Method and apparatus for objective assessment of gastrointestinal conditions based on images captured in the GI tract
Inventors
Wang, Kang-Huai • Wu, Chenyu • Wilson, Gordon C.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11948303-B2
Publication Date
2024-04-02
Expiration Date
Abstract
A method and apparatus of objective assessment of images captured from a human gastrointestinal (GI) tract are disclosed. According to this method, one or more images captured using an endoscope when the endoscope is inside the human gastrointestinal (GI) tract are received. Whether there is any specific target object is checked. When one or more specific target objects in the images are detected: areas of the specific target objects in the images are determined; an objective assessment score is derived based on the areas of the specific target objects in a substantial number of images from the images; where the step of detecting the specific target objects is performed using an artificial intelligence process.
Core Innovation
A method is provided for objective assessment of images captured from a human gastrointestinal (GI) tract. The method receives one or more images captured using an endoscope when the endoscope is inside the human GI tract, and detects whether any specific target object is in the one or more images using an artificial intelligence process. When one or more specific target objects are detected, the method determines area sizes of the one or more specific target objects in the one or more images and derives an objective assessment score based on the area sizes.
The method uses travelled distance by the endoscope in the human GI tract to decide a normalization factor on the area sizes or a sum of the area sizes of the one or more specific target objects in the one or more images. In another scoring approach, the area sizes of the one or more specific target objects are scaled according to respective object distances of the one or more specific target objects in the one or more images, and scaled area sizes are used for deriving the objective assessment score.
In such approaches, the objective assessment score is derived from a ratio of a first sum of the area sizes scaled according to respective object distances and a second sum of entire area sizes of the one or more images scaled by distance information within the one or more images respectively. When one specific target object is away from a mucosa of the human GI tract, the respective object distance is derived from a first distance from a camera of the endoscope to the mucosa in a neighborhood area associated with the one specific target object.
The method can also exclude bad images, where the bad images comprise over-exposed images and/or under-exposed images and/or motion-smeared images, from deriving the objective assessment score. Additional embodiments include deriving sectional scores by partitioning the human GI tract and its associated examination images into sections and corresponding image groups to compute scores based on normalized summed area sizes of specific target objects.
Claims Coverage
The patent includes three independent method claims covering AI-based detection of specific target objects in endoscope images, area-size-based objective scoring, and different normalization/scaling strategies using travelled distance and object distance, including distance-to-mucosa in neighborhood regions.
Ai-based objective assessment of GI endoscope images via target-object area sizes
Receiving one or more images captured using an endoscope inside a human gastrointestinal (GI) tract; detecting whether any specific target object is in the one or more images using an artificial intelligence process; when one or more specific target objects are detected, determining area sizes of the one or more specific target objects; and deriving an objective assessment score based on the area sizes.
Distance-normalized objective scoring via object-distance scaling and area-size ratios
Receiving one or more images captured using an endoscope inside a human gastrointestinal (GI) tract; detecting whether any specific target object is in the one or more images using an artificial intelligence process; when one or more specific target objects are detected, determining area sizes; scaling the area sizes according to respective object distances and using scaled area sizes for deriving the objective assessment score; and deriving the objective assessment score from a ratio of a first sum of scaled area sizes and a second sum of entire area sizes scaled by distance information within the one or more images.
Neighborhood distance to mucosa for object-distance scaling of target-object area sizes
Receiving one or more images captured using an endoscope inside a human gastrointestinal (GI) tract; detecting whether any specific target object is in the one or more images using an artificial intelligence process; when one or more specific target objects are detected, determining area sizes; scaling the area sizes according to respective object distances and using scaled area sizes for deriving the objective assessment score; and when one specific target object is away from a mucosa of the human GI tract, deriving the respective object distance from a first distance from a camera of the endoscope to the mucosa in a neighborhood area associated with the one specific target object.
Across the independent claims, the scoring framework is grounded in AI-based detection of specific target objects, determination of their area sizes, and derivation of an objective assessment score. The key variations are normalization using travelled distance, scaling using object distances with a ratio of sums, and derivation of object distance from a distance to mucosa within neighborhood areas.
Stated Advantages
Documented Applications
Objective assessment of gastrointestinal conditions from endoscope image sequences by mapping an objective assessment score to bowel preparation cleanness condition categories.
Objective assessment of Crohn disease condition based on detected target objects including inflamed organ tissues and erosion or ulcers in the distal ileum and colon.
Interested in licensing this patent?