Stable dog itching model construction method and multi-dimensional evaluation method

Abstract

The invention provides a stable dog itching model construction method and a multi-dimensional evaluation method. The construction method comprises the following steps: S1, screening experimental dogs and recording basic information; s2, carrying out environment adaptation treatment on the experimental dogs; s3, preparing a sensitizer and configuring the concentration; s4, performing sensitization treatment on the experimental dogs for the first time; s5, skin reaction monitoring and data recording after first-time sensitization; s6, performing enhanced sensitization treatment on the experimental dogs; s7, strengthening preliminary verification of a post-sensitization itching model; s8, inducing pruritus behaviors and quantitatively recording the pruritus behaviors; s9, continuously monitoring the stability of the model; and S10, obtaining the stable dog itching model. According to the method, a standardized operation guide from experimental design to result judgment is provided, the technical threshold of model construction and evaluation is reduced, and the comparability and repeatability of research data are improved.

Description

Technical Field

[0001] This invention relates to the field of veterinary experimental animal model construction technology, and in particular to a method for constructing a stable canine pruritus model and a multi-dimensional evaluation method. Background Technology

[0002] Pruritus in dogs is a core symptom of canine skin diseases (such as allergic dermatitis and parasitic skin diseases) and an important animal model for studying pruritus diseases in humans, such as atopic dermatitis. Due to the high similarity between canine and human skin structure and immune response mechanisms, canine pruritus models have become a key tool in the biomedical field for studying the pathogenesis of pruritus and developing anti-pruritus drugs. However, existing techniques for constructing and evaluating canine pruritus models have significant shortcomings, mainly in the following three aspects:

[0003] Poor stability of model construction: Traditional construction methods lack standardized procedures. For example, the selection of experimental dogs only focuses on the basic health status and does not specify a unified range for breed and age. The selection of sensitizers is limited (mostly only a single protein sensitizer is used), and does not cover common clinical causes such as chemical and natural allergens. There is no strict environmental adaptation period or enhanced sensitization process during construction, which leads to large fluctuations in the sensitization effect of the model. Some models have problems such as short-term itching followed by rapid subsidence or no obvious itching behavior, which cannot meet the needs of long-term experiments.

[0004] The evaluation of models is limited in scope: existing evaluations often focus on a single indicator, such as judging the effectiveness of a model by observing skin redness and swelling, or only counting the frequency of itching behavior. They do not connect the physiological and pathological basis with the inherent logic of behavioral performance. Although some models show itching behavior, serum immune indicators (such as specific IgE) are not significantly elevated, or there is insufficient infiltration of skin inflammatory cells (such as mast cells), resulting in a lack of reliable biological mechanism support for the model, and low credibility of subsequent drug evaluation or mechanism research results.

[0005] Lack of scenario adaptability: Traditional technologies do not take into account the differences in different research needs. For example, models used for drug efficacy evaluation need to have stable behavior and clear physiological targets at the same time, while models used for mechanism research need to highlight the traceability of pathological indicators. However, the existing evaluation system has not formulated differentiated standards for different scenarios, which leads to the model being unable to accurately match the research needs, resulting in the waste of resources that the model is qualified but cannot be used in the research. Summary of the Invention

[0006] This invention provides a method for constructing a stable canine pruritus model and a multi-dimensional evaluation method. It constructs a stable canine pruritus model with clear immune response, persistent pathological manifestations, and stable pruritus behavior, solving the problems of large fluctuations in sensitization effect and poor behavioral repeatability of traditional models. It establishes a multi-dimensional evaluation system covering physiological and pathological basis, behavioral stability, and application adaptability, avoiding the one-sidedness of single-indicator evaluation and clarifying the model's adaptability value for different research scenarios.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for constructing a stable canine pruritus model includes the following steps:

[0009] S1. Select healthy experimental dogs, record their basic information, and form an experimental dog basic information table;

[0010] S2. Transfer the experimental dogs selected in S1 to a standardized experimental kennel for 7 days to adapt. After the adaptation period, confirm that the condition of the experimental dogs is consistent with the basic condition recorded in the experimental dog basic information table.

[0011] S3. Prepare three sensitizing agents: ovalbumin, dinitrochlorobenzene, and house dust mite extract, and store them in a refrigerator at 4°C;

[0012] S4. The experimental dogs in S2 were grouped according to the sensitizing agents. After hair removal on the back of each group of experimental dogs, the corresponding first sensitization treatment was performed, with intradermal injection of ovalbumin and house dust mite extract, and dinitrochlorobenzene applied topically.

[0013] S5. After the first sensitization, monitor the redness and swelling of the sensitized site, the level of skin reaction, and scratching behavior for 7 consecutive days. After the monitoring is completed, select experimental dogs with a skin reaction level of 2 or above and no systemic adverse reactions to form a list of qualified experimental dogs for the first sensitization.

[0014] S6. For the experimental dogs in the list of qualified experimental dogs for the first sensitization, use the same sensitizing agent and hair removal area as in S4, and perform enhanced sensitization treatment at 1 / 2 of the sensitization dose in S4. After treatment, observe the skin reaction daily and record it in the same way as in S5.

[0015] S7. On the 3rd day after the enhanced sensitization treatment, venous blood was collected to detect the sensitizer-specific IgE. At the same time, skin tissue was taken from the sensitization site to make sections and count mast cells. Dogs with IgE increased by more than 2 times and mast cells ≥10 per high power field were screened to form a preliminary list of qualified experimental dogs.

[0016] S8. For the experimental dogs in the preliminary qualified experimental dog list, use the same sensitizing agent and dosage as in S6, and induce itching again in the original hair removal area. Record the behavior for 4 hours and quantify it to form a quantitative record table of itching behavior.

[0017] S9. Repeat the S8 induction and quantification records for 3 consecutive days, calculate the coefficient of variation of the total duration of itching behavior per hour for each experimental dog within 3 days, screen dogs with a coefficient of variation ≤10%, and form a list of experimental dogs with stable itching behavior.

[0018] S10. Confirm that the experimental dogs in the list of stable pruritus behavior meet the following criteria: serum IgE fold-to-fold ratio ≥2, mast cells ≥10 / high power field, skin reaction maintained at grade 2-3, and pruritus duration variation coefficient ≤10% over 3 days. These dogs are considered stable canine pruritus models.

[0019] In this instruction manual, the healthy experimental dogs in S1 are Beagles or Labrador Retrievers, aged 6 to 12 months, and weighing 8 to 12 kg. When recording the basic skin condition, the back, abdomen, and inner sides of the limbs of the experimental dogs are examined. After confirming that there is no redness, swelling, peeling, or damage, a high-resolution 300 dpi photo of the skin is taken and stored in JPG format. This photo is used as a comparison basis when monitoring skin reactions in S5 and S6.

[0020] In this instruction manual, ovalbumin and house dust mite extract in S3 are dissolved or diluted with sterile physiological saline, and dinitrochlorobenzene is dissolved with anhydrous ethanol to prepare concentrations of 10 mg / mL, 0.5%, and 5 μg / μL, respectively. The three prepared sensitizers are then placed into sterile reagent bottles and labeled with the sensitizer name, concentration, and preparation date.

[0021] In this manual, the environment of the standardized experimental kennel in S2 is as follows: temperature 22-25℃, humidity 50%-60%, and light cycle of 12h light / 12h darkness.

[0022] In this instruction manual, the criteria for determining the skin reaction level in S5 are as follows: Level 0 is skin color and texture consistent with the surrounding normal skin, without redness, swelling, erythema, or papules; Level 1 is only a light red erythema with indistinct borders, without swelling or skin lesions; Level 2 is a dark red erythema with clear borders, accompanied by slight swelling, without papules; Level 3 is obvious erythema and significant swelling, with ≥3 papules, some of which may be accompanied by slight exudation.

[0023] A multidimensional evaluation method for a stable canine pruritus model, employing the stable canine pruritus model construction method described in any one of the above-mentioned methods, includes the following steps:

[0024] (1) Evaluation data collection: In S1, S5-S6, S7, S8-S9, the serum IgE level, skin reaction grade, number of mast cell infiltration in the skin, and the total duration and frequency of itching behavior over 3 consecutive days were extracted from the experimental dogs after basal and enhanced sensitization.

[0025] (2) Physiological and pathological quantitative evaluation: Based on the collected evaluation data, the serum IgE fold ratio is calculated and standardized together with the mast cell count. Combined with the skin reaction grade quantitative value, the comprehensive physiological and pathological score is obtained by weighting according to the preset weight. If the comprehensive physiological and pathological score is <60, the model construction S4 step is returned to resensitization. If the comprehensive physiological and pathological score is ≥60, the next step is entered.

[0026] (3) Evaluation of pruritus behavior stability: The coefficient of variation of total duration is calculated using 3 days of behavioral data. The weight adjustment factor is calculated in combination with the comprehensive physiological and pathological score. Then, the comprehensive score of behavior intensity and severity and the stability score are calculated and weighted to obtain the behavior stability index. If the behavior stability index is <70, return to the S8 step of model construction for re-induction. If the behavior stability index is ≥70, proceed to the next step.

[0027] (4) Comprehensive suitability evaluation: The suitability of drug efficacy evaluation, mechanism research and toxicology testing is evaluated. The comprehensive physiological and pathological score and behavioral stability index are combined with the preset weights of the scenarios to obtain the comprehensive suitability score of each scenario.

[0028] (5) Application of evaluation results: The suitability is determined by the comprehensive suitability score for each scenario, and experimental application suggestions are output.

[0029] In this instruction manual, the preset weighting in step (2) is as follows: serum IgE fold ratio weight 0.4, mast cell infiltration number weight 0.3, and skin reaction grade quantification value weight 0.3.

[0030] In this instruction manual, the specific method for calculating the serum IgE fold ratio and standardizing it together with the mast cell count in step (2) is as follows: When standardizing the serum IgE fold ratio, the minimum value is set to 2 and the maximum value is 5. If the value exceeds the range, it is taken as 2 or 5. The standardized value is (actual fold ratio - 2) / (5 - 2) × 100. When standardizing the number of mast cells infiltrating the skin tissue, the minimum value is set to 10 cells / high power field and the maximum value is 30 cells / high power field. If the value exceeds the range, it is taken as 10 or 30. The standardized value is (actual cell count - 10) / (30 - 10) × 100.

[0031] In this instruction manual, the calculation method of the weight adjustment factor in step (3) is as follows: weight adjustment factor = 0.5 + 0.005 × comprehensive physiological and pathological score. When the weight adjustment factor exceeds the range of 0.5-1.5, it is taken as 0.5 or 1.5.

[0032] In this manual, the specific criteria for scene adaptability scoring in step (4) are as follows:

[0033] Drug efficacy evaluation scenario: 40 points for ≥50% relief rate of itching after oral administration of 10mg / kg loratadine, 30 points for repeatability of behavioral records ≥90%, and 30 points for no skin damage. Total score: 100 points.

[0034] Scenario for studying the pathogenesis of pruritus: 40 points for repeated sampling of skin tissue more than 3 times, 30 points for detection error of physiological and pathological indicators ≤10%, 30 points for model survival period ≥30 days, total score 100 points;

[0035] Skin toxicology testing scenario: Transdermal water loss rate ≤20g / ( 40 points are awarded for the appearance of itching within 1 hour after applying the sensitizing agent, 30 points are awarded for the absence of a systemic allergic reaction, and 30 points are awarded for the absence of a systemic allergic reaction. The total score is 100 points.

[0036] In summary, the present invention has at least the following beneficial effects:

[0037] 1. Significantly improves the stability and reliability of model construction: The construction method ensures the stability of the model from physiological basis to behavioral performance by strictly screening experimental dogs (clearly defining the breed, age, and health status range), standardizing environmental adaptation (controlling temperature, humidity, light, and feeding conditions), covering multiple types of sensitizers (including protein, chemical, and natural allergens), and stepwise sensitization treatment (initial sensitization + enhanced sensitization). This avoids the problems of fluctuating sensitization effects and non-reproducible itching behavior in traditional construction, enabling the model to be stably used in long-term experiments (such as long-term drug efficacy monitoring and mechanism dynamic research).

[0038] 2. Achieving comprehensive and precise evaluation of model quality: The multi-dimensional evaluation method utilizes three fusion algorithms to form a complete evaluation chain encompassing physiological and pathological aspects, behavioral stability, and scenario adaptation. The physiological and pathological index quantification algorithm verifies the reliability of the model's physiological basis from the perspectives of immune response, inflammation level, and skin manifestations, avoiding misjudgments based on accidental behavioral stability without a clear physiological mechanism. The pruritus behavior stability evaluation algorithm dynamically adjusts behavioral evaluation weights based on physiological basis, ensuring the scientific rigor of behavioral stability evaluation. The comprehensive adaptability evaluation algorithm establishes differentiated adaptation standards for different research scenarios, enabling on-demand evaluation. Compared to traditional single-indicator evaluations, this evaluation system can more accurately determine model quality, providing a reliable basis for subsequent research.

[0039] 3. Improve the model's scenario adaptability and resource utilization efficiency: By designing a scenario-based evaluation algorithm for comprehensive adaptability evaluation, the model's adaptability to different scenarios such as drug efficacy evaluation mechanism research and skin toxicology testing is clarified. This avoids the waste of resources in traditional techniques where the model is qualified but cannot match the research needs. Researchers can quickly screen suitable models based on the evaluation results and use them directly for corresponding research, reducing the cost of repeated model construction and validation, and improving the efficiency of biomedical research and development and veterinary clinical research.

[0040] 4. Promote the standardization and industrial application of canine pruritus models: This solution provides a unified technical standard for canine pruritus models by clarifying the operational specifications of each step of the construction process (such as environmental parameters, sensitizing agent treatment methods, and behavioral monitoring methods) and the quantitative logic of the evaluation system (such as indicator weights and adaptation standards). This solves the problem of inconsistent model construction methods and non-recognition of evaluation results in different laboratories, promotes canine pruritus models as a standardized experimental tool, and helps their widespread application in the biopharmaceutical industry (such as the development of anti-pruritus drugs). Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the method for constructing a stable canine pruritus model involved in this invention.

[0042] Figure 2 This is a flowchart illustrating the multi-dimensional evaluation method for the stable canine pruritus model involved in this invention.

[0043] Figure 3 This is a schematic diagram of the CADESI-4 skin lesion scoring table involved in this invention.

[0044] Figure 4 This is a schematic diagram of the pruritus PVAS scoring table involved in this invention.

[0045] Figure 5 This is a schematic diagram of the severity index of canine atopic dermatitis skin lesions involved in this invention. Detailed Implementation

[0046] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0047] like Figure 1 As shown, this embodiment provides a method for constructing a stable canine pruritus model, including:

[0048] S1: Screening and Basic Information Recording of Experimental Dogs; 1. Refinement of Screening Criteria: To ensure the consistency of experimental dogs' responses to sensitizers and the stability of the model, it is necessary to strictly screen healthy experimental dogs that have no history of skin diseases (by reviewing the experimental dogs' health records for the past 3 months and combining clinical examinations to exclude skin diseases such as eczema and dermatitis), no history of allergies (confirming that they have not been exposed to common allergens such as ovalbumin and house dust mites), and no systemic diseases (such as canine distemper, parvovirus, etc., confirmed through body temperature measurement, blood routine tests, and biochemical indicators (alanine aminotransferase, creatinine, etc.) analysis). The breeds were limited to Beagles and Labrador Retrievers (both are commonly used experimental breeds, with stable physiological states and good repeatability of sensitization responses, and no other common experimental breeds were omitted). The age range was limited to 6-12 months (at this stage, the dogs' immune systems are fully developed, and their ability to respond to sensitizers is stable, avoiding model fluctuations caused by incomplete immunity in young dogs or immune decline in older dogs). The weight was controlled between 8-12 kg (excessive weight difference will affect the absorption of sensitizer dosage; within this range, the dogs' body surface area is similar, which can ensure consistent sensitizer intensity).

[0049] 2. Basic Information Recording Standards: For qualified experimental dogs, a standardized basic information form for experimental dogs shall be used to record information. The form shall include the dog's unique identification number (e.g., 20xx-B-001, where 20xx is the year, B represents a Beagle, and 001 is the serial number), breed, age (accurate to the month, e.g., 8 months old), weight (accurate to 0.1 kg, measured using an electronic scale (accuracy 0.01 kg, brand such as Mettler Toledo PL602-L), fasting for 12 hours before measurement), sex (indicate male or female; if female, confirm that she is not pregnant or lactating), basal body temperature (rectal temperature measurement, using a veterinary electronic thermometer, measurement time 5 minutes, normal range 38.5-39.2℃), and basic skin condition (detailed examination of easily observable areas such as the back, abdomen, and inner sides of the limbs, recording no redness, swelling, peeling, damage, or dandruff, and taking high-resolution photos of the skin (300 dpi resolution, JPG format) for subsequent comparison).

[0050] Three mL of cephalic vein blood was collected from the forelimbs of the experimental dogs (collected after fasting for 4–12 hours). The serum was separated by centrifugation (3000 rpm, 10 minutes) and sent to a third-party laboratory to test the baseline serum-specific IgE level (such as ovalbumin-specific IgE, house dust mite-specific IgE, etc., to match the type of sensitizer used in subsequent steps). This data was included in the experimental dog's basic information form. The data in this form will serve as the core benchmark for determining whether the experimental dog's physical condition has become abnormal due to the treatment in subsequent steps, and must be cross-checked and signed by two researchers.

[0051] S2: Environmental Adaptation Treatment for Experimental Dogs; 1. Standardized Environmental Setup: Transfer the qualified experimental dogs selected in S1 to experimental dog kennels that meet the standards for experimental animal environments and facilities. The environmental parameters are precisely controlled as follows: temperature 22-25℃ (using a constant temperature air conditioning system, recording the temperature once per hour, with a fluctuation range not exceeding ±0.5℃), humidity 50%-60% (adjusted by linkage between dehumidifier and humidifier, with a hygrometer (accuracy ±2%) suspended at the middle height of the dog cage), and a light cycle of 12h light / 12h darkness (lighting period is 8:00-20:00, using LED cold light source, light intensity 200-300 lux, avoiding strong light stimulation to the dogs). The dogs were housed in individual cages, each measuring 1.5m × 1.0m × 1.2m (length × width × height). The bottom of the cage was covered with a non-slip rubber mat (5mm thick with a textured surface to prevent the dogs from slipping). Stainless steel food bowls and water dispensers were placed inside the cages (cleaned and disinfected daily to prevent bacterial growth). The experimental dogs were labeled with their identification numbers on the outside of the cages to ensure clear individual identification.

[0052] 2. Adaptation Period Management and Status Confirmation: The experimental dogs need to adapt to the above environment for 7 days. During the adaptation period, they should be fed standard dog food (commercially available complete experimental dog food, such as Royal Canin PR27, which meets the standards for complete pet food dogs, with nutritional components listed as crude protein ≥22%, crude fat ≥8%, and crude fiber ≤5%, calculated based on body weight, i.e., 20g / kg, divided into 2 equal portions) at 7:00 and 19:00 daily. They should have free access to purified water disinfected by ultraviolet light (the water in the water bottle should be changed daily to ensure water quality). The experimental dogs' status should be recorded at 9:00 and 15:00 daily: appetite (record whether the amount of food consumed at each meal reaches more than 90% of the recommended amount; if it is less, mark it as decreased appetite), mental state (observe whether they stand up voluntarily, wag their tails, and have normal responses to sound or food stimuli, without curling up, lethargy, or other abnormalities), and defecation (record the shape of the feces (normally striped), color (brownish-yellow), and frequency (1-2 times per day), without diarrhea (watery stools) or constipation (hard stools, difficulty in defecation)). After the adaptation period ends, the experimental dogs' body temperature and skin condition are measured again and compared with the data in the experimental dogs' basic information table recorded in S1. If there is no difference (body temperature fluctuation ≤0.3℃, no new skin abnormalities), the subsequent steps can be carried out. If there is a difference, the adaptation period needs to be extended by 2 to 3 days until the condition returns to the basic level.

[0053] S3: Preparation and Concentration of Sensitizers; 1. Selection and Source of Sensitizers: Three different types of sensitizers were selected to cover common clinical causes of itching and ensure model applicability: Ovalbumin (OVA): A protein-based sensitizer with a clear sensitization mechanism (inducing humoral immunity to produce specific IgE), resulting in a high model construction success rate. Sourced from Sigma-Aldrich, catalog number A5503, analytical purity (≥98%); Dinitrochlorobenzene (DNCB): A chemical sensitizer that can induce contact hypersensitivity reactions, with a rapid onset of action. Sourced from Solarbio, catalog number D8040, analytical purity (≥99%); House dust mite extract: A natural allergen that simulates clinical canine house dust mite allergy-induced itching, closely resembling the actual disease scenario. Sourced from Solarbio, catalog number D1001, analytical purity (each milligram of extract contains ≥10 μg of the main house dust mite allergen, Der p1).

[0054] 2. Concentration preparation and quality control: Preparation of ovalbumin solution: In a biosafety cabinet (e.g., Haier HR40-IIA2), take 1g of ovalbumin powder and add 100mL of sterile physiological saline (0.9% sodium chloride injection, brand such as Huaren Pharmaceutical). Stir with a magnetic stirrer (300rpm) for 30 minutes until completely dissolved to prepare a concentration of 10mg / mL. After preparation, the concentration was measured at 280 nm using a UV spectrophotometer (Thermo Scientific NanoDrop 2000) (absorbance A280 = 0.14 corresponds to 1 mg / mL), ensuring the concentration error was ≤5%. For the dinitrochlorobenzene solution: Take 0.5 g of dinitrochlorobenzene powder and add 100 mL of anhydrous ethanol (analytical grade, purity ≥99.7%, brand such as Sinopharm Group). First, add a small amount of anhydrous ethanol to grind the powder until no particles remain, then gradually add ethanol to bring the volume to 100 mL, preparing a 0.5% (w / v) concentration. Avoid ethanol evaporation during stirring. After preparation, seal and let stand for 10 minutes. If no precipitation is observed, it is ready for use. For the house dust mite extract solution: Take 5 mg of house dust mite extract lyophilized powder and add 1 mL of sterile physiological saline. Use a pipette to repeatedly pipette 10 times to ensure uniform dissolution, preparing a concentration of 5 μg / μL. It should be used within 30 minutes after dissolution to avoid a decrease in extract activity. After all sensitizing agents are prepared, they are placed into sterile glass reagent bottles (50mL, graduated). The bottles are labeled with the name of the sensitizing agent, concentration, preparation date, and the person who prepared them. They are then stored in a 4℃ refrigerator (Haier BCD-216SDN) away from light and for no more than 7 days. Before each use, the bottles must be shaken well and observed to ensure there is no turbidity or precipitation. If any abnormality occurs, the bottles must be discarded immediately and re-prepared.

[0055] S4: Initial sensitization treatment of experimental dogs; 1. Grouping and subject confirmation: Select experimental dogs from S2 that have adapted to the environment and are in normal condition. Use a randomized block design to group them, ensuring balanced conditions in each group: divide them into ovalbumin group, dinitrochlorobenzene group, and house dust mite extract group, with each dog corresponding to one sensitizing agent. Each group should have at least 3 dogs (to meet statistical replication requirements and reduce the impact of individual differences on the results). When grouping, ensure that the sex ratio of experimental dogs in each group is consistent (e.g., each group consists of 2 males and 1 female) and the weight difference does not exceed 1 kg. Record the basic information table number of each group of experimental dogs to form an experimental dog grouping table, which shall be reviewed and confirmed by the person in charge of the experiment.

[0056] 2. Sensitization Procedure Details; Hair Removal Pretreatment: One day before sensitization, all experimental dogs underwent hair removal on both sides of the spine (1-2 cm from the spine, one on each side) using a pet-specific electric shaver (Andis ProClip AGC2, size 10 blade). A 1mm distance was maintained from the skin during shaving to avoid abrasions. The hair removal area was precisely controlled to 5cm x 5cm (boundaries were marked on the skin surface with a marker). After hair removal, the area was wiped with sterile saline to remove residual hair. Once the skin was checked for damage, it was marked as the sensitization site. Ovalbumin Treatment: Using a 1mL sterile syringe (brand such as BD, needle size 26G x 13mm), 10mg / mL ovalbumin solution was drawn. Four injection points were marked in a triangular pattern at the sensitization site (1cm apart). During injection, the needle was inserted into the skin at a 15° angle, and 0.1mL of the solution was injected. The diameter of the skin lesions was observed to be approximately 0.5cm. For the m-shaped wheal group: confirm that the medication is injected intradermally and not subcutaneously; if a wheal does not form, repeat the injection. After injection, gently press the injection site with a sterile cotton swab for 10 seconds to prevent medication leakage. For the dinitrochlorobenzene group: use a sterile defatted cotton swab (8cm×10cm) to apply 0.5% (w / v) dinitrochlorobenzene solution evenly in a clockwise direction to the sensitized area, ensuring that the medication completely covers the 5cm×5cm area. After application, wait 5 minutes for the ethanol to evaporate (to prevent the dog from ingesting the medication by licking it), and then put the dog back in its cage. For the house dust mite extract group: the procedure is the same as for the ovalbumin group. Draw 5μg / μL of house dust mite extract solution and inject 0.1mL intradermally at each of the 4 points to ensure the formation of a wheal.

[0057] 3. Postoperative observation and recording: After treatment, the experimental dogs were returned to the feeding cage in S2. The skin condition of the sensitized site was observed once at 0.5h, 1h, 2h, and 4h. Thereafter, the dogs were observed once each in the morning (9:00), noon (15:00), and evening (21:00) to record whether redness and swelling (skin raised above the surrounding normal skin, with the diameter measured), erythema (skin ranging from light red to dark red, with the clarity of the border recorded), and papules (raised skin lesions with a diameter of 1-3mm, with the number recorded). At the same time, the dogs were observed to see if they exhibited abnormal behaviors such as restlessness or frequent shaking of their bodies. The observation results were filled in the first post-sensitization observation form. The data in this form will serve as the initial basis for skin reaction monitoring in step S5.

[0058] S5: Skin Reaction Monitoring and Data Recording After Initial Sensitization; 1. Monitoring Cycle and Tool Preparation: Starting from the day of sensitization treatment (S4, designated D0), monitoring will continue for 7 days (until D7), with fixed times of 9:00 AM and 5:00 PM daily to avoid monitoring errors caused by time differences. Tools required include: a 0.02mm precision vernier caliper (Mitutoyo 500-196-30, calibrated before use), a skin reaction level assessment table, a high-definition camera (same as S1, used to take daily photos of skin condition), and a skin reaction monitoring record sheet.

[0059] 2. Monitoring Indicators and Judgment Criteria: Measurement of Redness and Swelling Diameter: Measure the longest and shortest diameters of the red and swollen area at the sensitized site using calipers, and record the average value (accurate to 0.1 mm) as the diameter of the redness and swelling for that day. If the redness and swelling are irregular in shape, multiple measurements are required, and the most representative value should be taken. Skin Reaction Grade Judgment: Grades are strictly divided according to the following standards: Grade 0: Skin color and texture are completely consistent with the surrounding normal skin, with no redness, swelling, erythema, papules, or any abnormalities; Grade 1: Only light red erythema appears, with blurred borders, no swelling or skin lesions, and the erythema area does not exceed the boundary of the sensitized site; Grade 2: The erythema deepens to dark red, with clear borders. Grade 1: Accompanied by slight swelling (skin thickness increased by 0.1-0.2 mm compared to normal, measured indirectly by calipers), no papules, erythema covering the entire sensitized area; Grade 2: Obvious erythema, significant swelling (skin thickness increased by more than 0.3 mm compared to normal), with ≥3 papules, some papules may be accompanied by slight exudation, erythema extending ≤0.5 cm beyond the boundary of the sensitized area; Preliminary scratching behavior record: Observe whether the experimental dog actively scratches the sensitized area during the monitoring period (scratching with front or hind paws, each scratch lasting ≥1 second), record the number of scratches per day, if the number of scratches per day is ≥3 times, it is marked as the appearance of preliminary itching behavior.

[0060] 3. Data Screening and Acceptance Criteria: After the monitoring period, the data from the 7-day skin reaction monitoring records are summarized and analyzed. First, dogs exhibiting systemic adverse reactions are excluded (e.g., lethargy, loss of appetite to below 70% of the recommended intake, diarrhea, body temperature exceeding 39.5℃, etc.). For the remaining dogs, their daily skin reaction level is recorded for 7 days. At least 5 days out of 7 days must have a level of 2 or higher (ensuring a sustained and stable sensitization reaction). Simultaneously, at least 3 days must show initial scratching behavior (indicating that the sensitization reaction has triggered itching-related behaviors). Dogs meeting the above criteria are entered into a list of dogs that have passed the initial sensitization. This list must include the dog's number, group, 7-day skin reaction level details, initial scratching behavior record, and the acceptance / disqualification result. The list must be jointly signed and confirmed by the data analyst and the experimental personnel. Dogs on this list will proceed to step S6.

[0061] In some embodiments, skin response monitoring may be combined with the CADESI-4 skin lesion scoring scale (reference). Figure 3 A more refined scoring system, based on Olivery T et al., Vet Dermatol. 2014, can be used to supplement the assessment of the distribution and severity of typical lesions associated with canine atopic dermatitis (such as erythema, papules, desquamation, lichenification, etc.). Simultaneously, the severity index of canine atopic dermatitis skin lesions (based on Rodrigues Gonçalves et al., Ciência Rural. 2018) can be referenced, quantifying the severity of lesions into discrete (1 point), moderate (2 points), accentuated (3 points), and negative (0 points). This system, combined with the original skin reaction grading criteria, forms a more comprehensive dermatopathological assessment system. During daily skin reaction monitoring, in addition to measuring the diameter of redness and swelling and determining the skin reaction grade, the CADESI-4 skin lesion scoring form should be filled out simultaneously. Eight types of lesions, including erythema, papules, desquamation, exudation, crusting, and lichenification, should be scored for each of the six areas: head, neck, trunk, forelimbs, hindlimbs, and perineum. Each type of lesion is scored from 0 to 3 points according to its severity, and the total score for the entire body's skin lesions is obtained. Simultaneously, based on the canine atopic dermatitis skin lesion severity index, the core lesions (such as erythema and papules) at the sensitization site are graded and quantified, and this data, along with the original skin reaction grade data, is entered into the skin reaction monitoring record form, providing richer evidence for subsequent physiological and pathological evaluation.

[0062] S6: Enhanced sensitization treatment of experimental dogs; 1. Timing and purpose of treatment: Enhanced sensitization was performed on the first day after the end of S5 monitoring (i.e., the 8th day after the first sensitization, D8). At this time, a certain amount of sensitizing antibodies had been produced in the experimental dogs. Enhanced sensitization can further activate the immune system, enhance the intensity and stability of the sensitization reaction, lay the foundation for subsequent induction of persistent itching behavior, and avoid model failure due to insufficient first sensitization reaction.

[0063] 2. Treatment Method and Consistency Control: Sensitive Area Preparation: Check the sensitized areas on the back marked in S4. If new hair has grown in the hair removal area (length exceeding 0.5cm), shave again with the original model electric shaver. The prepared hair removal area should be consistent with the 5cm×5cm area of ​​the initial sensitization. Remark the edges with a marker to ensure complete overlap of sensitized areas each time. If there is no obvious hair growth on the skin, simply wipe away dirt with sterile saline solution. Sensitizer Dosage Adjustment: To avoid severe skin damage due to excessive dosage, the sensitizing dose is adjusted to 1 / 2 of the initial sensitization dose, as follows: Ovalbumin Group: Use a 1mL sterile syringe to draw... Take 10 mg / mL ovalbumin solution and inject 0.05 mL at each of the four injection points at the sensitization site. The injection procedure is the same as in S4 to ensure the formation of a wheal. For the dinitrochlorobenzene group: dilute the 0.5% (w / v) dinitrochlorobenzene solution prepared in S3 with anhydrous ethanol at a ratio of 1:1 (take 1 mL of the original solution and add 1 mL of anhydrous ethanol, mix thoroughly) to prepare a concentration of 0.25% (w / v). Use immediately after dilution and apply evenly to the sensitization site with a sterile cotton swab. The application area is the same as the first time. For the house dust mite extract group: take 5 μg / μL of house dust mite extract solution and inject 0.05 mL at each point for a total of 4 points. The procedure is the same as in S4.

[0064] 3. Reaction Recording and Abnormal Handling: After enhanced sensitization, observe the skin reaction daily at the monitoring frequency of S5 (9:00 AM and 5:00 PM), recording the diameter of redness and swelling, and the skin reaction grade. The recording method should be completely consistent with S5 to ensure data comparability. If the following abnormalities occur, immediate action is required: Skin reaction grade drops to grade 1: Investigate whether it is due to insufficient sensitizer concentration or skin tolerance. If the concentration is confirmed to be correct, enhanced sensitization can be performed again after 24 hours (at the same dose); Severe skin damage or oozing (reaction grade exceeding grade 3): Immediately stop sensitization treatment, clean the affected area with sterile saline, apply 2% mupirocin ointment (brand such as Federal Pharmaceuticals), change the dressing twice daily, and assess whether to continue the experiment after the skin heals. If the reaction grade cannot recover to grade 2 after healing, the dog should be culled. The skin reaction records after enhanced sensitization will serve as an important reference for the initial validation of the S7 step model; therefore, the records must be accurate and complete.

[0065] S7: Preliminary Validation of the Post-Sensitization Itching Model; 1. Sample Collection and Operation Procedures: Samples were collected on the 3rd day (D11) after S6 sensitization, when the sensitization reaction had reached a stable peak, making the sample test results more representative: Serum Sample Collection: Before collection, the experimental dogs were fasted for 4-12 hours (to avoid food residue affecting serum components). The dogs were secured to the experimental table (using a special dog restraint frame to prevent the dogs from struggling). The cephalic vein area (5cm in diameter) of the forelimb was disinfected with a 75% ethanol cotton ball. After the ethanol evaporated, a 2mL sterile syringe (24G×19mm needle) was used for venipuncture, and 3mL of blood was slowly drawn (to avoid hemolysis caused by drawing too quickly). The blood was injected into a blood collection tube (5mL, brand BD) containing EDTA-K2 anticoagulant. The blood collection tube was gently inverted 5 times to ensure thorough mixing of the blood and anticoagulant. After collection, the puncture site was pressed with a sterile cotton swab for 5 minutes to stop bleeding. After observing no bleeding, the dogs were returned to their cages. Immediately place the blood collection tubes into a 4°C centrifuge (Eppendorf 5810R) and centrifuge at 3000 rpm for 10 minutes. After centrifugation, use a sterile pipette to aspirate the supernatant serum (avoiding aspiration of red blood cells) and dispense into 0.5 mL sterile centrifuge tubes (0.5 mL per tube). Label each tube with the experimental dog number, group, and collection date, and store in a -80°C freezer (Thermo Scientific Forma 900) for later testing. Skin tissue sample collection: After collecting serum, local anesthetize the sensitized area of ​​the experimental dog. Subcutaneously inject 2% lidocaine injection solution (brand such as Shandong Hualu) around the sensitized area (0.1 mL per point, for a total of 4 points, to form an anesthetic area). Confirm the anesthesia effect after 10 minutes (no reaction when the skin is gently pinched with forceps). Using sterile surgical scissors (10cm, such as B. Braun), cut a 0.5cm x 0.5cm piece of skin tissue along the skin texture (deep enough to reach the subcutaneous fat layer, ensuring complete epidermis and dermis). Immediately place the tissue piece into a specimen bottle containing 4% paraformaldehyde fixative (preparation ratio: 4g paraformaldehyde powder dissolved in 100mL PBS buffer, pH 7.4, such as Solarbio). The volume of fixative should be 10 times the volume of the tissue piece to prevent autolysis. After collection, apply sterile gauze to the wound for 5 minutes to stop bleeding. After hemostasis, apply 2% mupirocin ointment and lightly bandage with a sterile bandage (tight enough to allow one finger to be inserted). Change the bandage daily and observe the wound healing to ensure there is no infection (such as redness, swelling, or pus).

[0066] 2. Indicator Detection and Third-Party Collaboration: Serum-Specific IgE Detection: Frozen serum samples were sent to a veterinary testing institution with CNAS accreditation (such as Beijing Zhongjian Weikang Biotechnology Co., Ltd.) for enzyme-linked immunosorbent assay (ELISA). The testing instrument was a Thermo Scientific Multiskan FC microplate reader. The testing process was strictly carried out according to the kit instructions (brand such as R&D Systems, catalog number DY1684). The detection index was the specific IgE of the corresponding sensitizer (OVA-IgE for the ovalbumin group, DNCB-IgE for the dinitrochlorobenzene group, and Der p1-IgE for the house dust mite extract group). During testing, the baseline serum sample of the experimental dogs collected in step S1 (collected within 24 hours after passing the S1 screening, and stored under the same conditions as the current serum sample) should be added as a control. The ratio of the current serum IgE level to the baseline serum IgE level should be calculated. Skin tissue pathological examination: Skin tissue blocks fixed for 24 hours were sent to the same testing institution for pathological section preparation and staining: The tissue blocks were subjected to gradient dehydration (70% ethanol 1h → 80% ethanol 1h → 90% ethanol 1h → 95% ethanol 2h → anhydrous ethanol 2h), xylene clearing (twice, 30 minutes each time), paraffin wax immersion at 60℃ (twice, 1h each time), paraffin embedding, 4μm thick sectioning, mounting, dewaxing to water, staining with 0.1% toluidine blue solution for 10 minutes, differentiation solution (0.5% acetic acid solution) for 2 minutes, dehydration and clearing, and mounting with neutral resin. The pathological sections were observed by a senior pathologist under an optical microscope (model Olympus BX53). A 10×40x objective lens was selected, and five high-power fields were randomly selected in the dermal region. The number of mast cells in each field was counted (mast cells appear purplish-red after toluidine blue staining, and there are coarse granules in the cytoplasm). The average value of the five fields was taken as the number of mast cell infiltrations in the sample.

[0067] 3. Validation Criteria and Qualification Screening: Based on the test results, the following validation criteria were established: Serum-specific IgE: Ratio ≥ 2.0 (i.e., the serum IgE level is 2 times or more higher than the baseline state, indicating that the immune system has produced a clear sensitization response); Skin mast cell infiltration: Average ≥ 10 cells / high-power field (mast cell activation and histamine release is a key step in the occurrence of itching; this value indicates that the local skin inflammation response meets the model requirements). Simultaneously, it must be confirmed that the experimental dogs' wounds healed well (no infection, no obvious scarring), and the skin reaction grade remained at 2-3. Experimental dogs meeting all the above conditions were included in the preliminary qualified experimental dog list. The list must be accompanied by a copy of the serum test report, pathological slide photographs, and counting results, and jointly confirmed by the testing institution and the experimental team. Experimental dogs on this list proceeded to step S8.

[0068] S8: Induction and Quantitative Recording of Itching Behavior; 1. Induction Treatment and Parameter Control: Itching behavior was induced on the second day after sample collection in S7 (D13). At this time, the experimental dogs' wounds had basically healed, avoiding interference from wound pain on the behavior recording. The induction treatment involved re-application of a low dose of sensitizer at the original sensitization site, at half the dose of the enhanced sensitization in S6 (to further reduce the risk of skin damage while ensuring that itching behavior was triggered), as follows: Ovalbumin group: 0.025 mL of 10 mg / mL ovalbumin solution was injected intradermally at each of the four injection points at the sensitization site; Dinitrochlorobenzene group: 0.125% (w / v) dinitrochlorobenzene solution (diluted from 0.25% solution with anhydrous ethanol at a 1:1 ratio) was applied to the sensitization site; House dust mite extract group: 0.025 mL of 5 μg / μL house dust mite extract solution was injected intradermally at each of the four injection points at the sensitization site. The personnel and time of the induction treatment (fixed at 9:00 AM) are the same as those in S4 and S6 to ensure consistency in operation.

[0069] 2. Behavioral Recording Environment and Tools: After induction treatment, the experimental dogs were transferred to observation cages equipped with video monitoring. The environmental parameters (temperature, humidity, and light) of the observation cages were completely identical to those of the S2 rearing cages to avoid environmental changes affecting the dogs' behavior. A high-definition camera (Hikvision DS-2CD3T46WD-I3, 1080P resolution, 25fps) was installed 1.5m in front of the observation cage. The camera was connected to a computer and running the Noldus EthoVision XT16 video analysis software (with behavior recognition and quantification functions). The software preset sensitization area regions (marked with coordinates corresponding to the actual sensitization areas) to accurately identify behaviors targeting the sensitization areas.

[0070] 3. Definition and Quantification of Itching Behavior: Itching behavior is clearly defined as follows: Scratching: The experimental dog raises its front or hind paw, touches the sensitized area with the paw, and performs a reciprocating scratching motion. There is significant skin displacement during scratching, and the duration is ≥1 second (excluding cases where the paw briefly touches the skin); Licking: The experimental dog lowers its head, touches the sensitized area with its tongue, and performs continuous licking motions. The tongue's contact time with the skin is ≥2 seconds (excluding brief licking or licking of non-sensitized areas); Biting: The experimental dog lowers its head, touches the sensitized area with its teeth, and performs a slight chewing motion. The teeth do not cause skin damage (avoiding damaging bites), and the duration is ≥1 second. Quantification and Recording Process: Records are continuously recorded for 4 hours after induction treatment (9:00-13:00). Each hour is divided into one observation period. Within each period, records are recorded using a combination of automated software analysis and manual verification. Software analysis: By setting behavioral parameters (such as paw movement trajectory, head tilt angle, and contact time with the sensitized area), the system automatically identifies and counts the number of times (times / hour) and the duration of each behavior (seconds / time). Manual review: Two trained researchers simultaneously watch the video and verify the software's identification results. If the difference between the software and human identification results is ≤5%, the software result prevails; if the difference is >5%, the video is re-analyzed to determine the final data. Data aggregation: The total duration of itching behavior in each time period is calculated (total scratching time + total licking time + total biting time, unit: seconds / hour). All data is entered into the Quantitative Record Table of Itching Behavior, which must include the experimental dog's number, group, observation period, number and duration of each behavior, total duration, and the signatures of the recorder and reviewer.

[0071] In some embodiments, the quantification of pruritus behavior may include a subjective rating dimension, using a modified Visual Analogue Scale (PVAS) (see reference). Figure 4This study, cited from Cosgrove SB et al., Veterinary Dermatology. 2013, provides a supplementary evaluation. The scale is a 10-centimeter straight line, with the ends marked as normal dog (0 cm, 0 points) and extremely severe itching (10 cm, 10 points), respectively. Six descriptive terms for itching (ranging from mild / occasional to extremely severe itching) are distributed at 2-centimeter intervals in the middle. Researchers mark the points based on the dogs' itching behavior, measure the distance from the marked point to the 0-point end, and convert the distance into a numerical score of 0-10. This score is then combined with objective data on the total duration and frequency of itching behavior to improve the comprehensiveness of behavioral quantification. After the itching behavior recording is completed, two trained researchers independently use a modified version of the PVAS scale to subjectively rate the dogs' itching severity. Before rating, the use of the scale must be standardized (e.g., "mild itching" corresponds to 2-4 cm, "moderate itching" corresponds to 4-6 cm, etc.). If the difference between the two ratings exceeds 1 cm, they must jointly watch the behavioral video to reach a consensus. The PVAS score results, along with the total duration, total frequency, and duration of each instance of itching behavior statistically analyzed by the software, are entered into the itch behavior quantitative record table to form a dual quantitative system of "objective behavioral data + subjective score".

[0072] S9: Continuous monitoring of model stability; 1. Monitoring cycle and consistency assurance: Starting from the day of S8 induction treatment (D13), continuous monitoring will be conducted for 3 days (until D15). Induction treatment will be performed once daily using the S8 method (dosage, operator, and time will remain completely unchanged), and 4 hours of pruritus behavior recording will be completed. A quantitative record sheet of pruritus behavior will be generated daily. To ensure stable monitoring results, the following variables must be controlled: Induction treatment: The sensitizing agent must be shaken well before each use, and the dosage must be accurately aspirated using a pipette (error ≤ 0.001 mL). The sensitization site markings must remain unchanged; Behavioral recording: Camera position, software parameters, and manual review personnel must be fixed to avoid recording errors caused by changes in equipment or personnel; Dog status: Before each daily monitoring, the appetite, mental state, and skin condition of the experimental dogs must be checked to ensure there are no abnormalities (consistent with the baseline state in S2). If any abnormalities are found, monitoring will be suspended until the dogs recover.

[0073] 2. Stability Assessment Indicators and Calculation Methods: The total duration of itching behavior per hour is used as the core indicator. The model stability is assessed by calculating the coefficient of variation. The specific steps are as follows:

[0074] Calculate the daily average: Take the average of the total duration of itching behavior in the four observation periods of each day. For example, the average value of D13 M1 = (T13-1 + T13-2 + T13-3 + T13-4) / 4 (where T13-1 is the total duration of the first period of D13, and so on). Similarly, calculate the average value of D14 M2 and the average value of D15 M3.

[0075] Calculate the total mean and standard deviation: Total Mean Standard deviation (The denominator is the sample size - 1, i.e., the degrees of freedom).

[0076] Calculate the coefficient of variation: The coefficient of variation CV = (S / M average) × 100%. This coefficient reflects the degree of fluctuation in the total duration of itching behavior within 3 days. The smaller the CV, the better the model stability.

[0077] 3. Screening and Confirmation of Stable Dogs: A coefficient of variation ≤10% is set as the stability standard (determined based on preliminary experimental data; the model can meet the needs of subsequent experiments under this standard). Dogs meeting this standard are screened, and it is also necessary to confirm that their skin reaction level remains at 2-3 and there are no systemic adverse reactions. Qualified dogs are entered into a stable pruritus behavior experimental dog list. The list must include the dog's number, group, total duration of pruritus at each time period over 3 days, daily average, overall average, standard deviation, coefficient of variation, and stability assessment results. This list is jointly reviewed and signed by the statistician and the experiment leader, and is the core basis for model construction in step S10.

[0078] S10: Obtain a stable canine pruritus model; based on the list of experimental dogs with stable pruritus behavior formed in S9, each experimental dog on the list is finally qualified and must meet all of the following conditions (all are required): 1. Continuously qualified skin reaction: From the first sensitization in S4 to the end of monitoring in S9, the skin reaction grade is always maintained at level 2-3, without continuous decline or severe damage; 2. Immune indicators meet the standards: In step S7, the serum specific IgE level is 2 times or more higher than the baseline state, the number of mast cells infiltrating the skin tissue is ≥10 per high-power field, and the immune response is clear and stable; 3. Stable pruritus behavior: The coefficient of variation of the total duration of pruritus behavior over 3 days in step S9 is ≤10%, and the pruritus manifestation is repeatable and without drastic fluctuations; 4. Good overall condition: There are no systemic diseases or severe stress reactions throughout the experiment, and the appetite, mental state, and defecation are always consistent with the baseline state in S2. For experimental dogs that meet all the requirements, a complete stable canine pruritus model file is established. This file includes all original record forms from S1 to S9 (basic information form for the experimental dog, grouping form, skin reaction monitoring record form, etc.), serum test reports, pathological slide photographs, video screenshots of pruritus behavior, and analytical data. The file is kept by designated personnel for subsequent experimental traceability and verification. At this point, the experimental dog is considered a qualified stable canine pruritus model, suitable for subsequent scientific experiments such as research on the pathogenesis of canine pruritus and evaluation of the efficacy of antipruritic drugs.

[0079] A multi-dimensional evaluation method based on a stable canine pruritus model;

[0080] I. Overall Framework and Design Logic of the Solution; This multi-dimensional evaluation method takes the model foundation to application value as its core logic. Around the entire process of constructing a stable canine pruritus model, it designs three integrated algorithm modules: a physiological and pathological index quantification evaluation algorithm (A1), a pruritus behavior stability evaluation algorithm (A2), and a comprehensive fitness evaluation algorithm (A3). These three modules do not operate independently but form a collaborative closed loop through data transfer, weight interaction, and result fusion. The process of the multi-dimensional evaluation method for the stable canine pruritus model is as follows: Figure 2 As shown. 1. Data Source Connection: The input data for all algorithms comes directly from specific steps in the stable canine pruritus model construction method, ensuring a deep integration between evaluation and model construction processes and avoiding data disconnect. 2. Algorithm Interaction Logic: The output of A1 (comprehensive physiological and pathological score) serves as a weight adjustment factor for A2, directly affecting A2's weight allocation for stability and strength indicators. The results of A1 and A2 together serve as the core input of A3, which, combined with the application scenario requirements, outputs the final suitability score. 3. Hierarchical Evaluation Objectives: A1 focuses on whether the model's physiological basis is reliable, A2 focuses on whether the model's behavioral performance is stable, and A3 focuses on whether the model is suitable for specific research needs, achieving full-dimensional coverage from basic to application and solving the one-sidedness problem of single-dimensional evaluation.

[0081] II. Algorithm 1 (A1): Quantitative Evaluation Algorithm for Physiological and Pathological Indicators; The reliability of a stable canine pruritus model primarily depends on whether the physiological and pathological basis conforms to the pathogenesis of pruritus. The occurrence of pruritus is directly related to immune system activation (elevated serum IgE), local skin inflammation (mast cell infiltration), and pathological skin damage (redness, swelling, papules). The core objective of A1 is to transform these three dispersed physiological and pathological indicators into a unified comprehensive physiological and pathological score. To determine whether a model has a reliable biological basis from the perspective of etiology and pathology, and to provide a physiologically reasonable basis for subsequent behavioral evaluation.

[0082] Step 1: Construction of the A1 model for the quantitative evaluation algorithm of physiological and pathological indicators;

[0083] 1.1 Indicator Selection; In the method for constructing a stable canine pruritus model, the physiological and pathological data mainly focus on steps S1 (basal serum), S5 (skin reaction), S6 (enhanced sensitization reaction), and S7 (serum and tissue detection). Therefore, A1 selects the following three core indicators to ensure coverage of the entire chain of immune response-local inflammation-skin manifestation:

[0084] Indicator 1: fold change in serum specific IgE levels ( Source: Baseline serum collected in step S1 (); S7 step enhanced sensitization followed by serum collection and testing. Reason for selection: IgE is a key antibody mediating type I hypersensitivity reactions, and the original model requires... This indicator directly reflects whether the model has established an effective immune sensitization response, which is the core immune basis for the occurrence of itching.

[0085] Indicator 2: Number of mast cells infiltrating skin tissue ( Source: In step S7, skin tissue was taken from the sensitized site, pathologically sectioned, stained with toluidine blue, and the number of mast cells per high-power field was counted under a microscope. Reason for selection: The activation and release of histamine and other inflammatory mediators by mast cells is a crucial step in the perception of itching; the original model required… The index of one high-power field directly reflects the degree of local skin inflammation and serves as a key pathological bridge connecting the immune response and itching behavior.

[0086] Indicator 3: Quantitative value of skin reaction level ( (Source: Monitoring for 7 consecutive days after the initial sensitization in step S5, and daily monitoring after enhanced sensitization in step S6, recording the condition of skin redness, swelling, erythema, and papules, and converting them into quantitative values ​​according to the original model standards (0 = 0, 1 = 33, 2 = 66, 3 = 100). Reason for selection: Skin reaction is a direct pathological manifestation of the pruritus model. The original model requires the reaction level to be maintained at 2-3. This indicator can quickly determine whether the skin pathological damage is persistent and stable, avoiding an invalid model where the immune indicators are qualified but the skin has no reaction.)

[0087] 1.2 Formula Structure Design; Due to the different dimensions and value ranges of the three indicators (e.g. Multiples For the number, (This is a quantified value), it needs to be passed first. The function is standardized to [0, 100], then weights are assigned according to the importance of the indicators, and finally a comprehensive score is obtained by weighted summation. The formula is as follows:

[0088] (1)

[0089] in, , , These are the weighting coefficients for the serum-specific IgE level index, the skin mast cell infiltration number index, and the skin reaction grade index in A1, respectively.

[0090] The specific calculation formula is as follows:

[0091] (2)

[0092] Standardized parameter settings are based on (fully matching the original model standard): Original model requirements (Pass line) Based on preliminary experimental data, the immune response is saturated when the multiple reaches 5, therefore, the threshold is set as follows: =2, =5; for Original model requirements (Pass threshold) Preliminary experiments showed that exceeding 30 per high-power field resulted in excessive skin inflammation, which could lead to model death. Therefore, the threshold was set as follows: =10, =30; for The values ​​have already been quantized to [0, 100] according to the original model levels, requiring no further standardization; they can be directly substituted into the formula. Weight constraints: This ensures that the weighting distribution conforms to the normalization principle and avoids an imbalance in the scoring caused by an excessively high weight for a certain indicator.

[0093] Step 2: A1 Model Training; After the model is built, it needs to be trained to determine its performance. , , Specific values, to ensure It can accurately reflect the physiological reliability of the model and avoid the bias caused by subjective weight setting.

[0094] 2.1 Training data collection; Sample size selection: 100 stable canine pruritus models that have been validated through step S10 of the original model construction method were selected (covering two breeds, Beagle and Labrador Retriever, aged 6-12 months, weighing 8-12kg, to ensure sample diversity); Data collection content: For each model, the following data were extracted from the corresponding step of the original model construction: 1. and (Based on serum test reports from steps S1 and S7); 2. (Derived from the pathological slide counting record in step S7); 3. (Derived from the skin reaction monitoring table of steps S5 and S6, using the average quantitative value 3 days after enhanced sensitization); Manual scoring collection: Five veterinarians with more than 10 years of experience in canine dermatology research were invited to independently score the physiological reliability of each model. The scoring criteria include: the intensity of the immune response ( ), degree of inflammation ( Skin reaction stability The scoring range is [0, 100], and the average score of 5 veterinarians is taken as the artificial reference standard for the model.

[0095] 2.2 Weight Calculation Method; Training Objective: To find a set of weights. , , , making Calculated Compared with human rating The mean square error (MSE) is minimized, that is:

[0096] (3)

[0097] Substitute formula (1) into formula (3) and combine... Given the constraints, construct a system of linear equations. Solve using matrices (using MATLAB's `lsqlin` function or Excel's Solver tool) to obtain the trained weight coefficients (example results; actual calculations should be based on real data): =0.4 (serum IgE has the highest weight because the immune response is the root cause of itching). =0.3 (mast cells are a key pathological link). =0.3 (skin reaction is a direct manifestation).

[0098] 2.3 Training, Validation, and Optimization; Validation Dataset: Twenty newly constructed stable canine pruritus models were selected (without overlap with the training samples, and consistent breed, age, and weight distributions). Data was collected using the same method, and human ratings were obtained. ; Validation metrics: Calculate the validation set and coefficient of determination (Reflecting the degree of linear correlation between the two), if This indicates that the A1 score is highly consistent with the expert evaluation, and the model training is successful; Optimization strategy: If It needs to be readjusted. of , Parameters (such as expansion) (Extend the training sample size) or increase the training sample size (up to 150 animals), repeating steps 2.1-2.2 until... .

[0099] Step 3: A1 Model Application; After successful A1 training, it can be used for the physiological and pathological evaluation of the newly constructed stable canine pruritus model. The specific application process is as follows:

[0100] 3.1 Input Data Acquisition; Taking a newly constructed beagle pruritus model (number 2024-B-001, age 8 months, weight 10kg) as an example, the input data is obtained from the following steps of the original model construction method: After passing the screening in step S1, serum was collected and tested. =3.2 IU / mL; Blood was drawn on day 3 after the enhanced sensitization step S7, and the results were obtained. =8.5 IU / mL→ ; Step S7: Skin tissue section counting; mast cell counts in 5 high-power fields were 12, 15, 14, 13, and 16 respectively → average. =14 per high-power field of view; S6 steps enhanced the skin reaction levels 3 days after sensitization, which were grade 2, grade 2, and grade 3, with quantified values ​​of 66, 66, and 100 respectively, representing the average value. .

[0101] 3.2 Calculation 1. Standardized calculation (according to formula 2): ; ; .

[0102] 2. Substitute into Formula 1 to calculate. : (The results here are low, indicating that the physiological basis of this model is weak.)

[0103] 3.3 Physiological Basis Evaluation and Application of Results; based on The values ​​are used to divide the physiological basis of the model into three levels. The evaluation criteria are formulated by combining the original model's pass line with industry experience: Excellent ( Strong immune response ( The degree of inflammation is moderate. ), skin reaction stable ( The physiological basis is reliable, and it can be directly entered into the A2 behavioral assessment; good ( ): Qualified immune response ( ), inflammation level reaches standard ( ), skin reaction stable ( The physiological basis is basically reliable, and behavioral stability should be the focus of the A2 assessment; unqualified ( Insufficient immune response ) or low degree of inflammation ( ) or unstable skin reaction ( The physiological basis is unreliable, and steps S4-S7 of the original model (resensitization and detection) need to be repeated until... In this example If the physiological basis is deemed unqualified, the process must be repeated in step S4 of the original model, with the dosage of the sensitizer adjusted (e.g., the ovalbumin concentration increased from 10 mg / mL to 15 mg / mL), and the subsequent steps must be completed again.

[0104] In some embodiments, physiological and pathological quantitative evaluation can be incorporated into the CADESI-4 skin lesion scoring scale (reference). Figure 3The quantitative results of the study incorporate the distribution range of skin lesions (e.g., head, trunk, limbs, etc.) and severity scores into a comprehensive calculation with preset weights (e.g., 0.2). This, along with the serum IgE fold change ratio (weight 0.4), mast cell infiltration count (weight 0.3), and skin reaction grade quantification value (weight 0.1), forms a weighted system to further improve the accuracy of physiological and pathological evaluation. The severity of skin lesions is quantified according to the grading standard of the canine atopic dermatitis skin lesion severity index to ensure data consistency.

[0105] III. Algorithm 2 (A2): Stability Evaluation Algorithm for Itching Behavior; The core value of a stable canine pruritus model lies in its repeatable pruritus behavior. If the behavior fluctuates too much (e.g., frequent scratching on one day, no scratching on another), it cannot be used for drug efficacy evaluation or mechanism research. The core objective of A2 is to combine the behavioral monitoring data from steps S8-S9 of the original model with the output of A1. (Physiological basis), quantitatively evaluate the stability of the model's behavior, and output. (Itching behavior stability index). Simultaneously, by introducing... As a weighting adjustment factor The logic is to prioritize stability evaluation based on a better physiological foundation, thus avoiding misjudgments of individuals with poor physiological foundations but occasional stable behavior.

[0106] Step 1: A2 model construction;

[0107] 1.1 Indicator Selection; In the original model, steps S8-S9 recorded itching behavior through video monitoring. The core behavioral data included duration, frequency, and duration of each occurrence, which respectively reflect the total intensity, frequency, and severity of itching. Therefore, A2 selected the following three indicators, and introduced the coefficient of variation to reflect stability: Indicator 1: Coefficient of variation of the total duration of itching behavior over 3 days (… Source: Behavioral records from the first 4 hours after induction in step S8, and behavioral records from the 4 hours after two consecutive days of induction in step S9, totaling 3 days. , , Calculate the average total duration of itching behavior over 3 days. with standard deviation Later obtained Reason for selection: The coefficient of variation is a dimensionless indicator reflecting the degree of data fluctuation, which the original model requires. This indicator directly reflects behavioral stability and is the core evaluation indicator of A2. Indicator 2: Average frequency of total itching behavior over 3 days ( ); Source: Same as above, from 3 days Calculate the average value. Reason for selection: Frequency reflects the frequency of itching occurrence. If the frequency is too low (e.g., <5 times / hour), even with good stability, it cannot meet the experimental requirements (e.g., behavioral changes are difficult to observe after drug intervention), and needs to be used as a supplementary indicator of intensity. Indicator 3: The average duration of a single itching behavior over 3 days ( ); Source: Same as above, from 3 days ( Calculate the average value. Reason for selection: The duration of a single occurrence reflects the severity of itching. If the duration is too short (e.g., <2 seconds / occurrence), it may indicate that the model has a weak response to the sensitizer and should be used as a supplementary indicator of severity.

[0108] 1.2 Formula Structure Design; The scoring logic for A2 prioritizes stability, with intensity and severity as secondary considerations, but also takes into account physiological factors ( ) will affect weight allocation; the better the physiological basis ( The higher the level, the more reliable the physiological and pathological mechanism driving the behavior; in this case, more emphasis should be placed on stability. ); Physiological basis ( If the severity is low, the weights of the intensity and severity indicators need to be appropriately increased to avoid overlooking models with average physiological basis but stable behavior and sufficient intensity. Therefore, the A2 formula is divided into two parts: first through... Calculate the weight adjustment factor Then substitute into the master formula to calculate. .

[0109] 1. Weighting adjustment factor The calculation formula is as follows:

[0110] (4)

[0111] Formula design basis: When (Physiological basis is not up to standard): Reduce the weight of stability indicators to avoid overestimating false stability due to poor physiological basis; when (Physiological foundation qualified): Follow Linear increase, each point improved , Increasing by 0.005 results in better physiological function and greater stability; when >200 (theoretical extreme value, practically impossible to achieve): To avoid overweighting the intensity indicator and causing it to be ignored.

[0112] 2. Stability index of pruritus behavior Main calculation formula: (5)

[0113] in: (Stability score): Based on Calculation, the original model requires ,therefore:

[0114] (6)

[0115] when When =0, =100; When =0.1, =90; (Combined score of intensity and severity): Fusion and Standardize first, then take the average:

[0116] (7)

[0117] Standardized parameter settings (based on the original model's behavioral data range): : The minimum value is 5 times / hour (lower limit of behavioral intensity), and the maximum value is 20 times / hour (upper limit of behavioral intensity; exceeding this may indicate model overreaction). =5, =20; : The minimum value is 2 seconds per instance (lower limit of severity), and the maximum value is 10 seconds per instance (upper limit of severity; exceeding this value may indicate pain caused by skin lesions rather than itching). =2, =10.

[0118] Step 2: A2 model training; the core of A2 training is the validation of the introduction. back, Can it more accurately reflect behavioral stability and avoid evaluation bias caused by unreasonable weight adjustment factor design?

[0119] 2.1 Training Data Collection; Sample Size Selection: The 100 stable canine scratching models used in A1 training were adopted (to avoid sample differences affecting training results); Data Collection Content: For each model, the following behavioral data were extracted from steps S8 and S9 of the original model construction: 1.3 days , , (Total duration of itching for 4 hours per day as recorded by video surveillance); 2.3 days , , (Total daily frequency of behaviors as counted by video analytics software); 3. Calculated by A1 (A1 training and verification have been passed to ensure the reliability of the physiological baseline score); Manual score collection: Five experts with more than 8 years of experience in canine behavioral research were invited to independently score the behavioral stability of each model. The scoring criteria include: fluctuation in total behavior duration ( ), behavioral intensity ( ), behavioral severity ( ), and also refer to (Physiological basis), scoring range [0,100], and the average score of 5 experts is taken as the reference standard.

[0120] 2.2 Training Objectives and Validation (Comparison with and without) (Scoring effect); Training objective: to validate the introduction back, and Whether the correlation is significantly improved. The specific steps are as follows: 1. None Calculation of the control group: fixed =1 (considering only stability and ignoring physiological basis), substitute into formulas (5)-(7) to calculate the model for each model. 2. Yes The experimental group was calculated according to formula (4). Then substitute into formulas (5)-(7) to calculate. 3. Correlation comparison: Calculate separately and of ,as well as and of 4. Qualification Standard: If and This indicates the introduction The subsequent scoring is more accurate, and A2 training is considered satisfactory; if or The coefficients in formula (4) need to be adjusted (e.g., change 0.005 to 0.006) and the calculation needs to be recalculated. And verify until the pass / fail criteria are met. Example result: Assume =0.72, =0.83, which indicates the introduction of back, The correlation with expert scores has significantly improved, indicating that A2 training is satisfactory.

[0121] Step 3: A2 Model Application; After the A2 model is successfully trained, it can be applied to behavioral data from the new model and A1. To evaluate its behavioral stability, the specific application process is as follows:

[0122] 3.1 Input Data Acquisition; Taking another newly constructed Labrador Retriever itching model (number 2024-L-002, age 10 months, weight 11kg) as an example, first obtain the data through A1 evaluation. =75.0 (good physiological foundation), and then obtain the following behavioral data from steps S8 and S9 of the original model: Step S8 (day 1 induction): =125 seconds / hour =9 times / hour→ seconds / time (more than) (Counted as 10); S9 Step 1 (Day 2 induction): seconds / hour times / hour → Seconds / time (calculated as 10); S9 step, day 2 (day 3 induction): seconds / hour times / hour → Seconds per tick (in 10 increments).

[0123] 3.2 Calculation (Calculate step by step, combining A1) );

[0124] 1. Calculation (Stability Indicators): seconds per hour;

[0125] seconds per hour;

[0126] (<0.1, meets the stability criteria of the original model).

[0127] 2. Calculation (Weight adjustment factor, based on A1) =75.0): because Substitute into formula (4): (Good physiological foundation, therefore) (Higher, with an emphasis on stability).

[0128] 3. Calculation (Stability score): Because Substitute into formula (6): .

[0129] 4. Calculation (Combined score of intensity and severity): Calculated : times / hour; Seconds per time (all calculated at the maximum limit);

[0130] Standardized calculation: ; Substitute into formula (7): .

[0131] 5. Substitute into formula (5) to calculate. : .

[0132] 3.3 Evaluation of Behavioral Stability and Application of Results; based on The value of is used to divide the model's behavioral stability into three levels, and the evaluation criteria are combined with the original model stability criteria ( ) and experimental requirements formulation: Excellent ( Stability score ( Intensity and severity scores The behavior is stable and of sufficient intensity, making it suitable for demanding experiments (such as long-term drug efficacy evaluation); good ( Stability score ( Intensity and severity scores Their behavior is generally stable and they can be used in routine experiments (such as short-term mechanism studies); Unqualified ( Stability score ( (or intensity and severity score) If the behavior fluctuates greatly or the intensity is insufficient, the original model steps S8-S9 need to be repeated (adjusting the induction dose or extending the observation time) until... In this example The behavior was deemed to be stable and can proceed to the A3 comprehensive suitability evaluation.

[0133] In some embodiments, the assessment of pruritus behavior stability can incorporate PVAS subjective scoring data (see reference). Figure 4 The coefficient of variation (CVV) of the PVAS scores over three consecutive days was calculated and combined with the CVV of the total duration of itching behavior at a 1:1 weight to form a composite CVV. This composite CVV was then combined with the combined scores of behavior intensity and severity, as well as the stability score, to calculate the behavior stability index. The CVV calculation method is consistent with that of the CVV of the total duration of itching behavior, that is, derived from the average and standard deviation of the three-day scores to ensure co-validation of subjective and objective behavioral data.

[0134] V. Algorithm 3 (A3): Comprehensive Adaptability Evaluation Algorithm; The ultimate value of a stable canine pruritus model lies in meeting specific research needs. For example, a model used for evaluating the efficacy of antipruritic drugs needs to possess both a reliable physiological basis and stable behavioral performance, while a model used for studying the pathogenesis of pruritus focuses more on the clarity of physiological and pathological indicators. The core objective of A3 is to integrate the features of A1. (Physiological basis) and A2 (Behavioral stability) Based on the needs of different application scenarios, output a comprehensive adaptability score for each scenario. This allows us to determine whether the model is suitable for the specific research task and to achieve on-demand evaluation.

[0135] Step 1: A3 model construction;

[0136] 1.1 Full Enumeration of Application Scenarios; Based on the mainstream research applications of the stable canine pruritus model and combined with the design goals of the original model construction method, the following three core application scenarios are enumerated, and the key requirements of each scenario are clarified:

[0137] Scenario 1 (k=1): Evaluation of the efficacy of antipruritic drugs; Research objective: To test the relieving effect of candidate drugs (such as antihistamines and anti-inflammatory drugs) on canine pruritus. Key requirements: ① Reliable physiological basis ( High), ensuring clear drug targets (e.g., IgE-mediated immune responses, mast cell activation); ② High behavioral stability ( (High), ensuring that behavioral changes before and after drug intervention are measurable; ③ Sensitivity to positive drugs (such as loratadine) (scenario fit index) (The core consideration) is to avoid misjudgment that the drug is ineffective but the model itself does not respond.

[0138] Scenario 2 (k=2): Research scenario on the pathogenesis of pruritus; Research objective: To explore the molecular mechanisms of canine pruritus (such as the role of inflammatory mediators and nerve signal transduction). Key requirements: ① Clear physiological and pathological indicators ( High, especially and ① Ensure traceability of mechanism research; ② Skin tissue samples can be repeatedly obtained (scenario adaptability index) The core considerations are: ① to avoid the inability to take multiple samples due to excessive skin damage; ② behavioral performance does not need to be extremely stable ( The requirements are lower than in scenario 1, because mechanism studies focus more on physiological and pathological changes.

[0139] Scenario 3 (k=3): Dermatological toxicology testing scenario; Research objective: To evaluate the sensitization and pruritus-inducing risk of new cosmetics and topical medications on canine skin. Key requirements: ① Sensitive behavioral response ( High, especially and ① Ensure that itching caused by toxicological irritation can be observed quickly; ② Ensure normal skin barrier function (scenario suitability index) The core considerations are to avoid affecting toxicological results due to skin damage; ③ It is sufficient to meet the physiological requirements ( The requirements are lower than those in scenarios 1 and 2, because toxicology tests focus more on the response to exogenous stimuli.

[0140] 1.2 Formula Structure Design; The core of A3 is scenario-based weighted fusion, assigning specific requirements to different scenarios. , , Different weights result in a comprehensive scene-specific adaptability score. The formula is as follows:

[0141] (8)

[0142] in: , , A3 respectively The weighting coefficient of the (comprehensive physiological and pathological score), The weighting coefficient of the (itch behavior stability index) The weighting coefficients for (scene suitability score); satisfying The weighting is determined by the key requirements of the scenario (e.g., scenario 1 requires...). and Both are relatively high, Scenario 2 requires Higher values ​​k=1,2,3: correspond to the three application scenarios mentioned above; The suitability score for the k-th scenario ([0,100]) is determined by domain experts based on scenario-specific requirements. The specific scoring criteria are as follows:

[0143] Scene k <![CDATA[C A3,k Scoring criteria (out of 100 points, points deducted based on the degree of satisfaction) k=1 1. Itch relief rate ≥50% with positive drug (10mg / kg loratadine orally) (40 points); 2. Behavioral record repeatability ≥90% (30 points); 3. No skin lesions (30 points) k=2 1. Skin tissue can be sampled more than 3 times (40 points); 2. The error in the detection of physiological and pathological indicators is ≤10% (30 points); 3. The model survival time is ≥30 days (30 points). k=3 1. Normal skin barrier function (transepidermal water loss ≤20g / (m²·h), 40 points); 2. Itching occurs within 1 hour after exogenous stimulation (30 points); 3. No systemic allergy (30 points).

[0144] Step 2: A3 Model Training; The core of A3 training is to determine the optimal model for each scenario using large sample data. , , Specific values, to ensure It can accurately reflect the model's suitability for the scenario.

[0145] 2.1 Training Data Collection; Sample Size Selection: For each scenario, select 50 stable canine pruritus models that have been used in the study of that scenario (a total of 150 models, some models can be used across scenarios but need to be re-evaluated); Data Collection Content: For each model, collect the following data: 1. (Based on A1 evaluation results); 2. (Based on A2 evaluation results); 3. (The scores are derived from the model's suitability ratings in the corresponding scenarios by domain experts; the average score is calculated after independent ratings from 5 domain experts). Human suitability ratings were collected by inviting 5 senior researchers in the corresponding scenarios (e.g., pharmacologist for scenario 1, pathologist for scenario 2, and toxicologist for scenario 3) to independently rate the scenario suitability of each model. The scoring range is [0, 100], and the final average value is taken as the reference standard.

[0146] 2.2 Calculation of scene weights; For each scene k, construct a least-squares optimization objective:

[0147] (9)

[0148] Substitute formula (8) into formula (9), and combine... The constraints are used to obtain the scene-specific weights through matrix solving (example results; actual calculations should be based on real data):

[0149] Scene k Scene Name <![CDATA[α A3,k (P) A1 Weight) <![CDATA[β A3,k (B) A2 Weight) <![CDATA[γ A3,k (C) A3,k Weight) k=1 Drug efficacy evaluation 0.35 (The physiological basis is important) 0.35 (Behavioral stability is important) 0.3 (Positive drug sensitivity is important) k=2 Pathogenesis research 0.5 (Core physiological indicator) 0.2 (Behavioral stability, minor) 0.3 (Tissue sampling is important) k=3 Skin toxicology test 0.2 (As long as the physiological basis is qualified) 0.5 (Behavioral Sensitivity Core) 0.3 (Skin barrier is important)

[0150] 2.3 Training and Validation; For each scenario, 10 new models were selected as the validation set, and the training and validation were performed. and of If all scenarios If the A3 training is successful, then the training is qualified; otherwise, adjust the weight coefficients of the corresponding scenario and retrain until it is qualified.

[0151] Step 3: A3 Model Application (Scenario-Specific Calculation) (To determine suitability); after the A3 model has passed training, its suitability can be evaluated for different scenarios. The specific application process is as follows:

[0152] 3.1 Input Data Acquisition (Fusing A1 and A2 Results with Scene-Specific Data); using the Labrador Retriever model with good behavioral stability from step 3.2 (number 2024-L-002, , Taking this as an example, we will obtain data for each of the three scenarios. :

[0153] Scenario 1 (k=1, drug efficacy evaluation): 1. After oral administration of 10mg / kg loratadine, the total duration of itching decreased from 121.67 seconds / hour to 58 seconds / hour → relief rate = (121.67-58) / 121.67≈52% (≥50%, 40 points); 2. The behavior record was repeated 3 times, and the coefficient of variation of the total duration was 0.02 (≤0.05, repeatability ≥90%, 30 points); 3. No skin damage (30 points) → .

[0154] Scenario 2 (k=2, pathogenesis study): 1. After 3 skin tissue samplings, the wounds healed within 3 days (≥3 samplings, 40 points); 2. Repeat detection error = 5% (≤10%, 30 points); 3. Model lifespan has reached 45 days (≥30 days), 30 points → .

[0155] Scenario 3 (k=3, Dermatological Toxicology Test): 1. Transdermal water loss rate = 18 g / (m²·h) (≤20, 40 points); 2. Itching occurs 45 minutes after topical application of the sensitizing agent (≤1 hour, 30 points); 3. No systemic allergic reaction (30 points) → .

[0156] 3.2 Scenario-based Calculation (Combined with scenario-specific weights); 1. Scenario 1 (k=1, drug efficacy evaluation, , , ): 2. Scenario 2 (k=2, pathogenesis research) , , ): 3. Scenario 3 (k=3, skin toxicology test) , , ): .

[0157] 3.3 Scenario Adaptability Evaluation and Application Suggestions; based on The value of is used to categorize the model's adaptability to each scenario into three levels: suitable, basically suitable, and unsuitable. The evaluation criteria are formulated based on scenario requirements and experimental reliability requirements.

[0158] Scene k <![CDATA[Adapt (Q A3,k ≥ threshold)]]> <![CDATA[Basic adaptation (threshold 1 ≤ Q A3,k < threshold 2)]]> <![CDATA[Not adapted (Q A3,k <threshold 1)]]> Evaluation Results and Application Recommendations for This Example k=1 ≥80 <![CDATA[60≤Q A3,1 < 80]]> <60 <![CDATA[Q A3,1 =80.79, suitable, can be used for long-term efficacy evaluation of antipruritic drugs, and it is recommended to focus on monitoring behavioral changes after drug intervention.<!-- 18 --> ]]> k=2 ≥75 <![CDATA[55≤Q A3,2 < 75]]> <55 <![CDATA[Q A3,2 =81.53, suitable, can be used for molecular-level research on the pathogenesis of pruritus; it is recommended to prioritize the analysis of immune and inflammatory markers. k=3 ≥80 <![CDATA[60≤Q A3,3 < 80]]> <60 <![CDATA[Q A3,3 =80.07, compatible, suitable for toxicological testing of topical skin products; it is recommended to monitor rapid behavioral responses after exogenous irritation.

[0159] In some embodiments, the scenario-specific adaptability score can be combined with scale data (see reference). Figure 3 and Figure 4Optimized standards: Drug efficacy evaluation scenario: Add PVAS score relief rate ≥50% (20 points), adjust the original standard weight to pruritus relief rate ≥50% after oral administration of loratadine (30 points), behavioral record repeatability ≥90% (25 points), no skin damage (25 points), and the total score remains 100 points; Pruritus pathogenesis research scenario: refine the physiological and pathological index detection error ≤10% to serum IgE detection error ≤10% and CADESI-4 score error ≤10% to ensure the comprehensiveness of pathological index evaluation; Skin toxicology testing scenario: Add PVAS score ≥6 points (moderate and above pruritus) (20 points), adjust the original standard weight to transdermal water loss rate ≤20g / (m²・h) (30 points), pruritus behavior within 1 hour after topical sensitization (30 points), and no systemic allergic reaction (20 points) to improve the specificity of scenario adaptability evaluation. For example, during the data collection phase, additional CADESI-4 total skin lesion score and PVAS score data for three consecutive days need to be extracted; in the physiological and pathological quantitative evaluation, the CADESI-4 total score is standardized as "(actual score / full score) × 100" and then included in the weighted calculation; in the evaluation of pruritus behavior stability, the comprehensive coefficient of variation = (coefficient of variation of total behavior duration + coefficient of variation of PVAS score) / 2, ensuring that the stability evaluation covers both objective behavior and subjective feelings; in the comprehensive suitability evaluation, the scoring criteria for each scenario are implemented according to the optimized criteria of the fused scale, improving the matching degree between the evaluation results and the actual research needs. See the specific lesion severity scores below. Figure 5 Severity index of skin lesions in canine atopic dermatitis. Figure 5 Information cited from Rodrigues Gonçalves, Bárbara & Dantas Matos, Bruna & Faleiro, Mariana Batista & Emmanuel Arnhold, & Chediak Matos, Moema & Iglesias Santin, Ana & Moura, Veridiana. (2018). Correlation between clinical findings, mast cell count and interleukin 31immunostaining in the skin of dogs with atopic dermatitis.CiênciaRural.48.10.1590 / 0103-8478cr20180004.

[0160] Figure 3Validation of the Extent andSeverity of Canine Atopic Dermatitis (ICADA) by the CADESI-4 Diagnostic Surveillance System Olivry T,Saridomichelakis M,Nuttall T,Bensignor E,Griffin CE,Hill PB; Index (CADESI)-4,a simplified severity scale for assessing skin lesions of atopic dermatitis in dogs.Vet Dermatol.2014 Apr;25(2):77-85,e25.doi:10.1111 / vde.12107.Epub 2014 Jan 25.PMID:24461108.

[0161] Figure 4 Antibodies to PVAS: Cosgrove SB,Wren JA,Cleaver DM,Martin DD,Walsh KF,Harfst JA,Follis SL,King VL,Boucher JF,Stegemann M R.Efficacy and safety of oclacitinib for the control of pruritus and associated skin lesions in dogs with canine allergic dermatitis[J].Veterinary Dermatology,2013,24(5):479-e1

[0162] The Visual Analogue Scale (VAS) uses a 10-centimeter straight line as its base, with extreme states (such as "none" and "most intense") marked at both ends. Patients mark corresponding positions on the line based on their subjective feelings, and healthcare professionals measure the distance from the marked point to the 0-point end (accurate to millimeters), converting it into a numerical score of 0-10. Besides its common use in pain assessment, it can quantify various subjective feelings, such as "0 points no anxiety - 10 points extreme anxiety" to assess anxiety level, "0 points no fatigue - 10 points extreme fatigue" to measure physical fatigue, "0 points no itching - 10 points unbearable itching" to assess the intensity of skin itching, and even be used to quantify indicators such as quality of life ("0 points completely dissatisfied - 10 points very satisfied") and appetite ("0 points no appetite - 10 points excellent appetite"), demonstrating its universality and flexibility in converting abstract feelings into objective values.

[0163] This study used the modified Visual Analogue Scale (VAS) score adopted by Apok in 2013. The VAS scale consists of a 10 cm long straight line containing six descriptive words for itching, evenly distributed at 2 cm intervals, with 0 cm representing "normal dog" and 10 cm representing "extreme pruritus".

[0164] Overall evaluation process:

[0165] 1. Data Preparation Stage: Extract data from steps S1, S5, S6, S7, S8, and S9 of the stable canine pruritus model construction method. , , , , , Wait for all input data;

[0166] 2. A1 evaluation stage: Substitute into formulas (1)-(2) to calculate ,like If the value is <60, return to step S4 of the original model for resensitization; if the value is ≥60, proceed to A2 evaluation.

[0167] 3. A2 evaluation stage: Substitute into formulas (4)-(7) to calculate ,like If the value is less than 70, return to step S8 of the original model for re-induction; if the value is greater than or equal to 70, proceed to evaluation in step A3.

[0168] 4. A3 Evaluation Stage: Invite domain experts to score. Substitute into formula (8) to calculate It outputs adaptability evaluation and application suggestions according to the scenario;

[0169] 5. Report generation stage: Integrate the calculation process, results and evaluation conclusions of A1, A2 and A3 to form a multi-dimensional evaluation report of the stable canine pruritus model, which is used for experimental design and model quality control.

[0170] In some embodiments, the multi-dimensional evaluation method for the stable canine pruritus model incorporates the Bayesian Network (BN) algorithm, forming a four-algorithm synergistic fusion system. The core logic is as follows: 1. Data Association: BN uses the input indicators of A1 and A2 as observation nodes, the outputs of A1 and A2 as latent variable nodes, and the scenario requirements of A3 as evidence nodes to construct a probabilistic inference network; 2. Dynamic Weight Adjustment: The posterior probability output of BN serves as the weight correction factor for A2 and A3, achieving a closed loop of uncertainty quantification → dynamic weight allocation → evaluation result optimization; 3. Bidirectional Interaction: The output of A1 influences the prior probability of BN, and the output of BN corrects the weights of A2; the output of A2 updates the posterior probability of BN, and the output of BN further corrects the weights of A3; the results of A3 feed back into the parameter optimization of BN, forming a pairwise interactive fusion relationship.

[0171] (a) Model construction of Bayesian Network (BN) algorithm;

[0172] 1. Network Structure Design (Nodes and Causal Relationships); The core of Batch Normalization (BN) is to express the causal relationships between variables using a Directed Acyclic Graph (DAG). Nodes and edges are defined as follows: Observation Node (Input): (Serum IgE fold change), M (mast cell count), S (skin response), CV (coefficient of variation in behavior) (Mean frequency of behavior) (Average duration of a single event) (Scene adaptability); Latent variable nodes (intermediate output): (A1 comprehensive physiological score) (A2's behavioral stability index); Target node (final output): (Scene adaptability score for A3); Edge relationships: (Physiological indicators affect physiological scores); (Physiological scores and behavioral indicators together influence behavioral stability); (Physiological, behavioral, and scene adaptability all contribute to the final adaptability).

[0173] 2. Definition of Conditional Probability Table (CPT); CPT is used to quantify the probabilistic relationships between nodes, for example... Indicates that in a given M, S The conditional probability. All nodes are discretized into three levels (low / medium / high) according to their value range, as shown in the example below: Low (2-3), Medium (3-4), High (4-5); Low (0-60), Medium (60-80), High (80-100); CPT Medium (When all physiological indicators are high, the probability of a high physiological score is 80%).

[0174] (ii) Fusion of A1 and BN: Prior probability update; A1 output As a latent variable in BN, BN is determined by the input index of A1 ( (M, S) Update The probability distribution provides a basis for uncertainty correction for A2.

[0175] Model construction and calculation for A1; Step 1: Indicator standardization; standardization: ( M standardization: ( S does not need to be standardized (it is already [0,100]).

[0176] Step 2: Calculation of comprehensive physiological and pathological scores ;in , , A1 respectively The weighting coefficients of M, S, and M; weights =0.4, =0.3, =0.3, obtained by fitting the training data.

[0177] 2. BN's probability update of A1; BN updates the probability of A1 based on the input index of A1 using Bayes' theorem. Posterior probability: ;

[0178] Historical data statistics Prior distribution (e.g.) ); Conditional probabilities defined in CPT (e.g.) ); Marginal probability of the evidence (normalization constant). Output: (like ,express The probability of it being high is 75%.

[0179] (III) Integration of A2 and BN: Dynamic weight adjustment; A2 weight adjustment factor Originally only relied on After integrating BN, through Correction ,reflect The impact of uncertainty on behavioral evaluation.

[0180] 1. Initial model construction for A2; Step 1: Calculation (Initial weight adjustment factor);

[0181] ;

[0182] Step 2: Calculate the stability score and strength score; Stability score ( (Otherwise, it is 0); Intensity score ; , .

[0183] 2. BN's weight adjustment for A2 (interactive formula); BN outputs the weight adjustment factor for A2. (based on (correction factor) ; The closer to 1, The closer the value is to 1.2, the stronger the stability weight; conversely, the closer it is to 0.8, the weaker the stability weight. The final calculation of A2 after fusion: The final weight adjustment factor of A2 after incorporating BN. (Normalization ensures) ).

[0184] (iv) Integration of A3 and BN: Optimization of scene weights; Scene-specific weights of A3 ( , , Originally fixed, after incorporating BN, through posterior probability Dynamic adjustments reflect the impact of the reliability of behavioral stability on scene adaptability.

[0185] 1. Initial model construction for A3; Step 1: Scene adaptability scoring Drug efficacy evaluation (k=1): based on scores such as positive drug remission rate and behavioral repeatability; Mechanism study (k=2): based on scores such as tissue sampling feasibility and indicator stability; Toxicological testing (k=3): based on scores such as skin barrier function and reaction sensitivity. Step 2: Calculation of initial comprehensive suitability score; ;in , , A3 respectively Initial weighting coefficients Initial weighting coefficients Initial weight coefficients; Initial weights: when k=1 =0.35, =0.35, =0.3; when k=2 =0.5, =0.2, =0.3; when k=3 =0.2, =0.5, =0.3.

[0186] 2. BN's weight optimization for A3 (interactive formula); BN is based on With CV, , Relationship, output posterior probability :

[0187] ;

[0188] Final weights and calculations of A3 after fusion: Adjusted weights (must be normalized to a sum of 1): ; ; Final overall fit score: .

[0189] (v) Feedback optimization of BN and A3 (closed-loop fusion); A3 output Feedback to BN, updating its parameters (Conditional probability table), optimized using maximum a posteriori estimation (MAP):

[0190] ;

[0191] Let be the prior distribution of the parameters. It is the likelihood function, updated based on the new evaluation results.

[0192] Model Training Process (Four-Algorithm Collaborative Training); 1. Training Data Preparation: Select historical data from 150 stable canine pruritus models, including: A1 Input: M, S; Output: And veterinary manual scoring; A2 input: CV, , Output: And human scoring by behavioral science experts; A3 input: Output: 2. A1 training; optimization using least squares method. , , ,make Minimize the mean squared error (MSE) with veterinary manual scoring, requiring 3. A2 training; verification before and after merging BN. The correlation with behavioral expert ratings requires fusion. (Higher than before fusion) 4. Batch Normalization (BN) training; learning using maximum likelihood estimation. (Conditional probability table), using 10-fold cross-validation to ensure network inference accuracy (prediction). , (Accuracy ≥ 80%). 5. A3 training; optimize initial weights for different scenarios. , , This makes the fused BN Ratings by domain experts .

[0193] Core Role and Integration Value: 1. The core role of Bayesian Networks (BN): Quantifying the uncertainty of A1 and A2 indicators (e.g., High but 1. Low conflict), balancing the influence of each indicator through probabilistic reasoning to avoid the arbitrariness of a single threshold judgment; 2. The value of integrating A1 and BN: transforming the absolute scores of physiological indicators into probability distributions, providing a confidence basis for adjusting the weights of A2 (e.g., When the probability is high but low, its impact on A2 is weakened; 3. The value of integrating A2 and BN: By correcting the scenario weights through the reliability probability of behavioral indicators, the evaluation of A3 is made more consistent with the actual performance of the model (e.g., When the value is high but fluctuates greatly, reduce its weight in scenario adaptability); 4. Overall integration value: form a closed loop of indicator → score → probability → correct score, so that the evaluation results remain robust when there is data noise and individual differences, and improve the reliability of the model in drug development and mechanism research.

Claims

1. A method for constructing a stable model of canine pruritus, characterized by, Includes the following steps: S1. Select healthy experimental dogs, record their basic information, and form an experimental dog basic information table; S2. Transfer the experimental dogs selected in S1 to a standardized experimental kennel for 7 days to adapt. After the adaptation period, confirm that the condition of the experimental dogs is consistent with the basic condition recorded in the experimental dog basic information table. S3. Prepare three sensitizing agents: ovalbumin, dinitrochlorobenzene, and house dust mite extract, and store them in a refrigerator at 4°C. S4. The experimental dogs in S2 were grouped according to the sensitizing agents. After hair removal on the back of each group of experimental dogs, the corresponding first sensitization treatment was performed, with intradermal injection of ovalbumin and house dust mite extract, and dinitrochlorobenzene applied topically. S5. After the first sensitization, monitor the redness and swelling of the sensitized site, the level of skin reaction, and scratching behavior for 7 consecutive days. After the monitoring is completed, select experimental dogs with a skin reaction level of 2 or above and no systemic adverse reactions to form a list of qualified experimental dogs for the first sensitization. S6. For the experimental dogs in the list of qualified experimental dogs for the first sensitization, use the same sensitizing agent and hair removal area as in S4, and perform enhanced sensitization treatment at 1 / 2 of the sensitization dose in S4. After treatment, observe the skin reaction daily and record it in the same way as in S5. S7. On the 3rd day after the enhanced sensitization treatment, venous blood was collected to detect the sensitizer-specific IgE. At the same time, skin tissue was taken from the sensitization site to make sections and count mast cells. Dogs with IgE increased by more than 2 times and mast cells ≥10 per high power field were screened to form a preliminary list of qualified experimental dogs. S8. For the experimental dogs in the preliminary qualified experimental dog list, use the same sensitizing agent and dosage as in S6, and induce itching again in the original hair removal area with the sensitizing agent. Record the behavior for 4 hours and quantify it to form a quantitative record table of itching behavior. S9. Repeat the S8 induction and quantification records for 3 consecutive days, calculate the coefficient of variation of the total duration of itching behavior per hour for each experimental dog within 3 days, screen dogs with a coefficient of variation ≤10%, and form a list of experimental dogs with stable itching behavior. S10. Confirm that the experimental dogs in the list of stable pruritus behavior meet the following criteria: serum IgE fold-to-fold ratio ≥2, mast cells ≥10 / high power field, skin reaction maintained at grade 2-3, and pruritus duration variation coefficient ≤10% over 3 days. These dogs are considered stable canine pruritus models.

2. The method of claim 1, wherein the stable dog pruritus model is constructed by administering the at least one of the pruritogenic substance and the at least one of the antipruritic substance to the dog. In S1, the healthy experimental dogs were Beagles or Labradors, aged 6–12 months, and weighing 8–12 kg. When recording the baseline skin condition, the back, abdomen, and inner sides of the limbs of the experimental dogs were examined. After confirming that there was no redness, swelling, peeling, or damage, a high-resolution 300 dpi photo of the skin was taken and stored in JPG format. This photo was used as a comparison basis for monitoring skin reactions in S5 and S6.

3. The method of claim 1, wherein the stable dog pruritus model is constructed by administering the at least one of the pruritogenic agent and the at least one of the anti-pruritic agent to the dog. In S3, ovalbumin and house dust mite extract were dissolved or diluted with sterile physiological saline, and dinitrochlorobenzene was dissolved with anhydrous ethanol to prepare concentrations of 10 mg / mL, 0.5%, and 5 μg / μL, respectively. The three prepared sensitizing agents were then placed into sterile reagent bottles and labeled with the name, concentration, and preparation date of the sensitizing agent.

4. The method for constructing a stable canine pruritus model according to claim 1, characterized in that, The environment of the standardized experimental kennel in S2 is as follows: temperature 22-25℃, humidity 50%~60%, and light cycle of 12h light / 12h darkness.

5. The method for constructing a stable canine pruritus model according to claim 1, characterized in that, The criteria for determining the skin reaction level in S5 are as follows: Level 0: The skin color and texture are consistent with the surrounding normal skin, without redness, swelling, erythema, or papules; Level 1: Only light red erythema with indistinct borders appears, without swelling or skin lesions; Level 2: The erythema is dark red with clear borders, accompanied by slight swelling, without papules; Level 3: The erythema is obvious, the swelling is significant, and there are ≥3 papules, some of which may be accompanied by slight exudation.

6. A multi-dimensional evaluation method for a stable canine pruritus model, characterized in that, The method for constructing a stable canine pruritus model according to any one of claims 1 to 5, wherein the multi-dimensional evaluation method for the stable canine pruritus model includes the following steps: (1) Evaluation data collection: In S1, S5-S6, S7, S8-S9, the serum IgE level, skin reaction grade, number of mast cell infiltration in the skin, and the total duration and frequency of itching behavior over 3 consecutive days were extracted from the experimental dogs after basal and enhanced sensitization. (2) Physiological and pathological quantitative evaluation: Based on the collected evaluation data, the serum IgE fold ratio is calculated and standardized together with the mast cell count. Combined with the skin reaction grade quantitative value, the comprehensive physiological and pathological score is obtained by weighting according to the preset weight. If the comprehensive physiological and pathological score is <60, the model construction S4 step is returned to resensitization. If the comprehensive physiological and pathological score is ≥60, the next step is entered. (3) Evaluation of pruritus behavior stability: The coefficient of variation of total duration is calculated using 3 days of behavioral data. The weight adjustment factor is calculated in combination with the comprehensive physiological and pathological score. Then, the comprehensive score of behavior intensity and severity and the stability score are calculated and weighted to obtain the behavior stability index. If the behavior stability index is <70, return to the S8 step of model construction for re-induction. If the behavior stability index is ≥70, proceed to the next step. (4) Comprehensive suitability evaluation: The suitability of drug efficacy evaluation, mechanism research and toxicology testing is evaluated. The comprehensive physiological and pathological score and behavioral stability index are combined with the preset weights of the scenarios to obtain the comprehensive suitability score of each scenario. (5) Application of evaluation results: The suitability is determined by the comprehensive suitability score for each scenario, and experimental application suggestions are output.

7. The multi-dimensional evaluation method for a stable canine pruritus model according to claim 6, characterized in that, The preset weighting in step (2) is as follows: serum IgE fold ratio weight 0.4, mast cell infiltration number weight 0.3, and skin reaction grade quantification value weight 0.

3.

8. The multi-dimensional evaluation method for a stable canine pruritus model according to claim 6, characterized in that, Step (2) The specific method for calculating the serum IgE fold ratio and standardizing it together with the mast cell count is as follows: When standardizing the serum IgE fold ratio, the minimum value is set to 2 and the maximum value is 5. If the value exceeds the range, it is taken as 2 or 5. The standardized value = (actual fold ratio - 2) / (5 - 2) × 100; When standardizing the number of mast cells infiltrating the skin tissue, the minimum value is set to 10 cells / high power field and the maximum value is 30 cells / high power field. If the value exceeds the range, it is taken as 10 or 30. The standardized value = (actual cell count - 10) / (30 - 10) × 100.

9. The multi-dimensional evaluation method for a stable canine pruritus model according to claim 6, characterized in that, Step (3) The weight adjustment factor is calculated as follows: weight adjustment factor = 0.5 + 0.005 × comprehensive physiological and pathological score. When the weight adjustment factor exceeds the range of 0.5-1.5, it is taken as 0.5 or 1.

5.

10. The multi-dimensional evaluation method for a stable canine pruritus model according to claim 6, characterized in that, The specific criteria for scene adaptability scoring in step (4) are as follows: Drug efficacy evaluation scenario: 40 points for ≥50% relief rate of itching after oral administration of 10mg / kg loratadine, 30 points for repeatability of behavioral records ≥90%, and 30 points for no skin damage. Total score: 100 points. Scenario for studying the pathogenesis of pruritus: 40 points for repeated sampling of skin tissue more than 3 times, 30 points for detection error of physiological and pathological indicators ≤10%, 30 points for model survival period ≥30 days, total score 100 points; Skin toxicology testing scenario: Transdermal water loss rate ≤20g / ( 40 points are awarded for the appearance of itching within 1 hour after applying the sensitizing agent, 30 points are awarded for the absence of a systemic allergic reaction, and 30 points are awarded for the absence of a systemic allergic reaction. The total score is 100 points.

Images