KR102494384B1 - Method, apparatus and system for providing baby health diagnosis solution by using diaperstool image - Google Patents

Abstract

A method, apparatus, and system for providing a baby health diagnosis solution using diaper feces images are disclosed. A method for providing a baby health diagnosis solution using a diaper stool image according to an embodiment includes acquiring a diaper stool image of an identifier, detecting a stool object and a color code from the diaper stool image, and based on the color code, a stool object. Preprocessing is performed, and the preprocessed credit object is processed to generate the result of determining the credit status of the identifier.

Description

Method, apparatus and system for providing baby health diagnosis solution using diaper feces image

The following embodiments relate to a method, apparatus, and system for providing a baby health diagnosis solution using a diaper feces image.

As background art related to the embodiments, 　Korean Patent Laid-Open Publication 　KR 10-2020-0141776 A discloses a method and apparatus for managing health of infants and young children using stool images. Specifically, in the preceding literature, time-sequentially collecting at least one or more of vital sign data and hospital record data along with stool images for a predetermined time period, and deriving stool condition information through analysis of each collected stool image The step of generating and storing stool condition information for determination by normal state, disease state, and health state in a user database based on stool state information and acquired data, and then, obtaining stool images of infants, Thereafter, the step of generating stool condition information through analysis of the acquired stool image, and the stool state information and stool state information and Determining whether feces condition information matched within a predetermined threshold range exists in the user database, and, as a result of the determination, if there is matched stool condition information, generating infant condition determination information; Disclosed is a stool image-based health management method for infants and young children, which includes transmitting state determination information to pre-registered user-side terminals.

Through this, the preceding literature uses at least one or more of hospital record data, feeding record data, and vital sign data along with stool images of infants to generate stool condition information for determining each condition of infants, and then use the stool image based on this. By determining the condition of infants and young children, it is possible to help in health management of infants and young children.

In addition, Korean Patent Laid-Open Publication No. KR 10-2020-0119135 A discloses a stool condition analysis method using a neural network module and a system therefor. Specifically, the prior literature provides temperature information and humidity information for a predetermined time period, respectively, from a temperature sensor, a humidity sensor, a gyro sensor, a gas sensor, and an image sensor installed at a mounting hole of an automatic excreta disposal device mounted on a patient's body. , a processing unit configured to collect the patient's posture information, gas information, and image information and perform a pre-processing step on the collected information, and a temperature value that is input information output from the processing unit after the pre-processing step is performed; Disclosed is a server including a neural network module that receives a humidity value, a posture value, a gas concentration value, a gas generation frequency, and an image value and outputs a stool condition value of the patient.

Through this, the prior literature can provide a method for analyzing a patient's stool condition using a neural network module and a server therefor.

However, prior literatures obtain a diaper stool image of an identifier, detect a stool object and color code from the diaper stool image, preprocess the stool object based on the color code, and process the preprocessed stool object to determine the stool status of the identifier. It does not disclose the technical features that produce the result.

Republic of Korea Patent Publication KR 10-2020-0141776 A Republic of Korea Patent Publication KR 10-2020-0119135 A

Embodiments attempt to generate a stool state determination result of an identifier using a diaper stool image.

Embodiments attempt to pre-process a stool object using a color code included in a diaper stool image.

Embodiments attempt to learn the state of fecal objects using a neural network.

A method for providing a baby health diagnosis solution according to an embodiment of the present invention includes acquiring a diaper feces image of an identifier; detecting a stool object and a color code from the diaper stool image; pre-processing the feces object based on the color code; and processing the pre-processed feces object to generate a feces condition determination result of the identifier.

As an embodiment, the detecting of the color code may include obtaining training diaper stool images; extracting training color codes from the training diaper stool images; obtaining first labels indicating predefined color information for the training color codes; generating training outputs corresponding to the training color codes by applying the training color codes to a pretrained first neural network; and training the first neural network based on the training outputs and the first labels.

As an embodiment, the pre-processing of the feces object may include flattening a curve of the color code; detecting colors of the color code; generating a color correction value by comparing predefined color information for the colors of the color code with the detected colors; and correcting the color of the stool object using the color correction value.

As an embodiment, the pre-processing of the feces object may include calculating the size of the feces object using predefined size information of the color code.

As an embodiment, the step of generating the stool state determination result is the state of stool using the preprocessed stool object - the state of the stool is diarrhea constipation, normal, very, very, slightly, normal, slightly watery, classifying - including at least one of thin and very thin; and generating health state information of the identifier according to the classified stool condition.

As an embodiment, personal information of the identifier - the personal information includes at least one of age, gender, and allergy information of the identifier - and food intake information - the food intake information includes breastfeeding status, type of baby food, and food to be eaten including at least one of: receiving; obtaining training diaper stool images; extracting training stool objects from the training diaper stool images; acquiring first labels representing states of the training fecal objects according to the personal information and food information of the identifier; generating training outputs corresponding to the training stool objects by applying the training stool objects to a pretrained first neural network; and training the first neural network based on the training outputs and the first labels.

As an embodiment, transmitting the preprocessed stool object to an expert terminal; and receiving feedback information on the preprocessed stool object from the expert terminal.

As an embodiment, transmitting a hospital reservation request when a visit to the hospital is required according to the feedback information; and making a hospital reservation based on the hospital reservation request.

As an embodiment, assigning a weight to each state of the classified feces; obtaining first diaper stool images for a period preset for the identifier; extracting stool conditions from each of the first diaper stool images; Calculating a digestibility index by multiplying and summing the assigned weights for each of the opponents of the extracted feces; and calculating a ranking of digestibility of the identifier based on the digestibility index.

As an embodiment, calculating a first body age index of the identifier based on the generated stool condition determination result; calculating a second body age index of the identifier using the personal information; calculating a third body age index of the identifier by using the food intake information; assigning a weight to each of the first to third body age indices; calculating an average of values obtained by multiplying the first to third body age indices by weights assigned to the first to third body age indices; and generating the calculated average as a final body age index.

As an embodiment, the obtaining of the diaper feces image may include receiving a first part diaper feces image from a first user terminal; Receiving a second part diaper stool image from a second user terminal to which a second user account is logged in; generating a first diaper feces image by combining the first part diaper feces image and the second part diaper feces image; Obtaining a first ratio, which is a ratio occupied by the first part diaper stool image in the first diaper stool image, and obtaining a second ratio, which is a ratio occupied by the second part diaper stool image in the first diaper stool image ; When it is determined that the first ratio is greater than the second ratio, information of the first user account is registered as first author information of the first diaper feces image, and information of the second user account is registered as information of the first diaper feces image. registering the feces image as second author information; If it is confirmed that the first ratio is smaller than the second ratio, the information of the second user account is registered as the first author information of the first diaper feces image, and the information of the first user account is registered as the first diaper feces image. registering the feces image as second author information; confirming whether information about the original author of the first diaper feces image matches any one of information of the first user account and information of the second user account; If it is determined that the original author information of the first diaper feces image, the information of the first user account, and the information of the second user account do not all match, the first diaper stool image is modified to the second diaper stool image. Step of judging the generated secondary work; When it is determined that the first ratio is greater than the second ratio, information of the first user account is registered as first author information of the first diaper feces image, and information of the second user account is registered as information of the first diaper feces image. Registering information of a third user account registered as the original author of the first diaper stool image and registering information of a third user account registered as the original author of the first diaper stool image as second author information of the stool image as third author information indicating a third author of the first diaper stool image; and if it is determined that the first ratio is smaller than the second ratio, register information of the second user account as first author information of the first diaper stool image, and register information of the first user account as information of the first diaper feces image. and registering information of the third user account as second author information of the diaper feces image, and registering information of the third user account as third author information of the first diaper feces image.

A computer program according to an embodiment of the present invention may be stored in a computer readable recording medium in order to execute the method for providing a baby health diagnosis solution on a computer.

An apparatus for providing a baby health diagnosis solution according to an embodiment of the present invention includes a memory for obtaining and storing diaper stool images corresponding to identifiers; And a processor for detecting a stool object and a color code from the diaper stool image, preprocessing the stool object based on the color code, and generating a stool state determination result of the identifier by processing the preprocessed stool object. can do.

Embodiments analyze diaper feces images including color codes based on a learning model to standardize diaper feces images that change according to diaper photographing environments and camera specifications, and prevent analysis distortion of feces images. Due to this, it is possible to more accurately and quickly provide health status information according to color and shape from the baby's feces by analyzing the diaper feces image.

1 is a diagram for explaining a system for providing a baby health diagnosis solution using a diaper feces image according to an embodiment.
2 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.
3 is a diagram for explaining an apparatus for providing a baby health diagnosis solution according to an embodiment.
4 is a flowchart of a method for providing a necessary baby health diagnosis solution according to an embodiment.
5 is an exemplary diagram of a configuration of a device according to an embodiment.
6 to 14 are application implementation examples of a baby health diagnosis solution according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only, and may be modified and implemented in various forms. Therefore, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Although terms such as first or second may be used to describe various components, such terms should only be construed for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element.

It should be understood that when an element is referred to as being “connected” to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle.

Terms used in the examples are used only for descriptive purposes and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

The embodiments may be implemented in various types of products such as personal computers, laptop computers, tablet computers, smart phones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, machines learn and judge on their own. The more AI systems are used, the higher the recognition rate and the more accurate understanding of user preferences. Conventional rule-based smart systems are gradually being replaced by deep learning-based AI systems.

Artificial intelligence technology consists of machine learning and element technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as deep learning to mimic functions such as recognition and judgment of the human brain. It consists of technical fields such as understanding, inference/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology for recognizing and applying/processing human language/characters, including natural language processing, machine translation, dialogue systems, question and answering, voice recognition/synthesis, and the like. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, space understanding, image improvement, and the like. Inference prediction is a technique of reasoning and predicting logically by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology for controlling the autonomous driving of a vehicle and the movement of a robot, and includes motion control (navigation, collision, driving), manipulation control (action control), and the like.

In general, in order to apply machine learning algorithms to real life, learning is performed in a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

1 is a diagram for explaining a system for providing a baby health diagnosis solution using a diaper feces image according to an embodiment.

As shown in FIG. 1, the system 100 for providing a baby health diagnosis solution using a diaper feces image includes a plurality of user terminals 110-1, 110-2, ..., 110-n, and a baby health diagnosis solution. It may include a device 120, a database 130 and a network N. According to one embodiment, the database 130 is illustrated as being configured separately from the device for providing a baby health diagnosis solution 120, but is not limited thereto, and the database 130 may be included in the device for providing a baby health diagnosis solution 120. . For example, the apparatus 120 for providing a baby health diagnosis solution may include a plurality of neural networks for performing a machine learning algorithm.

The apparatus 120 for providing a baby health diagnosis solution is an identifier (eg, 110-1) from any one of the plurality of user terminals 110-1, 110-2, ..., 110-n. For example, an image of feces in a diaper of an infant) may be acquired. According to an embodiment, the apparatus 120 for providing a baby health diagnosis solution may acquire diaper feces images for a preset period of time (eg, 3 days, 1 week, 1 month, etc.).

The apparatus 120 for providing a baby health diagnosis solution may detect a stool object and a color code from the acquired diaper stool image. According to one embodiment, color patches may be attached to various locations on the diaper, and may be located around the diaper separately from the diaper.

The apparatus 120 for providing a baby health diagnosis solution may perform pre-processing of the detected stool object based on the detected color code. According to an embodiment, the baby health diagnosis solution providing device 120 flattens the curve of the color code, detects colors of the color code, and provides predefined color information and detected colors for the colors of the color code. A color correction value may be generated by comparison, and the color of the feces object may be corrected using the color correction value. According to another embodiment, the apparatus 120 for providing a baby health diagnosis solution may calculate the size of a feces object using predefined size information of a color code.

The apparatus 120 for providing a baby health diagnosis solution may process the pre-processed stool object to generate a stool condition determination result of an identifier. According to an embodiment, the apparatus 120 for providing a baby health diagnosis solution may classify the state of feces using the pre-processed stool object, and generate health state information of an identifier according to the classified stool state. Here, the condition of stool may include at least one of diarrhea constipation, normal, very good, good, little, normal, slightly watery, watery, and very watery, but is not limited thereto.

The apparatus 120 for providing a baby health diagnosis solution may transmit the preprocessed stool object to the expert terminal and receive feedback information on the preprocessed stool object from the expert terminal. In addition, if a visit to the hospital is required according to the feedback information, the device 120 for providing a baby health diagnosis solution may transmit a hospital reservation request to the hospital server, and the hospital server may make a hospital reservation based on the hospital reservation request.

The baby health diagnosis solution providing device 120 assigns a weight to each of the classified stool states, obtains first diaper feces images for a preset period of time with respect to the identifier, and obtains feces from each of the first diaper stool images. The digestibility index is calculated by extracting the states, multiplying and summing the weights assigned to each of the opponents of the extracted feces, and calculating the digestibility ranking of the identifier based on the calculated digestibility index.

The apparatus 120 for providing a baby health diagnosis solution calculates a first body age index of the identifier based on the generated stool condition determination result, calculates a second body age index of the identifier using personal information, and consumes food information. A third body age index of the identifier is calculated using , a first weight is assigned to each of the first to third body age indexes, and each of the first to third body age indexes and the first to third body age indexes are weighted. An average of values multiplied by the assigned first weight may be calculated, and the calculated average may be generated as a final body age index.

The apparatus 120 for providing a baby health diagnosis solution receives a first part diaper stool image from the first user terminal 110-1 and receives a second part diaper stool image from the second user terminal to which the second user account is logged in. Receiving, combining the first part diaper stool image and the second part diaper stool image to generate a first diaper stool image, and obtaining a first ratio, which is a ratio occupied by the first part diaper stool image in the first diaper stool image, , Obtaining a second ratio, which is a ratio occupied by the second part diaper feces image in the first diaper feces image, and when it is confirmed that the first ratio is greater than the second ratio, the information of the first user account of the first diaper feces image Register as the first author information, register the information of the second user account as the second author information of the first diaper feces image, and if it is confirmed that the first ratio is smaller than the second ratio, the information of the second user account is registered as the second author information. 1 registers as the first author information of the diaper feces image, registers the information of the first user account as the second author information of the first diaper feces image, and registers the original author information of the first diaper feces image and the information of the first user account, and It is checked whether any one of the information of the second user account matches, and if it is determined that the original author information of the first diaper feces image, the information of the first user account, and the information of the second user account do not all match, the first It is determined that the diaper stool image is a secondary work created by modifying the second diaper stool image, and if it is confirmed that the first ratio is greater than the second ratio, the information of the first user account is changed to the first author of the first diaper stool image. information, the information of the second user account is registered as the second author information of the first diaper stool image, and the information of the third user account registered as the original author of the first diaper stool image is registered as the first diaper stool image. 3 Register as third author information indicating the author, and if the first ratio is smaller than the second ratio, the information of the second user account is registered as the first author information of the first diaper stool image, the information of the first user account is registered as the second author information of the first diaper stool image, and the information of the third user account is registered as the author information of the first diaper stool image. 3 Author information can be registered.

The baby health diagnosis solution providing device 120 receives the identifier's personal information and intake food information, acquires training diaper stool images, extracts training stool objects from the training diaper stool images, and identifies the identifier's personal information and intake. Obtaining first labels representing states of training stool objects according to food information, applying the training stool objects to a pre-learned first neural network, generating training outputs corresponding to the training stool objects, training outputs and Based on the first labels, a first neural network may be trained. Here, the personal information may include at least one of age, gender, and allergy information of the identifier, and the food intake information may include at least one of breastfeeding status, type of baby food, and food to be eaten, but is not limited thereto. A learning method for pre-processing of feces objects will be described later.

The network N is configured to perform wireless or wired communication between the plurality of user terminals 110-1, 110-2, ..., 110-n, the baby health diagnosis solution providing device 120, the database 130, and the like. can do. For example, the network may include long-term evolution (LTE), LTE Advanced (LTE-A), code division multiple access (CDMA), wideband CDMA (WCDMA), wireless broadband (WiBro), wireless fidelity (WiFi), Bluetooth ( Bluetooth), near field communication (NFC), global positioning system (GPS), or global navigation satellite system (GNSS). For example, the network may perform wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS). .

The database 130 may store various data. The data stored in the database 130 is obtained by at least one component of the plurality of user terminals 110-1, 110-2, ..., 110-n and the device for providing a baby health diagnosis solution 120, As data to be processed or used, it may include software (eg: a program). Database 130 may include volatile and/or non-volatile memory. As an embodiment, the database 130 includes diaper feces images obtained from a plurality of user terminals 110-1, 110-2, ..., 110-n, and feces formed in the baby health diagnosis solution providing device 120. Objects, color codes, stool condition determination results, etc. can be stored.

In the present invention, artificial intelligence (AI) means a technology that imitates human learning ability, reasoning ability, perception ability, etc., and implements it with a computer, and concepts such as machine learning and symbolic logic can include Machine learning (ML) is an algorithm technology that classifies or learns the characteristics of input data by itself. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the result of the analysis, and can make judgments or predictions based on the result of the learning. In addition, technologies that mimic the functions of the human brain, such as recognition and judgment, using machine learning algorithms, can also be understood as artificial intelligence. For example, technical fields such as linguistic understanding, visual understanding, inference/prediction, knowledge expression, and motion control may be included.

Machine learning may refer to processing that trains a neural network model using experience of processing data. Through machine learning, computer software could mean improving its own data processing capabilities. A neural network model is constructed by modeling a correlation between data, and the correlation may be expressed by a plurality of parameters. The neural network model derives a correlation between data by extracting and analyzing features from given data, and optimizing the parameters of the neural network model by repeating this process can be referred to as machine learning. For example, a neural network model may learn a mapping (correlation) between an input and an output with respect to data given as an input/output pair. Alternatively, even when only input data is given, the neural network model may learn the relationship by deriving a regularity between given data.

An artificial intelligence learning model or a neural network model may be designed to implement a human brain structure on a computer, and may include a plurality of network nodes having weights while simulating neurons of a human neural network. A plurality of network nodes may have a connection relationship between them by simulating synaptic activities of neurons that transmit and receive signals through synapses. In the artificial intelligence learning model, a plurality of network nodes can send and receive data according to a convolutional connection relationship while being located in layers of different depths. The artificial intelligence learning model may be, for example, an artificial neural network model, a convolutional neural network model (CNN), and the like. As an embodiment, the artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Ada-boost , Perceptron, Genetic Programming, Clustering, etc. may be used.

Of these, CNNs are a type of multilayer perceptrons designed to use minimal preprocessing. A CNN consists of one or several convolution layers and general artificial neural network layers on top, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize the input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back-propagation. CNNs are easier to train than other feedforward artificial neural network techniques and have the advantage of using fewer parameters.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have greatly improved many computer vision tasks. With huge data sets, such as those available for many tasks today, overfitting is not critical, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. To this end, a distributed, scalable implementation of deep neural networks can be used.

When a diaper is photographed using a smartphone application of the user terminal 110-1, the baby health diagnosis solution providing device 120 may analyze the diaper image and provide health condition information according to color and shape from the baby's feces. there is.

When a parent accessing the user terminal 110-1 inputs basic information such as the baby's birth date, gender, and breastfeeding status through the application, the device for providing a baby health diagnosis solution 120 detects a diaper based on the image taken. It is possible to provide customized stool condition guide information according to the analyzed stool condition, medical examination result and cause according to the stool condition, necessity of visiting a pediatrician, and number of days after birth.

In pediatrics, health conditions are already diagnosed from baby feces, and taking diapers to a specialist is recommended in pediatric diagnosis. Image-based medical diagnosis technologies, such as CT/MRI, are also developing with the application of CNN algorithms. If machine learning-based image processing technology is used, it is possible to train specialist diagnosis results for diaper stool images.

While the birth rate is declining, the market for baby products, infant food, and baby health is growing rapidly as the rich are more interested in childrearing. Until now, sensing-based services of body information such as ECG, oxygen content, nearby fine dust, and body temperature have been provided for baby health diagnosis.

A “color code”-based diaper stool diagnosis solution can attract investment from diaper manufacturers and create a business model through collaboration with pediatricians.

In order to correct image distortion, the baby health diagnosis solution providing device 120 attaches a color code to one side of the diaper, applies the image containing the color code to a pre-learned SegNet model and a ResNet model to classify stool conditions, However, it can provide preliminary information for rapid pediatric diagnosis.

After developing a diaper feces image diagnosis solution, a diaper purchase coupon can be presented as an incentive for taking pictures to secure users. From the diaper manufacturer's point of view, customer loyalty to diapers can be increased by providing baby health diagnosis information even if it costs money, and solution accuracy can be increased by additionally using the stool image obtained in this way for training.

Currently, it is possible to secure users and increase sales by presenting a baby health diagnosis solution as an additional service to provide a bowel movement alarm using a diaper sensor. By presenting the user's subscription service options as 1) diaper + 2) bowel movement alarm + 3) baby's health condition diagnosis, the subscription service using diapers as an item can be expanded.

Based on the preliminary results of the baby's health condition diagnosis, a nearby pediatrician is recommended, and a revenue model can be created to use stool images for pediatric diagnosis. A business settlement may be possible.

The health of a baby with limited communication can be diagnosed from the color and shape of the stool. If abnormal symptoms are found from the condition of the baby's stool, you may need an accurate diagnosis by promptly seeking treatment at the hospital.

Parents show a picture of the baby's stool to the pediatrician or take the diaper to the doctor for an accurate diagnosis, because the condition of the baby's stool, which varies depending on the number of days after birth, is a measure of health.

For example, in the first week after birth, the stool condition changes every day, after 5 days after birth, golden stool is excreted because of breast milk, and after a week, the stool condition becomes yellow or brown depending on whether or not breastfeeding. If the stool is gray, you may need to see a specialist with the soiled diaper. However, it is difficult for parents to determine whether to visit a hospital by looking at the state of feces containing a lot of health information, and it may be inconvenient to bring a diaper to the hospital.

The baby health diagnosis market is growing rapidly with the development of artificial intelligence technology. Until now, most of the services are diagnostic services using information collected using sensors (ECG, oxygen level, nearby fine dust, body temperature, etc.).

As the use of artificial neural networks such as CNN increases, the technology for analyzing medical information through images is also being advanced. The use cases of diagnosis by radiologists through machine learning-based MRI/CT analysis technology are increasing, and the popularization of medical diagnosis services is accelerating.

The device for providing a baby health diagnosis solution 120 may photograph feces containing baby health diagnosis information and analyze the photographed image with deep learning technology to provide preliminary and auxiliary baby health-related information prior to expert judgment.

In a situation where the smart diaper market is also expanding and services that combine traditional diapers with creative IT technology are being launched one after another, if a baby health diagnosis service is launched without a separate sensor installed from feces directly related to baby health, it can be used at the appropriate time when medical care is needed. You can connect parents and pediatricians.

There is a medical classification system called the Bristol stool scale that diagnoses health conditions such as chronic gastrointestinal diseases through the shape and color of adult stool. Below is the Bristol Stool Chart.

The shape and color of a baby's feces change depending on the number of days of birth and breastfeeding, and the digestion ability is incomplete in the process of forming teeth and digestive organs. It can be important to provide parents.

As the maturity of baby health diagnosis solutions increases, it may be possible to build a classification system suitable for deep learning-based image processing technology by acquiring a large amount of images.

The device for providing a baby health diagnosis solution 120 may secure a new market in the baby health market by implementing a solution for preliminary diagnosis of a health condition based on baby feces information.

The device for providing a baby health diagnosis solution 120 may provide information necessary for childcare in a form suitable for the baby based on information about the baby, such as the number of months after birth, as well as analysis of the feces image of the diaper. For example, information such as food suitable for the number of months, bowel habit training method, and sensory training method may be provided.

The device for providing a baby health diagnosis solution 120 may obtain an image of the baby's stool and continuously monitor the health status of the baby by tracking the history.

Parents can determine whether a pediatric treatment is required through analysis of the baby's feces image, and if treatment is required, the device 120 for providing a baby health diagnosis solution can promote prompt treatment by connecting to a nearby hospital reservation service.

The apparatus 120 for providing a baby health diagnosis solution may provide information necessary for an accurate diagnosis by providing a diaper stool photographing result during a hospital diagnosis.

In order to improve the baby health diagnosis result solution, an application for preprocessing can be implemented in a standard that can be trained by taking a diaper feces image.

According to a 2019 paper on stool image classification (Augmenting Gastrointestinal Health: A Deep Learning Approach to Human Stool Recognition and Characterization in Macroscopic Images, https://arxiv.org/ftp/arxiv/papers/1903/1903.10578.pdf), 650 It is stated that the accuracy of the identification and classification model is approximately 70% or higher when training about 10 images, and an accuracy of 99.4% can be secured when training 25200 images.

It is possible to employ two pediatricians to perform labeling of the obtained diaper feces images, and to construct three classification models of constipation/normal/diarrhea first.

First, we label 500 training images, and based on the labeled training images, we can train a model to identify stool shapes with SegNet.

The apparatus 120 for providing a baby health diagnosis solution may train a model for performing color-based classification using ResNet on feces identified based on SegNet.

The apparatus 120 for providing a baby health diagnosis solution may implement a diaper feces image analysis solution by utilizing SegNet and ResNet learned from labeled images.

The apparatus 120 for providing a baby health diagnosis solution may implement an application equipped with an image distortion correction function by utilizing a diaper feces image to which a “color code” is attached.

Distortion may occur due to image conditions that change depending on the imaging environment of the diaper feces image or the specifications of the image sensor. Analysis results can improve the accuracy of diagnosis.

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

2 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.

As shown in FIG. 2 , the learning device may train the neural network 124 for pre-processing of a stool object included in a diaper stool image. According to one embodiment, the learning device may be a separate subject different from the baby health diagnosis solution providing device 120, but is not limited thereto.

The neural network 124 includes an input layer 122 into which training samples are input and an output layer 126 which outputs training outputs, and may be learned based on differences between the training outputs and labels. Here, the labels may be defined based on color codes and identification information corresponding to the color codes. The neural network 124 is connected to a group of a plurality of nodes, and is defined by weights between connected nodes and an activation function that activates the nodes.

The learning device may train the neural network 124 using a gradient descent (GD) technique or a stochastic gradient descent (SGD) technique. The learning device may use a loss function designed by outputs and labels of the neural network.

The learning device may calculate a training error using a predefined loss function. The loss function may be predefined with labels, outputs and parameters as input variables, where the parameters may be set by weights in the neural network 124. For example, the loss function may be designed in a mean square error (MSE) form, an entropy form, and the like, and various techniques or methods may be employed in an embodiment in which the loss function is designed.

The learning device may find weights affecting the training error using a backpropagation technique. Here, weights are relationships between nodes in neural network 124 . The learning device may use the SGD technique using labels and outputs to optimize the weights found through the backpropagation technique. For example, the learning device may update weights of a loss function defined based on labels, outputs, and weights using the SGD technique.

According to an embodiment, the learning device may obtain training diaper stool images and extract training color codes from the training diaper stool images. The learning device may obtain prelabeled information (first labels) for each of the training color codes, and may obtain first labels indicating color information predefined in the training color codes.

According to an embodiment, the learning device may generate first training feature vectors based on appearance features, pattern features, and color features of training color codes. Various methods may be employed to extract features.

According to an embodiment, the learning device may obtain training outputs by applying the first training feature vectors to the neural network 124 . The learning device may train the neural network 124 based on the training outputs and the first labels. The learning device may train the neural network 124 by calculating training errors corresponding to training outputs and optimizing a connection relationship of nodes in the neural network 124 to minimize the training errors. The apparatus 120 for providing a baby health diagnosis solution may extract a color code from a diaper feces image using the neural network 124 that has been trained.

According to another embodiment, the learning device may obtain training diaper stool images and extract training color codes from the training diaper stool images. The learning device may obtain pre-labeled information (second labels) for each of the training color codes, and obtain second labels indicating predefined rectangular information for the training color codes.

According to another embodiment, the learning device may generate second training feature vectors based on appearance features, pattern features, and color features of the training color codes. Various methods may be employed to extract features.

According to another embodiment, the learning device may obtain training outputs by applying the second training feature vectors to the neural network 124 . The learning device may train the neural network 124 based on the training outputs and the second labels. The learning device may train the neural network 124 by calculating training errors corresponding to training outputs and optimizing a connection relationship of nodes in the neural network 124 to minimize the training errors. The apparatus 120 for providing a baby health diagnosis solution may flatten the curvature of the color code extracted from the diaper feces image by using the neural network 124 that has been trained.

According to another embodiment, the learning device may receive the identifier's personal information and food intake information. For example, the personal information may include at least one of age, gender, and allergy information of the identifier, and the food intake information may include at least one of breastfeeding status, type of baby food, and food eaten, but is not limited thereto. .

According to another embodiment, the learning device may obtain training diaper stool images and extract training stool objects from the training diaper stool images. The learning device may obtain pre-labeled information (third labels) for the training stool objects, respectively, and may obtain third labels indicating the status of the training stool objects according to the personal information of the identifier and the food information consumed. there is.

According to another embodiment, the learning device may generate third training feature vectors based on appearance features, pattern features and color features of the training fecal objects. Various methods may be employed to extract features.

According to another embodiment, the learning device may obtain training outputs by applying the third training feature vectors to the neural network 124 . The learning device may train the neural network 124 based on the training outputs and the third labels. The learning device may train the neural network 124 by calculating training errors corresponding to training outputs and optimizing a connection relationship of nodes in the neural network 124 to minimize the training errors. The apparatus 120 for providing a baby health diagnosis solution may generate a stool state determination result of an identifier from a diaper stool image using the neural network 124 after learning has been completed.

3 is a diagram for explaining an apparatus for providing a baby health diagnosis solution according to an embodiment.

As shown in FIG. 3 , the apparatus 120 for providing a baby health diagnosis solution may include one or more processors 122 , one or more memories 124 , and/or a transceiver 126 . As an example, at least one of these components of the device for providing a health diagnosis solution for baby 120 may be omitted or another component may be added to the device for providing a health diagnosis solution for baby 120 . Additionally or alternatively, some of the components may be integrated and implemented, or implemented as a singular or plural entity. At least some of the internal and external components of the baby health diagnosis solution providing device 120 use a bus, general purpose input/output (GPIO), serial peripheral interface (SPI) or mobile industry processor interface (MIPI). They are connected to each other through and can exchange data and/or signals.

One or more processors 122 may control at least one component of the apparatus 120 for providing a baby's health diagnosis solution connected to the processor 122 by driving software (eg, commands, programs, etc.). In addition, the processor 122 may perform operations such as various calculations, processing, data generation, and processing related to the present invention. In addition, the processor 122 may load data or the like from one or more memories 124 or store data in one or more memories 124 .

As described above, the one or more processors 122, via the transceiver 126, a plurality of user terminals (110-1, 110-2, ..., 110-n) of any one of the user terminal (eg, 110 Diaper stool images for a preset period from -1) may be received in real time or non-real time in the form of digital packets.

One or more processors 122 may detect stool objects and color codes using diaper stool images received from the user terminal 110 - 1 through the transceiver 126 . According to one embodiment, one or more processors 122 may detect stool objects and color codes from diaper stool images using a machine learning algorithm, such as deep learning.

One or more processors 122 may pre-process the feces object based on the detected color code. According to an embodiment, a curvature of the detected color code is flattened, colors of the color code are detected, predefined color information for the colors of the color code is compared with the detected colors to generate a color correction value, The color of the feces object may be corrected using the color correction value.

One or more processors 122 may process the pre-processed stool object to generate a stool status determination result of the identifier. According to one embodiment, the processor 122 may generate the stool condition determination result of the identifier as diarrhea, constipation, normal, very good, good, little, normal, slightly weak, weak, very weak, etc. Not limited.

One or more memories 124 may store various data. Data stored in the memory 124 is data acquired, processed, or used by at least one component of the device for providing a baby health diagnosis solution 120, and may include software (eg, commands, programs, etc.) can Memory 124 may include volatile and/or non-volatile memory. In the present invention, commands or programs are software stored in the memory 124, and the operating system, application and/or application for controlling resources of the device for providing a health diagnosis solution for baby 120 is the device for providing a health diagnosis solution for baby 120 It can include middleware that provides various functions to applications so that they can utilize the resources of the system.

The one or more memories 124 store diaper stool images received from the above-described plurality of user terminals 110-1, 110-2, ..., 110-n through the network N, and are formed by the one or more processors 122. Feces objects, color codes, feces state determination results, etc. may be stored. Additionally, the one or more memories 124 may store instructions that, when executed by the one or more processors 122, cause the one or more processors 122 to perform operations.

As an embodiment, the baby health diagnosis solution providing device 120 may further include a transceiver 126 . The transceiver 126 is wireless or wired between a plurality of user terminals 110-1, 110-2, ..., 110-n, the baby health diagnosis solution providing device 120, the database 130, and/or other devices. communication can be performed. For example, the transceiver 126 may transmit eMBB (enhanced mobile broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (long-term evolution), LTE-A (LTE Advance), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth (Bluetooth), NFC ( Wireless communication may be performed according to a method such as near field communication (GPS), global positioning system (GPS), or global navigation satellite system (GNSS). For example, the transceiver 126 may perform wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS). there is.

As an embodiment, one or more processors 122 control the transceiver 126 to receive information from a plurality of user terminals 110-1, 110-2, ..., 110-n and the device for providing a baby health diagnosis solution 120. can be obtained. Information acquired from the plurality of user terminals 110 - 1 , 110 - 2 , ... , 110 - n and the apparatus 120 for providing a baby health diagnosis solution may be stored in one or more memories 124 .

As an example, the baby health diagnosis solution providing device 120 may be various types of devices. For example, the baby health diagnosis solution providing device 120 may be a portable communication device, a computer device, or a device according to one or more combinations of the above devices. The device 120 for providing a baby health diagnosis solution of the present invention is not limited to the devices described above.

Various embodiments of the baby health diagnosis solution providing device 120 according to the present invention may be combined with each other. Each embodiment may be combined according to the number of cases, and the embodiments of the device 120 for providing a baby health diagnosis solution made in combination also fall within the scope of the present invention. In addition, internal/external components of the apparatus 120 for providing a baby health diagnosis solution according to the present invention described above may be added, changed, replaced, or deleted according to embodiments. In addition, internal/external components of the above-described device for providing a baby health diagnosis solution 120 may be implemented as hardware components.

4 is a flowchart of a method for providing a necessary baby health diagnosis solution according to an embodiment.

Although process steps, method steps, algorithms, etc. are described in a sequential order in the flowchart of FIG. 4, such processes, methods and algorithms may be configured to operate in any suitable order. In other words, the steps of the processes, methods and algorithms described in the various embodiments of the invention need not be performed in the order described herein. Also, although some steps are described as being performed asynchronously, in other embodiments some of these steps may be performed concurrently. Further, illustration of a process by depiction in the drawings does not mean that the illustrated process is exclusive of other changes and modifications thereto, and that any of the illustrated process or steps thereof may be one of various embodiments of the present invention. It is not meant to be essential to one or more, and it does not imply that the illustrated process is preferred.

As shown in Fig. 4, in step S410, the diaper feces image of the identifier is obtained. For example, referring to FIGS. 1 to 3 , the apparatus 120 for providing a baby health diagnosis solution is any one of a plurality of user terminals 110-1, 110-2, ..., 110-n For example, a diaper stool image may be obtained from 110-1) through the network N.

In step S420, a feces object and a color code are detected from the diaper feces image. For example, referring to FIGS. 1 to 3 , the apparatus 120 for providing a baby health diagnosis solution may detect a stool object and a color code from the diaper stool image obtained in step S410.

In step S430, the stool object is pre-processed based on the color code. For example, referring to FIGS. 1 to 3 , the apparatus 120 for providing a baby health diagnosis solution may pre-process the stool object detected in step S410 based on the color code detected in step S410. According to an embodiment, the baby health diagnosis solution providing device 120 flattens the curve of the detected color code, detects the colors of the color code, and provides predefined color information and detected color information about the colors of the color code. It is possible to perform preprocessing of the feces object by comparing the colors to generate a color correction value and correcting the color of the feces object using the color correction value.

In step S440, the pre-processed feces object is processed to generate a stool status determination result of the identifier. For example, referring to FIGS. 1 to 3 , the apparatus 120 for providing a baby health diagnosis solution may process the feces object preprocessed in step S430 to generate a stool condition determination result of an identifier. According to one embodiment, the baby health diagnosis solution providing device 120 generates diarrhea, constipation, normal, very good, little, normal, slightly thin, thin, very thin, etc. as the stool condition determination result of the identifier can do.

5 is an exemplary diagram of a configuration of a device according to an embodiment.

Device 401 according to an embodiment includes a processor 402 and a memory 403 . The device 401 according to an embodiment may be the above-described server or terminal. The processor may include at least one device described above with reference to FIGS. 1 to 4 or may perform at least one method described above with reference to FIGS. 1 to 4 . The memory 403 may store information related to the above method or store a program in which the above method is implemented. Memory 403 may be volatile memory or non-volatile memory.

The processor 402 may execute a program and control the device 401 . Codes of programs executed by the processor 402 may be stored in the memory 403 . The device 401 may be connected to an external device (eg, a personal computer or network) through an input/output device (not shown) and exchange data.

6 to 14 are application implementation examples of a baby health diagnosis solution according to an embodiment.

6 The same initial screen may be displayed on the display of the user terminal 110-1. An ID and password input field (①), a social network service (SNS) login button (②), and an input button (③) may be displayed on the initial screen. If you do not enter an ID and/or password and select the input button (③), the message “Please enter your ID and password” is displayed, and you enter an ID and/or password that does not match the preset ID and/or password. If you select the input button (③), the message “Please check your ID or password” may be displayed.

When the application is run for the first time, a member information input screen for service subscription may be displayed. When signing up for the service, you can select a member type, and in the case of a doctor member (a pediatric bowel specialist), you can use menus such as opinion registration and bowel information check. menu is available. When the service subscription is completed by entering information such as ID, password, and address, the application can be used.

As shown in FIG. 7 , a screen for capturing a diaper feces image may be displayed. On the photographing screen ①, a photographing button may be displayed in conjunction with the basic camera provided in the user terminal 110-1, and the phrase "Please photograph the color code and diaper visible" may be displayed. If you select the baby-shaped button (②) located at the top, the findings history screen may be displayed.

As shown in FIG. 8, a diaper feces image (①) photographed by a camera is displayed on the findings detail screen, but an image without image processing may be displayed. Information such as the number of defecation, the time of defecation, health condition, previous diet, baby's weight, and special matters can be entered in the information input box (②) of the captured diaper feces image. For example, a separate screen is displayed for the state of health, and it is selectable as very good, good, average, bad, or very bad. The previous meal is displayed on a separate screen, so you can select powdered milk, baby food, or rice. When all information is entered and the transmit button is selected, the diaper stool image captured by the camera is transmitted to the server, and image processing such as color correction can be performed using the neural network existing in the server. The image processed by the server may be transmitted to the baby health diagnosis solution providing device 120 and provided as a resultant image.

As shown in FIG. 9, the view detail screen displays past view details (①) of the member, and a button (②) to move to the shooting screen is displayed on the upper right, and a logout button (③) is displayed on the upper left. ) can be displayed.

As shown in FIG. 10, on the finding confirmation screen, an icon and number (defecation) for each type according to the defecation status (very, very, slightly, normal, slightly thin, thin, very thin) in the defecation status display box (①) Types A to G) may be displayed, and opinions written between doctors may be displayed in the doctor's opinion section (②).

As shown in FIG. 11, on the doctor information and reservation screen, the hospital name, address, phone number, etc. are displayed as subscription information of the doctor member in the doctor information field (①), and the date selection field and time selection field are displayed in the hospital reservation field (②). etc. can be displayed.

As shown in FIG. 12, a summary information check box (①) is displayed on the defecation information confirmation screen, so that the doctor member can check the information of the baby member.

As shown in FIG. 13, a diaper feces image (①) photographed by the user terminal 110-1 and processed by the server may be displayed on the opinion registration screen. The diaper feces image may be displayed together with information such as the number of bowel movements, the time of bowel movements, health condition, previous diet, baby weight, and special matters. In the doctor's opinion field (②), a defecation type selection button and an opinion input field in which the doctor's opinion can be freely entered may be displayed together.

As shown in FIG. 14, a diaper feces image (①) photographed by the user terminal 110-1 and processed by the server may be displayed on the past history details screen. The diaper feces image may be displayed together with information such as the number of bowel movements, the time of bowel movements, health condition, previous diet, baby weight, and special matters.

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

Claims (1)

obtaining a diaper feces image of the identifier;
Personal information of the identifier—the personal information includes at least one of age, gender, and allergy information of the identifier—and food information consumed—the food information includes at least one of breastfeeding status, type of baby food, and food eaten receiving a ham-;
Detecting stool objects and color codes from training diaper stool images by using an artificial intelligence model pretrained to detect training stool objects and training color codes based on predefined color information from the training diaper stool images;
pre-processing the feces object based on the color code;
Transmitting the preprocessed feces object to an expert terminal;
Receiving feedback information about a stool state of the preprocessed stool object from the expert terminal;
assigning a weight to each of the stool conditions;
obtaining first diaper stool images for a period preset for the identifier;
Calculating a digestibility index by multiplying and summing the assigned weights for each of the stool conditions;
Calculating an digestibility ranking of the identifier based on the digestibility index;
calculating a first body age index of the identifier based on the stool condition;
calculating a second body age index of the identifier using the personal information;
calculating a third body age index of the identifier by using the food intake information;
assigning a first weight to each of the first to third body age indices;
calculating an average of values obtained by multiplying the first to third body age indices by a first weight assigned to each of the first to third body age indices; and
Generating the calculated average as a final body age index
including,
Detecting the color code
acquiring stool images of the training diaper;
extracting the training color codes from the training diaper stool images;
obtaining first labels indicating predefined color information for the training color codes;
generating training outputs corresponding to the training color codes by applying the training color codes to a pretrained first neural network; and
training the first neural network based on the training outputs and the first labels;
including,
The step of preprocessing the stool object is
flattening the curvature of the color code;
detecting colors of the color code;
generating a color correction value by comparing predefined color information for the colors of the color code with the detected colors; and
Correcting the color of the stool object using the color correction value
including,
How to provide baby health diagnosis solution.

Images