A system and method for dectecting against virus infection
Patent Information
- Application Number
- IN202011020103
- Authority / Receiving Office
- IN · IN
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-05-13
- Publication Date
- 2026-07-13
- Estimated Expiration
- 2040-05-13
AI Technical Summary
Current methods for detecting COVID-19, such as thermal scanners, are limited in accuracy and cannot diagnose the virus effectively, as high temperature is observed in few patients at a late stage, and do not assess other vital parameters essential for identifying virus carriers.
A system utilizing a combination of sensors and machine learning algorithms to remotely sense and analyze physiological parameters like body temperature, heart rate, respiratory rate, oxygen saturation, and hemoglobin levels, along with facial attributes, using AI and IoT technologies to quickly and accurately diagnose potential virus carriers from a safe distance.
The system enables high-accuracy, rapid detection of COVID-19 virus carriers within 40 seconds, helping to contain the spread of the disease by leveraging AI and IoT technologies for comprehensive health assessment and data analytics.
Abstract
Description
The present invention relates to the field of scanning health assessment in the view of viruses such as Severe acute respiratory syndrome (SARS) associate coronavirus or Covid-19. More specifically, a system and method for detecting / scanning the subject against the potential virus infection, with high accuracy and in short time.BACKGROUND OF THE INVENTION:The following background discussion includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.Epidemic Covid-19 has coerced the world to find different means of addressing issues and challenges put forth by this highly virulent disease in different ways of life. The world is forced to use body temperature sensors to scan / detect Covid-19 virus. These body scanners are being used to identify potential patients of Covid-19 who have prevented them from entering crowded establishments such as government offices and even residential housing societies. Many residential housing societies have also deployed these scanners and are keeping a check on visitors. Preferably, thermal scanners are used to detect an elevated Body Temperature and it is an effective tool to measure skin surface temperature and identify individuals with Elevated Body Temperature (EBT). US20130124039A1 patent provide a thermal scanning system more specifically discloses a devices or method for measuring brain temperature, brain function, metabolic function, hydrodynamic function, hydration status, hemodynamic function, body chemistry and the like. US 10,504,020 B2 patent provide the application interacts with peripherals coupled to the client and receives input from sensors coupled to the peripherals. The sensor data is provided to the server for processing, including by neural networks. The application includes a workflow defining a finite state machine that traverses states at least partially based on the response to sensor data. The server may provide dynamic reallocation of compute resources to resolve demand for classifier training job requests; use of jurisdictional certificates to define data usage and sharing; and data fusion. Applications include manufacturing verification, medical diagnosis and treatment, genomics and viral detection. This cited art is about computational load distribution between a client & server, whereas client is rather light-weight processor which is unable to perform compute intensive tasks when applying deep learning neural networks to the sensor data. It provisions powerful computing resources like GPUs, FPGAs and ASICs etc for compute intensive tasks. Considering the global outbreak of the Corona virus (COVID-19), society is deeply concerned about the spread of infection and seeking tools to contain and ultimately stop the spread of the virus. Covid-19 is highly contagious virus. Already, millions in the world are suffering across different continents. One major limitation in curbing the virus spread is lack of methods to detect people with virus. As an interim measure, people are using thermal scanners identify infected individuals. But this approach has many limitations as the high temperature is observed in very few Covid-19 patients and that too at a very late stage. Although no thermal cameras can detect or diagnose the Corona virus. Thermal scanner is used as an adjunct to other body temperature screening tools for detecting elevated skin temperature in high-traffic public places through quick individual screening.Despite knowing limitations of these thermal scanner / method, the whole world is using the scanner for scanning the potential patient because of lack of alternatives. The available diagnostic tool / scanner avoid scanning of other vital parameter which are essential for finding virus carrying human / COVID patent. Hence there is need of system for detection of COVID-19 / virus carrier individuals / subjects, by analysing certain vital parameters, including body temperature. Further system and method used for detection leverages latest of medical sciences along with advancements in technologies like computation & communication and able to quickly scans the visitor's vital parameters and appearance attributes, measuring / assessing all of them non-invasively, from a safe distance. OBJECTIVE OF THE INVENTION:Primary object of the present invention is to overcome the limitation associated with the prior art.Another object is to provide a novel system for overall health assessment of human carrying viruses. Another object of the present invention is to provide a system and method for scanning of COVID-19 or any such virus carrier individuals / subjects, by analyzing certain vital parameters, including body temperature. Yet another object of the present invention is to provide an apparatus to diagnose / identify a virus carrying human.Yet another object of the present invention is to provide a system for scanning virus carrying human which quickly scans the visitor's vital parameters and appearance attributes, measuring / assessing all of them non-invasively with maintaining a safe distance.Yet another object of the present invention is to provide an efficient system for scanning virus carrying human.Yet another object of the present invention is to provide a system which leverages the latest technologies of Artificial Intelligence, Machine vision and Machine Learning to decipher if visitor is a Corona virus carrier and reduce the chance of error in the diagnosis.Yet another object of the present invention is to provide a method for scanning / diagnosing a human carrying virus or corona virus by using the system as accordance to present invention. Yet another object of the invention is to provide a method to detect potential virus infected subjects from the healthy one, with high accuracy and in short time.SUMMARY OF THE INVENTIONThe present disclosure provides a system for detecting virus infection in a user comprising; a plurality of sensors configured to remotely sense physiological parameters of the user; an image processing device that continuously photographs the user’s body facial attributes to acquire a plurality of images; a machine vision engine electronically communication with the image processing device and the plurality of sensors, configured to extracts the attributes of the user and the physiological parameters; wherein the machine vision engine compute for processing the extracted facial attributes and physiological parameters and works in-tandem with a machine learning engine to invoke data-analytics function; In an embodiment the present disclosure provide that the machine vision engine computes the extracted facial attributes and physiological parameters via artificial intelligence-based computing technologies.In an embodiment the present disclosure provide that the physiological parameters of the user are body temperature, heart rate, respiratory rate, oxygen saturation level and haemoglobin level.In yet another embodiment the present disclosure provide that the facial attributes are sweating, stress signs & level, apparent age, obesity, gender etc.In still another embodiment the present disclosure provide that the plurality of sensors is from the group consisting of spectroscopy, RF and ultrasonic waves technology-based sensors etc.In yet another embodiment the present disclosure provide that the machine vision engine is in communication with a second sensor module having environment / ambience sensors configured to get ambient temperature, humidity, barometric pressure and Geo-locations sensors configured to extract parameters like altitude, longitude and latitude. In yet another embodiment the present disclosure provide that the machine vision engine is in synchronization with a cloud server to provide connectivity, computing and database services.In yet another embodiment the present disclosure provide that the machine learning engine monitors all the computations done by the Machine-vision engine and continuously correlates and regresses the physiological parameters via data analyticsThe present disclosure provides a method of operation of system for detecting virus infection in a user comprising: sensing remotely by a plurality of sensors physiological parameters of the user; acquiring by an image processing device a plurality of images of the user’s body facial attributes; extracting by a machine vision engine electronically communication with the image processing device and the plurality of sensors, the attributes of the user and the physiological parameters; wherein the machine vision engine compute for processing the extracted facial attributes and physiological parameters and works in-tandem with a machine learning engine to invoke data-analytics function for virus detection;DETAILED DESCRIPTION OF DRAWING:To further clarify advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings in which:Figure 1: illustrates an overview of system for scanning a human carrying virus / infection / Corona virus as accordance to embodiment of the present invention.Figure 2: illustrates about the machine vision engine.DETAILED DESCRIPTION OF THE INVENTION:For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof. Throughout the patent specification, a convention employed is that in the appended drawings, like numerals denote like components.The present invention provides a holistic approach for assessing subject’s health, and probability of covid19 infection thereof, by means of specific physiological (vital) parameters; pivoted on the available medical science research about the Covid19 virus and its effects. Leveraging on AI (Artificial Intelligence) and IoT (Internet of Things) technologies, the invented system can learn automatically from the latest in healthcare research. Additionally, the machine learning engine can learn from system’s own dataset; training from this data, it can refine the assessment criteria and also share such learnings / interpretations with other systems / operators to help them improve. Present invention provides a robust / holistic virus infection detection system in which tasks are distributed based on focus to maintain accuracy and comprehensive database (which could be used by civic systems / authorities for various other applications). In present invention, the client is self-sufficient for its computational needs and depends on server for data and communication needs. Server, being the central node in the network of multiple clients, does certain computational tasks which are common to all the clients.Referring to figure 1, as depicted, any subject appearing before the present invention system detecting virus infection is scanned and the subject’s vital parameters are measured. At first, sensor module adjusts itself according to subject’s specifics like height, standing orientation etc. The subject is restricted to maintain a safe distance from the instrument. Once aligned, the sensor array and camera are triggered to start collecting samples / measurements / pics. The thermal scanner feeds in IR spectroscopy to gather body temperature. Similarly, other sensors capture heart rate, respiratory rate, oxygen saturation, haemoglobin and blood pressure. Environment / ambience sensors are also read in parallel to get ambient temperature, humidity, barometric pressure. Geo-locations parameters like altitude, longitude and latitude are also considered. Appropriate correction factors are applied to the measurement / analysis, as per these location specific parameters. The camera feed is captured in parallel and used for two purposes. To computes the oxygen saturation and haemoglobin level in-sync with the radar sensor. And to extract subject’s other attributes like sweating, stress signs & level, apparent age, gender and make-up etc. Machine vision engine attached to the camera extracts these physiological parameters and physical aspects of the subject. Image recognition function also works in-tandem with the camera and machine vision; while synching up with the cloud all the time, it performs two-pronged action. Firstly it pushes the image-signature to the cloud and also fetches blockchain entry (if exists) from the cloud. The cloud server provides overall connectivity, computing and database services on need basis. It maintains subject’s data across the places she visits over time. It could also connect with any existing public / govt database and leverage the subject’s information / data (from the other database) for performance improvements in terms of better accuracy and traceability of the subject. It can also enhance the public / govt databased by sharing useful / relevant information about the subject. It also fetches the latest information relevant for the purpose like medical research, virology information or any other empirical data. Cloud server records and analyses such information / data to realign / reconfigure the system such that its coefficients / configuration-matrices are updated for better, more accurate measurement & analysis. The core machine vision engine computes or processes all these data inputs of vitals, facial attributes and environmental data in accordance with the medical research available, including virology and any empirical data about virus impact. Compute engine / machine vision engine works in-tandem with a machine learning engine and invokes data-analytics function to formulate health metric-based assessment of the subject and concludes if she has virus infection or not.If an individual is found to be virus carrier, the instrument can immediately notify the subject herself and also to all stakeholders like the security, govt, authorities; for necessary compliance. It can notify via local siren / hooter, mobile app, SMS or even a call as per pre-configured settings.Referring to figure 2, machine vision engine is the amalgamation of hardware and software. When the camera, working as a sensor input for the system, provides RGB video feed of 720p at 30 FPS. Machine Vision engine, as depicted in the figure 2, processes this image, generates pseudo reflectance spectroscopy data, by converting it into VIBGYOR stream, leveraging on pre-set (RGB-to-VIBGYOR) conversion matrices (stored in the database). Video samples are taken from subject’s forehead and palpebral conjunctiva, both; under a broad-band lighting source. The 2 types of samples are analyzed and processed to compute the BR (Breathing Rate), HR (Heart rate), SpO2 (Blood oxygen saturation level), SpHb (Haemoglobin level). Amplitude variations of RGB colors in temporal domain are transformed into frequency domain using DFT (Discrete Fourier Transform). Peaks in the resultant frequency domain spectrum are computed as HR and BR by applying digital filters. Compute engine assesses AC (pulsatile) and DC (non-pulsatile) components from reflected-spectrum and correlates it with the blank-white image (taken earlier) to find accurate SpO2 and SpHb levels. The system keeps on recording blank-white images at regular intervals and uses it for auto-calibration.The Machine Learning engine monitors all the computations done by the Machine-vision engine and aligns it with the C-band radar sensor data. Radar sensor provides real-time physical changes in the subject due to breathing-cycle and systole-diastole of heart. Pulsatile / AC component of camera signals is temporally correlated with radar inputs to yield any artifacts or measurement or anomalies in the subject’s physiology. . Machine-vision engine also converts the RGB stream into HSV and YCbCr formats, which are used to get skin texture, sweating, stress signs etc. These signals are analyzed to find skin-texture signatures and signs of sweating. Facial expressions are analyzed to find signs of anxiety or stress. Machine-learning engine, continuously correlates and regresses the sensor’s data by data analytics. Based on these regressions analysis it keeps on updating the matrices used for “RGB-to-VIBGYOR conversion” and the “blood / hemoglobin spectroscopy data” in the run time. It also syncs with the cloud server for any changes to factored in.The configurations matrices used for “RGB-to-VIBGYOR” conversion and “Blood spectroscopy data” are provided by cloud server, which maintains a golden copy. Machine-vision engine uses cloud services to fetch it regularly; to maintain updated copy in use. Cloud server keeps on syncing and learning from the data provided by various systems installed at different places. It computes the various coefficients / configuration values used and updates the configuration matrices based on inputs / learnings from these different sources.In an aspect of the present invention, there is provided a system for detecting / scanning the subject against the potential virus infection from the healthy one, with high accuracy and in short time.In an embodiment, the system comprises plurality of sensor having a first sensor module and a second sensor module. The first sensor module is configured to sense plurality of parameter of the subjects where said first sensor module measures body temperature, heart rate, respiratory rate, oxygen saturation and hemoglobin of the subject. In an embodiment, the subject appearing in front of the system is scanned and the vital parameters like body temperature, heart rate, respiratory rate, oxygen saturation and hemoglobin of the subject are measured. The plurality of sensor adjusts itself according to subject’s specifics like height, body structure, standing orientation etc. The subject is restricted to maintain a safe distance from the instrument.In an embodiment, first sensor module comprises spectroscopy, RF and ultrasonic waves technology based sensors and camera in the instrument sense and measure various physiological (vital) parameters of the subject. These parameters are measured using different non-invasive sensors from a distance of >1m (between sensors and the subject).In an embodiment, a second sensor module comprises environment / ambience sensors which are read in parallel to get ambient temperature, humidity, barometric pressure. Geo-locations parameters like altitude, longitude and latitude are also taken into account. In an embodiment, the system comprises of a camera connected with plurality of sensors and configured to collecting samples / measurements / pics once subject is aligned and system triggered to start. The thermal scanner feeds in IR spectroscopy to gather body temperature. Similarly, other sensors capture heart rate, respiratory rate, oxygen saturation, hemoglobin and blood pressure.In an embodiment, the camera feed is captured in parallel, and used to extract subject’s other attributes like sweating, stress signs & level, apparent age, obesity, gender and make-up etc. Machine vision engine attached to the camera extracts these physical aspects of the subject.In an embodiment, the image recognition function also works in-tandem with the camera and machine vision; while synching up with the cloud all the time, it performs two pronged action. Firstly it pushes the image-signature to the cloud and also fetches block chain entry (if exists) from the cloud.In an embodiment, the cloud server provides overall connectivity, computing and database services on need basis. It maintains subject’s data across the places she visits over time. It could also connect with any existing public / govt database and leverage the subject’s information / data (from the other database) for performance improvements in terms of better accuracy in assessment and traceability of the subject. It can also enhance the public / govt data based by sharing useful / relevant information about the subject.In an embodiment, compute engine works in-tandem with the machine vision engine and invokes data-analytics function to formulate health metric (as per formulation explained) based assessment of the subject and concludes if she has virus infection or not.In an embodiment, in case of an individual is found to be virus carrier, the system can immediately notify the subject and also to the security, govt, authorities and all stakeholders for necessary compliance. It can notify via local siren / hooter, mobile app, SMS or even a call as per pre-configured settings.In an embodiment, system comprises artificial intelligence-based computing technologies augmented analysis by learning from the data and accelerated assessment through advance computing techniques.In an embodiment, the system and method could be used in many forms like standalone device or as an attachment to a (pre-existing) computer or maybe part of a robotic system or drone etc.The system of the present invention can quickly diagnose an individual if he is Covid-19 infected with at high accuracy. The present invention system detects virus carrier quickly (in less than 40 seconds) and helps in containing spread of the disease. In another aspect of the present invention, there is provided a method assessment / scanning the health of a subject or individual against virus infection with the help of the system as described above. The invention is further described by following non limiting example.EXAMPLE 1In an exemplary embodiment, the invention identifies a first set of vital parameters which can be interpreted / analyzed to formulate a unique metric about health of the subject. Measurement of the health metric for individual is used as criteria for detecting virus infection. These parameters are: body temperature, heart rate, respiratory rate, oxygen saturation level and hemoglobin level; the fundamental five parameters of the disclosed formulation. Sub-metric calculation for these fundamental five parameters, based on the empirical data and the normal range is defined as in the table below:VITAL PARAMETER SUB-METRICBODY TEMPERATURE (T) ?T = 5 T – 395HEART RATE (HR) ?HR = HR / 4 +75RESPIRATORY RATE (RR) ?RR = RR / 2 + 85OXYGEN SATURATION (SPO2) ?SPO2 = SPO2HAEMOGLOBIN (HB) ?HB = 1.82 HB + 68.18Based on above sub-metrices, overall health metric is calculated as variance of all sub-metrices; depicted below: ?h1 = {?(?i - 95)2} / 5Such measured health metric is further honed with the blood pressure, sweating, stress signs and environmental data to reach a final health metric as a measure of health assessment of the subject. The invention apply machine learning as semi-supervised learning machine with the above fundamental five parameters and ?h1 health metric as the labelled data and rest of the facial analysis and environmental data, along with any more data it can retrieve from the third party data-bases like public data-base or local Govt database, as unlabeled data. The invention provides a system to measure this set of vitals in non-invasive and contactless manner. The system comprises a union of sensors based on Spectroscopy, Radio frequency, ultrasound techniques and video camera. Robotics component makes sensing adaptive to subject’s physique, environment and any such local aspects; it enhances accuracy and tempo of testing.APPLICATIONS: As the world is struggling with the Covid-19 epidemic, it has to mend it’s approach significantly in many ways of life. One such big change is expected in the way security / surveillance is maneuvered and managed today – a check of virus carriers seems imperative & inevitable. Mere thermal scanning doesn’t suffice the requirement. The disclosed instrument could be used in all the public places where people can carry virus and transmit (knowingly or unknowingly); places like Airports, Hotels, Shopping malls, Offices, Commercial spaces, or even Hospitals. It could check the incomers for virus infections and provide access to the “good” cases only. It can work along with existing security / surveillance checkpoints as an additional health-based checkpoint or as a standalone / new checkpoint, wherever necessary.
Claims
We Claim:1.A system for detecting virus infection in a user comprising;a plurality of sensors configured to remotely sense physiological parameters of the user;an image processing device that continuously photographs the user’s body facial attributes to acquire a plurality of images;a machine vision engine electronically communication with the image processing device and the plurality of sensors, configured to extracts the facial attributes of the user and the physiological parameters;wherein the machine vision engine compute for processing the extracted facial attributes and physiological parameters and works in-tandem with a machine learning engine to invoke data-analytics function for determining viral infection; 2. The system as claimed in claim 1, wherein the machine vision engine computes the extracted facial attributes and physiological parameters via artificial intelligence-based computing technologies.
3. The system as claimed in claim 1, wherein the physiological parameters of the user are body temperature, heart rate, respiratory rate, oxygen saturation level and haemoglobin level.
4. The system as claimed in claim 1, wherein the facial attributes are sweating, stress signs & level, apparent age, obesity, gender etc.
5. The system as claimed in claim 1, wherein the plurality of sensors is from the group consisting of spectroscopy, RF and ultrasonic waves technology-based sensors etc.
6. The system as claimed in claim 1, wherein the machine vision engine is in communication with a second sensor module having environment / ambience sensors configured to get ambient temperature, humidity, barometric pressure and Geo-locations sensors configured to extract parameters like altitude, longitude and latitude.
7. The system as claimed in claim 1, wherein the machine vision engine is in synchronization with a cloud server to provide connectivity, computing and database services.
8. The system as claimed in claim 1, wherein the machine learning engine monitors all the computations done by the Machine-vision engine and continuously correlates and regresses the physiological parameters via data analytics9. A method of operation of system for detecting virus infection in a user comprising Sensing remotely by a plurality of sensors physiological parameters of the user;acquiring by an image processing device a plurality of images of the user’s body facial attributes;extracting by a machine vision engine electronically communication with the image processing device and the plurality of sensors, the attributes of the user and the physiological parameters;the machine vision engine compute for processing the extracted facial attributes and physiological parameters and works in-tandem with a machine learning engine to invoke data-analytics function for virus detection;
