Oral pharmaceutical preparation taste evaluation system and taste evaluation method

Abstract

To provide a novel taste evaluation system and a taste evaluation method for assessing taste interactions arising from a combination of bitterness of an active ingredient contained in an oral pharmaceutical preparation and flavorings and sweeteners.SOLUTION: A taste evaluation system for an oral pharmaceutical preparation according to one embodiment of the present invention includes an input layer that measures actual measured values of saltiness, sourness, sweetness, umami, and bitterness generated from an active ingredient, flavorings, and sweeteners contained in the oral pharmaceutical preparation, an intermediate layer that calculates analyzed values of saltiness, sourness, sweetness, umami, and bitterness generated by interactions between the active ingredient and the flavorings and sweeteners by means of a neural network trained using actual performance values of arbitrary flavorings and sweeteners as teacher data based on the measured values, and an output layer that outputs the analyzed values.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a taste measurement system and a taste measurement method for a pharmaceutical oral formulation, and in particular to a novel taste measurement system and a taste measurement method for evaluating the interaction between the bitterness of an active ingredient contained in a pharmaceutical oral formulation and the taste caused by a combination of a flavoring agent and a sweetener. [Background technology]

[0002] Taste is a sensation perceived in response to ingested substances. The five basic tastes are salty, sour, sweet, umami, and bitter. Among these five tastes, the bitterness of active ingredients in oral pharmaceutical formulations significantly impacts medication compliance. Adding sweeteners or flavorings to pharmaceutical formulations can reduce the bitterness. Sweeteners themselves have a strong sweetness that suppresses bitterness. On the other hand, flavorings primarily suppress unpleasant tastes such as bitterness through their aroma. While the bitterness of oral pharmaceutical formulations is typically evaluated through sensory testing, reproducibility is difficult due to variations in the individual tester and their physical condition. Furthermore, achieving reproducibility requires specially trained personnel to conduct the sensory testing, which is costly and time-consuming. Furthermore, for pharmaceutical oral formulations containing highly active active ingredients, human sensory testing is undesirable from an ethical standpoint.

[0003] Taste sensors are being introduced as a method for safely and objectively evaluating the taste, such as bitterness, of oral pharmaceutical preparations. For example, Non-Patent Document 1 discloses an objective taste evaluation method for quantifying the bitterness of oral pharmaceutical preparations using a taste sensor, due to issues such as reproducibility and ethics. Patent Document 1 also discloses a taste sensor that mimics human taste by quantitatively evaluating the saltiness, sourness, sweetness, umami, and bitterness of a sample, but has not been used to evaluate samples of oral pharmaceutical preparations containing flavorings. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 4240003 [Non-patent literature]

[0005] [Non-Patent Document 1] Pharmacia, Vol. 51, No. 2, p. 130 "Scientifically studying the taste of medicines using taste sensors" Summary of the Invention [Problem to be solved by the invention]

[0006] One object of the present invention is to provide a novel taste measurement system and method for evaluating the interaction between the bitterness of an active ingredient contained in a pharmaceutical oral formulation and the taste produced by a combination of a flavoring agent and a sweetener. [Means for solving the problem]

[0007] According to one embodiment of the present invention, there is provided a taste measurement system for oral pharmaceutical preparations, which includes: an input layer that measures actual measured values ​​of saltiness, sourness, sweetness, umami, and bitterness caused by the active ingredient, flavoring, and sweetener contained in the oral pharmaceutical preparation; an intermediate layer that calculates analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness caused by the interaction between the active ingredient, flavoring, and sweetener from the actual measured values ​​using a neural network that has been trained using actual values ​​of any flavoring and sweetener as training data; and an output layer that outputs the analytical values.

[0008] In the taste measurement system for pharmaceutical oral preparations, the performance values ​​may be the results of sensory tests using any combination of flavoring and sweetening agents.

[0009] In the taste measurement system for pharmaceutical oral formulations, the flavor used in the sensory test may be one selected from the group consisting of yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, and yogurt flavors.

[0010] In the taste measurement system for a pharmaceutical oral formulation, the flavor contained in the pharmaceutical oral formulation may be the flavor used in the sensory test.

[0011] In the taste measurement system for pharmaceutical oral formulations, the sweetener used in the sensory test may be one selected from the group consisting of thaumatin, stevia, sucralose, aspartame, saccharin, acesulfame K, refined white sugar, lactose, and D-mannitol.

[0012] In the taste measurement system for a pharmaceutical oral preparation, the sweetener contained in the pharmaceutical oral preparation may be the sweetener used in the sensory test.

[0013] In the taste measurement system for a pharmaceutical oral formulation, the active ingredient may have a bitter taste.

[0014] Furthermore, according to one embodiment of the present invention, there is provided a method for measuring the taste of a pharmaceutical oral formulation, comprising: measuring actual values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from the active ingredient, flavoring, and sweetener contained in the pharmaceutical oral formulation; calculating analytical values ​​of the saltiness, sourness, sweetness, umami, and bitterness resulting from the interaction between the active ingredient and the flavoring and sweetener from the actual measured values ​​using a neural network that has been trained using actual values ​​of any flavoring and sweetener as training data; and outputting the analytical values. [Effects of the Invention]

[0015] According to one embodiment of the present invention, a novel taste measurement system and method are provided for evaluating the interaction between the bitterness of an active ingredient contained in a pharmaceutical oral formulation and the taste produced by a combination of a flavoring agent and a sweetening agent. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a diagram showing the configuration of a taste measurement system according to one embodiment of the present invention. [Figure 2]FIG. 1 is a diagram showing a combination of training samples for a taste measurement system according to one embodiment of the present invention. [Figure 3] FIG. 1 is a diagram showing the analysis results of Example 1 using the taste measurement system according to one embodiment of the present invention. [Figure 4] FIG. 1 is a diagram showing a radar chart of Example 1 using a taste measurement system according to one embodiment of the present invention. [Figure 5] The results of the sensory test for Reference Example 1 are shown below. [Figure 6] 1 shows the analysis results of Comparative Example 1 using the taste measurement system before learning the teacher data. [Figure 7] 10 shows a radar chart of Comparative Example 1 using a taste measurement system before learning training data. [Figure 8] 1 shows the analysis results of bitterness scores for Example 1 and Comparative Example 1 using the taste measurement system before and after learning of training data. [Figure 9] FIG. 10 is a diagram showing the analysis results of Example 2 using the taste measurement system according to one embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing a radar chart of Example 2 using a taste measuring system according to one embodiment of the present invention. [Figure 11] The results of the sensory test for Reference Example 2 are shown below. [Figure 12] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. [Figure 13] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. [Figure 14] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. [Figure 15] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. [Figure 16] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. [Figure 17] 1 shows training data learned by a taste measurement system according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] A taste measurement system and a taste measurement method according to one embodiment of the present invention will be described below. However, the taste measurement system and the taste measurement method of the present invention should not be interpreted as being limited to the description of the following embodiments and examples.

[0018] The inventors' research has shown that the analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from active ingredients and sweeteners contained in pharmaceutical oral formulations measured using a conventional taste measurement system correlate with the values ​​measured in human sensory tests. On the other hand, the analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from active ingredients, flavors, and sweeteners contained in pharmaceutical oral formulations did not correlate with the values ​​measured in human sensory tests. In other words, because flavors suppress unpleasant flavors with their aroma components, it was found that simply evaluating samples using a taste sensor using a conventionally known taste measurement method cannot evaluate the unpleasant flavor suppression effect of the flavor. Therefore, the neural network of the taste measurement system was trained using the values ​​measured in human sensory tests of selected flavors and sweeteners as training data. The analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness produced by the active ingredients, flavors, and sweeteners contained in pharmaceutical oral preparations, measured by a taste measurement system equipped with a neural network trained with the performance values ​​of flavors and sweeteners, were correlated with the performance values ​​of human sensory tests.It was found that the neural network trained with the performance values ​​of flavors and sweeteners can mimic the taste interactions produced by the combination of flavors and sweeteners.

[0019] (Taste measurement system) 1 is a diagram showing the configuration of a taste measurement system according to one embodiment of the present invention. The taste measurement system 1 includes an input layer 10, an intermediate layer 20, and an output layer 30.

[0020] The input layer 10 includes a sensor 100. The sensor 100 acts as a taste bud. The sensor 100 electrochemically reacts with each of the components that contribute to saltiness, sourness, sweetness, umami, and bitterness, and outputs a quantitative electrical signal. The components that contribute to saltiness may be, for example, NaCl, KCl, LiCl, etc., with NaCl and KCl being preferred. The components that contribute to sourness may be, for example, H derived from hydrochloric acid, acetic acid, citric acid, malic acid, succinic acid, etc. + are preferred. Sweetening ingredients may include glucose, sucrose, fructose, maltose, glycine, aspartame, etc., with glucose and sucrose being preferred. Umami ingredients may include glutamate, inosinic acid, guanylic acid, etc., with glutamate being preferred. Bittering ingredients may include caffeine, quinine, tannin, phenylalanine, Mg2+, etc., with caffeine being preferred. Each tasting ingredient may be one or more types. Measuring each of these ingredients allows for the quantification of saltiness, sourness, sweetness, umami, and bitterness, respectively. In the input layer 10, a sensor 100 is used to measure the actual values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from the active ingredient, flavoring, and sweetener contained in the pharmaceutical oral formulation. The active ingredient contained in the pharmaceutical oral formulation may have a bitter taste.

[0021] The intermediate layer 20 includes a neural network 200. For example, RBFN can be used as the neural network 200. For example, an improved RBFN described in Patent Document 1 is preferably used as the neural network 200. The improved RBFN is a neural network that is specialized for chemical data analysis and has enhanced generalization capabilities. The improved RBFN uses a basis auto-optimization algorithm to improve the estimation accuracy for chemical data compared to conventional methods. Furthermore, by adding a weight suppression term to the network's evaluation function, the generalization capabilities are improved compared to conventional methods.

[0022] The neural network 200 first calculates the concentration of each component from the actual measurement values ​​obtained by the sensor 100. For this purpose, the neural network 200 may be trained by inputting the actual measurement values ​​and the actual concentrations of each taste component. The relationship between the actual measurement values ​​and the actual concentrations of each taste component can be obtained by measuring the actual measurement values ​​with each sensor for a mixture containing each taste component at a known concentration.

[0023] Next, neural network 200 calculates analytical values ​​indicating the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans from the concentration of each component. To this end, neural network 200 may be trained by inputting the concentration of each component and the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans. The relationship between the concentration of each component and the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans can be obtained by sensory testing.

[0024] In the sensory test, a panel of people tastes or samples five standard samples each containing salty, sour, sweet, umami, and bitter tastes, and then tastes or samples multiple learning samples. The panel of people then sensorily evaluates the strength of each of the salty, sour, sweet, umami, and bitter tastes, and preferably compares the strength of each taste with that of the standard samples to quantify the results.

[0025] The sweetness test sample used in the sensory test may be, for example, one of sweeteners selected from the group consisting of thaumatin, stevia, sucralose, aspartame, saccharin, acesulfame K, refined white sugar, lactose, and D-mannitol. However, the sweetness test sample is not particularly limited.

[0026] FIG. 2 is a diagram showing a combination of training samples for a taste measurement system according to one embodiment of the present invention. The training samples used in the sensory test may be, for example, a combination of a sweetener and a flavoring. The flavoring to be combined with one of the sweeteners may be, for example, one selected from the group consisting of yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, yogurt, apple, melon, blueberry, grape, white grape, muscat, milk, chocolate, caramel, vanilla, mango, matcha, coffee, cherry, fruit mix, and menthol flavorings. The flavoring to be combined with one of the sweeteners is preferably, for example, one selected from the group consisting of yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, and yogurt flavorings. However, the flavoring to be combined with one of the sweeteners is not particularly limited.

[0027] In this embodiment, the sensory test is preferably performed by having a panelist taste or sample five standard samples each independently containing salty, sour, sweet, umami, and bitter flavors, and then taste or sample a sweetness test sample each independently containing one of nine sweeteners, and perform a sensory evaluation of the strength of each of the salty, sour, sweet, umami, and bitter flavors, which are then compared to the intensity of each of the standard samples and quantified.In this embodiment, the sensory test is preferably further performed by having a panelist taste or sample five standard samples each independently containing salty, sour, sweet, umami, and bitter flavors, and then taste or sample 10 test samples obtained by combining the sweeteners with 10 flavorings, and perform a sensory evaluation of the strength of each of the salty, sour, sweet, umami, and bitter flavors, which are then compared to the intensity of each of the standard samples and quantified.

[0028] More preferably, the standard samples consist of two types of standard samples with different concentrations for each taste, and the sensory evaluation of the training samples is quantified based on a five-point rating for each taste: (1) not perceptible at all, (2) perceived as weaker than the low-concentration standard sample, (3) perceived as equivalent to the low-concentration standard sample, (4) perceived as intermediate between the low-concentration standard sample and the high-concentration standard sample, and (5) perceived as equivalent to or stronger than the high-concentration standard sample. Each of these five-point ratings may be further subdivided.

[0029] Neural network 200 can learn the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans due to interactions between flavorings and sweeteners by inputting the results of sensory tests using nine sweetness training samples containing any of nine sweeteners and 90 training samples of sweetener-flavoring combinations, each of which is a combination of one of the nine sweeteners with any of ten flavorings. Furthermore, by having neural network 200 learn the concentration of each component and the standard deviation of the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans, it can also calculate the variability in the intensities of saltiness, sourness, sweetness, umami, and bitterness perceived by humans.

[0030] The output layer 30 outputs analytical values ​​that indicate the strength of saltiness, sourness, sweetness, umami, and bitterness that are perceived by a person as analyzed in the intermediate layer.

[0031] In the taste measurement system of the present invention, the neural network 200 is trained to learn the intensities of saltiness, sourness, sweetness, umami, and bitterness that humans perceive due to the interactions between flavors and sweeteners, making it possible to more accurately mimic human taste. As specifically demonstrated in the examples below, surprisingly, the system of the present invention was also able to reproduce the taste illusions that humans perceive, such as the increase or decrease in the bitterness of an active ingredient contained in a pharmaceutical oral formulation, by changing the combination of flavors and sweeteners.

[0032] (Taste measurement method) A method for measuring the taste of a sample using the above-described taste measurement system will be described below.

[0033] To assess the taste of a sample, any sample is prepared, which may contain the active ingredient and one sweetener contained in a pharmaceutical oral dosage form, or one flavoring and one sweetener.

[0034] A sample is input to the input layer 10 of the taste measurement system 1. The input layer 10 uses sensors 100 to measure the actual values ​​of saltiness, sourness, sweetness, umami, and bitterness in the sample.

[0035] The actual measurement values ​​obtained by the sensor 100 are input to the intermediate layer 20 of the taste measurement system 1. The intermediate layer 20 uses a neural network 200 to calculate the concentration of each component from the obtained actual measurement values.

[0036] The intermediate layer 20 further uses the neural network 200 to calculate analytical values ​​that indicate the strength of saltiness, sourness, sweetness, umami, and bitterness perceived by humans from the obtained concentrations of each component.

[0037] The output layer 30 outputs analytical values ​​that indicate the strength of saltiness, sourness, sweetness, umami, and bitterness that are perceived by a person as analyzed in the intermediate layer.

[0038] In the taste measurement method of the present invention, the neural network 200 is trained to learn the intensities of saltiness, sourness, sweetness, umami, and bitterness that humans perceive due to the interactions between flavors and sweeteners, making it possible to more accurately mimic human taste. As specifically demonstrated in the examples below, surprisingly, the method of the present invention was also able to reproduce the taste illusions that humans perceive, such as the increase or decrease in the bitterness of an active ingredient contained in a pharmaceutical oral formulation, by changing the combination of flavors and sweeteners. [Example]

[0039] The taste measurement system and method for oral pharmaceutical preparations according to the present invention will now be described in more detail with reference to specific examples and test results.

[0040] (1) Sensor Na is a component that gives saltiness + , and Cl - , H as a component that gives sour taste + We have fabricated sensors for sweet components such as glucose and sucrose, umami components such as glutamate, and bitter components such as caffeine. + Sensor, Cl - Sensor, H + The sensors were ion-selective electrodes, whereas the glucose, sucrose, and glutamate sensors were enzyme electrodes.

[0041] (2) Learning training data In order to simulate the taste interaction caused by the combination of flavoring and sweetening agents, a sensory test was conducted by six panelists, and the results were studied. The sensory test by the panelists was conducted as follows.

[0042] As standard samples for the five tastes of saltiness, sourness, sweetness, umami, and bitterness, five types of samples containing each taste independently were prepared, each with two different concentrations, as shown in Table 1. [Table 1]

[0043] The sweetness samples used for the study included thaumatin (Sunsweet® T / Sanei Gen F.F.I. Co., Ltd.) 0.5 mg / 10 mL, stevia (stevioside / Tokyo Chemical Industry Co., Ltd.) 1.7 mg / 10 mL, sucralose (sucralose / Sanei Gen F.F.I. Co., Ltd.) 1.0 mg / 10 mL, aspartame (aspartame / Ajinomoto Co., Inc.) 3.3 mg / 10 mL, saccharin Na (saccharin sodium hydrate BP / Daiwa Kasei Co., Ltd.) 1.7 mg / 10 mL, acesulfame K (Sunet Pharm Grade Type D (fine powder) / MC Food Specialties Co., Ltd.) 3.0 mg / 10 mL, refined white sugar (refined white sugar / Fujifilm Wako Pure Chemical Corporation) 300 mg / 10 mL, and lactose (Pharmatose® 200 mL / FrieslandCampina DMV). Nine types of ethanol were prepared and used: 1000 mg / 10 ml of ethanol (BV) or 500 mg / 10 ml of D-mannitol (Mannit P / Mitsubishi Corporation Life Sciences Co., Ltd.).

[0044] The sweetener and flavor test samples were prepared by combining the sweetness test samples with flavors. The flavors combined with each of the sweetness test samples were 10 types: yuzu (SGW-0008 / Shiono Koryo Co., Ltd.), pineapple (SGW-0018 / Shiono Koryo Co., Ltd.), cider (SGW-0029 / Shiono Koryo Co., Ltd.), strawberry (SGW-0013 / Shiono Koryo Co., Ltd.), peach (SGW-0020 / Shiono Koryo Co., Ltd.), orange (SGW-0003 / Shiono Koryo Co., Ltd.), lemon (SGW-0005 / Shiono Koryo Co., Ltd.), grapefruit (SGW-0004 / Shiono Koryo Co., Ltd.), banana (SGW-0019 / Shiono Koryo Co., Ltd.), and yogurt (SGW-0033 / Shiono Koryo Co., Ltd.). The concentration of all the fragrances was adjusted to 8.5 mg / 10 ml.

[0045] Each panelist tasted the standard sample and memorized each basic taste. Then, they tasted the sweetness training material and the training samples of 10 types of sweeteners and flavorings, and rated the intensity of the five basic tastes for each training sample. Each taste was rated on a five-point scale and described numerically. All samples were at room temperature, and participants rinsed their mouths with water before tasting the next sample. The average and standard deviation of the obtained values ​​were calculated.

[0046] By inputting the results of the sensory test into the neural network 200, it was possible to make the intermediate layer 20 learn the intensities of saltiness, sourness, sweetness, umami, and bitterness that humans perceive as being caused by the interaction of flavors and sweeteners. Figures 12 to 14 show the average values ​​learned by the neural network 200, and Figures 15 to 17 show the standard deviations learned by the neural network 200.

[0047] (3) Example 1 Using the taste measurement system according to this embodiment, saltiness, sourness, sweetness, umami, and bitterness resulting from the active ingredients, flavorings, and sweeteners contained in a pharmaceutical oral preparation were analyzed.

[0048] Teneligliptin hydrobromide was used as the active ingredient in a pharmaceutical oral formulation, and sucralose was used as the sweetener. Yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, or yogurt flavorings were added, and the salty, sour, sweet, umami, and bitter tastes resulting from the active ingredient, flavoring, and sweetener contained in the pharmaceutical oral formulation were analyzed. Samples were prepared by adding 31 mg of teneligliptin hydrobromide, 1.4 mg of sucralose, and 12.5 mg of flavoring to 15 mL of water, mixing, and analyzing the samples using a taste measurement system (Taste Sensor Leo, OISSY Inc.) after training with training data. Figure 3 shows the analysis results of Example 1 using the taste measurement system according to this embodiment. Figure 4 shows a radar chart of Example 1 using the taste measurement system according to this embodiment.

[0049] (4) Reference example 1 The sensory test evaluated the tastes produced by the active ingredients, flavors, and sweeteners contained in the pharmaceutical oral formulations.

[0050] Teneligliptin hydrobromide was used as the active ingredient in a pharmaceutical oral formulation, and sucralose was used as the sweetener. Yuzu, pineapple, cider, strawberry, or peach flavoring was added, and the taste resulting from the active ingredient, flavoring, and sweetener in the pharmaceutical oral formulation was evaluated. Samples were prepared by adding 31 mg of teneligliptin hydrobromide, 1.4 mg of sucralose, and 12.5 mg of flavoring to 15 mL of water, mixing, and then a sensory test was conducted by two panelists. To comprehensively evaluate the five basic tastes, a three-point rating system was used, ranging from "good" to "bad." Figure 5 shows the sensory test results for Reference Example 1.

[0051] (5) Comparative Example 1 Using a taste measurement system before training data, we analyzed the salty, sour, sweet, umami, and bitter tastes produced by the active ingredients, flavorings, and sweeteners contained in pharmaceutical oral preparations.

[0052] Teneligliptin hydrobromide was used as the active ingredient in a pharmaceutical oral formulation, and sucralose was used as the sweetener. Yuzu, pineapple, cider, strawberry, or peach flavoring was added, and the salty, sour, sweet, umami, and bitter tastes resulting from the active ingredient, flavoring, and sweetener contained in the pharmaceutical oral formulation were analyzed. The sample was prepared by adding 31 mg of teneligliptin hydrobromide, 1.4 mg of sucralose, and 12.5 mg of flavoring to 15 mL of water, mixing, and adjusting the ratio. The sample was analyzed using a taste measurement system (Taste Sensor Leo, OISSY Inc.) before training with training data. Figure 6 shows the analysis results of Comparative Example 1 using the taste measurement system before training with training data. Figure 7 shows a radar chart of Comparative Example 1 using the taste measurement system before training with training data.

[0053] (6) Comparison Results Between Example 1, Reference Example 1, and Comparative Example 1 In Example 1, strawberry and peach, which were perceived as bitter in the sensory test of Reference Example 1, were different from yuzu and pineapple, which were not perceived as bitter, and in particular, strawberry had a difference of 0.2 or more (a difference that 95% or more people could recognize) compared to yuzu and pineapple, confirming a correlation with the sensory test of Reference Example 1. Furthermore, a correlation was confirmed between the analytical results using another sweetener (saccharin) and the sensory test results.

[0054] In Comparative Example 1, strawberry and peach, which were perceived as bitter in the sensory test of Reference Example 1, could not be distinguished from yuzu and pineapple, which were not perceived as bitter, and peach in particular had the lowest bitterness score. It was found that Comparative Example 1 did not correlate with the sensory test of Reference Example 1.

[0055] 8 shows the analysis results of the bitterness scores for Example 1 and Comparative Example 1. In Comparative Example 1, the bitterness scores of strawberry and peach, which were perceived as bitter in the sensory test of Reference Example 1, were lower than those of Example 1. Furthermore, the sourness scores of all the fruits except for peach in Comparative Example 1 were higher than those of Example 1, resulting in different analysis results before and after learning the training data.

[0056] (7) Example 2 The taste measurement system according to this embodiment was used to analyze the salty, sour, sweet, umami, and bitter tastes produced by the active ingredients, flavorings, and sweeteners contained in other pharmaceutical oral preparations.

[0057] Lacosamide was used as the active ingredient in a pharmaceutical oral formulation, and sucralose was used as the sweetener. Yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, or yogurt flavorings were added, and the salty, sour, sweet, umami, and bitter tastes resulting from the active ingredient, flavoring, and sweetener contained in the pharmaceutical oral formulation were analyzed. Samples were prepared by adding 50 mg of lacosamide, 1.4 mg of sucralose, and 12.5 mg of flavoring to 15 mL of water, mixing, and adjusting the ratio. Analysis was performed using a taste measurement system (Taste Sensor Leo, OISSY Inc.) after training with training data. Figure 9 shows the analysis results of Example 2 using the taste measurement system according to this embodiment. Figure 10 shows a radar chart of Example 2 using the taste measurement system according to this embodiment.

[0058] (8) Reference example 2 The sensory test evaluated the tastes produced by the active ingredients, flavors, and sweeteners contained in the pharmaceutical oral formulations.

[0059] The active ingredient in the pharmaceutical oral formulation was lacosamide, and the sweetener was sucralose. Yuzu, strawberry, cider, banana, orange, or lemon flavoring was added, and the taste resulting from the active ingredient, flavoring, and sweetener in the pharmaceutical oral formulation was evaluated. Samples were prepared by adding 50 mg of lacosamide, 1.4 mg of sucralose, and 12.5 mg of flavoring to 15 mL of water, mixing them, and then a sensory test was conducted by two panelists. To comprehensively evaluate the five basic tastes, a five-point scale was used: 5 (delicious), 4 (better than sweetener alone), 3 (similar to sweetener alone), 2 (less than sweetener), and 1 (unpleasant). Figure 11 shows the sensory test results for Reference Example 2.

[0060] (9) Comparison Results Between Example 2 and Reference Example 2 In Example 2, orange, lemon, and yuzu, which were better than those containing only the sweetener in the sensory test of Reference Example 2, had lower values ​​(results in less perceived bitterness) than the results of the sweetener alone, so a correlation with the sensory test of Reference Example 2 was confirmed. [Explanation of symbols]

[0061] 1: Taste measurement system, 10: Input layer, 20: Middle layer, 30: Output layer, 100: Sensor, 200: Neural network

Claims

1. an input layer for measuring actual measured values ​​of saltiness, sourness, sweetness, umami, and bitterness caused by active ingredients, flavors, and sweeteners contained in a pharmaceutical oral formulation; an intermediate layer that calculates analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from interactions between the active ingredient and the flavoring and sweetening agent using a neural network that has been trained using actual values ​​of any flavoring and sweetening agent as training data from the actual measured values; an output layer that outputs the analysis value; A taste measurement system for a pharmaceutical oral formulation, comprising:

2. The taste measurement system for pharmaceutical oral formulations according to claim 1 , wherein the performance values ​​are the results of sensory tests using any combination of flavoring and sweetening agent.

3. 3. The taste measurement system for a pharmaceutical oral formulation according to claim 2, wherein the flavor used in the sensory test is one selected from the group consisting of yuzu, pineapple, cider, strawberry, peach, orange, lemon, grapefruit, banana, and yogurt flavors.

4. The taste measurement system for a pharmaceutical oral formulation according to claim 3 , wherein the flavoring contained in the pharmaceutical oral formulation is the flavoring used in the sensory test.

5. 5. The taste measurement system for a pharmaceutical oral formulation according to claim 4, wherein the sweetener used in the sensory test is one selected from the group consisting of thaumatin, stevia, sucralose, aspartame, saccharin, acesulfame K, refined sucrose, lactose, and D-mannitol.

6. The taste measurement system for a pharmaceutical oral formulation according to claim 5 , wherein the sweetener contained in the pharmaceutical oral formulation is the sweetener used in the sensory test.

7. The taste measurement system for a pharmaceutical oral formulation according to claim 6, wherein the active ingredient has a bitter taste.

8. The actual values ​​of saltiness, sourness, sweetness, umami, and bitterness generated by the active ingredients, flavors, and sweeteners contained in the pharmaceutical oral preparation are measured, From the measured values, analytical values ​​of saltiness, sourness, sweetness, umami, and bitterness resulting from the interaction between the active ingredient and the flavor and sweetener are calculated using a neural network that has been trained using the actual values ​​of any flavor and sweetener as training data; outputting the analysis value; A method for measuring the taste of a pharmaceutical oral formulation, comprising:

Images