Method and device for predicting allergic reaction likelihoods

Abstract

Disclosed is a method and a device for predicting allergic reaction or irritation likelihoods. In particular, the present application provides a method and an electronic user device for predicting a l

Description

[2] Currently, around 10% - 30% of the global population have some kind of allergy and this appears to be increasing. For example, the percentage of children diagnosed with allergic rhinitis and eczema have both trebled over the last 30 years. Figures released from the UK’s National Health Service (NHS) in 2022 by The Medicines and Healthcare products Regulatory Agency showed that there were nearly 26,000 admissions in 2022-23, more than double the 12,361 admissions in 2002-03 (a 108% increase) for food-related anaphylaxis and other adverse reactions. [3] Allergies and symptoms caused by similarly appearing ailments such as food intolerance, and irritation caused by irritants, cost health care services an enormous amount each year. In the UK alone, allergic diseases across all ages cost the NHS an estimated £900 million a year, mostly through prescribed treatments in primary care, representing 10% of the GP prescribing budget. [4] Allergies and symptoms caused by similarly expressed ailments such as food intolerance and irritation caused by irritants also take a massive tole on the individual. Almost a third of allergy sufferers have had to change their lifestyles to reduce their allergic reactions. Actions range from keeping their home extra clean to using special bedding. [5] A large part of the expense of treating allergies and symptoms caused by similarly expressed ailments is that it is hard to pinpoint the exact cause or causes of the symptoms. This results in people experiencing a sense of powerlessness to identify their problem. It also results in very long waiting lists for allergy testing in systems like the NHS, and large amounts of capital being spent on treating people’s symptoms rather than the root cause of their symptoms. [6] People interact with so many different products and substances in their day to day lives that it is near impossible for them to identify what may be causing their issues without rigorous and expensive testing. Complexity is added to the issue of discovering what allergens and irritants are causing people issues when you consider that they might be allergic to or irritated by an array of different substances or that the symptoms they are experiencing may be caused by a combination of products and substances rather than just the one. [7] Current tests to discover irritants and allergens, such as patch testing are incredibly work-intensive for medical professionals, intrusive for patients and, oftentimes, the tests are inconclusive or do not cover a great enough range of potential causes of symptoms to be effective. The tests may reveal some of the causes of symptoms but not all of them, or cause patients to think an irritant or allergen is not an issue for them because the irritant or allergen only causes problems for them when it interacts with another substance. [8] The present applicant has identified the need for an improved method for predicting a likelihood that a product causes, or has caused, an allergic reaction in a user. Summary [9] Ina first approach of the present techniques, there is provided a computer-implemented method, performed by an electronic user device, for predicting a potential allergen or irritant for a user, the method comprising: receiving, via a user interface, information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated; obtaining, in response to the received information, user personal data comprising known allergens or irritants of the user; calculating a correlation between the obtained user personal data and the potential allergen or irritant; and predicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user.

[10] The term “potential allergen or irritant” used herein means any matter formed of at least one component. For example, the potential allergen or irritant may be, or may be formed of, any of: a single substance, more than one substance, a single ingredient, more than one ingredient, a composition, a mixture, a chemical substance, and a biological substance (e.g. pollen). The potential allergen or irritant may be a medical product (e.g. paracetamol, cough syrup, etc.), a household product (e.g. laundry detergent, beeswax, etc.), a healthcare or sanitary product (e.g. shampoo, soap, etc.), a food item (including individual food items such as a fruit, vegetable, or grain, and prepared food items, such as potato chips / crisps, biscuits / cakes, pasta sauce, etc.), or a drink product. It will be understood these are nonlimiting and non-exhaustive examples of potential allergens or irritants.

[11] The mechanisms for symptoms caused by irritants and allergens are different. Allergen symptoms are caused by the immune response, while irritants cause irritation or inflammation to the individual directly.

[12] The term “user personal data” used herein means data relating to the user of the electronic user device. In other words, the user personal data relates to personal characteristics of the user. For example, the user personal data may relate to the age, sex and / or race of the user. It will be appreciated that the user personal data may relate to any number of personal characteristics that are relevant to the allergies of the user.

[13] Advantageously, the present techniques provide a method for predicting a likelihood that a potential allergen or irritant may cause an allergic reaction in or irritation to a user based on known allergens and / or irritants of the user. For example, the user may have known allergens or irritants, such as tree and grass pollen. The present techniques provide a way of predicting whether the user may have an allergic reaction to or experience irritation from a new potential allergen or irritant, such as honey. This is advantageous because the user may not have a concrete idea, prior to having consumed a product before, whether the product is likely to cause them an allergic reaction. However, there are often similarities between the underlying composition of different products (e.g. constituent proteins) which are the underlying cause of an allergic reaction. That is, there may be a correlation between the user personal data and the product for which a likelihood is to be calculated. For example, there may be a strong correlation between known allergens, such as pollen, and a product such as honey. The correlation is then used to predict the likelihood of an allergic reaction in a user.

[14] Furthermore, for potential allergens or irritants comprising multiple ingredients or constituents, it can be difficult to determine which of the ingredients is most likely to cause an allergic reaction or irritation using existing techniques. For example, for a product like shampoo comprising ingredients such as coconut oil and sorbitan, it can be difficult to determine whether the product will cause an allergic reaction, and also which of the coconut oil or sorbitan may be responsible for the reaction. Another advantage of the present techniques is that the likelihood of each of the ingredients of a composite product may be calculated.

[15] Therefore, in some cases, receiving information about a product comprises receiving information about a potential allergen or irritant comprising a single ingredient or constituent. That is, the present techniques may be applied to comparatively simple cases where the potential allergen or irritant is not a composite product, and instead comprises a single ingredient or constituent For example, the potential allergen or irritant may be sorbitan. In such cases, the method further comprises: outputting, via the user interface, the likelihood that the single ingredient or constituent causes an allergic reaction in or irritation to the user.

[16] In other cases, receiving information about a potential allergen or irritant comprises receiving information about a potential allergen or irritant comprising at least two ingredients or constituents. That is, the potential allergen or irritant may be a composite product (e.g. shampoo), and may comprise multiple ingredients or constituents. In such cases, calculating a correlation comprises calculating a correlation between the obtained user personal data and each of the at least two ingredients or constituents; and predicting a likelihood comprises predicting a likelihood that each of the at least two ingredients or constituents causes an allergic reaction or irritation. Therefore, the method may further comprise outputting, via the user interface, a list of the likelihood of each of the at least two ingredients or constituents causing an allergic reaction in or irritation to the user. That is, the present techniques advantageously present information relating to multiple potential allergens or irritants in a way that is clear and comprehensible for the user.

[17] In addition to providing a way of predicting whether an allergic reaction or irritation may occur in the future, the present techniques also provide a way of using past information to predict the reaction likelihood. Therefore, predicting the likelihood may further comprise: prior to the receiving, requesting, via the user interface, an indication on whether the user has experienced an allergic reaction or irritation in a predefined period of time. That is, the user may be prompted to specify whether or not an allergic reaction or irritation has already occurred at some point in the past. The predefined period of time may be any appropriate period of time which a user may be interested in assessing, such as one day, two days, one week, two weeks, etc.

[18] In some cases, receiving the indication comprises receiving, via the user interface, an indication that the user has not experienced an allergic reaction or irritation in a predefined period of time; and requesting, via the user interface, information about a potential allergen or irritant a user wishes to evaluate; wherein receiving information about a potential allergen or irritant for which a likelihood is to be calculated comprises receiving any one or more of: an allergen name; an irritant name; a product name; an image of a product; a product identifier; a product barcode (including a QR code or 2D matrix barcode); and a link to a product on a website. That is, the user may indicate that they have not experienced an allergic reaction or irritation in the predefined period of time, and the user may wish to evaluate a potential allergen or irritant as a precautionary measure to help them determine whether or not a potential allergen or irritant may be an allergen or irritant.

[19] In other cases receiving the indication comprises, receiving, via the user interface, an indication that the user has experienced an allergic reaction or irritation in a predefined period of time; and requesting, via the user interface, information about at least one potential allergen or irritant the user has interacted with over the predefined period of time; wherein receiving information about a potential allergen or irritant for which a likelihood is to be calculated comprises receiving at least one potential allergen or irritant the user has interacted with over the predefined period of time. That is, the user may indicate that they have experienced a past allergic reaction or irritation, and the user may wish to evaluate which, of the potential allergens or irritants they have interacted with over the predefined period of time, may have been responsible for the allergic reaction or irritation.

[20] As noted above, user personal data corresponds to data relating to characteristics of the user. Therefore, obtaining user personal data may comprise, in addition to known allergens and / or irritants, obtaining any one or more of the following for the user: medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; family locations of birth; location of birth of the user; location in which the user lives; history of smoking; weight; and pregnancy status.

[21] The correlation being calculated serves as an indicator of the connection between personal characteristics of the user and the potential allergen or irritant. The correlation may be calculated based on population data corresponding to personal data of a plurality of other users and their allergic reactions to or irritations caused by the potential allergen or irritant. That is, existing data relating to the reactions of a wider population of other users to the potential allergen or irritant may be used to infer a correlation between personal characteristics of the user and the potential allergen or irritant. For example, there may be many other users who are male and aged 25 to 35 with an existing pollen allergy who also had an allergic reaction to the potential allergen or irritant. This may be used to infer whether the personal characteristics of the user and the potential allergen or irritant are correlated. Therefore, calculating a correlation between the ol‘ ' srsonal data and the potential allergen or irritant may comprise: obtaining population data comprising personal data of a plurality of other users and their reactions to the potential allergen or irritant; and correlating the personal data of other users and their reactions to the potential allergen or irritant.

[22] Obtaining population data may comprise selecting a set of users with at least one characteristic in common with the user. That is, the population data relating to the set of users which are most similar to the user is likely to be the most relevant to the prediction of the likelihood. This is advantageous because, rather than using the population data of the entire set of users, only a sub-portion of the most relevant users are selected.

[23] Selecting a set of users may comprise selecting users based on any one or more of the following characteristics: geographic location; medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; location of birth; users location of birth; location in which the user lives; history of smoking; users weight; and pregnancy status. That is, the set of users may be selected based on multiple characteristics. For example, certain allergies, such as pollen allergies may be correlated with geographic location (areas with high levels of pollen), whereas other allergies (such as a peanut allergy) may be correlated with genetics.

[24] In a second approach of the present techniques, there is provided an electronic user device for predicting a potential allergen or irritant for a user, comprising: a user interface for receiving information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated; and at least one processor coupled to memory for: obtaining, in response to the received information, user personal data comprising known allergens or irritants of the user; calculating a correlation between the obtained user personal data and the potential allergen or irritant; and predicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user.

[25] The electronic user device may be a smartphone, tablet, laptop, computer or computing device. That is, the electronic user device may be any appropriate device for personal use via a user interface and which is capable of performing the above-mentioned prediction.

[26] The features described above with respect to the first approach apply equally to the second approach, and for the sake of conciseness are not repeated.

[27] The user interface may comprise a display screen, and the at least one processor is arranged to output, on the display screen, the likelihood that the potential allergen or irritant causes an allergic reaction in or irritation to the user.

[28] The electronic device may further comprise at least one image capture device for capturing any one or more of: an allergen name; an irritant name; a product name; an image of a product; a product identifier; a product barcode; and a link to a product on a website.

[29] In a third approach of the present techniques, there is provided a system for predicting a potential allergen or irritant for a user, the system comprising: a server comprising storage for storing user personal data for a plurality of users of the system, the user personal data comprising known allergens or irritants of each user; and a plurality of electronic user devices used by the plurality of users, each user device of a user comprising: a user interface for receiving information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated; and a communication module for communicating with the server; wherein the server further comprises at least one processor coupled to memory for: receiving, via the communication module of a user device, information about a product for which a likelihood is to be calculated and information identifying a user of the user device; obtaining, from the storage using the received information identifying a user, user personal data comprising known allergens or irritants of the user; calculating a correlation between the obtained user personal data and the potential allergen or irritant; predicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user; and transmitting, to the user device, the predicted likelihood.

[30] The features described above with respect to the first and second approaches apply equally to the third approach, and for the sake of conciseness are not repeated.

[31] Ina related approach of the present techniques, there is provided a computer-readable storage medium comprising instructions which, when executed by a processor, causes the processor to carry out any of the methods described herein.

[32] As will be appreciated by one skilled in the art, the present techniques may be embodied as a system, method or computer program product. Accordingly, present techniques may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[33] Furthermore, the present techniques may take the form of a computer program product embodied in a computer readable medium having computer readable program code embodied thereon. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

[34] Computer program code for carrying out operations of the present techniques may be written in any combination of one or more programming languages, including object oriented programming languages and conventional procedural programming languages. Code components may be embodied as procedures, methods or the like, and may comprise subcomponents which may take the form of instructions or sequences of instructions at any of the levels of abstraction, from the direct machine instructions of a native instruction set to high-level compiled or interpreted language constructs.

[35] Embodiments of the present techniques also provide a non-transitory data carrier carrying code which, when implemented on a processor, causes the processor to carry out any of the methods described herein.

[36] The techniques further provide processor control code to implement the abovedescribed methods, for example on a general purpose computer system or on a digital signal processor (DSP). The techniques also provide a carrier carrying processor control code to, when running, implement any of the above methods, in particular on a non-transitory data carrier. The code may be provided on a carrier such as a disk, a microprocessor, CD- or DVD-ROM, programmed memory such as non-volatile memory (e.g. Flash) or read-only memory (firmware), or on a data carrier such as an optical or electrical signal carrier. Code (and / or data) to implement embodiments of the techniques described herein may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as Python, C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language such as Verilog (RTM) or VHDL (Very high speed integrated circuit Hardware Description Language). As the skilled person will appreciate, such code and / or data may be distributed between a plurality of coupled components in communication with one another. The techniques may comprise a controller which includes a microprocessor, working memory and program memory coupled to one or more of the components of the system.

[37] It will also be clear to one of skill in the art that all or part of a logical method according to embodiments of the present techniques may suitably be embodied in a logic apparatus comprising logic elements to perform the steps of the above-described methods, and that such logic elements may comprise components such as logic gates in, for example a programmable logic array or application-specific integrated circuit. Such a logic arrangement may further be embodied in enabling elements for temporarily or permanently establishing logic structures in such an array or circuit using, for example, a virtual hardware descriptor language, which may be stored and transmitted using fixed or transmittable carrier media.

[38] In an embodiment, the present techniques may be realised in the form of a data carrier having functional data thereon, said functional data comprising functional computer data structures to, when loaded into a computer system or network and operated upon thereby, enable said computer system to perform all the steps of the above-described method. Brief description of the drawings

[39] Implementations of the present techniques will now be described, by way of example only, with reference to the accompanying drawings, in which:

[40] Figure 1 is a flowchart of steps in a method for predicting a likelihood that a potential allergen or irritant causes an allergic reaction in or irritation to a user;

[41] Figure 2 is a block diagram of an electronic user device for predicting a likelihood that a potential allergen or irritant causes a reaction in a user;

[42] Figure 3 is a block diagram of a system for predicting a likelihood that a potential allergen or irritant causes a reaction in a user;

[43] Figure 4 is a flowchart of example steps to predict the likelihood that a potential allergen or irritant causes a reaction in the user; and

[44] Figure 5 is a flowchart of steps in an alternative example to predict the likelihood that a potential allergen or irritant causes a reaction in the user. Detailed description of the drawings

[45] Broadly speaking, embodiments of the present techniques provide a method and a device for predicting allergic reaction or irritation likelihoods. In particular, the present application provides a method and an electronic user device for predicting a likelihood that a potential allergen or irritant causes an allergic reaction in or irritation to a user.

[46] The present techniques provide a technology that looks to empower both individual users and health care professionals to identify potential allergens, irritants and infections. Such allergens, irritants and infections may be expressed on the skin and / or eyes of an individual. This technology could later be applied to identifying the causes, of a range of other allergies expressed in different ways.

[47] Existing techniques, such as those described in CN110767309A and CN110767309B, suggest using an app for tracking the food people eat, the time they eat it, and the symptoms of the allergic reactions they have. The app collates the data from all users and then uses a correlation algorithm to help to identify an allergen causing an allergic reaction of an individual user. This is far from ideal as the technology only looks at food consumed by the user. It does not take into account enough information to provide a useful idea of what is causing a person to have an allergic reaction or even if the symptoms the person is showing is an allergic reaction or just the effects of an irritant. These existing techniques only take into account food causing problems for the user. In contrast, the present techniques take into account any product or substance the user encounters that has the potential to induce symptoms. The present techniques also collect additional information on the user such as geography, race or health issues, amongst a vast list of other items that would better allow for correlations to be drawn to allow for more accurate predictions of the likelihood of a substance or product causing irritation or an allergic reaction.

[48] The present techniques advantageously provide a more holistic healthcare app to gather substantial amounts of data on users, the food they eat, and the products and substances they interact with as well as personal information, such as health issues, family health issues history, age, sex, gender, geographic location, genetic and epigenetic information, pregnancy status, location of birth; users location of birth; location in which the user lives; history of smoking; users weight; and information on the gut microbiome.

[49] The present techniques obtain large amounts of user information, for example, the food they eat, the products they use and health conditions they have, and use this information to identify trends in the information that may be used to inform a user about potential substances and products that may be an irritant or an allergen. The present techniques may be implemented via a software application or ‘app’, which is designed to be used by individual users who wish to find out about their allergies or potential allergies. The app will inform the user of the percentage likelihood a particular product or substance causing them to have a reaction based on the data. The information will also be used to help to identify products and substances that irritated the users skin or eyes or caused them to have a reaction and give them the percentage likelihood of each product or substance that they have interacted with before having the reaction, causing the reaction based on the data. The present techniques could also potentially be used to identify locations in which substances or products are present that may cause the user to have a reaction. The present techniques could also potentially be used to identify substances or products that are causing users to develop allergies. The present techniques could additionally be used to help users find products they need for every day life that will not cause them to have an allergic reaction.

[50] The present techniques provide an individual user with information on substances and products they are likely to be allergic to or irritated by, and also substances that have or may have caused an allergic reaction or irritation to them in the past, based on a large dataset of their own user data and the data of lots of other different users. The user will receive a percentage likelihood of having had an allergic reaction to a substance or product or receive a percentage likelihood of being allergic to a substance or product.

[51] When the user creates an account on the app they will enter into the app an initial category of data, the personal category, this describes features such as the users family history of illnesses, know allergens and irritants, location, age, gender, sex, current health problems, location of birth; users location of birth; location in which the user lives; history of smoking; weight; and personal history of health problems.

[52] When using the app the app will acquire a second category of data from the user, the interaction category. This will be information such as foods the user has consumed, products the user has used, substances like the pollen the user has been around and substances the user may have interacted with such as perfume. Here interaction means that the substance may have had the opportunity to interact with the users body, for example when perfume droplets come into contact with the skin, or are breathed in. To make it easier for the user to use the app, the user will either be able to scan a product to enter information about it into the app, or to scan a products barcode.

[53] When a user has an allergic reaction or reaction like symptoms the user will notify the app. The app will provide the user with a two lists. One list will show a list of products and substances the user has interacted over the week (or any other suitable predefined period of time) previous to the reaction. Each item on the list be scored with a likelihood of the reaction being caused by it. The likelihood will be calculated by looking at both categories of user data and comparing them with data from all other users with one or more item in the personal category in common with the user who has had the reaction. The list will be sent to the user in descending order of likelihood of a product or substance having caused the reaction.

[54] List Two will be just for constituent parts of product so users can see the which parts of products are likely to cause them the most problems based on the data. List Two will be generated in the same way as List One except products will be broken down into their constituent parts. For example, if a user used a shampoo that contained coconut oil and sorbitan, list two would not give the likelihood of a reaction to the shampoo but it would show the user the likelihood of having a reaction to coconut oil or sorbitan. Once again, this list will be in descending order of likelihood of a reaction.

[55] The more data from users the app has access to the more accurate the information given to users will be.

[56] The app will also be able to scan products to tell the user the likelihood of a food or product causing an allergic reaction and the likelihood of each ingredient in the food causing the user to have a reaction.

[57] The app will also tell users when a reaction may be caused by a fungal, viral, microbial or parasitic infection based on their location data and if they have come into contact with other users or significant number of users with one of these types of infections.

[58] The features of the present techniques mentioned above are now described in more detail.

[59] Identifying Allergens and Irritants based on large amounts of user data.

[60] This app will allow users to identify products and substances which are potentially the cause of an allergic reaction or skin or eye irritation to the user. These allergic reactions and skin and eye irritation will be referred to as a reaction.

[61] The app will do this by allowing users to use an input system on an electronic device in which they can enter in a series of data points of personal information which will be stored in a database. Going forward this kind of information will be referred to as personal category data points. Personal category data points include; health issues, medical conditions, genetics, epigenetics, gut microbiome, age, sex, race geographic location, pre-identified allergies, family history of medical conditions, location of birth, users location of birth , location in which the user lives, history of smoking, users weight, and pregnancy,

[62] The users will also record data on substances and products they interact with, this can mean they consume them, use them or are adjacent to them to them in such a way that a chemical reaction might be caused. For example, a user may consume an apple, use a deodorant or may be in the presence of mould or pollen. All of this information will be recorded in a database. This information will be referred to as interaction category data points going forward. Interaction category data points include; consumed or used food products, used household cleaning products, used skin care products, used toiletries, used medication, symptoms of the reaction, clothing, water provider, the presence of mould in the users accommodation, user exercise data, user images of their own allergic reaction symptoms, interaction with animals, user interaction with metals, used supplements, used vitamin tablets,

[63] Some datapoints may be placed in both the personal data category and the interaction data point category and used twice in calculations one time as a point in the personal data point category and a second time the interaction category. These types of data point include, medication used, the presence of mould in a household currently or for a prolonged period previously, nutritional deficiencies.

[64] Additional data collected includes; date and time of product use, date of an allergic reaction, quantity of a product used by a user.

[65] If a user experiences a reaction they will log it on the app, and may also note the symptoms and the time of reaction in the electronic input device. Once logged two lists will be produced. List one in which all of the products and substances a user has interacted with over the week previous before the reaction, that may have caused the reaction. Then the other list, List Two will only list substances and constituent parts of products that may have caused the reaction.

[66] List One will not include products or substances the user has interacted with longer than a week ago, providing that within that week period after interacting with the product or substance the user did not experience a reaction. The remaining products and substances the user interacted with will be added to List One and assigned a percentage likelihood chance of having caused the reaction. The list will also include the percentage likelihood of combinations of products and substances having caused the allergic reaction.

[67] List Two will list substances and constituent parts of products that the user may have had an allergic reaction to and the percentage likelihood of every item on the list or a combination of items on the list having caused the reaction.

[68] The percentage likelihoods of a product, substance or constituent part having caused the reaction will be calculated by using the data collected from the user in question and all other users.

[69] The prediction of an allergic reaction likelihood will now be described in more detail in relation to Figure 1. Figure 1 is a flowchart of steps in a method for predicting a likelihood that a potential allergen or irritant causes an allergic reaction in or irritation to a user. The method may be performed by an electronic user device. The method comprises receiving, via a user interface, information about a potential allergen or irritant for which a likelihood is to be calculated (S100); obtaining, in response to the received information, user personal data comprising known allergens and / or irritants of the user (S102); calculating a correlation between the obtained user personal data and the potential allergen or irritant (S104); and predicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction in or irritation to the user (S106).

[70] At step S102, obtaining user personal data may comprise, in addition to known allergens and / or irritants, obtaining any one or more of the following for the user: medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; location of birth; users location of birth ; location in which the user lives; history of smoking; users weight; and pregnancy.

[71] At step S104, calculating a correlation between the obtained user personal data and the potential allergen or irritant may comprise: obtaining population data comprising personal data of a plurality of other users and their reactions to the potential allergen or irritant; and correlating the personal data of other users and their reactions to the potential allergen or irritant. Obtaining population data may comprise selecting a set of users with at least one characteristic in common with the user. Selecting a set of users may comprise selecting users based on any one or more of the following characteristics: medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; parents location of birth; users location of birth; location in which the user lives; history of smoking; users weight; and pregnancy status.

[72] As noted above, the allergic reaction likelihood may be predicted for a potential allergen or irritant comprising a single or multiple ingredients (or constituents). In both cases, the method may further comprise an outputting step (S108).

[73] In cases where the potential allergen or irritant comprises a single ingredient or constituent, the step S100 of receiving information about a potential allergen or irritant comprises receiving information about a potential allergen or irritant comprising a single ingredient or constituent. The outputting step S108 comprises outputting, via the user interface, the likelihood that the single ingredient or constituent causes an allergic reaction in the user.

[74] In cases where the potential allergen or irritant comprises two or more ingredients, the step S100 of receiving information about a potential allergen or irritant comprises receiving information about a potential allergen or irritant comprising at least two ingredients or constituents. Additionally, step S104 of calculating a correlation comprises calculating a correlation between the obtained user personal data and each of the at least two ingredients or constituents; and step S106 of predicting a likelihood comprises predicting a likelihood that each of the at least two ingredients or constituents causes a reaction.

[75] Figure 2 is a block diagram of an electronic user device for predicting a likelihood that a potential allergen or irritant causes an allergic reaction in or irritation to a user. The electronic user device 100 comprises a user interface 102 for receiving information about a potential allergen or irritant for which a likelihood is to be calculated; and at least one processor 104 coupled to memory 106 for: obtaining, in response to the received information, user personal data comprising known allergens and / or irritants of the user; calculating a correlation between the obtained user personal data and the potential allergen or irritant; and predicting, using the correlation, a likelihood that the potential allergen or irritant causes a reaction in the user. The electronic user device 100 may be a smartphone, tablet, laptop, computer or computing device. That is, the electronic user device 100 may be any appropriate device for personal use via a user interface.

[76] The user interface 102 may comprise a display screen 108, and the at least one processor 104 is arranged to output, on the display screen 108, the likelihood that the potential allergen or irritant causes a reaction in the user.

[77] The electronic device 100 may further comprise at least one image capture device 110 for capturing any one or more of: an allergen name; an irritant name; a product name; an image of a product; a product identifier; a product barcode; and a link to a product on a website.

[78] Figure 3 is a block diagram of a system for predicting a likelihood that a potential allergen or irritant causes an allergic reaction in or irritation to a user. The system 200 comprises a server comprising storage 204 for storing user personal data 206 for a plurality of users of the system, the user personal data comprising known allergens and / or irritants of each user; and a plurality of electronic user devices 212i, ..., 212n used by the plurality of users, each user device of a user comprising: a user interface (UI in Figure 5) 214 for receiving information about a product for which a likelihood is to be calculated; and a communication module (Comm, module in Figure 5) 216 for communicating with the server 202; wherein the server 202 further comprises at least one processor 208 coupled to memory for 210: receiving, via the communication module 216 of a user device, information about a potential allergen or irritant for which a likelihood is to be calculated and information identifying a user of the user device. For simplicity of illustration, only one electronic user device of the plurality of electronic user devices is shown as comprising the user interface 214 and the communication module 216.

[79] There are several ways the percentage likelihood for each item can be calculated. Below are two, non-limiting, example calculation methods.

[80] Calculation 1: This calculation is to work out the percentage likelihood that substance X may have caused a reaction experienced by User Y.

[81] A = Total number of data points in the users personal data category which can be linked to X. A personal data category data point can be linked to X if any user who is not User Y has interacted with substance X and has a personal category data point in common with User Y.

[82] Bi = The percentage of users, not including User Y, who have data point i in common with User Y, who have interacted with X and have logged that it causes them to have some sort of reaction, i could be any personal data point.

[83] Bi = The percentage of users, not including User Y, who have data point 1 in common with User Y, who have interacted with X and have logged that it causes them to have some sort of reaction. 1 could be any personal data point.

[84] C =The percentage likelihood of user Y being allergic to substance X.

[85]

[86] Calculation 2: This is a calculation for substance X relating to a reaction exhibited by User Y. However, unlike in Calculation 1 we are considering that some data points that can be related to product X may not have enough data entries to be accurate and thus should not be treated with the same confidence. To mitigate this problem we will multiply the Bi value of different personal category data points by the exponential function, e, to the power of negative D in order to diminish their weighting in the calculation.

[87] E = number of data entries for a personal category data point that relate to Bi. For example, If there are 6 users who have logged personal data point i and have also interacted with substance X, with or without a logged reaction, E would equal 6.

[88] T = the threshold value at which we do not lower the weighting of the data of a given personal data point.

[89] When E <T, D = 1 - , when E >T, D = 0

[90] A = Total number of data points in the users personal data category which can be linked to X. A personal data category data point can be linked to X if any user who is not User Y has interacted with substance X and has a personal category data point in common with User Y.

[91] Bi = The percentage of users, not including User Y, who have a personal data point i in common with user Y, who have interacted with X and have logged that it causes them to have some sort of reaction, i could be any personal data point.

[92] Bi = The percentage of users, not including User Y, who have data point 1 in common with user Y who have interacted with X and have logged that it causes them to have some sort of reaction. 1 could be any personal data point.

[93] C =The percentage likelihood of User Y being allergic to substance X.

[94] C = SB?(Ble~P+B2e~D+B3e~D......Bne~°)

[95] D could be multiplied by an additional coefficient to alter the weighting of Bi.

[96] There is another version of this calculation in which Bi = the percentage of users, not including User Y, who have a personal category data point i in common with user Y, and who have interacted with X and have a had some sort of reaction or have been caused irritation. The difference being that not all users being related to X, when calculating Bi, have logged X as an allergen or irritant in their personal data category.

[97] When combined the sum of the likelihoods of a user being allergic to or irritated by each product that they have interacted with over the previous week before the allergic reaction may exceed 100%. This is not a problem as it is important for the user to understand the issues they are experiencing may be caused by more than one substance or product.

[98] However, the app will never say to a user that a product or substance is either 100% or 0% causing the problems they are experiencing this could be considered medical advice, the aim of the app is only to indicate what may be causing the issues the user is experiencing and help them and any medical advisors supporting them such as general practitioners and dermatologists to identify the causes of the irritation or allergic reactions the user is experiencing. For instance if the result of the calculation was that user Y was 100% likely to be allergic to product X the app would tell them there is a 99% chance of them being allergic to the product. If the calculation result showed that there was a 0% chance that user Y was allergic to or irritated by product X then the app would tell them there was a 1% chance they are allergic to the product. This is because regardless of how large the database is there will always be some level of uncertainty as to how likely the user is to be allergic to or irritated by a particular product or substance and hence, we do not want to imply certainty by using 100% or 0% as a result.

[99] Once the calculation has been carried out for each substance and product on List One and each constituent part on List Two. The user will be shown the lists of products and substances and the list of constituent parts that may have caused the reaction. At the top of the list will be the products and substances and constituent parts that the users reaction was most likely to be caused by and at the bottom of the list will be the substances and products and constituent parts that are the least likely to have caused the users reaction. The list will display the product name and the percentage likelihood that the user is allergic to or irritated by the product. The list may or may not also show an image of the product to improve the user experience and make it clearer what each product is.

[100] Identifying when irritation or an allergic reaction may have been caused by multiple products, substances or constituent parts interacting with each other.

[101] Some allergic reactions are caused by the products and substances or their constituent parts interacting with each other. As it is harder to know when an allergic reaction is caused by two substances we will have to take into account when a user does not appear to be allergic to some substances. To do this we will use data users have entered into the database when they have not had an allergic reaction. If a user has not had a reaction, within two weeks, this could be set to be any arbitrary length of time, from a data entry about a product or substance a user has interacted with then when the user does have a reaction next, the product or substance the user has not had a reaction to previously will not be included on the list of potential allergens and irritants. This means causes of irritation or allergic reactions that are less common such as irritation and allergic reactions caused by the interaction of multiple products, substances or constituent parts will be higher on the list shown to the user.

[102] This over longer periods of time will improve the data we have on interacting irritants and allergens that cause the user irritation or allergic reactions due to these interactions. This is because as the data on interacting substances or products gets closer to the top of users’ lists, they are more likely to test the hypothesis that their irritation or allergic reaction is caused by a combination of substances and hence add it to their list of known irritants or allergens.

[103] However, due to the chance of sensitisation to allergens or irritants that have not previously been an issue to the user it is important to return excluded potential irritants or allergens to the list of potential reaction causes at some threshold time after they have been taken off the list. This time could initially be set arbitrarily and altered using data to be a smaller amount for particular products and substances depending on how many people develop allergies or irritation to them and at what frequency.

[104] The above-mentioned features of the present techniques directed to the prediction of a likelihood based on whether or not a user has experienced a previous reaction in a predefined period of time will now be described in relation to Figures 4 and 5.

[105] As noted above, the method may further comprise prior to the receiving, requesting, via the user interface, an indication on whether the user has experienced an allergic reaction in a predefined period of time. That is, in some cases, the indication is that the user has experienced an allergic reaction in the predefined period of time, and in other cases, the indication is that the user has not experienced an allergic reaction in the predefined period of time. Both of these cases will now be described in relation to Figures 4 and 5, respectively.

[106] Figure 4 is a flowchart of example steps to predict the likelihood that a potential allergen or irritant causes a reaction in the user. In this example, the indication is that the user has not experienced a reaction in a predefined period of time. The method comprises: receiving, via the user interface, an indication that the user has not experienced a reaction in a predefined period of time (S200); and requesting, via the user interface, information about a potential allergen or irritant a user wishes to evaluate (S202). The method then continues to step S100 of Figure 1. However, in this case, the receiving information step S100 of Figure 1 comprises receiving any one or more of: a product name; an image of a product; a product identifier; a product barcode; and a link to a product on a website.

[107] Figure 5 is a flowchart of steps in an alternative example to predict the likelihood that a potential allergen or irritant causes a reaction in the user. In this alternative example, the indication is that the user has experienced a reaction in a predefined period of time. The method comprises receiving, via the user interface, an indication that the user has experienced a reaction in a predefined period of time (S300); and requesting, via the user interface, information about at least one potential allergen or irritant the user has interacted with over the predefined period of time (S302). The method then continues to step S100 of Figure 1. However, in this case, the receiving information step S100 of Figure 1 comprises receiving at least one potential allergen or irritant the user has interacted with over the predefined period of time.

[108] Identifying if a product or substance may be a potential problem for the user and the likelihood of it being a problem.

[109] When a user is buying a new product online or in a shop that another user has already uploaded information about into the database, they will be able to scan the product with a camera on their electronic device or enter in the name of the product into the device manually. This will result in the electronic device calculating the percentage likelihood of them being allergic to or irritated by the product or substance and presenting to them this information.

[110] The user will be presented with two sets of information, one which is the likelihood that they are allergic to the product or substance. This will be calculated using Calculations 1 or 2 where the product or substance is substance X.

[111] The other is a list of the products or substances, constituent parts. For example ingredients listed as being contained in food products or chemicals listed as being contained in cleaning products or perfumes. Each constituent part will be listed and the likelihood the user is allergic to or may be irritated by the constituent part will be displayed beside the name of the constituent part. The likelihood of each constituent part causing the user to have a reaction will be calculated independently using Calculations 1 or 2 and then compiled into a list for the user to view.

[112] Data Entry

[113] Data can either be entered by the user manually or alternatively for ease of use the user may be able to scan the labels of products and substances that they are either interacting with or looking to see if they may have a reaction to them.

[114] Current phones are capable of scanning writing and extracting the words. Users can use this method to quickly scan a products contents and add that information to the database. They can also use this to scan the name of the product and the brand.

[115] Alternatively it may be possible for users to simply scan the barcode of a product to extract the same information.

[116] If product labels of different languages are scanned the data will automatically be translated to the language the database operates in.

[117] Confirming data entry on product or substance contents is correct if carried out manually or in the in case of incorrect scans.

[118] To confirm data entry is correct the average entry of the data will be taken and used as the contents of a product or substance. For example if 20 users scan the ingredients of product X or enter in the ingredients of product X manually and 18 users enter data stating that product X contains ingredients A, B and C while the other two users enter data saying product X contains only ingredients B and C, we will presume that product X contains ingredients A, B and C.

[119] However, this will only be true within individual countries as different countries may have products of the same name which contain different ingredients. So the contents of product X in the UK would be decided only by users that have recorded that they are based in the UK and the contents of product X in China would be determined by data entered only by users who have recorded that they are based in China.

[120] Identifying if a product contains a substance not acknowledged on the label.

[121] Many products in different areas of the world contain chemicals not listed on their label. This may cause issues when collecting user data as users may miss potential allergens or irritants when inputting information to the database about products they have interacted with. To overcome this issue we will look to see if users who are allergic to a particular substance or product, for example substance x, are allergic to a product, in this example product y, at a large scale. If z percentage of users who are allergic to x are also allergic to product y, we will presume y contains substance x for every user going forward.

[122] Percentage z must be set to be higher than the average likelihood percentage a user has of being allergic to a substance or product before they declare they are allergic to it by adding this information to the database as a personal category. For example, if the average percentage likelihood of a user being allergic to a product or substance is 5% before they declare they are allergic to it, then z must be set to be higher than 5%. If in this instance z was set to be 5% or lower it would be likely that the user was not only allergic to substance x but also product y or a constituent of product y.

[123] Incentivising users to add data to the database for their benefit and the benefit of other users.

[124] Compared to the prior art users are far more incentivised to add data to the database using this technology. This is firstly because it takes advantage of an increased amount of user data, looking at all products and substances a user may interact with, not just food, and it also considers that users symptoms may be because of irritants not allergens. This makes the data more accurate and thus more likely to help the user and convince them to keep using the app.

[125] Secondly the use of scanning technology to quickly read the ingredients, labels and names of products to add their information to the database makes it much easier to use the app. This added convenience will make users more likely to use the app and thus the data more accurate further increasing the likelihood users will use the app.

[126] Identifying allergens and irritants based on location

[127] The app can also be used to track where people are having an allergic reaction and then use that information to work out what caused the allergic reaction. We can do this by tracking the location of the user using the app / input electronic device. If the user has some sort of allergic reaction or skin or eye irritation then locations the user has been and the routes they have taken to get to them will be logged. If multiple other users have crossed paths with the user that has had some sort of reaction and have also had a reaction then the location where multiple users have had a reaction can be flagged as the potential cause of the reaction. We can then analyse the user data to see what they are allergic to and compare this with the data of all other users who have had an allergic reaction and also visited the same location or similar location. We can then work out the percentage chance of each product or substance the collection of users is allergic to being the cause of the reaction by taking the number of users allergic to each product or substance and dividing that by the total number of users who have had the allergic reaction. If the percentage chance of a product causing someone to have a reaction in a problem location reaches a threshold value, then the product that’s reached the threshold value could be added to a users list of potential reaction causes if they had a reaction and had moved through the problem location. For example, if the threshold value was set at 9% and 10 out of 100 of the users who have had a reaction and visited location X are allergic to sorbitol, we would add sorbitol to the lists of potential allergens and irritants of all users who had a reaction after moving through location X.

[128] Parameters that would have to be determined for the above calculation to be carried out would be, the exact radius from the user that would be considered to be the same location, the minimum number of users that would have to have had a reaction to something within the same location as other users for the location to be considered a problem area.

[129] This could be useful for determining what pollens people are allergic to. It could also help those who run establishments such as shops, gyms, bars and restaurants work out if customers are being caused to have an allergic reaction or irritation due to some cleaning product or scent released.

[130] This will also allow the users allergen and irritant percentage likelihoods to take into account weather, humidity, pollution warnings, UV and other location based data without the user having to input this into the database manually.

[131] This data could also be very useful when identifying occupational allergies (OA) if many people working in the same location are experiencing some kind of reaction. Occupational allergies refer to those disorders or conditions that are caused by exposure to substances in the work environment and in whose pathogenesis allergic factors are determinant.

[132] One of the main types of OA is contact dermatitis. Contact dermatitis is associated with long-term consequences as it has been shown that 30-80% of affected individuals remain symptomatic even after quitting their job.

[133] Follow-up studies of workers with OA have consistently documented that the condition is associated with a high rate of prolonged unemployment, ranging from 14% - 69% and a reduction in work-derived income in 44% - 74% of affected workers.

[134] Identifying Allergens and Irritants based on partners, family members and housemates accounts

[135] Users will be able to use the input device to link to the accounts of people who they are frequently in close proximity with such as partners, family members and housemates. When users do this the data of the products and food the collective group of users has used will be added to the list of products and substances that each user has interacted with in the last two weeks. This means if one or more of the users have an allergic reaction, they will be more likely to be able to figure out what is causing their allergic reaction as they will have more data to work with. For example if user x didn’t realise that their partner was wearing a perfume, they would not record that they interacted with that data point. However, if their partner who wore the perfume recorded that they had used it and linked accounts, then the partner who had the reaction would be able to assess the likelihood that the perfume caused the reaction. This improves the data for each user as it makes the data in the database more accurate especially if an individual user is able to better pin point the cause of their reaction and confirm it with a doctor before adding the information to the database.

[136] Another reason this improves the collection of data is because some users will be more diligent than others in recording important data to go into the database. For example, if only one of two flatmates who use the app records that their accommodation contains mould, the less diligent flatmate will still be able to identify if mould is a potential cause of an allergic reaction or irritation. The less diligent flatmate could then confirm if the mould is a problem with a doctor and if it is indeed a problem, they could add this information to the database.

[137] Identifying Allergen and Irritant information based on proximity to other users for a given amount of time

[138] An alternative way to add data about who users interact with frequently is to use their location. We can look at the proximity of users to each other and if they are within range x of each other for y amount of time over a period of z, weeks I months then we can consider the accounts to be linked without linking the accounts officially, we will term this, semi-linked. Users with a semi-linked account could be sent a virtual messages to check if they are linked to the other account, the same way facebook may suggest potential friends.

[139] To predict a suitable range for x we could look at the average proximity of users with linked accounts when they remain stationary and are nearby to each other. To predict y we can look at the average time linked users spend within x proximity of each other in a day. Finally z will be set as a threshold of days or weeks within which users are within x proximity of each other for y number of hours in a day.

[140] This location information could also improve the data produced when linked accounts are no longer in close proximity to each other. For example, if two users whose accounts are linked are not in close proximity to each other for over a day information gathered about the products and substances one of the others interact with during that time should not be added to the list of the other user, especially in the case of a potential reaction. This data could be improved by simply asking the users via a virtual message if they are in contact with the other user and how long they expect to not be in contact for.

[141] Mapping of skin infections

[142] The above description of using users location to identify potential allergens and irritants can also be applied to identifying if a user may have contracted a viral, bacterial, fungal and parasitic infection.

[143] If a user has been in contact with another user, or significant number of users who have logged that they have a virus or infection that expresses symptoms on the skin or eyes then the user logging that they have had a reaction will be notified that this may be a cause of the reaction they are experiencing and that they should see a GP.

[144] Identifying Allergens and Irritants based on genetic and epigenetic information and identifying the gene or gene expression / lack of gene expression causing the problem.

[145] It is possible to also add users genetic information to the database, this could be obtained through partnering with companies like AncestryDNA or through using a Genome sequencing service like Novogene. Alternatively we could ask users to attend a lab in which we take a sample of human DNA or ask them to post a sample of human DNA to a lab. This sample could be in the form of human tissue, saliva or blood amongst other forms.

[146] Many different platforms could be used to analyse the sample, here we will describe how the Next Generation Sequencing platform could be used.

[147] Once a sample of user DNA has been collected you must isolate the DNA from the selected sample, this can be carried out via genomic DNA isolation kits which can be bought.

[148] Then a library must be prepared. This is carried out by fragmenting the DNA into smaller pieces. You take a full genome, fragment it into many parts and add bar codes, which are indexes for single sequences of DNA, the barcode DNA sequences are already known, at the end of the barcode there is a sequence which allows for the entire fragment to be hybridised to a flow cell. This goes into a flow cell which is input into the next generation sequencing platform

[149] The Next Generation Sequencing platform used may be a nexseq or a novaseq sequencer.

[150] Then the data generated by the platform used must be analysed. First check the data is good by carrying out a fast quality check. The software for this is called Fast QC. Multiplex if needed due to sequencing different genomes. Align sequence fragments to a reference genome.

[151] At this point you know a user’s genome. This can then be added to the database to improve the data given to the user.

[152] Each gene sequence can be considered to be a personal category data point that may be related to an allergen or an irritant that could potentially be problematic to the user. The gene sequence of a particular user gene can then be added the users data profile and be used to predict the likelihood of them having a problem with a particular allergen or irritant using the same calculation as above, Calculations 1 and 2

[153] Next, we look at epigenome sequencing, this is more complex because there are different ways to carry this out. Most of the process is the same, however, additional chemical conversions have to be carried out beforehand. To detect an epigenetic mark, you would take two samples of the same genome, one is treated chemically, for example for 5-methylcytosine (5MC), one of the genome samples would be treated with sodium bisulfite. The downstream steps for both samples are then the same. Comparing the genome sample data will then tell you where the 5MC genetic mark is present.

[154] This is interesting because were 5MC is present generally 5MC correlates with inactive gene expression. You can use similar methods for other epigenetic marks.

[155] This information would be used to improve the use of the genetic information gathered above. For example, if a genetic mark such as 5MC, which indicates a gene is unlikely to be expressed, was found to be related to gene x then gene x would be removed from the user profiles personal data point category as since it is probably not expressed in the user, it is unlikely to to increase or decrease the likelihood of the user having an issue with any irritant or allergen.

[156] There are some exceptions to these rules in which epigenetic markers like 5MC do not mean a gene will not be expressed. If it is discovered that the presence of an epigenetic marker does not mean that a gene will not be expressed, then regardless of the epigenetic marker related to that specific gene it should be added to the data profile of users.

[157] Identifying Allergens or irritants based on the gut microbiome and suggesting how the gut microbiome could be improved to decrease the impact of allergens and irritants.

[158] Data on users gut microbiomes can also be gathered and added to the database. This is information on the bacteria, viruses and parasites that may live within your body. This information could be collected through working with companies such as Healthpath for example. Alternatively this data could be gathered in a lab and then added to the database.

[159] To generate this data in a lab we would first require a faecal sample from a user. This sample would undergo next generation sequencing as described above.

[160] From here we use taxonomic profiling in which the data generated using the next generation sequencing platform can be compared to a reference database of information relating to the genomic information of species that may be found in the microbiome.

[161] Statistical analysis can then be used to identify concentrations of species within the gut microbiome.

[162] This information will have to be used differently to the genomic information in the database. Here we will have personal category data points for the user which not only include species which are present in the microbiome in large amounts but also profile categories which identify when species are not present in the microbiome or are present in small amounts.

[163] Suggesting products that may work for people based on the data in the database

[164] Using the database we can suggest products users are unlikely to be allergic to that they require for day to day life. To do this we would require users to label potential products and substances that they interact with, with a product type category, a category separate from the personal data point category and the interaction category. For example, head and shoulders shampoo would be labelled shampoo and dove for men deodorant would be labelled as a deodorant. Using this data if a user said they required a shampoo, we would calculate the likelihood of the user being allergic to all items labelled shampoo in the database using Calculations 1 or 2. These products would then be displayed in a list starting with the items the user is least likely to be allergic to or irritated by and descending towards the products the user is most likely to be allergic to or irritated by.

[165] The process could be made more useful to the user by limiting the products displayed in the list to those that had been used by users in their geographic location making it more likely that the product will be available in their geographic area and that the labels of the product will be written in a language the user speaks.

[166] Identifying causes of sensitisation

[167] Once there is a significant amount of data in the database it will be possible to start making predictions about sensitising agents that may be causing allergies to develop. Over w amount of time if x amount of users who have all been in contact with substance or product y develop an allergy to z, then y may be a potential sensitising agent. Deciding when number x is significant enough to be investigated is subjective. However, if x is considered to be significant then the correlation between y and z may be useful for scientists looking to investigate sensitising agents and help to direct them with their research in order to decide if y is or is not a sensitising agent.

[168] Stress as a cause and how to measure this

[169] Stress is widely considered to be a trigger for skin reactions. However, because people often change their behaviour when stressed for example, they may drink an increased amount or alcohol or eat and increased amount of sugary foods, it is hard to say exactly what impact stress has on any individual when it comes to skin or eye irritation. To take into account users stress the app will ask them to score their daily stress level on a scale, for example between 1 and 10. The app may describe the feeling of stress level 1 and stress level 10. The average stress of the individual the week before an allergic reaction would be calculated by taking the mean average from each of the 7 days in that week and adding to their dataset under the product or substance category. This will help to identify how much of a factor stress is, in causing people to experience allergy like symptoms or skin or eye irritation. It may also identify the subjective level of stress at which people begin to experience allergy like symptoms or skin or eye irritation.

[170] Hours of sleep in relation to skin and eye allergies and irritation

[171] The database can be improved by recording the length and quality of the individuals sleep. Initially users could be asked to record the number of hours of sleep they had each night the week before an allergic reaction. The number of hours of sleep would be averaged out and rounded to the nearest whole number and added to the database as a product or substance category data point for that user. This will help to identify how much of a factor lack of sleep is in causing people to experience allergy like symptoms or skin or eye irritation. It may also identify the level of sleep at which people begin to experience allergy like symptoms or skin or eye irritation. This data may be improved by using data collected by health trackers and sleep trackers like Whoop.

[172] Considering Autoimmune Diseases, Pregnancy and Puberty

[173] If a person already knows they have an autoimmune disease, if the are pregnant or at an age where they may be undergoing puberty, they should be notified by the app that this may be the cause the of the symptoms they are experiencing. However, the app could still be useful for them to identify allergens or irritants that may exacerbate their condition.

[174] Different Languages

[175] The user will be able to select a language for the app to operate in. When information is input by the user into the database it will be translated from the users language into the language the database operates in. When information generated using the data is returned to the user it will be translated back into the language they selected to operate the app in.

[176] Those skilled in the art will appreciate that while the foregoing has described what is considered to be the best mode and where appropriate other modes of performing present techniques, the present techniques should not be limited to the specific configurations and methods disclosed in this description of the preferred embodiment Those skilled in the art will recognise that present techniques have a broad range of applications, and that the embodiments may take a wide range of modifications without departing from any inventive concept as defined in the appended claims.

Claims

1. A computer-implemented method, performed by an electronic user device, for predicting a potential allergen or irritant for a user, the method comprising:receiving, via a user interface, information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated;obtaining, in response to the received information, user personal data comprising known allergens or irritants of the user;calculating a correlation between the obtained user personal data and the potential allergen or irritant; andpredicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user.

2. The method as claimed in claim 1 wherein receiving information about a potential allergen or irritant comprises receiving information about a potential allergen or irritant comprising a single ingredient or constituent.

3. The method as claimed in claim 2 further comprising:outputting, via the user interface, the likelihood that the single ingredient or constituent causes an allergic reaction or irritation in the user.

4. The method as claimed in claim 1 wherein:receiving information about a potential allergen or irritant comprises receiving information about a potential allergen or irritant comprising at least two ingredients or constituents;calculating a correlation comprises calculating a correlation between the obtained user personal data and each of the at least two ingredients or constituents; andpredicting a likelihood comprises predicting a likelihood that each of the at least two ingredients or constituents causes an allergic reaction or irritation.

5. The method as claimed in claim 4 further comprising:outputting, via the user interface, a list of the likelihood of each of the at least two ingredients or constituents causing an allergic reaction or irritation in the user.

6. The method as claimed in claim 5 wherein the outputting comprises ordering the at least two ingredients or constituents in the list in descending order of likelihood.

7. The method as claimed in any preceding claim further comprising, prior to the receiving, requesting, via the user interface, an indication on whether the user has experienced an allergic reaction or irritation in a predefined period of time.

8. The method as claimed in claim 7, further comprising:receiving, via the user interface, an indication that the user has not experienced an allergic reaction or irritation in a predefined period of time; andrequesting, via the user interface, information about a potential allergen or irritant the user wishes to evaluate;wherein receiving information about a potential allergen or irritant for which a likelihood is to be calculated comprises receiving any one or more of: an allergen name; an irritant name; a product name; an image of a product; a product identifier; a product barcode; and a link to a product on a website.

9. The method as claimed in claim 7, further comprising:receiving, via the user interface, an indication that the user has experienced an allergic reaction or irritation in a predefined period of time; andrequesting, via the user interface, information about at least one potential allergen or irritant the user has interacted with over the predefined period of time;wherein receiving information about a potential allergen or irritant for which a likelihood is to be calculated comprises receiving at least one potential allergen or irritant the user has interacted with over the predefined period of time.

10. The method as claimed in any preceding claim wherein obtaining user personal data comprises, in addition to known allergens, obtaining any one or more of the following for the user: medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; birthplace of the user’s biological parents; birthplace of the user; geographic location of user residence; history of smoking; weight; and pregnancy status.

11. The method as claimed in any preceding claim wherein calculating a correlation betweenthe obtained user personal data and the potential allergen or irritant comprises:obtaining population data comprising personal data of a plurality of other users and their allergic reactions to or irritations caused by the potential allergen or irritant; andcorrelating the personal data of other users and their allergic reactions to or irritations caused by the potential allergen or irritant.

12. The method as claimed in claim 11 wherein obtaining population data comprises selecting a set of users with at least one characteristic in common with the user.

13. The method as claimed in claim 12 wherein the selecting comprises selecting users based on any one or more of the following characteristics: geographic location; medical history; genetic data; epigenetic data; age; sex; race; geographic location; family medical history; gut microbiome data; birthplace of the user’s biological parents; birthplace of the user; geographic location of user residence; history of smoking; weight; and pregnancy status.

14. An electronic user device for predicting a potential allergen or irritant for a user, comprising:a user interface for receiving information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated; andat least one processor coupled to memory for:obtaining, in response to the received information, user personal data comprising known allergens or irritants of the user;calculating a correlation between the obtained user personal data and the potential allergen or irritant; andpredicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user.

15. The electronic user device as claimed in claim 14 wherein the user interface comprises a display screen, and the at least one processor is arranged to output, on the display screen, the likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user.

16. The electronic user device as claimed in claim 14 or 15 further comprising:at least one image capture device for capturing any one or more of: an allergen name; an irritant name; a product name; an image of a product; a product identifier; a product barcode; and a link to a product on a website.

17. A system for predicting a potential allergen or irritant for a user, the system comprising: a server comprising storage for storing user personal data for a plurality of users of the system, the user personal data comprising known allergens or irritants of each user; anda plurality of electronic user devices used by the plurality of users, each user device of a user comprising:a user interface for receiving information about a potential allergen or irritant for which a likelihood of causing the user to experience an allergic reaction or irritation is to be calculated; anda communication module for communicating with the server;wherein the server further comprises at least one processor coupled to memory for: receiving, via the communication module of a user device, information about a product for which a likelihood is to be calculated and information identifying a user of the user device;obtaining, from the storage using the received information identifying a user, user personal data comprising known allergens or irritants of the user;calculating a correlation between the obtained user personal data and the potential allergen or irritant;predicting, using the correlation, a likelihood that the potential allergen or irritant causes an allergic reaction or irritation in the user; andtransmitting, to the user device, the predicted likelihood.

Images