Medical image data
AssigneesSiemens Healthcare GmbH, New York University
InventorsJin-hyeong Park, Sasa Grbic, Matthias Fenchel, Esther Raithel, Dana Lin
Life of this patent
Estimating the term…
Abstract
Medical image data is received at a data processing system, which is an artificial intelligence-based system. An identification process is performed at the data processing system to identify a subset of the medical image data representing a region of interest including one or more target tendons. A determination process is performed at the data processing system to determine one or more characteristics relating to one or more abnormalities of the one or more target tendons. Abnormality data is output, the abnormality data relating to the one or more abnormalities and being based on the one or more characteristics.
Claims15
A method for processing medical image data, the method comprising:
- receiving medical image data at a data processing system, the data processing system being an artificial intelligence-based system;
- identifying, by the data processing system, a region of interest including one or more target tendons;
- cropping, by the data processing system, the medical image data to the region of interest;
- determining, by the data processing system, from the cropped medical image data one or more characteristics relating to one or more abnormalities of the one or more target tendons;
- and
- outputting abnormality data relating to the one or more abnormalities, the abnormality data being based on the one or more characteristics,
- wherein the data processing system is a neural network system comprising a neural network for performing the identifying and the determining, the neural network having been trained to perform both the identifying and the determining using a single training process.
A non-transitory computer readable storage medium storing one or more neural networks trained to determine one or more characteristics relating to one or more abnormalities relating to one or more target tendons in medical image data and storing instructions that, when executed by a processor, cause the processor to:
- receive medical image data;
- provide the medical image data to the one or more neural networks to perform:
- an identification process to identify a region of interest including one or more target tendons;
- a cropping process to crop the medical image data to the region of interest;
- a determination process comprising processing the cropped medical image data to determine one or more characteristics relating to one or more abnormalities of the one or more target tendons;
- and
- output abnormality data relating to the one or more abnormalities, the data being based on the one or more characteristics,
- wherein the one or more neural networks comprise a neural network for performing the identifying process and the determining process, the neural network having been trained to perform both the identifying and the determining using a single training process.
An apparatus for processing medical image data, the apparatus comprising:
- one or more processors;
- and
- a memory storing:
- one or more neural networks trained to determine one or more characteristics relating to one or more abnormalities relating to one or more target tendons in medical image data;
- and
- instructions that, when executed by the one or more processors, cause the one or more processors to:
- receive medical image data;
- provide the medical image data to the one or more neural networks to:
- identify a region of interest in the medical image data including one or more target tendons;
- crop the medical image data to the region of interest;
- determine, from the cropped medical image data, one or more characteristics relating to one or more abnormalities of the one or more target tendons;
- and
- output abnormality data relating to the one or more abnormalities, the data being based on the one or more characteristics,
- wherein the one or more neural networks comprise a neural network configured to identify the subset of the medical image data and to determine the one or more characteristics, the neural network having been trained both to identify and to determine using a single training process.
Dependencies are inferred from the claim text.