System and method for detecting anomalies in a tissue imaged in-vivo
AssigneeGiven Imaging Ltd.
InventorsEli Horn, Igor Zingman
Life of this patent
Estimating the term…
Abstract
A system and method for detecting an anomaly in a tissue captured or imaged by an imaging system are provided. A uniform region in an image representing a tissue may be defined by identifying a group of pixels associated with a predefined variance from an imaging parameter value. An adaptive size window may be used to define a sub-region in a uniform region. An anomaly may be detected by identifying, in the sub-region, a set of pixels associated with imaging parameter values indicative of an anomaly. Other embodiments are described and claimed.
Claims23
A method of detecting an anomaly in an image captured in vivo by an in-vivo imaging system, the method comprising:
- segmenting an image captured in vivo into valid tissue zones for anomaly detection, and non-valid tissue zones excluded from anomaly detection, wherein each valid tissue zone is a portion of the image, wherein each pixel in the valid tissue zone has a pixel intensity value that is greater than a predetermined intensity reference value, and wherein each pixel in the valid tissue zone is associated with a respective set of imaging parameter values, the set of imaging parameter values having a deviation from a corresponding reference value no greater than a predefined deviation;
- identifying, in the valid tissue zones, individual pixels whose color characteristics deviate from reference color parameters representing normal tissue, and classifying said pixels as anomaly pixels;
- and
- analyzing each identified valid tissue zone by:
- (i) identifying a group of the individual anomaly pixels in the valid tissue zone;
- (ii) if the group of anomaly pixels in the valid tissue zone satisfies predetermined anomaly criteria, determining that a region of the zone containing the group of anomaly pixels is an anomaly region;
- and
- repeating steps (i) and (ii) for all anomaly pixels identified in the valid tissue zone.
A system for detecting anomalies in an image captured in vivo by an in-vivo imaging system, the system comprising:
- a workstation comprising a memory and a processor, the memory to store an image captured by the in-vivo imaging system, and the processor to:
- segment an image captured in vivo into valid tissue zones for anomaly detection, and non-valid tissue zones excluded from anomaly detection, wherein each valid tissue zone is a portion of the image, wherein each pixel in the valid tissue zone has a pixel intensity value that is greater than a predetermined intensity reference value, and wherein each pixel in the valid tissue zone is associated with a respective set of imaging parameter values, the set of imaging parameter values having a deviation from a corresponding reference value no greater than a predefined deviation;
- identify, in the valid tissue zones, individual pixels whose color characteristics deviate from reference color parameters representing normal tissue, and
- classifying said pixels as anomaly pixels;
- and
- analyze each identified valid tissue zone by:
- (i) identifying a group of the individual anomaly pixels in the valid tissue zone;
- (ii) if the group of anomaly pixels in the valid tissue zone satisfies predetermined anomaly criteria, determining that a region of the zone containing the group of anomaly pixels is an anomaly region;
- and
- repeating steps (i) and (ii) for all anomaly pixels identified in the valid tissue zone.
A system for detecting anomalies in an image captured in vivo by an in-vivo imaging system, the system comprising:
- a workstation comprising a memory and a processor, the memory to store an image captured by the in-vivo imaging system, and the processor to:
- determine, for at least a subset of the pixels in the image, if an associated imaging parameter value is indicative of an anomaly by comparing the associated imaging parameter value to a reference value;
- define a uniform region in the image by identifying a set of pixels associated with a respective set of imaging parameter values, the set of imaging parameter values having a deviation from the reference value no greater than a predefined deviation;
- define a sub-region in the uniform region;
- and
- if a predefined portion of pixels included in the sub-region is associated with an imaging parameter value indicative of an anomaly as determined in the determining operation, then determine an anomaly is present in a tissue segment associated with the sub-region, wherein
- the in-vivo imaging system captures a set of images of the tissue and wherein the reference value is associated with a normal tissue and is determined adaptively based on at least a subset of the captured set of images.
Dependencies are inferred from the claim text.