Nutrient Digestibility Assistant and Its Computer-Implemented Algorithm

Abstract

A computer-implemented method for analyzing fecal matter, the computer-implemented method including obtaining visual information of the fecal matter, providing the visual information to an artificial intelligence (AI) model, processing the visual information using the AI ​​model to determine multiple scales of the fecal matter, outputting the multiple scales of the fecal matter, receiving at least one input for adjusting the multiple scales of the fecal matter, and providing suggestions based on the adjusted multiple scales of the fecal matter.

Description

[Technical Field]

[0001] The present invention relates to a nutrient digestibility assistant, in particular to a method, device and computer-implemented algorithm for analyzing waste products. [Background technology]

[0002] In the healthcare field, for example, for care recipients such as humans (e.g., infants or the elderly) and other animals (e.g., dogs, cats, chickens, cows, sheep, goats, etc.), technological advances make it possible to track the development of care recipients and assist caregivers in quickly detecting certain abnormalities.

[0003] More specifically, with regard to human infant nutrition, it is important to detect whether there are any abnormalities in the digestive system or whether the infant's body is absorbing all the necessary nutrients. Analyzing stool patterns to assess digestive system performance is known to provide excellent insight. Scales for comparing stool to a set of stool analysis scale scores can help classify stool types and draw conclusions from them. Examples of such scales are the Bristol Stool Form Scale (BSS) and the Amsterdam Stool Scale. The BSS, consisting of seven images of different stool consistency levels, allows for an objective assessment of stool consistency (from scale 1, indicating a hard lump, to scale 7, indicating a watery stool). The BSS can also be used to characterize the stool of infants and other young children.

[0004] For example, in babies, healthcare professionals (HCPs) typically ask caregivers about the consistency of their infant's stool, but these questions are difficult for most parents to answer. When parents are asked to keep a record of their infant's stool consistency, it is difficult for them to identify the stool consistency and associated stool analysis scale score that matches their child's stool.

[0005] It would be desirable to have a system that allows parents and caregivers to track their baby's stool pattern, i.e., stool consistency, frequency, and color, in real time. It would further be desirable for the stool pattern to be tracked in an objective and consistent manner regardless of which caregiver (parent, grandparent, nanny, or daycare) is changing the diaper or helping the child use the potty or toilet chair. It would also be desirable to have a system that, based on the observed stool pattern, provides an indication that all is normal, reassuring the parents and caregivers, or that the infant's stool pattern is not as expected and a recommendation to see an HCP.

[0006] The above desired scenario also applies to other types of caregivers and care recipients, such as pet owners, veterinarians or farmers of cows, sheep, goats, chickens, etc. The animals may also be at other stages other than infancy.

[0007] Nowadays, portable computing devices, such as smartphones, tablet computers, or other portable devices with mobile applications (apps), can facilitate normal daily activities for users, which can also be applied to tracking stool patterns. Programs or apps are known that allow users to introduce or capture images of stool and manually select a score on a stool analysis scale that best suits the stool on the image. Furthermore, programs or apps are known that allow automatic detection of stool color using color recognition techniques. Summary of the Invention [Problem to be solved by the invention]

[0008] However, the accuracy of recognition needs to be improved. Furthermore, other excretions, such as urine, can be additionally analyzed to improve the health analysis results for better given cure suggestions. [Means for solving the problem]

[0009] The present invention relates to a method, an apparatus and a computer-implemented algorithm for analyzing fecal matter.

[0010] The invention is defined by the claims.

[0011] In the following, the invention will be discussed in more detail with reference to the accompanying drawings. [Brief explanation of the drawings]

[0012] [Figure 1] A computer-implemented method for analyzing fecal matter is presented. [Figure 2] 10 shows an example of a user interface for adjusting the scale. [Figure 3] Indicates the equipment. [Figure 4] 1 illustrates a computer-implemented method for providing meal suggestions. DETAILED DESCRIPTION OF THE INVENTION

[0013] Embodiments of the present disclosure are described below with reference to the accompanying drawings. However, the embodiments of the present disclosure are not limited to the particular embodiments, but should be construed as including all modifications, variations, equivalent apparatus and methods, and / or alternative embodiments of the present disclosure.

[0014] As used herein, the terms "have," "may have," "include," and "may include" indicate the presence of the corresponding feature (e.g., a value, function, operation, or element such as a component) and do not exclude the presence of additional features.

[0015] As used herein, the terms "A or B," "at least one of A and / or B," or "one or more of A and / or B" include all possible combinations of the items listed therewith. For example, "A or B," "at least one of A and B," or "at least one of A or B" means (1) including at least one A, (2) including at least one B, or (3) including both at least one A and at least one B.

[0016] As used herein, terms such as "first" and "second" may modify various elements and do not limit the corresponding elements, regardless of the order and / or importance of the corresponding elements. These terms may be used to distinguish one element from another. For example, a first printed form and a second printed form may refer to different printed forms, regardless of the order or importance. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the present invention.

[0017] When an element (e.g., a first element) is "operably or communicatively coupled" or "connected" to another element (e.g., a second element), it will be understood that the element may be directly coupled to the other element, and that there may be intervening elements (e.g., third elements) between the element and the other element. Conversely, when an element (e.g., a first element) is "directly coupled" or "directly connected" to another element (e.g., a second element), it will be understood that there are no intervening elements (e.g., third elements) between the element and the other element.

[0018] As used herein, the phrase "configured to" may be used interchangeably with "suitable for," "capable of," "designed to," "adapted to," "made to," or "capable of," depending on the context. The term "configured to" does not necessarily mean "specially designed" at the hardware level. Instead, the phrase "device configured to" may mean that the device is "capable of" in conjunction with other devices or components in certain contexts.

[0019] The terms used in describing various embodiments of the present disclosure are intended to describe particular embodiments and are not intended to limit the present disclosure. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise. All terms used herein, including technical or scientific terms, have the same meaning as commonly understood by those skilled in the relevant art unless otherwise defined. Terms defined in commonly used dictionaries should be interpreted as having the same or similar meaning as in the context of the relevant art, and should not be interpreted as having an ideal or exaggerated meaning unless clearly defined herein. Depending on the context, even terms defined in this disclosure should not be interpreted as excluding embodiments of the present disclosure.

[0020] Animal waste (e.g., feces and urine excreted by an animal's body) can reveal many signs of the health of an animal, such as a human infant or other animal. However, there is a need for an easy and accurate method for waste analysis. The present invention relates to a waste analysis method, an apparatus, and a computer-implemented algorithm therefor.

[0021] Figure 1 illustrates a computer-implemented method for analyzing fecal matter. Some steps in Figure 1 may be performed in a different order (e.g., step 107 may be performed again after step 104 and / or after step 106), merged (e.g., steps 101, 102, and 103), or omitted (e.g., steps 101, 102, or 105).

[0022] In step 101, visual information of the excrement is obtained. The visual information may be at least one of video information, image information, three-dimensional information, and thermal image information. Although images may be used as examples of visual information throughout this specification, it should be understood that images are merely an example and other forms of visual information are also encompassed by the present invention. In step 101, for example, an image or video may be captured (which may or may not be displayed, and which may or may not be stored in memory) by, for example, a camera, a thermal imaging camera, retrieved from a memory, received from an external device via a telecommunications unit, or by other means. The visual information may have therein the excrement to be analyzed.

[0023] The waste may be feces or urine from any animal body, such as babies, adults, humans, dogs, cats and other animals. The waste may include at least one of urine and feces.

[0024] The visual information may be one of diaper visual information, diaper visual information, container visual information (e.g., toilet bowl, potty chair, etc.), litter visual information, flush toilet visual information, grass visual information, ground visual information, etc., on which excrement may be present or in which excrement may be present.

[0025] In step 102, it may be determined whether excrement is present in the visual information, which may be determined by a pre-trained artificial intelligence (AI) model or image / visual information recognition algorithm.

[0026] If it is determined that there is no excrement in the visual information, the method can be stopped, or a warning can be output to the user (via a screen or speaker) to remind the user to capture or change the visual information, after which step 102 can be performed again with the new visual information. Decision step 102 can be omitted, for example, it can be assumed that there is always excrement in the visual information.

[0027] In step 102, if it is determined that excrement is present in the visual information, the method may further determine the composition of the excrement, for example, whether the excrement is only feces, only urine, or both. This determination of composition may be omitted.

[0028] In step 103, the method provides the visual information to an artificial intelligence (AI) model. In step 104, the AI ​​model processes the visual information and can determine multiple scales of waste. The waste scales can include a color scale, a consistency scale, a quantity scale, etc.

[0029] For example, if the waste comprises stool, the plurality of scales for the waste may include at least one of a stool color scale and a stool consistency scale. The color scale may include a standard color scale or a subset of a standard color scale. The consistency scale may include watery, soft, formed, and hard based on the BSS or any other stool consistency standard.

[0030] If the waste product includes urine, the plurality of scales for the waste product may include a urine color scale, which may include a standard color scale or a subset of a standard color scale.

[0031] The AI ​​model may be pre-trained with training visual information (e.g., images), which may be processed by a loss function. The loss function may include blending two original visual information (e.g., images) together according to different transparency levels, randomly generating blending coefficients multiple times, and training the AI ​​model according to the data distribution during pre-training. Such a pre-training method may improve the prediction accuracy of the AI ​​model. Furthermore, automatic data augmentation may be performed on the training visual information before it is used to train the AI ​​model, and such augmented training visual information may also improve prediction accuracy. The AI ​​model may also be trained using a semi-supervised training method.

[0032] Before processing the visual information using the AI ​​model in step 104, the visual information may be preprocessed using, for example, at least one of inversion, lightening, darkening, and cropping. Such preprocessing may help the AI ​​model to more accurately predict scale.

[0033] In step 105, the multiple scales of the stool determined by the AI ​​model are output, for example, displayed or output with sound to indicate the scale. For example, the instructions for the color scale can be text, a color image, and / or color text corresponding to the color scale. For example, the text "dark brown" can be displayed, which can correspond to the dark brown stool determined by the AI ​​model, or the text "stool" in dark brown can be displayed, or the text "dark brown" in dark brown can be displayed, or a dark brown overlay image can be displayed. Similarly, the instructions for the consistency scale can be text or an overlay image indicating the consistency scale.

[0034] In the present invention, augmented reality (AR) technology can be used. For example, when multiple scales of excrement are output, at least one virtual excrement overlay object (e.g., virtual excrement overlay, overlay virtual stool image, virtual stool icon, stereoscopic image, multimedia information, etc.) can be displayed, and the multiple scales of the virtual excrement can be displayed according to the determined multiple scales of the excrement. The virtual excrement (i.e., the virtual excrement overlay object) can be displayed next to the excrement in the visual information according to the determined multiple scales of the excrement.

[0035] For example, if the excrement in the visual information includes feces, an overlay virtual feces object / icon (e.g., a feces-shaped icon, a feces image with a predetermined sharpness, a copy image of the identified feces, or a combination thereof) may be displayed next to the captured visual information, e.g., next to the identified feces. In this example, the overlay virtual feces object / icon may be displayed according to the determined multiple scales of the feces, i.e., multiple scales of the virtual feces may be displayed according to the determined multiple scales of the excrement.

[0036] Taking a color scale as an example, if the determined color scale of the stool in the visual information (e.g., an image) is dark brown, the overlay virtual stool object may be displayed in a corresponding dark brown color. In a consistency scale, if the determined consistency of the stool is watery, the overlay virtual stool object may indicate that the stool is watery, for example, by including a water drop icon in the overlay virtual stool object or by having a smooth stool surface in the overlay virtual stool object.

[0037] In particular, AR technology involving virtual waste overlay objects can encourage a user (e.g., a caregiver) to identify the difference between the scale of waste in real-world visual information (i.e., a captured image that may be displayed) and the scale of waste determined by an AI model (using the virtual waste overlay object).

[0038] At step 106, at least one input can be received from a user to adjust the waste scales. This facilitates improving accuracy in determining the final / actual waste scale. The input can be received via a touchscreen displaying the input receiving component, or by physical button input on the device, or by any other means known to those skilled in the art.

[0039] For example, when virtual excrement is displayed, the user can easily see the difference between the color / consistency in the real visual information (i.e., the obtained visual information) and the color / consistency in the virtual excrement. The user can then adjust the scales of the excrement, and the virtual excrement can be displayed accordingly. The user can stop inputting when there is little visible difference between the displayed virtual excrement and the real excrement in the obtained visual information.

[0040] Thus, if the received at least one input is configured to adjust the scales of the waste, the scales of the virtual waste may be displayed according to the adjusted scales of the waste. In this manner, the predicted scale of the AI ​​model may be further adjusted by the user so that the actual scale of the waste can be accurately determined.

[0041] If the excrement includes feces, the plurality of scales of the virtual excrement may include at least one of a virtual stool color scale and a virtual stool consistency scale. If the excrement includes urine, the plurality of scales of the virtual excrement may include a virtual urine color scale.

[0042] Similarly, if the excrement includes stool, the adjusted plurality of scales for the excrement may include at least one of a virtual stool color scale and a virtual consistency scale, and if the excrement includes urine, the adjusted plurality of scales for the excrement may include a virtual urine color scale.

[0043] Furthermore, the adjusted scales of waste may be more accurate than the scales originally determined by the AI ​​model, and the adjusted scales of waste in step 105 may be used to retrain the AI ​​model so that the accuracy of the AI ​​model may be further improved.

[0044] In step 107, suggestions based on the adjusted multiple scales of waste (and / or some additional information, such as recorded dietary information and at least one of sex, age, disease records, weight and body length) are provided, e.g., displayed or provided with a sound.

[0045] For example, the suggestions may include at least one health suggestion, which may include at least one of dietary nutritional recommendations for the baby to improve the baby's bowel function and health, at least one immune system indicator for the baby, and nutritional guidance for nursing mothers.

[0046] The suggestions may include at least one meal plan, which may be further based on dietary status and other additional information, such as at least one of gender, age, medical history, weight, and body length, and the dietary status may be at least one of dietary preferences and allergen information of the care recipient who produced the waste.

[0047] The suggestions may further include a digestive health score calculated based on the adjusted scales and / or additional information.

[0048] The suggestion may further include, or may only include, comparable data of cases having the same excrement scale as the adjusted multiple scales of excrement in the obtained visual information. For example, the adjusted multiple scales of excrement may indicate that the baby (who has excreted) is in an abnormal / unhealthy condition. Then, relevant data may be output about the frequency and / or proportion of other babies having the same abnormal condition. This can better indicate to the caregiver (e.g., parents) how urgent / serious the baby's situation / condition is. If the baby's abnormality is rather common, the caregiver can feel reassured; otherwise, they will know that they need to contact a doctor immediately.

[0049] Such suggestions may be generated according to at least one lookup table in a database cross-referenced with the scale and / or additional information. For example, the lookup table may be specific to a 4-month-old breastfed human infant, where all possible values ​​of the scale are included, with each combination of values ​​corresponding to a particular suggestion. For example, dark brown, dry stools may result in a suggestion to provide the infant with more fluids every few hours, while light-colored urine may result in a suggestion to provide the infant with less fluids. Meal plans may be included in the suggestions, e.g., specific recipes for meals from a database may be cross-referenced with the scale (and / or additional information) and suggested.

[0050] Such suggestions can be provided by a suggestion AI model that can be pre-trained to provide suggestions according to the scale of waste and / or additional information, such as recorded dietary information and at least one of gender, age, disease records, weight and body length.

[0051] As an example, if the adjusted color scale indicates stool having a red, maroon, black, or light color, a warning may be displayed to attract the user's attention. Under such conditions, suggestions may be further based on the food record by checking whether the food record contains the cause of the red, maroon, black, or light color.

[0052] Figure 4 illustrates a computer-implemented method (that may be implemented by a device) for providing one or more dietary suggestions based on multiple scales of a user's waste, for example. Figure 4 may be combined with the method of Figure 1; for example, Figure 4 may be considered a more detailed method for step 107 of Figure 1.

[0053] In step 401, multiple scales of the user's waste can be obtained, which can be from steps 101-105 or steps 101-106 in FIG. 1, or can be input directly into the device by the user.

[0054] In step 402 (optional step), a digestive health status of the user (producing waste) can be determined based on the multiple scales of waste output. This determination can be further based on user information, such as age, geographic location, weight, gender, body mass index, food records, etc. The digestive health status can be provided along with a score indicating a health level of digestive development. If the user is an infant (i.e., the infant as the user producing waste), the digestive health status can be, for example, the infant's digestive maturity, indicating how developed the infant's digestive system is, which can be determined based on the multiple scales of waste output and the infant's user information. Step 402 is optional and can be omitted in the method of FIG. 4; for example, the method of FIG. 4 can include only steps 401 and 404 or only steps 401, 403, and 404.

[0055] In step 403 (optional step), the user's dietary record can be obtained, for example, from a user input on the device, or from a pre-stored database in the device, or from a server. The dietary record can include the user's dietary history information (for waste production) over a predetermined past period, for example, dietary information for the past week or three days. The dietary information can include meal times, amounts eaten, types of food and drink, etc. This dietary information can be used when determining digestive health in step 402 above. Step 403 is optional and can be omitted in the method of FIG. 4 , i.e., step 402 (determining digestive health) can proceed directly to step 404 (determining dietary suggestions).

[0056] In step 404, one or more dietary suggestions may be determined and / or output to the user via the device, which may be based on at least one of the multiple scales of waste, the digestive health status, and the dietary record. For example, if the waste has a small, spherical shape, it may indicate constipation, in which case the suggestion may be to increase dietary fiber intake. If the stool is loose, it may suggest reducing dietary fiber or other foods that tend to moisten the intestines, such as bananas. Another example is if the digestive health status may indicate early development of the infant's digestive system, the suggestion may include making the mother's breast milk the infant's primary food for a certain period of time in the future and delaying other supplemental foods until the infant's digestive system is fully developed as expected. As a further example, if the dietary record indicates that the user (who produces waste) has a habit of consuming the same or similar diet (e.g., a high-fiber diet, a high-protein diet, etc.) for a certain period of time in the past, the suggestion may instead be the user's dietary habits.

[0057] FIG. 2 is an example of a user interface for adjusting waste scaling on an exemplary device 200.

[0058] Device 200 may be a smartphone, tablet, smart TV, laptop, or any other computing device. Device 200 may include a screen 201, which may be a touchscreen for both displaying and receiving input. Device 200 may also include several physical input buttons.

[0059] 2, an image 202 (as an example of obtained visual information) is displayed on a screen 201, and part or all of the image 202 is displayed. In the example of image 202, there is a diaper 203. There is excrement (204 and 205) on the diaper 203. In this example, both feces 204 and urine 205 (or urine stains) are on the diaper 203.

[0060] In steps 103 and 104, image 202 may be processed by an AI model to determine the scale of the waste. In this example, the determined scale of the waste may be displayed in bar format, for example using bars 208, 209 and 210, with the determined scale values ​​indicated by indicators 2081, 2091 and 2101 on bars 208, 209 and 210, respectively. Optionally, other forms of output may be used, such as a text indication, a color indication, a spinning wheel icon, an arrow pointing to a particular scale, etc.

[0061] In step 106, input for adjusting the scale value can be received. To better adjust the scale, virtual waste overlay images 206 and 207 (which may also be referred to as waste icons) are displayed on the screen 201. The stool icon 206 and the urine icon 207 may be the same shape as those in the diaper, or may simply be representative icons (e.g., the stool icon is a cartoon stool shape, and the urine icon is a water droplet). The initial display colors of the waste icons correspond to the waste color scale determined by the AI ​​model. For example, the initial color of the stool icon 206 may be the stool color scale of the stool 204 on the image 202 determined by the AI ​​model, and similarly, the initial color of the urine icon 207 may be the urine color scale of the urine 205 on the image 202 determined by the AI ​​model. The initial positions of the indicators 2091 and 2101 of the bars 208 and 209 correspond to the initial color scale of the stool icon 206 and the urine icon 207, respectively. The stool 204 may also have a consistency scale determined by the AI ​​model, as well as a color scale, and indicated by a consistency bar 208 and indicator 2081.

[0062] Here, the user can see the displayed colors of the stool icon 206 and the urine icon 207, as well as the colors of the stool 204 and the urine 205 in the image, and also the stool and urine in the actual diaper. After comparing them, the user may decide to adjust / correct the colors by moving the indicators 2091 and 2101 on the bars 209 and 210, respectively. When the indicators 2091 and 2101 are moved, the colors of the stool icon 206 and the urine icon 207 are changed according to the colors currently indicated by the indicators 2091 and 2101, while the colors of the stool 204 and the urine 205 remain unchanged. The user may decide to stop changing the colors when the color difference between the stool icon 206 and the stool 204 (or actual stool) and the color difference between the urine icon 207 and the urine 205 (or actual urine) are minimized, i.e., when the colors are approximately the same (e.g., no longer distinguishable with the naked eye). The adjusted / corrected color scale (and / or other scale information and / or additional information) is used to generate suggestions in step 107 so that the suggestions are more accurate. Furthermore, the corrected colors can be used to retrain the AI ​​model so that future color predictions can be more accurate.

[0063] The consistency of the stool can be adjusted in a similar manner to the color. In the example of FIG. 2, the consistency of the stool is indicated by the bar 208 and the indicator 2081. The stool icon 206 itself may display cracks to indicate how dry the stool is (e.g., even in a portion of a soft stool if it is indicated as very dry). For example, the initial position of 2081 is according to a consistency scale determined by the AI ​​model, and the stool icon 206 displays cracks according to the determined scale. The user can compare the displayed consistency of the stool icon 206 with the stool 204 (or actual stool) on the image 202. If a difference is identified by the user, an input can be provided to the indicator 2081 to change the consistency of the stool icon 206 without changing the stool 204. A final consistency scale can be determined as the final adjusted consistency scale until the difference is minimized. The adjusted / corrected consistency scale (and / or other scale information and / or additional information) is used to generate a suggestion in step 107 so that the suggestion is more accurate. Furthermore, the corrected stiffness scale can be used to retrain the AI ​​model so that future stiffness predictions can be more accurate.

[0064] Alternatively, each element of waste (i.e., corresponding to each virtual waste overlay image / object) may be displayed one at a time along with a corresponding scale adjustment user interface (e.g., an input-accepting component such as the bar containing the indicator in FIG. 2). In the example of FIG. 2, stool icon 206 with its scale adjustment bar 208 and urine icon 207 with its scale adjustment bar 2101 are displayed within the same screen display. Alternatively, in the example of FIG. 2, a first screen display may display stool icon 206, bars 208 and 209 (including indicators 2081 and 2091) and image 202, and a second screen display may display urine icon 207, bar 210 (including indicator 2101) and image 202.

[0065] Each virtual waste overlay object / image can be displayed on the screen with a certain percentage of transparency along with a corresponding scale adjustment user interface, so that the background content on the screen (e.g., image) is not completely blocked.

[0066] The virtual waste overlay objects may be omitted, e.g., only the input acceptance components are displayed. For example, in the example of Figure 2, the feces icon 206 and the urine icon 207 may be omitted, and only the input acceptance components, i.e., bars 208, 209, and 210 containing indicators 2081, 209a, and 2101, may be displayed.

[0067] FIG. 3 shows a device 300 (e.g., the same as device 200 of FIG. 2) for implementing the present invention, such as a mobile phone, tablet, laptop, desktop, smartwatch, TV, etc.

[0068] The device 300 may include a processor 301, a display 302 (eg, the same as the screen 201 of FIG. 2), a communication unit 303, a memory 305, a camera 306, and other input / output units 307.

[0069] The processor 301 is configured to execute programs / instructions (e.g., as in the methods of Figures 1 and 4) stored in the memory 305 by controlling other components such as the display 302, the communication unit 303, the memory 305, the camera 306 and other input / output units 307.

[0070] Display 302 can be controlled by processor 301 to perform all display functions of the present invention, such as those in steps 105 and 107 and the example display screen of Fig. 2. Display 302 can be a touch screen that can receive input via an input receiving component displayed on display 302 in step 106 (e.g., in Fig. 2, the input receiving component includes bars 208, 209, and 210, each including indicators 2081, 2091, and 2101).

[0071] The communication unit 303 may be controlled by the processor 301 to perform all communication functions in the present invention. For example, when an external device 310 (e.g., a server) is used to perform some of the functions of the steps in Fig. 1 (e.g., the visual information in step 101 may be obtained from the external device 310, steps 102 and 103 may be performed by the device 300, steps 105 and 106 may be performed by the device 300, and suggestions may be generated by and output by the external device 310, or all steps may be performed on the user device 300, some steps may be performed on the user device 300, and the remaining steps may be performed on at least one or more external devices 310), messages may be transmitted via the communication unit 303. Optionally, a database used in the present invention, such as a look-up table for suggestions, may be stored in the external device 310 or the user device 300.

[0072] The memory 305 may be configured to store instructions and data for implementing the method of the present invention. For example, a lookup table for suggestions and derived visual information may also be stored in the memory 305. The device 300 may provide at least one entry for the user to review / view these data.

[0073] Camera 306 is configured to obtain visual information, eg, capture an image, as an example of step 101 .

[0074] Other input / output unit 307 may be configured to perform other input / output functions of the present invention, for example to receive user input for adjusting the scale.

[0075] In the present invention, at least a part of an apparatus (e.g., FIG. 3) or a method (e.g., FIG. 1 or FIG. 4 as a computer-implemented method) may be implemented as instructions stored in a non-transitory computer-readable storage medium, for example, in the form of a program module, a piece of software, a mobile app, and / or other form. When executed by a processor (e.g., processor 301), the instructions may enable the processor to perform corresponding functions according to the present invention. The non-transitory computer-readable storage medium may be memory 305.

[0076] A computer-implemented method for analyzing fecal matter includes obtaining visual information of the fecal matter, providing the visual information to an artificial intelligence (AI) model, processing the visual information using the AI ​​model to determine multiple scales of the fecal matter, outputting the multiple scales of the fecal matter, receiving at least one input for adjusting the multiple scales of the fecal matter, and providing a suggestion based on the adjusted multiple scales of the fecal matter.

[0077] Obtaining visual information of the waste may include at least one of capturing an image and determining whether the waste is present in the image.

[0078] The waste can be from any one of babies, adults, humans, dogs, cats and other animals.

[0079] The waste may include at least one of urine and feces.

[0080] The visual information may be at least one of video information, image information, three-dimensional information, and thermal image information, and the visual information may be at least one of diaper visual information, nappy visual information, container visual information, litter visual information, and flush toilet visual information.

[0081] When the waste comprises stool, the plurality of waste scales and the regulated plurality of waste scales may comprise at least one of a stool color scale and a stool consistency scale.

[0082] When the waste comprises urine, the plurality of waste scales and the plurality of adjusted waste scales may comprise a urine color scale.

[0083] When outputting the multiple scales of the excrement, at least one virtual excrement overlay object may be displayed, and the multiple scales of the virtual excrement may be displayed according to the determined multiple scales of the excrement, and / or if the at least one received input is configured to adjust the multiple scales of the excrement, the multiple scales of the virtual excrement may be displayed according to the adjusted multiple scales of the excrement.

[0084] When the excrement includes stool, the plurality of scales of the virtual excrement may include at least one of a color scale and a consistency scale of the virtual stool.

[0085] If the excrement includes urine, the multiple scales of the virtual excrement may include a color scale of virtual urine.

[0086] The adjusted scales of excrement can be used to retrain the AI ​​model.

[0087] The computer-implemented method may further include pre-training the AI ​​model with training visual information, where the training visual information may be processed by a loss function, where the loss function may include blending two original visual information together according to different transparency, randomly generating blending coefficients multiple times, and training the AI ​​model according to the data distribution during pre-training.

[0088] Before processing the visual information using the AI ​​model, the visual information may be pre-processed by including at least one of inversion, brightening, darkening, and cropping.

[0089] RandAugment can be used to perform automatic data augmentation on the training visual information, and / or the FixMatch semi-supervised training method can be used to train the AI ​​model.

[0090] The suggestions may include at least one health suggestion, which may include at least one of dietary nutritional recommendations to improve bowel function and health, at least one immune system indicator thereof, and nutritional guidance for nursing mothers.

[0091] The suggestions may include at least one meal plan that may be further based on dietary status and at least one of gender, age, medical history, weight, and body length, where the dietary status is at least one of dietary preferences and allergen information.

[0092] The proposal may include adjusted scales of waste and comparable data of cases with the same waste scale.

[0093] The suggestions may include a calculated digestive health score based on adjusted scales of waste output.

[0094] The suggestions may be provided via a suggestion AI model.

[0095] If the adjusted color scale indicates stool having a red, maroon, black, or light color, a warning may be displayed and / or suggestions may be further based on the food record by verifying whether the food record contains causes of the red, maroon, black, or light color.

[0096] The recommendation determination is further based on at least one of the recorded dietary information and at least one of sex, age, medical record, weight, and body length.

[0097] The device may be configured to perform the above method.

[0098] The storage medium may store instructions, and the instructions may be configured to cause a processor to perform the above-described methods.

Claims

1. 1. A computer-implemented method for analyzing fecal matter, comprising: Obtaining visual information about excrement; providing the visual information to an artificial intelligence (AI) model; and processing the visual information using the AI ​​model to determine multiple scales of the waste; outputting the plurality of scales of the excrement; receiving at least one input for adjusting the plurality of scales of the waste material; providing suggestions based on the adjusted plurality of scales of the waste; 10. A computer-implemented method comprising:

2. obtaining the visual information of the excrement, Capturing an image; determining whether excrement is present in the image; The computer-implemented method of claim 1 , comprising at least one of:

3. The computer-implemented method of claim 1 , wherein the waste is from any one of a baby, an adult, a human, a dog, a cat, and other animals.

4. The computer-implemented method of claim 1 , wherein the excrement includes at least one of urine and feces.

5. 5. The computer-implemented method of claim 1, wherein the visual information is at least one of video information, image information, three-dimensional information, and thermal image information, and the visual information is at least one of diaper visual information, nappy visual information, container visual information, litter visual information, and flush toilet visual information.

6. 6. The computer-implemented method of claim 1, wherein, when the excrement includes stool, the plurality of scales of the excrement and the adjusted plurality of scales of the excrement include at least one of a color scale and a consistency scale of the stool.

7. 7. The computer-implemented method of claim 1, wherein, when the excrement includes urine, the plurality of scales of the excrement and the adjusted plurality of scales of the excrement include a color scale of the urine.

8. When outputting the plurality of scales of the excrement, at least one virtual excrement overlay object is displayed, and the plurality of scales of the virtual excrement are displayed according to the determined plurality of scales of the excrement; and / or 8. The computer-implemented method of claim 1, wherein when the received at least one input is configured to adjust the plurality of scales of the excrement, the plurality of scales of the virtual excrement are displayed according to the adjusted plurality of scales of the excrement.

9. 9. The computer-implemented method of claim 8, wherein when the waste comprises stool, the plurality of scales of the virtual waste comprises at least one of a color scale and a consistency scale of the virtual stool.

10. The computer-implemented method of claim 8 or 9, wherein if the excrement comprises urine, the plurality of scales of the virtual excrement comprises a color scale of virtual urine.

11. The computer-implemented method of any one of claims 1 to 10, wherein the adjusted scales of the waste are used to retrain the AI ​​model.

12. 12. The computer-implemented method of claim 1, further comprising: pre-training the AI ​​model with training visual information, wherein the training visual information is processed by a loss function, the loss function comprising: blending two original visual information together according to different transparency; randomly generating blending coefficients multiple times; and training the AI ​​model according to data distribution during the pre-training.

13. 13. The computer-implemented method of claim 1, wherein prior to processing the visual information using the AI ​​model, the visual information is pre-processed by including at least one of inversion, lightening, darkening, and cropping.

14. 14. The computer-implemented method of claim 12 or 13, wherein RandAugment is used to perform automatic data augmentation on the training visual information and / or FixMatch semi-supervised training method is used to train the AI ​​model.

15. 15. The computer-implemented method of claim 1, wherein the suggestions include at least one health suggestion including at least one of dietary nutritional recommendations to improve bowel function and health, the at least one immune system indicator, and nutritional guidance for nursing mothers.

16. 16. The computer-implemented method of claim 1, wherein the suggestions include at least one meal plan further based on dietary status and at least one of gender, age, disease history, weight, and body length, and the dietary status is at least one of dietary preferences and allergen information.

17. The computer-implemented method of any one of claims 1 to 16, wherein the suggestions include comparable data of cases having the same scale of the excrement as the adjusted plurality of scales of the excrement.

18. The computer-implemented method of any one of claims 1 to 17, wherein the suggestion includes a calculated digestive health score based on the adjusted scales of the waste products.

19. 19. The computer-implemented method of any one of claims 1 to 18, wherein the suggestions are provided via a suggestion AI model.

20. 20. The computer-implemented method of any one of claims 1 to 19, wherein if the adjusted color scale indicates stool having a red, maroon, black or light color, a warning is displayed and / or the suggestion is further based on the food record by checking whether the food record contains causes of the red, maroon, black or light color.

21. 21. The computer-implemented method of any one of claims 1 to 20, wherein the determination of the suggestions is further based on at least one of recorded dietary information and at least one of gender, age, disease records, weight and body length.

22. Apparatus configured to carry out the method according to any one of claims 1 to 21.

23. A storage medium storing instructions, the instructions configured to cause a processor to implement any one of claims 1 to 21.

Images