Back to results
US 11,961,232 B2Granted
Automated sensing system for health detection
AssigneeInternational Electronic Machines Corp.
InventorsZahid F. Mian, Noreen F. Mian
Life of this patent
Estimating the term…
Abstract
A system and method for detecting medical conditions in individuals in crowded settings is described, including methods and approaches for addressing confounding issues such as variation due to external factors.
Claims20
A non-contact, non-intrusive, autonomous system for monitoring health conditions or parameters of persons in an area, said system comprising:
- a plurality of imaging devices including at least two imaging devices sensitive to radiation of distinct wavelength ranges, each with a particular field of view, wherein the fields of view of the at least two imaging devices are substantially co-registered at the area;
- a processor operatively connected to the imaging devices, the processor configured to autonomously monitor persons in the area by:
- receiving image data of the area acquired by the at least two imaging devices;
- identifying, for at least one person at least partially visible within the image data of the area acquired by each of the at least two imaging devices, facial image data corresponding to a face of the person, where the identifying includes extracting from the facial image data regions or elements of the image data that correspond to an area of interest on the face for the at least one person at least partially visible within the image data, where the extracting includes:
- identifying image data corresponding to a face of the at least one person from the image data of each of the at least two imaging devices;
- and
- identifying relevant components of the face in the facial image data of each of the at least two imaging devices;
- examining the identified facial image data for the at least one person acquired by each of the at least two imaging devices to determine at least one key parameter indicative of health, wherein the at least one key parameter indicative of health is determined from the facial image data of at least one of the relevant components of the face;
- and
- evaluating a need to send an alert in response to the determined at least one key parameter.
7 more elements
A non-contact, non-intrusive, autonomous method for determining medical health conditions or parameters of persons in an area, the method comprising at least one processor performing the following:
- receiving sensor data from sensors that cover the area, wherein each sensor is a non-contact, non-intrusive sensor configured to acquire the corresponding sensor data;
- extracting relevant data relating to at least one individual in the area from the sensor data, wherein the relevant data includes facial image data of the at least one individual;
- adjusting the relevant data using at least one compensation process relevant to the relevant data;
- determining a value of at least one health-related parameter for the at least one individual in the area using the adjusted relevant data;
- assessing the value of the at least one health-related parameter to determine if the value of the at least one health-related parameter is outside of threshold limits;
- and
- generating an alert in response to determining the value of the at least one health-related parameter is outside of threshold limits.
4 more elements
Dependencies are inferred from the claim text.