first imaging modality; sorting the first set of image data into a plurality of gates to generate a plurality of gated data sets; reconstructing each gated data set to generate a respective gated image for each gated data set; registering the respective gated images to generate a plurality of registered images; and generating a median image from the plurality of registered images, wherein each voxel of the median image is a respective median value of the corresponding voxels of the plurality of registered images.

Method and system for processing gated image data

A method for automatic mismatch correction is presented. The method includes identifying a feature of interest in a reference image volume and a target image volume. Furthermore, the method includes computing a cost matrix based on one or more pairs of image slices in the reference image volume and the target image volume. The method also includes identifying one or more longest common matching regions in the reference image volume and the target image volume based on the computed cost matrix. In addition, the method includes aligning the reference image volume and the target image volume based on the identified one or more longest common matching regions. A non-transitory computer readable medium including one or more tangible media, where the one or more tangible media include code adapted to perform the method for automatic mismatch correction is also presented. Systems and non-transitory computer readable medium configured to perform the method for automatic mismatch correction of image volumes are also presented.

Growing community of inventors

Bengaluru, India

Sheshadri Thiruvenkadam