A system and method for determining skin lesion scores in companion animals

Abstract

This disclosure relates to a method for determining a companion animal's skin lesion score using an evaluation module trained on pre-acquired metadata related to the companion animal, selected from a dataset having at least animal data and skin region data, the method comprising at least the steps of: receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the dataset; encoding the metadata into a metadata vector; operating the trained evaluation module on the metadata vector; generating a skin lesion score representing the companion animal's skin condition based on the evaluation module's analysis of the metadata vector; and evaluating whether the companion animal has atopic dermatitis based on the skin lesion score representing at least one skin condition of the companion animal.

Description

Description of Related Applications

[0001] This application claims the benefit of priority of European Patent Application No. 23161761.4, filed on March 14, 2023, the content of which is incorporated herein by reference in its entirety and to which priority is claimed.

Technical Field

[0002] The present disclosure relates to the health of companion animals, particularly to the study of the condition of atopic dermatitis. Various embodiments of the present disclosure broadly relate to machine learning-based techniques for monitoring the health of companion animals, and more particularly to systems and methods for determining a skin lesion score for a companion animal suspected of having the condition of atopic dermatitis.

Background Art

[0003] In most companion animals, healthy skin and hair indicate that the animal is generally in good health. Since the condition of the skin and hair of companion animals has such an important visual impact, particularly on their owners, evaluating the skin condition of animals has been an ongoing objective in the art.

[0004] Skin diseases suffered by companion animals are the single most common reason for companion animals to be taken to veterinary clinics. It is suggested that animals suffering from skin diseases account for 15% of the workload of such veterinary clinics. One of the main skin diseases contributing to this statistic is atopic dermatitis. Atopic dermatitis is a common skin disease with an inflammatory and itchy nature and a genetic predisposition. The diagnosis of atopic dermatitis can be complicated by variables such as the diversity of clinical symptoms, genetic factors, the extent of lesions, the stage of disease progression, secondary infections, and the similarity to other non-atopic skin diseases.

[0005] Accurately classifying skin lesions and effectively guiding companion animal owners can be challenging. Dermatology is a complex medical field because the skin can reflect a wide range of surgical and medical conditions. Diagnosing skin diseases typically requires clinical examinations, a time-consuming process demanding a deep understanding of dermatology. Furthermore, since only symptomatic treatment is usually prescribed for skin conditions, the underlying cause of the skin disease is often not addressed.

[0006] A smartphone application called "Atopy Index" has been developed based on the Canine Atopic Dermatitis Lesion Index (CADLI) and itchiness intensity. This application helps veterinarians easily and quickly assess the extent and severity index of canine atopic dermatitis (CADESI-4) before and after treatment. Companion animal owners can identify lesions by color-coding them on the application based on 20 different locations on the body surface and three types of lesions. Based on a validated severity scale, the application allows users to obtain cutaneous pathological indicators along with the severity index in just a few minutes, and track the total score and the progression of severity at each lesion site at each consultation. However, this application is not designed for the diagnosis of atopic dermatitis, but only as an aid for owner management.

[0007] Non-patent document 1 describes a method for evaluating the skin condition of companion animals using several criteria, including skin area data. Users need to use these criteria to recognize patterns of skin disease in patients, which vary depending on the animal breed.

[0008] Therefore, in the field of atopic dermatitis, there is still a need for improvements to existing methods and for the provision of efficient, simple, and reliable methods to help pet owners and / or veterinarians assess the skin condition of companion animals.

[0009] References to specific activities in this disclosure are for convenience only and are not intended to limit the disclosure. Those skilled in the art will recognize that the underlying concepts of the disclosed devices and methods may be used in any appropriate activity. This disclosure can be understood by referring to the following description and accompanying drawings, where identical elements are denoted by the same reference numerals.

[0010] The technical terms used below, even when used in conjunction with the detailed descriptions of specific examples herein, should be interpreted as broadly as reasonably possible. In fact, certain terms may even be emphasized below. However, any technical term intended to be interpreted restrictively is explicitly and specifically defined as such in this section of detailed descriptions. Both the general descriptions above and the detailed descriptions below are illustrative and descriptive only and do not limit the features as described in the claims. [Prior art documents] [Non-patent literature]

[0011] [Non-Patent Document 1] Favrot et al; A prospective study on the clinical features of chronic canine atopic dermatitis and its diagnosis; Vet Dermatol; 2010;21(l):23-31 [Overview of the Initiative]

[0012] According to certain aspects of this disclosure, a method and system for determining a skin lesion score in a companion animal are disclosed.

[0013] According to a first aspect, the Disclosure relates to a method for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the condition of atopic dermatitis in a companion animal, based at least on a set of pre-obtained metadata related to the companion animal, selected from a data set (A) including at least animal data and skin area data. a) A step of receiving metadata related to a companion animal, including two or more animal data and two or more skin area data listed in the data set (A), b) A step of encoding metadata into a metadata vector, c) A step of running a trained evaluation module on a metadata vector, the step of inputting the metadata vector into a predictive model, d) A step of generating a skin lesion score representing at least one skin condition of a companion animal based on the analysis of metadata vectors by a predictive model associated with the evaluation module, and e) A step of evaluating whether the companion animal has atopic dermatitis based on a skin lesion score representing at least one skin condition of the companion animal, Regarding methods that include at least the following.

[0014] By correlating animal data obtained from metadata with skin region data and using a pre-trained evaluation module, a score indicating at least one skin condition of the companion animal can be obtained. This disclosure thus enables general veterinarians to support dermatological diagnoses. This method can dermatological diagnostic hypotheses that best fit clinical symptoms, providing companion animals with the opportunity to receive more reliable and rapid diagnoses and treatments.

[0015] In the embodiment, metadata related to companion animals selected from data group (A) includes at least animal data and skin region data, - Animal data must include at least two of the following: (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering / spaying status, (vi) age, (vii) age of onset, (viii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xii) coat information, (xiii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis. - Skin region data includes at least two of the following: (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital, (viii) perioral, (ix) elbow flexor, (x) forefoot, (xi) tarsal flexor, (xii) hindfoot, and (xiii) nonspecific atopic regions.

[0016] In a particular embodiment, metadata related to a companion animal, selected from data set (A), includes at least animal data and skin region data. - Animal data may include, without limitation, other data such as sex, weight, spay / neuter status, age, physical condition, health status, habitat, coat information, activity level, and biological values ​​from biological samples, including at least (i) breed, (ii) species, (iii) lifestyle, (iv) age of onset, and (v) history of dermatitis. - Skin region data includes at least (i) the groin, (ii) the axilla, and (iii) nonspecific atopic areas, and may further include, without limitation, additional data such as the ventral chest, perineum / genitals, ventral neck, auricle, periorbital, perioral, elbow flexor, forefoot, tarsal flexor, and hindfoot.

[0017] In a particular embodiment, metadata related to a companion animal, selected from data set (A), includes at least animal data and skin region data. - Animal data must include at least (i) breed, (ii) species, (iii) lifestyle, (iv) age of onset, and (v) history of dermatitis. - Skin area data includes at least (i) the groin, (ii) the axillary region, and (iii) nonspecific atopic areas.

[0018] In certain embodiments, the method further comprises receiving an explanation of a pattern associated with a lesion of the companion animal and inputting the explanation into a prediction model. The step of generating a skin lesion score is further based on an analysis of the explanation by the prediction model.

[0019] In certain embodiments, the method further comprises calculating second-order or higher-order cross features based on the interaction between features among the metadata and inputting the second-order or higher-order cross features into a prediction model. The step of generating a skin lesion score is further based on an analysis of the second-order or higher-order cross features by the prediction model.

[0020] According to another aspect, the present disclosure provides a method of training an evaluation module to learn features representing the condition of atopic dermatitis in a companion animal, using at least a plurality of pre-acquired metadata related to the companion animal selected from a data group (A) including at least two of (i) breed, (ii) species, (iii) sex, (iv) body weight, (v) presence or absence of castration / contraception, (vi) age, (vii) age of onset, (viii) physical condition, (ix) health condition, (x) lifestyle, (xi) living environment, (xii) coat information, (xiii) activity level, (xiv) biological values from biological samples, (xv) age of onset, and (xvi) history of dermatitis, and at least including skin region data corresponding to two or more of (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitalia, (v) ventral neck, (vi) pinna, (vii) periorbital region, (viii) perioral region, (ix) flexor cubiti, (x) forefoot, (xi) flexor calcanei, (xii) hindfoot, and (xiii) non-specific atopic regions, the method comprising: - extracting at least one feature from each pre-acquired metadata; - associating at least the at least one feature with the skin condition of the animal, and - training the evaluation module to learn the association. Relates to a method including...

[0021] According to another aspect, the present disclosure is a device for determining a skin lesion score of a companion animal suspected of having the condition of atopic dermatitis, the device comprising at least (i) breed, (ii) species, (iii) sex, (iv) weight, (v) presence or absence of castration / contraception, (vi) age, (vii) age of onset, (viii) physical condition, (ix) health condition, (x) lifestyle, (xi) living environment, (xii) coat information, (xiii) activity level, (xiv) biological value from a biological sample, (xv) age of onset, and (xvi) history of dermatitis; and animal data including two or more of the foregoing, and at least (i) inguinal area, (ii) axillary area, (iii) ventral chest, (iv) perineum / genitalia, (v) ventral neck, (vi) pinna, (vii) periorbital area, (viii) perioral area, (ix) flexor cubital area, (x) forefoot, (xi) flexor calcaneal area, (xii) hindfoot, and (xiii) non-specific atopic area; and skin area data corresponding to two or more lesion or non-lesion skin areas, and comprising an evaluation module pre-trained to learn features representing the condition of atopic dermatitis in a companion animal based at least on a plurality of pre-acquired metadata related to the companion animal selected from a data group (A), the trained evaluation module operating on metadata related to the companion animal including two or more animal data and two or more skin area data listed in the data group (A) and configured to generate a skin lesion score representing at least one skin condition of the companion animal.

[0022] The present disclosure thus provides an efficient and reliable digital tool to assist veterinarians in diagnosis.

[0023] In another aspect, the disclosure includes at least two animal data, including (i) breed, (ii) species, (iii) sex, (iv) weight, (v) whether or not the animal is neutered or spayed, (vi) age, (vii) age of onset, (viii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xii) coat information, (xiii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis, and at least (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital, (viii) perioral, (ix) elbow flexor, (x) forefoot, and (xi) foot. A computer program product for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in a companion animal, based on at least multiple pre-acquired metadata related to the companion animal, selected from a data set (A) having at least skin region data corresponding to two or more lesional or non-lesional skin regions including the flexor rhizoids, (xii) hind legs, and (xiii) nonspecific atopic regions, wherein the computer program product includes a medium on which instructions readable by a processor are stored, and these instructions are a) Receive metadata related to companion animals, including two or more animal data and two or more skin region data listed in data set (A), b) Encode metadata into a metadata vector, c) Run a trained evaluation module on the metadata vector, which includes inputting the metadata vector into the predictive model. d) Based on the analysis of metadata vectors by a predictive model related to the evaluation module, generate a skin lesion score representing at least one skin condition of the companion animal. e) Based on a skin lesion score representing at least one skin condition of the companion animal, assess whether the companion animal has atopic dermatitis. This relates to computer program products that are configured in such a way.

[0024] In another aspect, the disclosure relates to a computer-readable medium containing a computer program product as defined above.

[0025] According to an exemplary embodiment, the present disclosure provides a computer implementation method for determining a skin lesion score in a companion animal, - A step of receiving metadata related to a companion animal, including two or more animal data and two or more skin region data listed in data set (A), - Steps to encode metadata into metadata vectors, - A step of running a trained evaluation module on a metadata vector, the step of inputting the metadata vector into a predictive model, - A step of generating a skin lesion score representing at least one skin condition of a companion animal based on the analysis of metadata vectors by a predictive model associated with the evaluation module. - A step of evaluating whether a companion animal has atopic dermatitis based on a skin lesion score representing at least one skin condition of the companion animal, and - A step of sending skin lesion scores and evaluations to the device, This concerns computer implementation methods, including those mentioned above.

[0026] In embodiments of the computer implementation method according to the present disclosure, the skin lesion score evaluates at least one other dermatitis condition, which may include at least one of the following: sarcoptic mange, demodectic mange, bacterial hyperthermia, Malassezia dermatitis, bacterial folliculitis, contact dermatitis, and / or mucocutaneous T-cell lymphoma.

[0027] In embodiments of the computer implementation method according to this disclosure, at least one other dermatitis condition may include at least one atopic dermatitis condition.

[0028] In the embodiments of this disclosure, the computer implementation method is - A step of reading a pre-stored skin score, - A step of comparing the pre-stored skin score with the skin lesion score, - A step to evaluate the effectiveness of previous treatments based on that comparison. It can further include

[0029] In the embodiments of this disclosure, the computer implementation method is - Steps to determine recommendations based on skin lesion scores, and - The step of sending those recommendations to the device, It can further include

[0030] In embodiments of the computer implementation method according to this disclosure, the recommendations may be at least one health recommendation, at least one nutrition recommendation, and / or at least one medical recommendation.

[0031] In embodiments of the computer implementation methods of this disclosure, at least one health recommendation may include at least one of at least one food, at least one pet service, at least one nutritional supplement, at least one ointment, at least one medicine, and / or at least one pet product.

[0032] In embodiments of the computer implementation method according to this disclosure, at least one nutritional recommendation may include at least one instruction for feeding a companion animal at least one nutritional supplement and / or at least one of at least one food.

[0033] In embodiments of the computer implementation method according to this disclosure, at least one medical recommendation may include at least one instruction for at least one ointment and / or at least one instruction for at least one drug.

[0034] In embodiments of the computer implementation method described herein, the trained evaluation module may include at least one supervised classifier machine learning model.

[0035] In the embodiment of the computer implementation method according to this disclosure, the training of the trained evaluation module is performed as follows: - A step of updating the weights of at least one supervised classifier machine learning model according to the association between the skin condition of an animal and at least one feature, This may further include:

[0036] It should be understood that both the summary above and the following explanation are illustrative and descriptive only, and do not limit the disclosed embodiments as described in the claims. [Brief explanation of the drawing]

[0037] The accompanying drawings, incorporated herein and constituting part thereof, illustrate various exemplary embodiments and, together with the description, serve to illustrate the principles of the disclosed embodiments. [Figure 1] Flowchart illustrating an exemplary method using a trained machine learning model to determine a skin lesion score in a companion animal suspected of having atopic dermatitis, according to one or more embodiments. [Figure 2] Examples of using the method by one or more embodiments, illustrating a user interface that enables data input, such as metadata for companion animals. [Figure 3] Examples of some of the most common lesion areas in dogs with atopic dermatitis in multiple breeds, according to one or more embodiments. [Figure 4] Examples of some of the most common lesion areas in dogs suffering from atopic dermatitis, according to one or more embodiments. [Figure 5] Examples of some of the most common lesion areas in cats suffering from atopic dermatitis, according to one or more embodiments. [Figure 6]An example of training the random forest classification algorithm used in the present invention. [Figure 7] Example of using the evaluation module according to the present invention [Modes for carrying out the invention]

[0038] definition In the detailed description herein, references to “embodiments,” “certain embodiments,” “one non-limiting embodiment,” and “in various embodiments” indicate that the embodiments described may include certain features, structures, or characteristics, but not all embodiments necessarily include those specific features, structures, or characteristics. Furthermore, such phrases do not necessarily refer to the same embodiments. Moreover, if certain features, structures, or characteristics are described in relation to one embodiment, it is considered within the knowledge of those skilled in the art that such features, structures, or characteristics will be affected in relation to other embodiments, whether explicitly stated or not. After reading this description, it will be clear to those skilled in the art how the disclosure may be implemented in alternative embodiments.

[0039] In general, technical terms can be understood at least partially from their use in context. For example, terms such as “and,” “or,” or “and / or” as used herein can have various meanings that may depend at least partially on the context in which they are used. Typically, when “or” is used to relate a list such as A, B, or C, it is intended to mean both A, B, and C in an inclusive sense, and A, B, or C in an exclusive sense. Furthermore, the term “one or more” as used herein can be used at least partially context-dependent to describe any feature, structure, or characteristic in a singular sense, or to describe a combination of features, structures, or characteristics in a plural sense.

[0040] Similarly, terms such as "a," "an," and "the" can sometimes mean singular use and sometimes plural use, depending at least partially on the context. Moreover, the term "based on" can be understood not necessarily to mean an exclusive set of factors, and again, depending at least partially on the context, it can allow for the presence of additional factors that are not necessarily explicitly mentioned.

[0041] The terms “having,” “including,” “containing,” and “comprising,” or similar terms, are interchangeable, and a person skilled in the art will recognize that these terms are unrestrictive. These terms are intended to imply non-exclusive inclusion; therefore, a process, method, article, or apparatus containing a list of elements does not include only those elements, but may include other elements not expressly enumerated or that are specific to the process, method, article, or apparatus.

[0042] As used herein, the terms “about” or “approximately” mean that a particular value is within an acceptable margin of error as determined by those skilled in the art, which will depend in part on how that value is measured or determined, i.e., on the limits of the measuring system.

[0043] As used herein, the terms “companion animal” or “pet” refer to animals that are not human, and may, for example, include, without limitation, companion animals. For example, companion animals may include canids, felids, dogs, cats, rabbits, hamsters, guinea pigs, rats, and / or rodents. Preferred companion animals as used herein are, without limitation, canids or felids, in particular dogs and cats, and especially dogs.

[0044] As used herein, the term “mammal” or “mammals” may include, for example, humans or animals without limitation. In particular, the term “animal” or “animals” may include, for example, ruminants, poultry, pigs, mammals, horses, mice, rats, rabbits, guinea pigs, hamsters, cattle, cattle, or dogs, preferably companion animals, i.e., cats and / or dogs.

[0045] As used herein, the term “adult” may include, for example, animals that have passed a certain age, have reached a stage of biological maturity, or have both, without limitation.

[0046] As used herein, the term “canine” refers to the following breeds: Afghan Hound, Airedale Terrier, Akita Inu, Alaskan Malamute, Basset Hound, Beagle, Belgian Shepherd, Bloodhound, Border Collie, Border Terrier, Borzoi, Boxer, Bulldog, French Bulldog, Bull Terrier, Bullmastiff, Cairn Terrier, Chihuahua, Chow Chow, Cocker Spaniel, Collie, Corgi, Dachshund, Dalmatian, Doberman, Shar Pei, English Setter, Fox Terrier, German Shepherd, Golden Retriever, Great Dane, Greyhound, Brussels Griffon, Irish Setter, Irish Wolfhound, King Charle This may include companion animals selected from recognized dog breeds (some of which are further subdivided), such as the Scudley Spaniel, Labrador Retriever, Lhasa Apso, Mastiff, Newfoundland, Old English Sheepdog, Papillon, Pekingese, Pointer, Pomeranian, Poodle, Pug, Rottweiler, Saint Bernard, Saluki, Samoyed, Schnauzer, Scottish Terrier, Shetland Sheepdog, Shih Tzu, Siberian Husky, Skye Terrier, Vizsla, Rhodesian Ridgeback, Staffordshire Terrier, Jack Terrier, Springer Spaniel, West Highland White Terrier, Whip Companion, Yorkshire Terrier, and Bichon Frise.

[0047] As used herein, the term “feline” may include, without limitation, animals including cheetahs, pumas, jaguars, leopards, lions, lynxes, ligers, tigers, panthers, bobcats, ocelots, smilodons, caracals, servals, and companion animals selected from cats. As used herein, cats include wild cats and domestic cats, most preferably domestic cats.

[0048] As used herein, “subpopulation” may include, for example, a group of one or more animals of a species, but not the entire species, without limitation. For example, a “subpopulation” can be defined in respect of one or more attributes of genotype and / or physiological status common to the members of a subpopulation of multiple members. In some embodiments, a subpopulation may be defined at least in part by a particular breed. In addition, for example, in the case of hybrid animals, a subpopulation may be defined at least in part by the pedigree background of a breed, which may be identified by knowledge of the parent breeds, phenotypic characteristics, genotypic evaluation, or genetic markers such as SNPs. In some embodiments, a subpopulation may be defined at least in part by physiological status.

[0049] As used herein, the terms “metadata” or “metadata file” may include, without limitation, data on companion animals, any one or combination of animal attributes, including, at a minimum, breed, species, activity level, medical history, reproductive status, age, sex, weight, spay / neuter status, biological values ​​from biological samples, physical condition, health status, lifestyle, habitat, coat information, or risk factors, and / or age of disease onset, history of past hotspots, urticaria, or angioedema, presence or absence of adrenal-responsive pruritus, excessive hair loss, scaling, or dryness, gastrointestinal symptoms, worsening of symptoms after grass walking, and / or a history of chronic and / or recurrent skin diseases or otitis externa.

[0050] According to this disclosure, the term “animal data” may include at least two of the following: (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering status, (vi) age, (vii) age of onset, (viii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xii) coat information, (xiii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis. In certain embodiments, animal data includes at least (i) breed, (ii) species, (iii) sex, (iv) age of onset, and (v) history of dermatitis. In some embodiments, the animal data includes at least (i) breed, (ii) species, (iii) sex, (iv) age of onset, and (v) history of dermatitis, and may further include, without limitation, other data such as sex, weight, whether or not the animal is neutered or spayed, age, physical condition, health status, habitat, coat information, activity level, and biological values ​​from biological samples.

[0051] According to this specification, “skin area data” may include at least two of the following: (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital, (viii) perioral, (ix) elbow flexor, (x) forefoot, (xi) tarsal flexor, (xii) hindfoot, and (xiii) nonspecific atopic areas.

[0052] - Skin area data includes at least (i) the groin, (ii) the axillary region, and (iii) nonspecific atopic areas, and may also include additional data such as the ventral chest, perineum / genitals, ventral neck, auricle, periorbital region, perioral region, elbow flexor region, forefoot, tarsal flexor region, and hindfoot.

[0053] According to this specification, the term "dermatological history" refers to the condition of dermatitis at that time, and means whether the dermatitis is chronic dermatitis, recurrent dermatitis, or a persistent underlying condition.

[0054] According to this specification, the term “non-specific atopic area” may include all skin areas other than (i) the groin, (ii) the axilla, (iii) the ventral chest, (iv) the perineum / genitals, (v) the ventral neck, (vi) the auricle, (vii) the periorbital region, (viii) the perioral region, (ix) the elbow flexor region, (x) the forefoot, (xi) the tarsal flexor region, or (xii) the hindfoot.

[0055] In addition, metadata may consist of answers to questions related to the aforementioned list of attributes, but is not limited to the following:

[0056] As used herein, the term “evaluation module” may include, for example, modules relating to learning models and learning algorithms that analyze data, particularly those used for classification and regression analysis, without limitation.

[0057] As used herein, the term “expert” may include, for example, any person who can identify skin lesions on the body surface of an animal, label the features associated with such lesions, determine such lesions, and / or associate them with the pathology of atopic dermatitis, other skin diseases, or other skin scars. Such an expert may, for example, be a veterinarian.

[0058] As used herein, the terms “biological sample” or “substance of biological origin” may include, for example, at least one of feces, urine, hair, blood, saliva, and tissue, without limitation.

[0059] For example, the terms “biological sample” or “bio-derived material” may refer to samples of tissue or fluid isolated from a subject, including, but not limited to, blood, plasma, serum, feces, urine, bone marrow, bile, cerebrospinal fluid, lymphatic tissue and lymph, skin samples, secretions from skin, respiratory tract, intestinal tract, and genitourinary system, tears, saliva, milk, blood cells, organs, and / or biopsy specimens. The terms “biological sample” or “bio-derived material” may also refer to samples of conditional cultures resulting from the growth of cells or tissues in culture media, such as samples of extracorporeal cell culture components containing recombinant cells or cellular components. The terms “biological sample” or “bio-derived material” may also refer to, for example, polypeptides or polynucleotides, or fragments of organisms or cells obtained from environmental sampling, such as airborne pathogens, without limitation.

[0060] As used herein, the terms “food,” “food composition,” “prescription diet,” or “foodstuff” may, for example, without limitation, refer to foodstuffs, prescription diets, dietary supplements, liquids and / or ingredients that may contain proteins, carbohydrates and / or crude fats. For example, the terms may also refer to auxiliary substances or additives, such as minerals, vitamins, and spices (see Merriam-Webster's Collegiate Dictionary, 10th edition, 1993, all of which are cited herein). Such food compositions or foods may or may not be nutritionally complete.

[0061] As used herein, “companion food” or “animal food” may include, for example, products manufactured by manufacturers of companion food, which are processed, semi-processed or unprocessed, and / or intended for consumption by companion animals after being placed on the market, in accordance with European Union (EU) Regulation 767 / 2009, without limitation.

[0062] As used herein, “training dataset” or “training data” may include one or more data sets for training a machine learning model. A training dataset may be collected via one or more client devices (e.g., by crowdsourcing) or from other sources (e.g., a database). In one non-limiting embodiment, a training dataset for pet wellness assessment may include data from both a treatment group and a control group.

[0063] Some non-limiting embodiments are described below with reference to block diagrams and operation diagrams of methods, processes, devices, and apparatus. Each block in a block diagram or operation diagram, and any combination of blocks within a block diagram or operation diagram, is understood to be implementable by analog or digital hardware and computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, dedicated computer, ASIC, or other programmable data processing device to modify its function as detailed herein, so that when executed via the processor of the computer or other programmable data processing device, the instructions perform the function / operation defined in the block diagram or operation block(s). In some alternative implementations, the functions / operations shown in the blocks may occur in a different order than that shown in the operation diagram. For example, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may be executed in reverse order depending on the functions / operations involved.

[0064] In some non-limiting embodiments, the term “server” should be understood to refer to a service point that provides processing, database, and communication functions. For example, without limitation, the term “server” may refer to a single physical processor with associated communication, data storage, and database functions, or to a network-connected or clustered group of processors such as an elastic computer cluster, and associated network storage devices, as well as operating software and one or more database systems and application software that support the services provided by the server. A server may be, for example, a cloud-based server, a cloud computing platform, or a virtual machine. While the configuration and functionality of a server can vary considerably, generally a server may include one or more central processing units and memory. A server may also include one or more mass storage devices, one or more power supplies, one or more wired or wireless network interfaces, one or more input / output interfaces, or one or more operating systems such as Windows Server, Mac OSX®, Unix, Linux®, or FreeBSD.

[0065] In some non-limiting embodiments, “network” should be understood as a network capable of connecting devices to exchange communications, such as between a server and a client device or other types of devices, including, for example, between wireless devices connected via a wireless network. Networks may also include, for example, mass storage devices such as network-attached storage (NAS), storage area networks (SANs), and other forms of computers or machine-readable media. Networks may include the internet, one or more local area networks (LANs), one or more wide area networks (WANs), wired connections, wireless connections, cellular communications, or any combination thereof. Similarly, subnetworks that may employ different architectures or conform to and be compatible with different protocols can also interoperate within larger networks. For example, various types of devices that provide interoperability for different architectures and protocols become available. As a helpful example, a router can provide links between otherwise separate and independent LANs.

[0066] Communication links or channels may include, for example, analog telephone lines such as twisted-pair wires and coaxial cables, full or partial digital lines including T1, T2, T3, or T4 type lines, Integrated Digital Networks (ISDN), Digital Subscriber Lines (DSL), wireless links including satellite links, or other communication links or channels well known to those skilled in the art. Furthermore, computer devices or other related electronic devices may be remotely connected to the network, for example, via wired or wireless lines or links.

[0067] In a non-limiting embodiment, "wireless network" should be understood as a network that connects client devices to the network. A wireless network can employ standalone ad-hoc networks, mesh networks, wireless LANs (WLANs), cellular networks, and the like. A wireless network may be configured to include systems such as terminals, gateways, and routers connected by radio link, which can move freely, move randomly, or arbitrarily self-organize, thus allowing the network topology to change, sometimes rapidly. A wireless network can further employ multiple network access technologies, including Wi-Fi, Long-Term Evolution (LTE), WLAN, wireless router (WR) mesh, or second-generation (2G), third-generation (3G), fourth-generation (4G), and fifth-generation (5G) cellular technologies. Network access technologies enable wide-area coverage for devices, such as client devices with varying degrees of mobility. For example, a network can be configured to provide radio frequency (RF) or wireless communication using one or more network access technologies such as Global System for Mobile Communications (GSM), Universal Mobile Telecommunications System (UMTS), General Packet Radio Service (GPRS), Enhanced Data GSM Environment (EDGE), 3GPP® LTE, LTE Advanced, Wideband Code Division Multiple Access (WCDMA®), Bluetooth®, and 802.11b / g / n. A wireless network may include substantially any type of wireless communication mechanism capable of transmitting signals between devices such as client devices and computer devices, between networks, or within a network. Computer devices can send and receive signals via wired or wireless networks, or process and store signals as physical memory states in memory, etc.For example, a computer device can function as a server and may include, for instance, a dedicated rack-mount server, a desktop computer, a notebook computer, a set-top box, or an integrated device combining two or more functions of the aforementioned devices. Servers can vary widely in configuration and functionality, but generally include one or more central processing units and memory. A server may also include one or more mass storage devices, one or more power supplies, one or more wired or wireless network interfaces, one or more input / output interfaces, or one or more operating systems.

[0068] As used herein, “machine learning model” generally encompasses instructions, data, and / or a model configured to receive an input and apply one or more weights, biases, classifications, or analyses to the input to produce an output. Examples of outputs include classifications of inputs, analyses based on inputs, designs, processes, predictions, or recommendations related to the input, or other appropriate types of outputs. Machine learning models are generally trained using training data, e.g., empirical data and / or samples of input data, which are fed into the model to establish, adjust, or modify one or more properties of the model, e.g., weights, biases, criteria for classification or clustering, etc. Embodiments of machine learning models can operate linearly, in parallel, through a network (e.g., an unrestricted supervised classifier machine learning model) or through any appropriate configuration.

[0069] The execution of a machine learning model may involve the deployment of one or more classifier machine learning techniques or models, such as linear models (e.g., including logistic regression), ensemble methods (e.g., including random forest classifiers), gradient boosting machines (GBMs), discriminant analysis (e.g., including partial least squares discriminant analysis), support vector classifiers, nearest neighbor methods (e.g., including K-nearest neighbor classifiers), Gaussian process classifiers, naive Bayes (e.g., including Gaussian naive Bayes classifiers), decision tree classifiers, neural network models (e.g., including multilayer perceptron classifiers), deep learning, and / or deep neural networks. Supervised training may be employed. For example, supervised training may involve providing training data and labels corresponding to the training data, such as true values. Any appropriate type of training may be used, such as probabilistic, gradient boosting, random seeding, recursive, epoch-based, or batch-based. As a use case, a machine learning model may determine a skin lesion score in a companion animal. A computer system may first be configured to receive companion animal data corresponding to the companion animal in question. For example, the companion animal data may include at least one metadata file. A trained evaluation module then analyzes or processes the companion animal data and generates a skin lesion score based on that analysis. The skin lesion score may, for example, indicate at least one skin condition of the companion animal, one of which may include at least one symptom of atopic dermatitis. The computer system may then be configured to send the skin lesion score to a device belonging to a veterinarian and / or the companion animal's caretaker. In another use case, a machine learning model, such as a trained evaluation module, may be trained to generate a skin lesion score. Training may involve utilizing training data. For example, the trained evaluation module may extract at least one feature from each of several pre-acquired metadata files. The trained evaluation module may then associate at least one feature with the companion animal's skin condition.This process may continue until an association is made with each of the previously acquired metadata. The trained evaluation module may then be further trained based on such associations.

[0070] While the above examples relate to skin lesions and the pathology of atopic dermatitis, it should be understood that the techniques described herein are applicable to the health analysis of any appropriate type of companion animal. Furthermore, it should be understood that the above examples are merely illustrative. The techniques and technologies described herein are applicable to any appropriate activity.

[0071] Method for determining skin lesion scores in companion animals In one aspect, the Disclosure relates to a method for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the condition of atopic dermatitis in a companion animal, based at least on a set of pre-acquired metadata related to the companion animal, selected from a dataset (A) having at least animal data and skin region data. a) A step of receiving metadata related to a companion animal, including two or more animal data and two or more skin area data listed in the data set (A), b) A step of running a trained evaluation module on the metadata received in step a), and c) A step of generating a skin lesion score representing at least one skin condition of the companion animal based on the evaluation module, Regarding methods that include at least the following.

[0072] In one embodiment, the companion animal is a canid or a feline.

[0073] In one embodiment, the companion animal is a dog or a cat.

[0074] In one embodiment, the companion animal is a dog.

[0075] Metadata related to companion animals In one embodiment, the metadata may include information about the animal's breed, its lifestyle, in particular whether it is primarily an indoor animal, the age at which atopic dermatitis develops, the fact that the condition has a chronic, recurrent, or permanent background, and information about at least one affected skin area, in particular the groin, axilla, or another area.

[0076] The data set (A) may further include information on reproductive status, risk factors, and / or at least one lesion area on the body surface of the companion animal, whether or not there is a history of past hotspot occurrences, a history of urticaria or angioedema, whether or not there is adrenocortical reactive pruritus, excessive hair loss, scaling or dryness, gastrointestinal symptoms, whether or not symptoms worsen after grass walking, or a history of chronic and / or recurrent skin disease or otitis externa.

[0077] Preferably, the skin region data can be selected from the head, particularly the ears, auricles, periocular region, perioral region and / or periocular region; the legs, particularly the forefoot, hindfoot and / or interdigital region, elbow flexor muscles, tarsal joint flexor surface and / or carpal joint extensor surface; and the torso, particularly the groin, axilla, chest / abdomen or neck, groin, axilla and / or perineal / genital region.

[0078] Thus, metadata can conveniently derive from medical history and animal characteristics. Using them in the method according to this disclosure can increase the specificity and sensitivity of the method. Metadata may consist of answers to questions asked by the owner and / or veterinarian.

[0079] The methods described herein may include the step of providing a pathological profile of a companion animal based on multiple metadata. In some embodiments, the pathological profile may include a readily searchable, centrally managed information record containing multiple metadata of the companion animal.

[0080] Evaluation module In a particular embodiment, the evaluation module can use a predictive model.

[0081] In a particular embodiment, the prediction model may include a supervised classifier machine learning model and a metadata coding module.

[0082] In a preferred embodiment, the model is a random forest classification algorithm trained as described below and shown in Figure 6. In such an embodiment, the entire dataset is divided into a training set and a test set. -Step 1: Select a random sample from the given data or training set. - Step 2: Build a decision tree for all training data. -Step 3: Perform voting by averaging the decision tree or by majority vote of the decision tree. - Step 4: Validate the model performance using the test set.

[0083] Other known classification algorithms may be used.

[0084] In other embodiments, the prediction model may include one or more neural networks.

[0085] In certain embodiments, training and test sets are used, but the predictive model can also be trained using a cross-validation method, which can be configured from 5 to a maximum of 600 (single-value cross-validation).

[0086] The metadata encoding module can be advantageously configured to convert its metadata, in particular, the answers to predefined questions, into a category vector. For a list of questions having a predefined set of answers, the metadata encoding module can advantageously use a binarization function to assign either 1 or 0 to each possible answer to generate a metadata vector.

[0087] In a particular embodiment, the evaluation module may further include an evaluation supervised classifier machine learning model.

[0088] The aforementioned supervised evaluation classifier machine learning model can advantageously utilize its metadata vector to generate a score representing the skin condition of a companion animal.

[0089] Each of these vectors can be associated with a companion animal and its known skin condition.

[0090] The supervised classifier machine learning model described above can be refined for training to minimize errors and generate scores relevant to test data or new received data. For example, the model may be further refined by adding other features not selected for the model.

[0091] The pattern of an animal's skin condition may be a variable dependent on the animal's breed. In certain embodiments, the predictive model can use a description of the disease pattern and the breeds susceptible to that disease. The description of the disease pattern and the breeds susceptible to that disease may be two independent variables. If one has an effect on the other, and such an effect is recorded in a database, the use of a random forest, which can model nonlinearity, would be helpful in modeling this effect. As a further improvement, one could consider adding cubic or quartic cross-feature interactions to generate new variables, which could then be used as additional inputs to the predictive model. In some embodiments, the computer implementation method may further include a step of providing 2nd or higher cross-feature interactions computed from selected features. In some embodiments of the computer implementation method, the cross-feature interactions may include cubic cross-feature interactions.

[0092] Skin lesion score The score may be a probability or percentage indicating a level of confidence that the lesion presents as a condition of atopic dermatitis, such as low, medium, or high.

[0093] In the embodiment, the score may be in the form of a numerical value, for example, 0 (no symptoms of dermatitis) or 1 (symptoms of dermatitis).

[0094] In another embodiment, the score may be in numerical form, for example, a value between 0 and 10. In another embodiment, the score may be in letter form, in particular, indicating that the skin condition of the companion animal is considered to belong to a certain type of skin condition, for example, group A, group B, group C, etc.

[0095] In yet another embodiment, the score may be a class indicating the skin condition of the companion animal, or a class indicating the pathological condition of dermatitis in the companion animal (atopic or other), or a combination thereof, for example, a class and the probability that the animal belongs to that class.

[0096] The method according to this disclosure may further include a step of providing a user interface with a score for the companion animal generated in step c).

[0097] The score can be communicated to the user by any appropriate means, such as being displayed on an electronic device screen, printed, or via speech synthesis.

[0098] The aforementioned score can be used as an input value for another program and / or may be combined with other information, such as clinical and / or biological data.

[0099] The method described herein may suggest a complementary analysis to further refine the score. For example, the score could be improved by adding new questions / input items. It may also provide suggestions for optimal dietary recommendations related to each score.

[0100] For companion animals, the score generated in step c) can further evaluate changes in the severity of the atopic dermatitis condition. In certain embodiments, to evaluate the changes in the severity of the atopic dermatitis condition, the size of each lesion (e.g., in millimeters) can be added, which can be normalized by the size of the pet, so that the proportion of the pet's body surface occupied by the lesions can be evaluated.

[0101] In an embodiment, if the score generated in step c) for the companion animal indicates that at least one skin lesion does not indicate atopic dermatitis in that companion animal, the score evaluates the condition of another dermatitis, such as scabies, demodicosis, bacterial hyperproliferation syndrome, Malassezia dermatitis, bacterial folliculitis, contact dermatitis, or mucocutaneous T-cell lymphoma. In such embodiments, different features and / or clinical criteria than those used for atopic dermatitis may be used to best describe the condition of the other dermatitis.

[0102] In embodiments where past scores associated with the same companion animal are stored, the method may further include a step of comparing the past scores with the scores generated in step c) in order to evaluate the therapeutic effect of the treatment adopted. According to another embodiment, the Disclosure provides a method for evaluating the nature of skin lesions in a companion animal suspected of having atopic dermatitis, using at least an evaluation module pre-trained to learn features representing the pathogenesis of atopic dermatitis in a companion animal, based at least on a plurality of pre-acquired metadata associated with the companion animal, selected from a data set (A) including at least animal data and skin region data, a) A step of receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data set (A), b) The step of operating the trained evaluation module on the metadata, and c) A step of generating a score that, based on the evaluation module, evaluates whether the metadata indicates a condition of atopic dermatitis in the companion animal, Regarding methods that include at least the following.

[0103] A method for training an evaluation module to learn the characteristics that represent the pathology of atopic dermatitis in companion animals. In another aspect, the Disclosure provides a method for training an evaluation module to learn features representing the pathology of atopic dermatitis in companion animals, using at least a plurality of pre-acquired metadata related to companion animals, selected from a dataset (A) including at least animal data and skin region data, - A step of extracting at least one feature from each previously acquired metadata, - A step of relating at least one of its features to the skin condition of an animal, - A step to train the evaluation module to learn that association, Regarding methods including

[0104] In certain embodiments, with respect to learning, not all data may be from companion animals with skin lesions; some data may be from animals without skin lesions, and some may be from animals with lesions due to diseases other than atopic dermatitis. Preferably, the global learning dataset may include data from animals with atopic dermatitis, animals with other skin diseases, and healthy animals.

[0105] In one embodiment, the evaluation module may include at least one supervised classifier machine learning model, and the method includes the step of updating the weights of the supervised classifier machine learning model according to the association between the at least one feature and the skin condition of the animal.

[0106] The evaluation module can be configured to assign weights to each input during the learning phase, and weight optimization is performed, in particular, by a "stochastic gradient descent" type solver.

[0107] In one embodiment, to improve the accuracy and reliability of the method according to the present disclosure, the features of atopic dermatitis for reference for future iterations of the method can be modified using a correlation between at least one feature and the skin condition of an animal.

[0108] The features previously described for determining skin lesion scores can be applied to the method for training the evaluation module, and vice versa.

[0109] Devices, computer implementation methods, and media The provided method may be a computer implementation method.

[0110] Therefore, in one embodiment, a method for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, and / or a method for training an evaluation module to learn features representing atopic dermatitis in a companion animal, can be performed either offline, i.e., uncontrolled by a device such as a computer-assisted system, or online, i.e., controlled by a computer-assisted system, for example, a system having means adapted to perform the steps of the method, including a device suitable for determining a skin lesion score of a companion animal suspected of having atopic dermatitis; or both offline and online.

[0111] Therefore, according to one embodiment, the present disclosure relates to a device for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, the device comprising an evaluation module pre-trained to learn features representing atopic dermatitis in a companion animal, based at least on a plurality of pre-acquired metadata relating to the companion animal, selected from a data set (A) including at least animal data and skin region data, The trained evaluation module is configured to operate on metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data set (A), and to generate a skin lesion score representing at least one skin condition of the companion animal.

[0112] The device may include an acquisition module for obtaining the metadata.

[0113] In another aspect, the disclosure relates to a device for training an evaluation module, by a method for training the evaluation module to learn features representing the pathological condition of atopic dermatitis in companion animals, as described above.

[0114] In another embodiment, the Disclosure relates to a computer program product for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the condition of atopic dermatitis in a companion animal, based at least on a set of pre-acquired metadata about the companion animal selected from a data set (A) including at least animal data and skin region data. The computer program product includes a medium on which instructions that can be read by the processor are stored, and these instructions are, a) Receive metadata related to companion animals, including two or more animal data and two or more skin region data listed in data set (A), b) Run the trained evaluation module on the metadata received in step a), c) Based on the evaluation module, generate a skin lesion score representing at least one skin condition of the companion animal. It is structured in this way.

[0115] In another embodiment, the present disclosure relates to a computer program product for training an evaluation module, in accordance with a method for training the evaluation module to learn features representing the pathological condition of atopic dermatitis in companion animals, as described above.

[0116] In another embodiment, the disclosure relates to a computer-readable medium storing one or both of the computer program products described above. Such a computer-readable medium may include, or may consist of, a physical embodiment of a data set containing one or more datasets that can be configured into one or more databases. This medium may, therefore, include, or may consist of a medium on which such data can be stored. Such a computer-readable medium may also include, or may consist of, multiple media, however, in such cases the media may be functionally linked.

[0117] The computer-aided systems of this disclosure may include one or more user interfaces that enable the input of input data. Such input data may include, for example, metadata related to companion animals.

[0118] All embodiments of this disclosure and the functional operations described herein may be implemented in computer software, firmware, hardware, or one or more combinations thereof, including digital electronic circuits, or structures disclosed herein and their structural equivalents. Embodiments of this disclosure may be implemented as one or more modules of computer program instructions encoded in a computer-readable medium to be executed by or control the operation of one or more computer program products, i.e., data processing devices.

[0119] Computer-readable media may be machine-readable memory devices, machine-readable memory boards, memory devices, material compositions that generate machine-readable propagating signals, or one or more combinations thereof. The term “data processing device” encompasses any device, apparatus, and machine for data processing, including, for example, a programmable processor, a computer, or multiple processors or computers. In addition to hardware, such apparatus may include code that constitutes the execution environment of the computer program, such as processor firmware, a protocol stack, a database management system, an operating system, or one or more combinations thereof. A propagating signal is an artificially generated signal produced to encode information for transmission to a suitable receiving device, such as a machine-generated electrical signal, an optical signal, or an electromagnetic signal.

[0120] Computer programs (also known as programs, software, software applications, scripts, or code) can be written in any form of programming language, including compiled and interpreted languages, and can be deployed in any form, including as standalone programs or as modules, components, subroutines, or other units suitable for use in a computer environment. Computer programs do not necessarily correspond to files in a file system. A program may be part of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), a single file dedicated to the program, or multiple collaborative files (e.g., a file that stores one or more modules, subprograms, or parts of code). Computer programs may be deployed to run on a single computer, or on multiple computers located in the same location or distributed across multiple locations and interconnected by a communication network.

[0121] Computers can be incorporated into other devices, such as mobile phones, personal digital assistants (PDAs), portable audio players, and Global Positioning System (GPS) receivers, to name a few. Computer-readable media suitable for storing computer program instructions and data include, for example, semiconductor memory devices such as EPROMs, EEPROMs, and flash memory elements; magnetic disks such as internal hard disks or removable disks; magneto-optical disks; and all forms of non-volatile memory, media, and storage devices, including CD-ROMs and DVD-ROMs.

[0122] To enable interaction with the user, embodiments of this disclosure can be implemented in a computer or smartphone equipped with a display device for displaying information to the user, and a keyboard and a pointing device or haptic device to which the user can provide input to the computer.

[0123] Embodiments of the present disclosure can be implemented in a computer system that includes, for example, a backend component as a data server, or a middleware component, such as an application server, or a frontend component, such as a graphical user interface or a client computer with a web browser on which a user can interact with an implementation of the disclosure, or one or more such combinations of backend, middleend, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication, such as a communication network. Examples of communication networks include local area networks (LANs) and wide area networks ("WANs") such as the Internet.

[0124] According to a seventh aspect, the present disclosure relates to a computer implementation method for determining a skin lesion score for a companion animal suspected of having atopic dermatitis, the method comprising the steps of receiving at least one metadata related to the companion animal, determining a skin lesion score for the companion animal based on this data, and determining recommendations for the pet owner based on the score. In a particular embodiment, the recommendations are nutritional recommendations. In another particular embodiment, the method may further include the step of transmitting the recommendations to the pet owner's portable device.

[0125] Based on the score, recommendations can be determined and sent to pet owners, veterinarians, researchers, and / or any combination thereof. Examples of recommendations include one or more health recommendations to prevent a companion animal from developing one or more diseases, conditions, or illnesses, and / or any combination thereof. Recommendations may include, for example, one or more of the following: foods, pet services, nutritional supplements, ointments, medications to improve pet wellness or health, pet products, and / or any combination thereof. In other words, recommendations may be nutritional recommendations. In some embodiments, nutritional recommendations may include instructions to give a companion animal one or more of the following: chewable products, nutritional supplements, foods, and / or any combination thereof. In some embodiments, recommendations may be medical recommendations. For example, medical recommendations may include instructions to apply a softening agent to a companion animal, instructions to administer one or more medications to a companion animal, and / or instructions to provide one or more medications to a companion animal. The term “pet product” may include, for example, any product, service, or device of any kind designed, manufactured, and / or intended for use by companion animals, without limitation. For example, a pet product could be toys, chew toys, food, clothing, collars, medicines, health monitoring devices, location tracking devices, and / or any combination thereof. Another example could include pet genetic testing or DNA testing services. The term “pet owner” may include any individual, organization, and / or group of individuals who own and / or are responsible for any aspect of the care of a companion animal. [Examples]

[0126] Various steps in an example of the method according to this disclosure for determining a skin lesion score in companion animals suspected of having atopic dermatitis are shown in Figure 1 and described in detail below.

[0127] This method uses an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in companion animals, based on at least multiple pre-acquired metadata related to companion animals. This metadata is selected from a dataset (A) having at least animal data and skin region data, and in this example, - Animal data must include at least (i) breed, (ii) species, (iii) lifestyle, (iv) age of onset, and (v) history of dermatitis. - Skin area data includes at least (i) the groin, (ii) the axillary region, and (iii) nonspecific atopic areas.

[0128] In step 10, companion animal data corresponding to the companion animal is received. This companion animal data includes multiple metadata related to the companion animal data, including two or more animal data and two or more skin area data listed in data set (A) as described above.

[0129] During step 11, the trained assessment module operates on at least one metadata of the companion animal. During step 12, a score representing the skin condition of the companion animal is generated based on that assessment module.

[0130] A more reliable assessment can be made by taking several optional steps, such as using more data, more tests, or more questions.

[0131] Figure 2 shows an example of using the method described herein with a smartphone application.

[0132] In this example, users of the method described herein are asked to identify their dog's breed from a list of breeds with a specific predisposition. Figure 2 shows a user interface in which the animal owner can input several pieces of data, namely metadata related to the companion animal, such as the known breed, how long ago the suspected dermatitis began, whether it is an indoor or outdoor living environment, whether the dermatitis is chronic or recurrent, or whether it is a persistent underlying condition. In this example, four animal data sets are used in this manner.

[0133] Pet owners are also asked to select the areas of the body surface that show the most characteristic lesions, such as the head, armpits, groin, or legs. As described earlier, all of this input data can be used to generate a score representing the animal's skin condition, which may be displayed on a subsequent interface. In this example, the user is asked to select areas from a given list of 12 areas to form skin area data.

[0134] The performance obtained in this example, which includes data from four animals and 15 skin regions, is summarized in the table below.

[0135] [Table 1]

[0136] In a preferred embodiment, the evaluation module uses a predictive model including a supervised classifier machine learning model and a metadata coding module, as shown in Figure 7.

[0137] For training the supervised classifier machine learning model in the evaluation module, at least 100 companion animals with skin lesions will be used, including healthy animals and animals with characteristics other than atopic dermatitis. Selection criteria can include different ages, behavioral patterns, all coat colors, age-related variations, and all hair densities and compositions (primary, intermediate, and secondary hairs). To build the most important database, the animal distribution needs to be as broad as possible.

[0138] In the illustrated example, the metadata encoding module encodes metadata, i.e., question q. i Answer to A i (n) is a category vector a1(1)....a i It is configured to convert to (n), where i is the number of questions and n is the number of predefined answers.

[0139] In the illustrated example, the evaluation supervised classifier machine learning model uses metadata vectors from the metadata coding module to generate scores representing the skin conditions of companion animals, associated with known animal skin conditions, based on pre-trained vectors.

[0140] Figure 3 shows some common areas in dogs prone to atopic dermatitis, categorized by several common breeds. It can be seen that breed influences the areas affected by the disease.

[0141] Figure 4 shows some of the most common areas in dogs suffering from atopic dermatitis.

[0142] Figure 5 shows some of the most common areas in cats suffering from atopic dermatitis.

[0143] Thus, the areas most commonly affected are generally the head, especially the ears, around the mouth and / or around the eyes; the limbs, especially the fore feet and / or interdigital regions, the flexor surfaces of the tarsal joints and / or extensor surfaces of the carpal joints; and the trunk, especially the groin, armpits, ventral and / or perineum.

[0144] In the above description relating to the exemplary embodiments of the Disclosure, it should be noted that, for the purpose of simplifying the Disclosure and making one or more of the various embodiments of the Invention easier to understand, various features of the Disclosure may be summarized in a single embodiment, drawing, or description thereof. However, this method of the Disclosure should not be interpreted as reflecting the intent of the Disclosure to include features beyond those expressly described in each claim. Rather, as reflected in the following claims, embodiments of the Invention consist of fewer features than all the features of a single previously disclosed embodiment combined. Therefore, the claims following the Detailed Description are expressly incorporated into that Detailed Description, and each claim stands alone as an independent embodiment of the Disclosure.

[0145] Furthermore, while some embodiments described herein include some features included in other embodiments but do not include others, combinations of features from different embodiments are intended to form different embodiments, as will be understood by those skilled in the art, within the scope of the disclosure. For example, in the following claims, any of the embodiments described in the claims may be used in any combination.

[0146] Therefore, although specific embodiments have been described, those skilled in the art will recognize that other modifications and further modifications can be made thereto without departing from the spirit of the disclosure. It is intended that all such changes and modifications fall within the scope of the disclosure. For example, functions can be added or removed from the block diagram, or operations can be swapped between function blocks. Steps can also be added or removed from the methods described within the scope of the disclosure.

[0147] The subject matter disclosed herein is illustrative and not limiting, and the accompanying claims are intended to encompass all modifications, improvements, and other practices that fall within the true spirit and scope of this disclosure. Therefore, to the maximum extent permitted by law, the scope of this disclosure should be determined by the broadest permissible interpretation of the following claims and their equivalents, and should not be limited or restricted by the foregoing detailed description. While various practices of this disclosure have been described, it will be apparent to those skilled in the art that many more practices are possible within the scope of this disclosure. Therefore, this disclosure should be limited only in light of the accompanying claims and their equivalents.

Claims

1. Animal data including at least two of the following: (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering / spaying status, (vi) age, (vii) age of onset, (viiii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xii) coat information, (iiii) activity level, (xiv) biological values ​​from biological samples, (xv) age of onset, and (xvii) history of dermatitis, and at least two of the following: (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital area. A method for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in a companion animal, based on at least multiple pre-acquired metadata related to the companion animal, selected from a data set (A) which includes at least skin region data corresponding to lesion or non-lesional skin regions, including two or more of the following: (viiii) perioral region, (ix) elbow flexor region, (x) forefoot, (xi) tarsal flexor region, (xi) hindfoot, and (xiiii) nonspecific atopic regions. a) A step of receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data group (A), b) A step of encoding the metadata into a metadata vector, c) A step of running the trained evaluation module on the metadata vector, which includes inputting the metadata vector into a predictive model. d) A step of generating a skin lesion score representing at least one skin condition of the companion animal based on the analysis of the metadata vector by the predictive model associated with the evaluation module, and e) A step of evaluating whether the companion animal has atopic dermatitis based on the skin lesion score representing at least one skin condition of the companion animal, A method that includes at least [something].

2. The steps include receiving a description of a pattern associated with the lesions of the companion animal, and A step of inputting the above description into the prediction model, It further includes, The method according to claim 1, wherein the step of generating the skin lesion score is further based on the analysis of the description by the predictive model.

3. A step of calculating a second-order or higher cross-features based on the interaction between the features of the metadata, and A step of inputting the above-order or higher cross-features into the prediction model, It further includes, The method according to claim 1 or 2, wherein the step of generating the skin lesion score is further based on the analysis of the second-order or higher cross-features by the predictive model.

4. The method according to any one of claims 1 to 3, wherein the data set (A) further includes information on reproductive status, risk factors, and / or information on at least one lesion area on the body surface of the companion animal, whether or not there is a history of past hotspot occurrences, a history of urticaria or angioedema, whether or not there is adrenocortical reactive pruritus, excessive hair loss, scaling or dryness, gastrointestinal symptoms, whether or not symptoms worsen after walking in grassland, or a history of chronic and / or recurrent skin disease or otitis externa.

5. The method according to any one of claims 1 to 4, further comprising the step of providing a pathological profile of the companion animal based on the plurality of metadata, wherein the pathological profile includes an information record containing the plurality of metadata of the companion animal.

6. The method according to any one of claims 1 to 5, wherein the prediction model includes at least a supervised classifier machine learning model.

7. The method according to any one of claims 1 to 6, further comprising the step of providing the skin lesion score for the companion animal to the user interface.

8. A step of determining the size of each lesion associated with the companion animal, A step of normalizing the respective sizes associated with each lesion by the size of the companion animal, and A step of evaluating the severity of the atopic dermatitis condition based on the skin lesion score and the normalized size associated with each lesion, The method according to any one of claims 1 to 7, further comprising:

9. Animal data including at least two of the following: (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering / spaying status, (vi) age, (vii) age of onset, (viiii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xi) coat information, (xiiii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis, and at least two of the following: (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, ( A method for training an evaluation module to learn features representing the pathology of atopic dermatitis in a companion animal, using at least a plurality of pre-acquired metadata related to a companion animal, selected from a data set (A) which includes at least skin region data corresponding to lesional or non-lesional skin regions, including two or more of the following: (v) ventral neck, (vi) auricle, (vii) periorbital, (viiii) perioral, (ix) elbow flexor region, (x) forefoot, (xi) tarsal flexor region, (xi) hindfoot, and (xiiii) nonspecific atopic regions, - A step of extracting at least one feature from each previously acquired metadata, - A step of associating at least one of the above features with the skin condition of an animal, - A step of training the evaluation module to learn the relationship, A method that includes this.

10. The method according to claim 9, wherein the evaluation module includes at least one supervised classifier machine learning model, and the method includes updating the weights of the supervised classifier machine learning model in accordance with the association between the at least one feature and the skin condition of the animal.

11. A device for determining a skin lesion score in a companion animal suspected of having atopic dermatitis, the device comprising at least two animal data including (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering status, (vi) age, (vii) age of onset, (viiii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xi) coat information, (iii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis, and at least (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital, (viiii) perioral, (ix) flexor elbow, ( A device comprising an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in a companion animal based on at least a plurality of pre-acquired metadata related to the companion animal, selected from a data set (A) which includes at least skin region data corresponding to lesion or non-lesional skin regions, including two or more of the forelimbs, (xi) tarsal flexor regions, (xi) hindlimbs, and (xiiii) nonspecific atopic regions, wherein the trained evaluation module operates on data of the companion animal which includes a plurality of metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data set (A), and is configured to generate a skin lesion score representing at least one skin condition of the companion animal.

12. The device according to claim 11, further comprising an acquisition module for acquiring data of the companion animal.

13. Animal data including at least two of the following: (i) breed, (ii) species, (iii) sex, (iv) weight, (v) neutering / spaying status, (vi) age, (vii) age of onset, (viiii) physical condition, (ix) health status, (x) lifestyle, (xi) habitat, (xi) coat information, (iii) activity level, (xiv) biological values ​​from a biological sample, (xv) age of onset, and (xvi) history of dermatitis, and at least two of the following: (i) groin, (ii) axilla, (iii) ventral chest, (iv) perineum / genitals, (v) ventral neck, (vi) auricle, (vii) periorbital, (viiii) perioral, (ix) elbow flexor, (x) forelimb, (xi) tarsal flexor, ( A computer program product for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in a companion animal, based on at least multiple pre-acquired metadata related to the companion animal, selected from a data set (A) which includes at least skin region data corresponding to lesion or non-lesional skin regions, including two or more nonspecific atopic regions, wherein the computer program product includes a medium on which instructions readable by a processor are stored, and these instructions are a) Receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data group (A), b) Encode the metadata into a metadata vector, c) Run the trained evaluation module on the metadata vector, which includes inputting the metadata vector into a predictive model. d) Based on the analysis of the metadata vector by the prediction model associated with the evaluation module, generate a skin lesion score representing at least one skin condition of the companion animal. e) Based on the skin lesion score representing at least one skin condition of the companion animal, evaluate whether the companion animal has atopic dermatitis. A computer program product that is configured in such a way.

14. A method for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the pathological condition of atopic dermatitis in a companion animal, based on at least multiple pre-acquired metadata related to the companion animal, selected from a dataset (A) including at least animal data and skin region data, a) A step of receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data group (A), b) the step of operating the trained evaluation module on the metadata received in step a), and c) A step of generating a skin lesion score representing at least one skin condition of the companion animal based on the evaluation module, A method that includes at least [something].

15. A method for training an evaluation module to learn features representing the pathology of atopic dermatitis in companion animals, using at least a plurality of pre-acquired metadata related to companion animals, selected from a dataset (A) which includes at least animal data and skin region data, - A step of extracting at least one feature from each previously acquired metadata, - A step of associating at least one of the above features with the skin condition of an animal, - A step of training the evaluation module to learn the relationship, A method that includes this.

16. A device for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, the device comprising an evaluation module pre-trained to learn features representing atopic dermatitis in a companion animal based on at least a plurality of pre-acquired metadata related to the companion animal, selected from a data set (A) including at least animal data and skin region data, the trained evaluation module is configured to operate on data of the companion animal including a plurality of metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data set (A), and to generate a skin lesion score representing at least one skin condition of the companion animal.

17. A computer program product for determining a skin lesion score of a companion animal suspected of having atopic dermatitis, using an evaluation module pre-trained to learn features representing the pathology of atopic dermatitis in a companion animal, based on at least multiple pre-acquired metadata related to the companion animal selected from a dataset (A) including at least animal data and skin region data, wherein the computer program product includes a medium on which instructions readable by a processor are stored, and these instructions are a) Receiving metadata related to the companion animal, including two or more animal data and two or more skin region data listed in the data group (A), b) Operate the trained evaluation module on the metadata received in step a), c) Based on the evaluation module, generate a skin lesion score representing at least one skin condition of the companion animal. A computer program product that is configured in such a way.

Images