Main Menu

This technology includes a novel method to train deep learning convolution neural network model to improve the signal-noise-ratio for the magnetic resonance (MR) imaging. The novelty lies on the fact that actual MR imaging physics information is used in the deep learning training. The resulting model achieves significant signal-to-noise ratio (SNR) improved for different acceleration factors in MR imaging. The resulting model can be used for many body anatomies (e.g., brain, heart, liver, spine, etc.) to significantly improve the SNR.

Nuclear medicine is the second largest source of medical radiation exposure to the general population after computed tomography imaging. Imaging modalities utilizing nuclear medicine produce a more detailed view of internal structure and function and are most commonly used to diagnose diseases such as heart disease, Alzheimer’s and brain disorders. They are used to visualize tumors, abscesses due to infection or abnormalities in abdominal organs.

About half of the per capita dose of radiation due to medical exposures is provided by computed tomography (CT) examinations. Approximately 80 million CTs are performed annually in the United States. CT scans most commonly look for internal bleeding or clots, abscesses due to infection, tumors and internal structures. Although CT provides great patient benefit, concerns exist about potential associated risks from radiation doses – especially in pediatric patients more sensitive to radiation.

The National Cancer Institute (NCI) seeks parties to license software for modeling the targeted delivery of anti-cancer agents in solid tumors.

The software models the permeability and concentration of intravenously administered antibody anti-cancer agent conjugates in solid tumors.  The models can be used to determine optimal dosing regimen of a therapeutic in a particular cancer type.  Thirty factors that affect delivery rates and efficiencies are analyzed as variables in generating the models.

Cancer diagnosis is based on the assessment of patient biopsies to determine the tumor type, grade, and stage of malignancy. The proliferative potential of tumors correlates to their growth and metastasis. Visually identifying and quantifying mitotic figures (MF) in cancer biopsy tissue can be used as a surrogate for proliferative activity in tumors.

Cancer diagnosis depends on the assessment of patient biopsies to determine tumor type, grading, and stage of malignancy. Pathologists visually review specimens and count mitotic figures (MF) in a variety of cancer types to help gauge aggressiveness, guide treatment, and inform patient prognosis. Current technology for recording MF counts in surgical pathology is lacking in objectivity, and enumeration of MF by microscopy can be error prone. In particular, a lack of systematic means for recording contributes to recognized variability.

This technology includes a method for performing cardiac imaging without the need for the patient to hold their breath. Free breathing pixel-wise myocardial T1 parameter mapping includes performing a free-breathing scan of a cardiac region at a plurality of varying saturation recovery times to acquire a k-space dataset; generating an image dataset based on the k-space dataset; and performing a respiratory motion correction process on the image dataset.

This technology includes a fully automatic 3D image processing system to segment the heart as well as other organs from contrast-enhanced cardiac computed tomography (CCT) images. Our method detects four cardiac chambers including left ventricle, right ventricle, left atrium, right atrium, as well as the ascending aorta and left ventricular myocardium. It also classifies noncardiac tissue structures in the CCT images such as lung, chest wall, spine, descending aorta, and liver.

This technology includes a computer-implemented method for determining magnetic field inversion time of a tissue species using a T1-mapping image, information about the region of interest, and a tissue classification algorithm. This method includes T1-mapping image comprising a plurality of T1 values within an expected range of T1 values for the tissue of interest. An image mask is created based on predetermined identification information about the tissue of interest. Next, an updated image mask is created based on a largest connected region in the image mask.

This technology includes a method to reduce scanning time while retaining high image quality during MRI scans. A reconstructed image is rendered from a set of MRI data by first estimating an image with an area which does not contain artifacts or has an artifact with a relatively small magnitude. Corresponding data elements in the estimated image and a trial image are processed, for instance by multiplication, to generate an intermediate data set.