Non-invasive non-contact system and method for evaluating primary and secondary hypertension conditions using thermal imaging
Inventors
Shivpure, Sameer Raghuram • THIRUVENGADAM, Jayanthi • Choda, Anuhya • CHODA, Gayathri
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12186110-B2
Publication Date
2025-01-07
Expiration Date
Abstract
System and method for measuring hypertension conditions of a subject is disclosed. The disclosed system and method includes thermal sensors for capturing thermal images and/or videos of a body part; and a processing engine to detect a predefined region of the body part in each frame of the captured images and/or videos. The processing engine segments one or more portions from the detected predefined region in each frame of the captured images and/or videos to identify a region of interest comprising arteries in the one or more segmented portions. Based on the identified region of interest, the engine extracts pixel values from each frame of the captured images and/or videos to determine parameters associated with a blood flow velocity and a blood pressure of the subject. Further a type of hypertension and a risk score for the hypertension condition based on the determined parameters using computational models are measured.
Core Innovation
The invention relates to a non-contact, non-invasive thermal-imaging system and method for measuring a hypertension condition of a subject. The system includes a set of thermal sensors for capturing one or more thermal images and/or thermal videos of at least one body part, and a processing engine to receive data packets associated with the captured thermal images/videos. On receipt of the data packets, the processing engine detects a predefined region in each frame, segments one or more portions from the detected predefined region, and identifies a region of interest comprising one or more arteries.
After identifying the region of interest, the processing engine segments the identified region of interest from each frame of the thermal images/videos. The processing engine then extracts one or more pixel values representing a set of biosignals from each frame based on the segmented region of interest. Using the extracted pixel values, the processing engine determines one or more parameters associated with blood flow velocity and blood pressure, and where the determined parameters are associated with blood density and a pulse rate.
The processing engine compares the determined one or more parameters with a predetermined set of reference parameters to measure a risk score for the hypertension condition. Based on the measured risk score, a primary hypertension condition and a secondary hypertension condition are determined. The risk score measurement is performed using computational models, including calculating signal parameters for corresponding arterial sections of healthy and hypertensive subjects, identifying parameters that correlate with complications using principal component analysis, determining patterns and differences using statistical methods and visualization, and training computational models with machine learning units including clustering models, logistic regression, random forest, and a neural network model.
Claims Coverage
The partial content provides two independent claims, a system claim and a method claim. Each independent claim includes the same core inventive pipeline of thermal data capture, predefined-region detection, arterial region-of-interest segmentation, pixel-value biosignal extraction, blood-flow and blood-pressure parameter determination, reference comparison to compute a hypertension risk score, and primary-versus-secondary hypertension determination using computational models trained with machine learning units.
Thermal-image hypertension measuring system
A system for measuring a hypertension condition that includes thermal sensors capturing thermal images and/or thermal videos and a processing engine that receives data packets, detects a predefined region in each frame, segments portions, identifies a region of interest comprising one or more arteries, segments the region of interest, extracts pixel values representing biosignals, determines parameters associated with blood flow velocity and blood pressure, compares the parameters with predetermined reference parameters, and measures a risk score, where primary and secondary hypertension conditions are determined based on the measured risk score.
Computational-model based risk score determination
Measurement of the hypertension risk score is performed using computational models whose generation includes calculating signal parameters for corresponding arterial sections of a healthy subject and a hypertensive subject; identifying parameters that correlate to complications associated with primary and secondary hypertension using principal component analysis; determining patterns and differences among the signal parameters between hypertensive and healthy subjects using statistical methods and visualization; and training the computational models using machine learning units comprising clustering models, logistic regression, random forest, and a neural network model on the set of parameters.
Thermal-imaging hypertension measuring method
A method for measuring a hypertension condition that includes capturing thermal images and/or thermal videos using thermal sensors; receiving data packets at a processing engine; detecting a predefined region in each frame; segmenting portions; identifying a region of interest comprising arteries; segmenting the region of interest; extracting pixel values representing biosignals from each frame; determining parameters associated with blood flow velocity and blood pressure; comparing the parameters with predetermined reference parameters; and measuring a risk score, where primary and secondary hypertension conditions are determined based on the measured risk score.
Computational-model based training and hypertension condition determination
The risk score measurement is performed using computational models, where model generation comprises calculating signal parameters for corresponding arterial sections of a healthy subject and a hypertensive subject; identifying parameters correlated to complications associated with primary and secondary hypertension using principal component analysis; determining patterns and differences among signal parameters between hypertensive and healthy subjects using statistical methods and visualization; and training computational models using machine learning units comprising clustering models, logistic regression, random forest, and a neural network model on the set of parameters.
Across the two independent claims, the invention is centered on deriving hypertension risk from thermal-image and video frames by detecting and segmenting a predefined region, isolating a region of interest comprising arteries, extracting pixel-value biosignals, determining blood-flow and blood-pressure-related parameters, comparing them to predetermined reference parameters, and computing a risk score using computational models trained with principal component analysis, statistical methods, visualization, and machine learning units.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?