Systems, methods, and computer readable media for using descriptors to identify when a subject is likely to have a dysmorphic feature
Inventors
Gelbman, Dekel • Gurovich, Yaron
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9579068-B2
Publication Date
2017-02-28
Expiration Date
Abstract
Systems, methods, and computer-readable media are disclosed for identifying when a subject is likely to be affected by a medical condition. For example, at least one processor may be configured to receive information reflective of an external soft tissue image of the subject. The processor may also be configured to perform an evaluation of the external soft tissue image information and to generate evaluation result information based, at least in part, on the evaluation. The processor may also be configured to predict a likelihood that the subject is affected by the medical condition based, at least in part, on the evaluation result information.
Core Innovation
The invention is an electronic system and a computer-implemented method for determining, from a series of pixels in time-lapse images of external cranio-facial soft tissue, whether a subject is likely to be affected by a genetic disorder. The system receives first pixel information reflective of a first external soft tissue image of the subject recorded at a first time and second pixel information reflective of a second external soft tissue image recorded at a second time, and each pixel information is derived from a relationship between pixels in a group of pixels.
The system analyzes the relationship between the pixels in each group of pixels, determines a time lapse between the original capture of the first external soft tissue image and the original capture of the second external soft tissue image, and compares the analysis of the first pixel information with the analysis of the second pixel information based on the determined time lapse. The system then predicts, based on the time lapse and the comparison, a likelihood that the subject is affected by the genetic disorder.
The described approach further supports obtaining the time lapse from image metadata, including geolocation data and a timestamp, or by using an age detection algorithm to estimate the subject’s age in the first and second external soft tissue images. The pixel-relationship analysis includes, in examples, anchored cells analysis, shifting patches analysis, and relative measurements analysis, and the implementation can compare multiple timepoints and use the resulting probability and severity changes in a time-lapse context.
Claims Coverage
The document includes three independent claims: an electronic system, a computer-implemented method, and a non-transitory computer-readable medium. Each independent claim centers on analyzing pixel relationships in time-lapse external cranio-facial soft tissue images, determining a time lapse between captures, comparing the analyses across time, and predicting a likelihood of being affected by a genetic disorder.
Time-lapse pixel relationships for genetic disorder likelihood prediction
Receive first pixel information reflective of a first external soft tissue image of the subject recorded at a first time, analyze a relationship between the pixels in a first group of pixels, receive second pixel information reflective of a second external soft tissue image recorded at a second time, analyze a relationship between the pixels in a second group of pixels, determine a time lapse between original captures, compare the analyses based on the determined time lapse, and predict a likelihood that the subject is affected by the genetic disorder.
Processing circuitry for time-lapse pixel relationship analysis and comparison
Receive, with processing circuitry, first and second pixel information from external soft tissue images recorded at different times, analyze the pixel relationships, determine a time lapse between original captures, compare the analyses based on the determined time lapse, and predict a likelihood that the subject is affected by the genetic disorder.
Non-transitory medium instructions for time-lapse prediction from pixel relationships
Instructions executed by at least one processor cause receiving first and second pixel information from external cranio-facial soft tissue images at different times, analyzing pixel relationships, determining a time lapse between original captures, comparing the analyses based on the determined time lapse, and predicting a likelihood that the subject is affected by the genetic disorder.
Across the independent claims, the core coverage is the same: derive pixel-group relationship information from two external cranio-facial soft tissue images at different times, analyze each pixel-group relationship, determine a time lapse between the captures, compare the analyses based on the time lapse, and predict a likelihood of being affected by a genetic disorder.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?