Method and system for predicting bladder urine volume

Abstract

To provide a method and a system for predicting bladder urine volume.SOLUTION: A method for predicting bladder urine volume is implemented by at least one processor, the method including the steps of: receiving an optical dataset associated with a specific user detected by a plurality of photodiodes, the plurality of photodiodes being configured to detect the intensity of light associated with light radiated to the skin located above the bladder of the specific user; estimating an optical characteristic value set for at least a part of the body of the specific user on the basis of the optical dataset; and estimating the bladder urine volume of the specific user using a urine volume estimation model on the basis of the estimated optical characteristic value set.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a method and system for predicting the amount of urine in the bladder, and more specifically, to a method and system for estimating physiological information based on optical information of the body.

Background Art

[0002] With the advent of an aging society, one in ten people aged 60 or over has urinary incontinence. In particular, patients with spinal cord injuries, dementia, stroke, urinary incontinence, nocturia, etc. may have difficulty determining the timing of their own urination and / or catheterization. If urination and / or catheterization are not performed at an appropriate time, it can lead to mild bladder dysfunction such as frequent urination, urinary incontinence, and urinary retention, as well as complications such as urinary tract infections, hydronephrosis, and vesicoureteral reflux.

[0003] For an accurate diagnosis of urinary incontinence, after a patient visits a hospital, the bladder function can be confirmed by measuring the amount of urine in the bladder through ultrasonic bladder volume examination, urodynamic study (UDS), etc. That is, the patient has to observe / diagnose the bladder function by visiting a hospital and measuring the amount of urine in the bladder only through expert examinations.

[0004] Alternatively, patients with urinary incontinence who have unclear urinary urgency or incomplete urination ability have to urinate and / or catheterize themselves or with the help of a caregiver in accordance with the guidelines of an expert diagnosed at a hospital and / or within the time intervals given to the patient. Due to various factors such as the patient's condition and drinking status on the day, the patient's urine output can deviate from the normal range of general urine output. At this time, if urination and / or catheterization are performed only depending on a certain time interval, problems such as difficulty in urination at an appropriate time, persistent urinary tract infections, and decreased kidney function may occur. Therefore, there can be various difficult problems such as restrictions on the patient's external activities and drinking restrictions in order to follow the medical guidelines instructing urination and / or catheterization at certain time intervals. Summary of the Invention [Problem to be solved by the invention]

[0005] The present disclosure provides a method and system (apparatus) for predicting bladder urine volume to solve the above problems. [Means for solving the problem]

[0006] The present disclosure may be embodied in numerous ways, including as a method, an apparatus (system), and / or a computer program stored on a computer readable storage medium, or a computer readable storage medium having a computer program stored thereon.

[0007] According to one embodiment of the present disclosure, a method for predicting bladder urine volume performed by at least one processor includes the steps of receiving a light data set associated with a specific user detected by a plurality of photodiodes, where the plurality of photodiodes are configured to detect light intensity associated with light irradiated onto skin located over the specific user's bladder, estimating a set of light characteristic values ​​for at least a portion of the specific user's body based on the light data set, and estimating the specific user's bladder urine volume using a urine volume estimation model based on the estimated set of light characteristic values.

[0008] According to one embodiment of the present disclosure, the urine volume estimation model is a deep learning-based model or a machine learning-based model trained on a plurality of training data sets, and the plurality of training data sets may include a pair of a particular user's actual urine volume and a set of light characteristic values ​​associated with the actual urine volume.

[0009] According to one embodiment of the present disclosure, the multiple learning data sets include a first learning data set and a second learning data set, and the first learning data set may include a pair of a first actual urine volume of a particular user and a first learning light characteristic value set associated with the first actual urine volume.

[0010] According to one embodiment of the present disclosure, the second training data set may include a pair of a second actual urine volume of a particular user and a second training light characteristic value set associated with the second actual urine volume.

[0011] According to one embodiment of the present disclosure, the second actual urine volume can be greater than the first actual urine volume.

[0012] According to one embodiment of the present disclosure, a teacher model is generated by learning multiple learning datasets, and the urine volume estimation model further learns a single or multiple additional learning datasets, where the additional learning dataset includes an additional learning urine volume and a pair of additional learning light characteristic value sets estimated by inputting the additional learning urine volume into the teacher model, and the additional learning urine volume can be greater than the first urine volume and smaller than the second urine volume.

[0013] According to one embodiment of the present disclosure, a urine volume estimation model may be trained by applying predefined weights to multiple training data sets.

[0014] According to one embodiment of the present disclosure, the first actual urine volume may correspond to a minimum urine volume of the particular user's bladder, and the second actual urine volume may correspond to a maximum urine volume of the particular user's bladder.

[0015] According to an embodiment of the present disclosure, the method may further include outputting a message associated with a urination recommendation if the estimated urine volume information is equal to or greater than a predetermined reference value.

[0016] According to one embodiment of the present disclosure, the urine volume estimation model further learns obesity information for learning, and the method further includes a step of receiving obesity information associated with the specific user, and the step of estimating the urine volume may include a step of estimating the urine volume using the urine volume estimation model based on the received obesity information and the set of light characteristic values.

[0017] A computer program stored on a computer-readable recording medium for executing a method according to an embodiment of the present disclosure on a computer.

[0018] According to one embodiment of the present disclosure, a user terminal includes a communication unit, a memory, and at least one processor coupled to the memory and configured to execute at least one computer-readable program contained in the memory, the at least one program including receiving a light data set associated with a particular user detected by a plurality of photodiodes, where the plurality of photodiodes are configured to detect light intensity associated with light irradiated to skin located on the particular user's bladder, estimating a light characteristic value set for at least a portion of the particular user's body based on the light data set, and estimating the particular user's bladder urine volume using a urine volume estimation model based on the estimated light characteristic value set. Effect of the Invention

[0019] Some embodiments of the present disclosure may provide physiological information to a user without the assistance of a medical professional, such as a doctor, and may be easy to use, providing convenience for the user and increasing accessibility to consumers through personalization.

[0020] According to some embodiments of the present disclosure, system parameters for a plurality of photodiodes included in a medical device can be uniformly calibrated. After performing one generation of calibration parameters, the medical device may not require additional calibration. That is, it is not necessary to perform calibration using a phantom, which can improve user convenience.

[0021] Some embodiments of the present disclosure can provide fast calculation and high accuracy, thereby enhancing user convenience.

[0022] According to some embodiments of the present disclosure, multiple light sources and multiple photodiodes can be used to provide physiological information about multiple regions. Physiological information can be provided not only for localized regions of the body, but also for a wide range of parts of the body. In addition, by providing physiological information about multiple regions, the state of organs contained within the body (e.g., the amount of urine stored in the bladder, the position of the bladder, etc.) can be specifically understood.

[0023] According to some embodiments of the present disclosure, a patient who does not feel the need to urinate can receive physiological information about his / her bladder and / or the amount of urine in the bladder in real time or periodically. Using the information provided, the patient can monitor the amount of urine stored in his / her bladder and urinate at the appropriate time.

[0024] According to some embodiments of the present disclosure, the urine volume estimation model can be provided for a user by learning the learning obesity level information. In addition, the urine volume estimation model uses a machine learning model or a deep learning model, which has relatively good support for app development, so that the invention according to the present disclosure can easily develop a mobile app for a wearable device. In addition, the machine learning model or the deep learning model is easy to re-learn, and the invention according to the present disclosure can perform an individualized estimation of the urine volume in the bladder. In addition, the urine volume estimation model is easy to maintain and improve, and has excellent model scalability and model versatility.

[0025] According to some embodiments of the present disclosure, when the amount of the learning data set is not large, a teacher model generates multiple additional learning data sets to perform data augmentation. The urine volume prediction model learns more data through data augmentation, and thus an automated bladder urine volume prediction method can be realized that is designed based on medical knowledge and diagnosis.

[0026] The effects of the present disclosure are not limited thereto, and other effects not mentioned therein should be clearly understood by a person having ordinary knowledge in the technical field to which the present disclosure pertains (referred to as an "ordinary engineer") from the description in the claims. [Brief description of the drawings]

[0027] BRIEF DESCRIPTION OF THE DRAWINGS Examples of the present disclosure will now be described with reference to the accompanying drawings, in which like reference numerals refer to like elements, without limitation, and in which: FIG. [Figure 1] FIG. 1 is a schematic diagram illustrating an example of a medical device for estimating physiological information according to an embodiment of the present disclosure. [Diagram 2] FIG. 1 is a schematic diagram showing a configuration in which an information processing system, a medical device, and multiple user terminals are communicatively connected to one embodiment of the present disclosure. [Diagram 3] 2 is a block diagram showing an internal configuration of a user terminal and an information processing system according to an embodiment of the present disclosure. FIG. [Figure 4] FIG. 4 is a diagram showing an example of an aspect in which diffuse light is detected using first to third photodiodes according to an embodiment of the present disclosure. [Diagram 5] FIG. 1 is a diagram illustrating an example of a process for estimating physiological information according to an embodiment of the present disclosure. [Figure 6] FIG. 13 is a diagram showing an example of generating calibration parameters using a calibration box according to an embodiment of the present disclosure. [Figure 7] 11 is a graph illustrating an example of a process for generating calibration parameters according to an embodiment of the present disclosure. [Figure 8] FIG. 13 is a diagram illustrating an example in which calibration parameters according to an embodiment of the present disclosure are applied. [Figure 9] FIG. 11 is a diagram illustrating an example of a learning process of an initial light characteristic value estimation model according to an embodiment of the present disclosure. [Figure 10] FIG. 1 illustrates an example of a medical device according to an embodiment of the present disclosure. [Figure 11] FIG. 13 is a diagram illustrating an example of estimating physiological information based on a plurality of optical data according to an embodiment of the present disclosure. [Figure 12] FIG. 13 illustrates an example of estimating physiological information based on multiple light scattering coefficient data maps according to one embodiment of the present disclosure. [Figure 13] FIG. 1 is a block diagram illustrating an example of a method for estimating bladder urine volume according to an embodiment of the present disclosure. [Figure 14] 1 is a graph showing an example of learning data according to an embodiment of the present disclosure. [Figure 15] FIG. 2 is a block diagram showing an example of a urine volume estimation model according to an embodiment of the present disclosure. [Figure 16] 11 is a graph showing a number of examples of a urine volume estimation model according to an embodiment of the present disclosure. [Figure 17] 1 is a flowchart for explaining a method for predicting a urine volume in the bladder according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] Hereinafter, specific contents for carrying out the present disclosure will be described in detail with reference to the accompanying drawings. However, in the following description, detailed descriptions of known functions and configurations will be omitted if there is a risk of unnecessarily obscuring the gist of the present disclosure.

[0029] In the accompanying drawings, the same or corresponding components are given the same reference numerals. In addition, in the following description of the embodiments, duplicated descriptions of the same or corresponding components may be omitted. However, even if the description of a component is omitted, it should not be intended that such a component is not included in a certain embodiment.

[0030] The advantages and features of the disclosed embodiments and the methods of achieving them will become apparent from the following examples, taken in conjunction with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below, and may be embodied in various different forms. However, the embodiments are provided only to complete the disclosure and to allow those skilled in the art to accurately recognize the category of the invention.

[0031] The terms used in this disclosure are briefly explained, and the disclosed embodiments are specifically described. The terms used in this disclosure are selected as currently widely used general terms as much as possible while taking into consideration the functions in this disclosure, but these may change depending on the intentions or precedents of engineers engaged in related fields, the emergence of new technologies, etc. In addition, in certain cases, there may be terms arbitrarily selected by the applicant, but the meanings of these will be described in detail in the description of the invention. Therefore, the terms used in this disclosure should be defined based on the meanings of the terms and the overall content of this disclosure, rather than simply the names of the terms.

[0032] In this disclosure, unless otherwise clearly specified in the context, singular expressions include plural expressions, and plural expressions include singular expressions. Throughout the specification, when a part "comprises" a certain element, this does not exclude other elements, and means that other elements may also be included, unless otherwise specified to the contrary.

[0033] In addition, the term "module" or "unit" as used in the specification means a software or hardware component, and the "module" or "unit" performs a certain function. However, the term "module" or "unit" is not limited to software or hardware. The "module" or "unit" may be configured to reside on an addressable storage medium and may be configured to execute one or more processors. Thus, by way of example, a "module" or "unit" may include components such as software components, object-oriented software components, class components, and task components, as well as at least one of a process, a function, an attribute, a procedure, a subroutine, a segment of program code, a driver, firmware, microcode, a circuit, data, a database, a data structure, a table, an array, or a variable. The components and "modules" or "units" may be combined into a smaller number of components and "modules" or "units" or the functions provided therein may be further separated into additional components and "modules" or "units".

[0034] According to one embodiment of the present disclosure, a "module" or a "unit" may be embodied with a processor and a memory. A "processor" should be broadly construed to include a general purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, and the like. In some environments, a "processor" may also refer to an application specific semiconductor (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), and the like. A "processor" may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors in conjunction with a DSP core, or any other such configuration. Additionally, a "memory" should be broadly construed to include any electronic component capable of storing electronic information. "Memory" may refer to various types of processor-readable media such as Random Access Memory (RAM), Read Only Memory (ROM), Non-Volatile Random Access Memory (NVRAM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, magnetic or optical data storage devices, registers, etc. Memory is said to be in electronic communication with a processor when the processor can read / write information from the memory or write information to the memory. Memory that is integrated into a processor is in electronic communication with the processor.

[0035] Furthermore, terms such as 1, 2, A, B, (a), (b), etc. used in the following examples are used only to distinguish one component from another, and are not intended to limit the essence, order, or procedure of the component.

[0036] In addition, in the following examples, when a certain component is "coupled," "coupled," or "connected" to another component, it should be understood that the components may be directly coupled or connected to each other, but that other components may also be "coupled," "coupled," or "connected" between each component.

[0037] In the present disclosure, "each of a plurality of A's" can refer to each of all the components included in the plurality of A's, or can refer to each of some of the components included in the plurality of A's.

[0038] Furthermore, the word "comprises" or "comprising" as used in the following examples does not exclude the presence or addition of one or more other components, steps, operations and / or elements to a referenced component, step, operation and / or element.

[0039] In the present disclosure, "diffuse reflectance" may refer to the ratio of the light intensity of a light source to the light intensity of the diffused light measured at a specific distance from the light source. Here, diffuse light may refer to light diffused from an object on which light is irradiated. For example, when light is irradiated onto a body, diffuse reflectance may refer to the ratio of the light intensity of the light source to the light intensity of the diffused light measured at a specific distance from the light source. Specifically, diffuse reflectance may be expressed as the following formula 1.

[0040]

number

[0041] In this disclosure, a "system parameter" may refer to a coefficient associated with the light detection of a photodiode. The system parameter may include a proportional coefficient and an intercept coefficient. The proportional coefficient and intercept coefficient of the system parameter may be understood from the following explanation.

[0042] When the optical data detected by the photodiode is a voltage value, it can be expressed as the following equation 2.

[0043]

number

[0044] Various embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings.

[0045] FIG. 1 is a schematic diagram illustrating an example of a medical device 100 for estimating physiological information according to an embodiment of the present disclosure. As illustrated in the figure, the medical device 100 may include a communication unit for transmitting and receiving with a user terminal 120. The medical device 100 may also include a plurality of photodiodes 112_1-112_20 and a plurality of light source groups 114_1-114_4. The medical device 100 may obtain light data related to a body using the plurality of photodiodes 112_1-112_20 and the plurality of light source groups 114_1-114_4. The user terminal 120 may receive the light data related to the body and estimate physiological information of a user based on the received light data. FIG. 1 illustrates the medical device 100 including 20 photodiodes 112_1-112_20 and four light source groups 114_1-114_4, but is not limited thereto. That is, the number of photodiodes and the number of light source groups included in the medical device 100 may be changed as necessary.

[0046] In one embodiment, the multiple photodiodes 112_1 to 112_20 and the multiple light source groups 114_1 to 114_4 may be arranged on one surface of the medical device 100. In this case, the medical device 100 may be attached to the body so that the one surface faces the body. As an example, the medical device 100 may be attached to the body so that the one surface faces the site where the bladder is located.

[0047] In one embodiment, each of the light source groups 114_1 to 114_4 may include six light sources having different wavelengths, but is not limited thereto. For example, the first light source group 114_1 may include first to sixth light sources. The second light source group 114_2 may include seventh to twelfth light sources. The third light source group 114_3 may include thirteenth to eighteenth light sources. The fourth light source group 114_4 may include nineteenth to twenty-fourth light sources. Each of the first to twenty-fourth light sources may be an LD (Laser Diode), an LED (Light-Emitting Diode), or an OLED (Organic Light-Emitting Diode). Also, each of the first to twenty-fourth light sources may emit continuous wave light.

[0048] In one embodiment, the light sources included in each of the light source groups 114_1 to 114_4 can be configured to emit light of different wavelengths. For example, the first to sixth light sources included in the first light source group 114_1 can emit light of different wavelengths. Furthermore, the seventh to twelfth light sources included in the second light source group 114_2 can emit light of different wavelengths. Furthermore, the thirteenth to eighteenth light sources included in the third light source group 114_3 can emit light of different wavelengths. Furthermore, the nineteenth to twenty-fourth light sources included in the fourth light source group 114_4 can emit light of different wavelengths.

[0049] Here, the light sources of the different light source groups can irradiate light of the same wavelength. For example, the first light source, the seventh light source, the thirteenth light source, and the nineteenth light source can irradiate light of the same wavelength. Similarly, the second light source, the eighth light source, the fourteenth light source, and the twentieth light source can irradiate light of the same wavelength. Also, the third light source, the ninth light source, the fifteenth light source, and the twenty-first light source can irradiate light of the same wavelength. Also, the fourth light source, the tenth light source, the sixteenth light source, and the twenty-second light source can irradiate light of the same wavelength. Also, the fifth light source, the eleventh light source, the seventeenth light source, and the twenty-third light source can irradiate light of the same wavelength. Also, the sixth light source, the twelfth light source, the eighteenth light source, and the twenty-fourth light source can irradiate light of the same wavelength.

[0050] In one embodiment, the multiple photodiodes 112_1 to 112_20 can detect light and generate light data. Specifically, the multiple photodiodes 112_1 to 112_20 can detect the light intensity of diffused light, which is light diffused from the body. Also, the multiple photodiodes 112_1 to 112_20 can detect diffused light associated with a light source irradiated by a light source included in the multiple light source groups 114_1 to 114_4. Also, each photodiode can detect the diffused light and measure a voltage value corresponding to the light intensity of the diffused light. At this time, when one light source is turned on, one photodiode can detect the diffused light.

[0051] In one embodiment, the user terminal 120 may send an optical data detection request to the medical device 100. The medical device 100 may operate the light source groups 114_1-114_4 and detect the photodiodes 112_1-112_20 in response to the optical data detection request. The process of operating the light source groups 114_1-114_4 and detecting the photodiodes 112_1-112_20 will be described in detail below with reference to FIG. 10. Alternatively, the medical device 100 may periodically detect optical data and transmit it to the user terminal 120 without receiving an optical data detection request from the user terminal 120.

[0052] In one embodiment, the medical device 100 can transmit a plurality of optical data detected through a plurality of photodiodes 112_1 to 112_20 to the user terminal 120. A processor included in the user terminal 120 can estimate physiological information based on the plurality of optical data. Here, the physiological information can be moisture (H 2 O), fat (fat) information, oxygenated hemoglobin (HbO 2The physiological information may include information about the blood pressure, blood glucose level, blood glucose level, and the like, information about the blood glucose level, information about deoxygenated hemoglobin (HHb), and bladder monitoring information (notification of urination time, notification of catheterization time, amount of urine in the bladder, etc.). A method for estimating physiological information based on a plurality of optical data will be described in detail later with reference to Figs. 4 to 12. Alternatively, the medical device 100 may directly estimate physiological information based on the optical data without transferring the optical data to the user terminal 120.

[0053] With this configuration, physiological information can be estimated based on the optical data acquired from the medical device 100. Additionally, the estimated physiological information can be provided to the user via the user terminal 120. In this manner, the invention according to the present disclosure can provide physiological information to the user without the support of a specialist such as a doctor. Furthermore, the invention according to the present disclosure can increase the convenience of the user through a simple method of use and increase the accessibility of consumers as a personal device.

[0054] 2 is a schematic diagram showing a configuration in which an information processing system 230 and a medical device 240 are connected to enable communication between a plurality of user terminals 210_1, 210_2, and 210_3 according to an embodiment of the present disclosure. As shown in the figure, the plurality of user terminals 210_1, 210_2, and 210_3 may be connected to the information processing system 230 and the medical device 240 that can provide a physiological information estimation service and / or a bladder urine volume prediction service via a network 220. Here, the plurality of user terminals 210_1, 210_2, and 210_3 may include terminals of users to which the physiological information estimation service and / or the bladder urine volume prediction service are provided.

[0055] According to one embodiment, the information processing system 230 may include computer-executable programs (e.g., downloadable applications) related to providing a physiological information estimation service, a bladder urine volume prediction service, etc., one or more server devices and / or databases that can store, provide and execute data, and one or more distributed computing devices and / or distributed databases based on a cloud computing service.

[0056] The physiological information estimation service and / or the bladder urine volume prediction service provided by the information processing system 230 may be provided to a user via a physiological information estimation service application, a bladder urine volume prediction service application, etc. installed in each of the multiple user terminals 210_1, 210_2, 210_3. For example, the information processing system 230 may provide information related to the physiological information estimation and / or the bladder urine volume prediction received from the user terminals 210_1, 210_2, 210_3 and / or the medical device 240, or perform corresponding processing, via the physiological information estimation service application, the bladder urine volume prediction service application, etc.

[0057] According to an embodiment, the information processing system 230 can estimate physiological information based on the light data, where the light data can be data measured by the medical device 240. The information processing system 230 can receive the light data directly from the medical device 240 or receive the light data via the user terminals 210_1, 210_2, 210_3. The information processing system 230 can provide the physiological information estimation result to the user terminals 210_1, 210_2, 210_3 and / or the medical device 240.

[0058] According to an embodiment, the information processing system 230 can estimate the bladder urine volume based on the light dataset, where the light dataset can be data measured by the medical device 240. The information processing system 230 can receive the light dataset directly from the medical device 240 or receive the light dataset via the user terminals 210_1, 210_2, 210_3. The information processing system 230 can provide the estimated bladder urine volume to the user terminals 210_1, 210_2, 210_3 and / or the medical device 240.

[0059] A plurality of user terminals 210_1, 210_2, and 210_3 can communicate with the information processing system 230 and the medical device 240 via the network 220. The network 220 can be configured to enable communication between the plurality of user terminals 210_1, 210_2, and 210_3, the information processing system 230, and the medical device 240. The network 220 can be composed of a wired network such as Ethernet (registered trademark), PLC (Power Line Communication), telephone line communication device, and RS-serial communication, a mobile communication network, a wireless network such as WLAN (Wireless LAN), Wi-Fi (registered trademark), Bluetooth (registered trademark), and ZigBee (registered trademark), or a combination thereof, depending on the installation environment. The communication method is not limited, and can include not only a communication method utilizing a communication network (e.g., a mobile communication network, a wired Internet, a wireless Internet, a broadcast network, a satellite network, etc.) that can include the network 220, but also a short-distance wireless communication between the user terminals 210_1, 210_2, and 210_3.

[0060] 2, the mobile phone terminal 210_1, the tablet terminal 210_2, and the PC terminal 210_3 are shown as examples of user terminals, but are not limited thereto, and the user terminals 210_1, 210_2, and 210_3 may be any computing device capable of wired and / or wireless communication and capable of installing and executing a physiological information estimation service application or a web browser. For example, the user terminal may include an AI speaker, a smartphone, a mobile phone, a navigation system, a desktop computer, a laptop computer, a digital broadcasting terminal, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), a tablet PC, a game console, a wearable device, an IoT (internet of things) device, a VR (virtual reality) device, an AR (augmented reality) device, a set-top box, and the like. Also, while FIG. 2 shows three user terminals 210_1, 210_2, and 210_3 communicating with the information processing system 230 and the medical device 240 via the network 220, this is not limited thereto, and a different number of user terminals may be configured to communicate with the information processing system 230 and the medical device 240 via the network 220.

[0061] 3 is a block diagram showing the internal configuration of a user terminal and an information processing system according to an embodiment of the present disclosure. The user terminal 210 can refer to any computing device capable of executing a physiological information estimation service application, a bladder urine volume prediction service application, and the like, and capable of wired / wireless communication, and can include, for example, a mobile phone terminal 210_1, a tablet terminal 210_2, and a PC terminal 210_3 of FIG. 2. As shown in the figure, the user terminal 210 can include a memory 312, a processor 314, a communication module 316, and an input / output interface 318. Similarly, the information processing system 230 can include a memory 332, a processor 334, a communication module 336, and an input / output interface 338. As shown in FIG. 3, the user terminal 210 and the information processing system 230 can be configured to communicate information and / or data via the network 220 using the respective communication modules 316 and 336. Additionally, the input / output device 320 may be configured to input information and / or data to the user terminal 210 and output information and / or data generated by the user terminal 210 via the input / output interface 318.

[0062] The memories 312 and 332 may include any non-transitory computer-readable recording medium. According to an embodiment, the memories 312 and 332 may include a permanent mass storage device such as a read only memory (ROM), a disk drive, a solid state drive (SSD), and a flash memory. As another example, a permanent mass storage device such as a ROM, an SSD, a flash memory, and a disk drive may be included in the user terminal 210 or the information processing system 230 as a permanent storage device separate from the memory. In addition, the memories 312 and 332 may store an operating system and at least one program code (e.g., a code for a physiological information estimation service application, a bladder urine volume prediction service application, etc., installed and run in the user terminal 210).

[0063] Such software components, etc. may be loaded from a computer-readable recording medium separate from the memory 312, 332. Such separate computer-readable recording medium may include a recording medium directly connectable to such a user terminal 210 and information processing system 230, and may include, for example, a computer-readable recording medium such as a floppy drive, a disk, a tape, a DVD / CD-ROM drive, and a memory card. As another example, the software components, etc. may be loaded into the memory 312, 332 via the communication module 316, 336, rather than a computer-readable recording medium. For example, at least one program may be loaded into the memory 312, 332 based on a computer program to be installed by a file provided via the network 220 by a developer or a file distribution system that distributes an installation file of an application.

[0064] The processor 314, 334 may be configured to process computer program instructions by performing basic arithmetic, logic, and input / output operations. The instructions may be provided to the processor 314, 334 by the memory 312, 332 or the communication module 316, 336. For example, the processor 314, 334 may be configured to execute instructions received by program code stored in a storage device, such as the memory 312, 332.

[0065] The communication modules 316, 336 may provide configurations and functions for the user terminal 210 and the information processing system 230 to communicate with each other via the network 220, and may provide configurations and functions for the user terminal 210 and / or the information processing system 230 to communicate with other user terminals or other systems (e.g., another cloud system, etc.). As an example, a request or data (e.g., optical data, a request to estimate physiological information, a request to estimate urine volume in the bladder, etc.) generated by the processor 314 of the user terminal 210 via a program code stored in a recording device such as the memory 312 may be transmitted to the information processing system 230 via the network 220 under the control of the communication module 316. Conversely, a control signal or command provided by the processor 334 of the information processing system 230 may be received by the user terminal 210 via the communication module 316 of the user terminal 210 via the communication module 336 and the network 220.

[0066] The input / output interface 318 may be a means for interfacing with the input / output device 320. As an example, the input device may include a device such as a camera including an audio sensor and / or an image sensor, a keyboard, a microphone, a mouse, etc., and the output device may include a device such as a display, a speaker, a haptic feedback device, etc. As another example, the input / output interface 318 may be a means for interfacing with a device having a configuration or function for performing input and output integrated into one, such as a touch screen. For example, when the processor 314 of the user terminal 210 processes an instruction of a computer program loaded in the memory 312, a service screen configured using information and / or data provided by the information processing system 230 or another user terminal may be displayed on the display via the input / output interface 318. In FIG. 3, the input / output device 320 is illustrated as not being included in the user terminal 210, but is not limited thereto, and may be configured integrally with the user terminal 210. Furthermore, the input / output interface 338 of the information processing system 230 may be a means for interfacing with a device (not shown) for input or output that may be connected to the information processing system 230 or may be included in the information processing system 230. In FIG. 3, the input / output interfaces 318, 338 are shown as elements configured separately from the processors 314, 334, but are not limited thereto, and the input / output interfaces 318, 338 may be configured to be included in the processors 314, 334.

[0067] The user terminal 210 and the information processing system 230 may include more components than those shown in FIG. 3. However, it is not necessary to explicitly show most of the conventional technical components. According to an embodiment, the user terminal 210 may be embodied to include at least a part of the above-mentioned input / output device 320. The user terminal 210 may further include other components such as a transceiver, a global positioning system (GPS) module, a camera, various sensors, and a database. For example, if the user terminal 210 is a smartphone, it may include components that are generally included in a smartphone, and may be embodied to further include various components such as an acceleration sensor, a gyro sensor, an image sensor, a proximity sensor, a touch sensor, an illuminance sensor, a camera module, various physical buttons, a button using a touch panel, an input / output port, and a vibrator for vibration.

[0068] When a program for a physiological information estimation service application, a bladder urine volume prediction service application, etc. is running, the processor 314 can receive text, images, videos, voice, and / or actions, etc. input or selected by an input device such as a touch screen, keyboard, camera including an audio sensor and / or image sensor, microphone, etc. connected to the input / output interface 318, and can store the received text, images, videos, voice, and / or actions, etc. in the memory 312 or provide them to the information processing system 230 via the communication module 316 and the network 220.

[0069] The processor 314 of the user terminal 210 may be configured to manage, process, and / or store information and / or data received from the input / output device 320, other user terminals, the information processing system 230, and / or multiple external systems. The information and / or data processed by the processor 314 may be provided to the information processing system 230 via the communication module 316 and the network 220. The processor 314 of the user terminal 210 may transfer and output information and / or data to the input / output device 320 via the input / output interface 318. For example, the processor 314 may display the received information and / or data on a screen of the user terminal 210.

[0070] The processor 334 of the information processing system 230 may be configured to manage, process and / or store information and / or data received from the plurality of user terminals 210 and / or the plurality of external systems. The information and / or data processed by the processor 334 may be provided to the user terminal 210 via the communication module 336 and the network 220.

[0071] JPEG2025080243000004.jpg99151

[0072] In one embodiment, the light source 420 may be one of the first to twenty-fourth light sources of the medical device 100 shown in Fig. 1. The first to third photodiodes 430_1, 430_2, 430_3 may be part of the multiple photodiodes 112_1 to 112_20 of the medical device 100 shown in Fig. 1. That is, from the explanation of the example of the light source 420, the first photodiode 430_1, the second photodiode 430_2, and the third photodiode 430_3, the calculation process of the normalized diffuse reflectance based on the light data detected by the multiple photodiodes can be understood.

[0073] JPEG2025080243000005.jpg127151

[0074] In one embodiment, the measured voltage value can be corrected using a calibration parameter. The corrected voltage value can be expressed as follows:

[0075]

number

[0076] In one embodiment, the system parameters may be different for each photodiode. Specifically, the proportional coefficients of the system parameters for each photodiode may be different due to the influence of manufacturing process errors, connected circuit devices, etc. As shown in Equation 3, the system parameters of each photodiode may be calibrated to be the same using the calibration parameters.

[0077] The normalized diffuse reflectance can mean a relative relationship of the diffuse reflectance of another photodiode based on the diffuse reflectance of a specific photodiode. Here, the normalized diffuse reflectance can be calculated based on a corrected voltage value. Specifically, the normalized diffuse reflectance can be understood by the following formula 4. Here, formula 4 can be derived from formulas 1, 2, and 3.

[0078]

number

[0079] JPEG2025080243000008.jpg54151

[0080] A process of calculating normalized diffuse reflectances of the light source 420 and the first to third photodiodes 430_1, 430_2, and 430_3 has been described in detail with reference to Fig. 4. A method of estimating physiological information based on the normalized diffuse reflectances of the light source 420 and the first to third photodiodes 430_1, 430_2, and 430_3 will be described in detail below with reference to Fig. 5.

[0081] FIG. 5 is a diagram illustrating an example of a process of estimating physiological information according to an embodiment of the present disclosure. The first optical data 510_1 may be data generated by a first photodiode. The second optical data 510_2 may be data generated by a second photodiode. The third optical data 510_3 may be data generated by a third photodiode. When measuring a voltage value corresponding to the light intensity detected by each photodiode, the multiple optical data 510_1, 510_2, 510_3 may be measured voltage values. For example, the first optical data 510_1 may be a measured voltage value of the first photodiode described above in FIG. 4. Similarly, the second optical data 510_2 may be a measured voltage value of the second photodiode described above in FIG. 4. Also, the third optical data 510_3 may be a measured voltage value of the third photodiode described above in FIG. 4.

[0082] In one embodiment, the correction unit 520 may calculate a plurality of corrected optical data 522_1, 522_2, 522_3 using the calibration parameters 512 based on the plurality of optical data 510_1, 510_2, 510_3. Specifically, the plurality of corrected optical data 522_1, 522_2, 522_3 may be calculated by correcting each of the plurality of optical data 510_1, 510_2, 510_3. For example, the corrected first optical data 522_1 may be a corrected voltage value of the first photodiode described above in FIG. 4. Similarly, the corrected second optical data 522_2 may be a corrected voltage value of the second photodiode described above in FIG. 4. Also, the corrected third optical data 522_3 may be a corrected voltage value of the third photodiode described above in FIG. 4. The process of correcting optical data using the calibration parameters 512 can be understood from the contents described above in FIG. 4.

[0083] In one embodiment, the diffuse reflectance calculation unit 530 can calculate a plurality of normalized diffuse reflectances 532_1, 532_2 based on a plurality of corrected optical data 522_1, 522_2, 522_3. For example, the normalized diffuse reflectance 532_1 of the second photodiode can be calculated based on the corrected first optical data 522_1 and the corrected second optical data 522_2. Similarly, the normalized diffuse reflectance 532_2 of the third photodiode can be calculated based on the corrected first optical data 522_1 and the corrected third optical data 522_3. The process of calculating the normalized diffuse reflectance can be understood from the content described above in FIG. 4.

[0084] In one embodiment, the light absorption coefficient and the light scattering coefficient can be estimated based on a plurality of normalized diffuse reflectances 532_1, 532_2. Here, the light absorption coefficient may be an optical coefficient of biological tissue for analyzing physiological components of the biological tissue according to the degree to which light is absorbed by the biological tissue for each wavelength. Also, the light scattering coefficient may be an optical coefficient indicating a structural characteristic of the biological tissue. For example, the adipose tissue of an obese patient having large fat cells scatters light relatively little, and the adipose tissue of a normal weight patient having small fat cells scatters light relatively much. As shown in the figure, an initial light characteristic value estimation model 540 and / or a numerical solver 550 can be used to estimate the light absorption coefficient and the light scattering coefficient.

[0085] In one embodiment, the initial light characteristic value estimation model 540 can estimate an initial light characteristic value for a specific region based on a plurality of normalized diffuse reflectances 532_1 and 532_2. Here, the specific region may be a body part associated with the second photodiode and the third photodiode. The initial light characteristic value may include an initial light scattering coefficient 542 and an initial light absorption coefficient 544. For example, the initial light characteristic value estimation model 540 may be an artificial neural network model (e.g., a deep learning-based model) that learns a plurality of light characteristic values ​​and normalized theoretical diffuse reflectances associated with the plurality of light characteristic values. The learning process of the initial light characteristic value estimation model 540 will be described in detail below with reference to FIG. 9.

[0086] In one embodiment, a numerical solver 550 can estimate a final light characteristic value based on an initial light characteristic value. In this case, the final light characteristic value can include a final light scattering coefficient 554 and a final light absorption coefficient 556. As an example, the numerical solver 550 can use the Levenberg-Marquardt algorithm. Specifically, the numerical solver 550 can estimate a final light characteristic value based on a theoretical formula 552 for diffuse reflectance by inputting an initial light characteristic value and a plurality of normalized diffuse reflectances 532_1 and 532_2 as initial values.

[0087] Here, the theoretical formula 552 for diffuse reflectance is as shown in the following formula 5.

[0088]

number

[0089] In one embodiment, the numerical solver 550 can receive an initial light characteristic value for the specific region and a plurality of normalized diffuse reflectances 532_1 and 532_2 as initial values, and estimate a final light characteristic value for the specific region based on a theoretical diffuse reflectance equation 552. Specifically, an initial light scattering coefficient 542 for the specific region, an initial light absorption coefficient 544 for the specific region, a normalized diffuse reflectance 532_1 of the second photodiode, and a normalized diffuse reflectance 532_2 of the third photodiode are input to the numerical solver 550 as a set, and a final light scattering coefficient 554 for the specific region and a final light absorption coefficient 556 for the specific region can be estimated.

[0090] In one embodiment, the physiological information estimator 560 can estimate physiological information 564 of a specific region based on the final light characteristic value. Specifically, the physiological information 564 can be estimated based on an extinction coefficient 562, a final light scattering coefficient 554, and a final light absorption coefficient 556. For example, the extinction coefficient 562 can be expressed as an extinction coefficient matrix as shown in Table 1 below.

[0091] [Table 1]

[0092] JPEG2025080243000011.jpg67151

[0093] In one embodiment, a pseudo inverse matrix of extinction coefficient matrix as shown in Table 1 can be calculated. Physiological information 564 based on the optical absorption coefficient for each wavelength can be calculated using the inverse matrix of the extinction coefficient matrix. Specifically, a process of calculating physiological information using the following Equation 6 in which the optical absorption coefficient for each wavelength is multiplied by the inverse matrix of the extinction coefficient matrix will be examined.

[0094]

number

[0095] JPEG2025080243000013.jpg67155

[0096] FIG. 5 shows an example of estimating physiological information 564 using light emitted from one light source, but is not limited thereto. For example, light is irradiated onto the body using a group of light sources having different wavelengths, and a group of photodiodes can detect the light intensity of the diffused light. Specifically, a group of light sources including a light source irradiating light of a first wavelength, a light source irradiating light of a second wavelength, a light source irradiating light of a third wavelength, a light source irradiating light of a fourth wavelength, a light source irradiating light of a fifth wavelength, and a light source irradiating light of a sixth wavelength is used. At this time, six final optical absorption coefficients 556 for a specific region can be estimated based on a plurality of optical data associated with the six wavelengths of light. Then, based on the six final optical absorption coefficients 556 for the specific region, the content of oxygenated hemoglobin, the content of deoxygenated hemoglobin, the content of water, and the content of fat for the specific region can be estimated using the inverse matrix of the absorption coefficient matrix.

[0097] In summary, one optical absorption coefficient for a specific region can be estimated based on three optical data. When six different wavelengths of light are used, six optical absorption coefficients for a specific region can be estimated based on three optical data associated with each wavelength, and four content information (oxygenated hemoglobin content, deoxygenated hemoglobin content, water content, and fat content) for the specific region can be estimated based on the six optical absorption coefficients.

[0098] 5 shows an example in which physiological information 564 is estimated using optical data detected by three photodiodes, but the present invention is not limited to this. For example, a number of photodiodes (e.g., 20) more than three photodiodes are used. In this case, physiological information 564 relating to a plurality of regions can be estimated.

[0099] The physiological information estimation process for the example of the medical device 100 shown in Fig. 1 can be understood from the method described in Fig. 4 and Fig. 5. The flow of the optical data structure for the example of the medical device 100 shown in Fig. 1 will be described in detail later with reference to Figs. 10 to 12.

[0100] 6 is a diagram showing an example of generating calibration parameters using a calibration box 610 according to an embodiment of the present disclosure. In an embodiment, a plurality of photodiode openings 612_1 to 612_20 and a plurality of light source group openings 614_1 to 614_4 may be formed on one surface of the calibration box 610. Each of the plurality of photodiode openings 612_1 to 612_20 may correspond to the positions of the plurality of photodiodes 112_1 to 112_20 included in the medical device 100 shown in FIG. 1. Also, each of the plurality of light source group openings 614_1 to 614_4 may correspond to the positions of the plurality of light source groups 114_1 to 114_4 included in the medical device 100 shown in FIG. 1. At this time, the medical device 100 can be placed in the calibration box 610 so that one side of the medical device 100 on which the multiple photodiodes 112_1 to 112_20 and the multiple light source groups 114_1 to 114_4 are arranged faces one side of the calibration box 610 on which the multiple openings 612_1 to 612_20 and 614_1 to 614_4 are formed.

[0101] The calibration box 610 is shown to have 20 photodiode openings 612_1 to 612_20 and four light source group openings 614_1 to 614_4, but is not limited thereto. That is, the number of openings formed in the calibration box 610 can be changed depending on the number of photodiodes and the number of light source groups included in the medical device 100 shown in FIG.

[0102] In one embodiment, the calibration box 610 may include a standard reflective object therein. The standard reflective object may diffuse (and / or reflect, hereinafter referred to as "diffusion") the irradiated light. In addition, optical information (e.g., diffuse reflectance by wavelength, etc.) regarding the standard reflective object may be predefined.

[0103] In one embodiment, before generating the calibration parameters, a look-up table (LUT) can be generated using the calibration box 610. In this case, the LUT can include relative relationship information regarding the light intensity of the diffused light between the multiple photodiode apertures 612_1 to 612_20. Specifically, the LUT can include information regarding the ratio of the light intensity of the diffused light that reaches each photodiode aperture.

[0104] As an example, the information contained in the LUT can be generated as follows: A first light source can be positioned in the first light source group aperture 614_1. Then, a specific photodiode can be positioned in the first photodiode aperture 612_1. Here, in order to eliminate the effect of the intercept coefficient of the system parameters on the specific photodiode, an offset can be set so that when no light is detected at the specific photodiode, the measured voltage value of the specific photodiode is zero.

[0105] JPEG2025080243000014.jpg96151

[0106] JPEG2025080243000015.jpg64151

[0107] [Table 2]

[0108] JPEG2025080243000017.jpg51151

[0109] In one embodiment, the LUT may store light intensity ratio information for each position of diffuse light for each wavelength of irradiated light in the form of a table. In this case, the LUT may be divided into a table associated with a first wavelength (e.g., a first light source) and a table associated with a second wavelength (e.g., a second light source) and stored. For example, when light of six different wavelengths (e.g., first to sixth light sources) is used, six tables may be generated for the six wavelengths, and 20 pieces of light intensity ratio information of diffuse light may be generated per table.

[0110] As can be seen from Equation 1, the light intensity of the diffuse light reaching each photodiode may be proportional to the light intensity of the light source. In addition, the light intensity of the diffuse light reaching other photodiodes may be proportional to the light intensity of the diffuse light reaching a specific photodiode using the information included in the LUT. In summary, the light intensity information of the diffuse light reaching each photodiode may be generated by correcting the light intensity information of the light source using the information included in the LUT. An example of a process of correcting the light intensity information of the light source using the information included in the LUT will be described in detail below with reference to FIG. 7.

[0111] In one embodiment, a calibration parameter can be generated based on the light intensity information of the light source corrected using the LUT. Specifically, a calibration parameter can be generated for each of the plurality of photodiodes. Here, the calibration parameter can include a proportionality coefficient and an intercept coefficient.

[0112] The process of generating the calibration parameters will be described based on the multiple photodiodes 112_1-112_20 and the multiple light source groups 114_1-114_4 shown in Fig. 1. The medical device 100 can be placed in a calibration box 610 so that one side of the medical device 100 on which the multiple photodiodes 112_1-112_20 and the multiple light source groups 114_1-114_4 are arranged faces one side of the calibration box 610 on which the multiple openings 612_1-612_20 and 614_1-614_4 are formed.

[0113] The first detection process may include a process in which the first photodiode detects diffuse light when the standard reflective object is irradiated with light from the first light source. At this time, the first light source included in the first light source group may be preferentially irradiated, but is not limited thereto, and one of the second to twenty-fourth light sources may be preferentially irradiated. The first detection process may include a process in which the second photodiode 112_2 detects diffuse light when the standard reflective object is irradiated with light from the first light source. That is, the first detection process may include a process in which all of the multiple photodiodes 112_1 to 112_20 detect diffuse light when light is irradiated by the first light source. Here, the light intensity of the first light source may be constant during the first detection process. The second detection process is similar to the first detection process, except that the light intensity of the first light source is changed in the first detection process. Similarly, the third detection process to the n-th detection process in which the light intensity of the first light source is changed may be continuously performed. Such a detection process may be performed several tens of times. As an example, the light intensity of the first light source can be continuously increased or decreased as the detection process progresses.

[0114] Through the continuous detection process, a measurement value graph of the light intensity of the light source can be generated for each photodiode. At this time, using the LUT information, a measurement value graph of the light intensity of the diffuse light can be generated for each photodiode. Then, based on the measurement value graph of the light intensity of the diffuse light, a trendline can be generated for each photodiode. Then, based on the equation of the generated trendline, calibration parameters can be generated. The process of generating the trendline will be described in detail in FIG. 7, and the process of generating the calibration parameters based on the equation of the trendline will be described in detail in FIG. 8.

[0115] 7 is a graph illustrating an example of a process for generating calibration parameters according to an embodiment of the present disclosure. For convenience of explanation, in FIG. 7, the explanation focuses on two photodiodes shown in FIG. 1, that is, the first photodiode 112_1 and the third photodiode 112_3.

[0116] JPEG2025080243000018.jpg74151

[0117] 6, the light intensity of the diffuse light reaching a particular photodiode may be expressed as proportional to the light intensity of the light source. For example, in the first detection process, if the light intensity of the light source is 16 mW, the relative value of the light intensity of the diffuse light reaching the third photodiode 112_3 may be 8 a.u.

[0118] JPEG2025080243000019.jpg93151

[0119] The second graph 720 is a graph of measurement values ​​based on the light intensity of the diffused light obtained by the first detection process to the fifth detection process. In the illustrated example, the light intensity of the light source may decrease as the detection process progresses. As a result, the light intensity of the diffused light reaching the third photodiode 112_3 may also decrease in proportion to the light intensity of the light source. Moreover, the light intensity of the diffused light reaching the first photodiode 112_1 may also decrease in proportion to the light intensity of the diffused light reaching the third photodiode 112_3.

[0120] JPEG2025080243000020.jpg64151

[0121] Similar to the above description of the trend lines of the first and third photodiodes, trend lines of the multiple photodiodes can be generated. Then, calibration parameters for the multiple photodiodes can be generated based on the equation of the generated trend lines. A specific process of generating the calibration parameters will be described in detail below with reference to FIG. 8.

[0122] 8 is a diagram illustrating an example of applying the calibration parameters according to an embodiment of the present disclosure. A first graph 810 is a graph showing trend lines 812, 814, 816, 818 of multiple photodiodes before using the calibration parameters. The x-axis is the light intensity of the diffuse light, and the y-axis is the measured voltage value. Using the first graph 810, the calibration parameters can be generated according to the following formula (7).

[0123]

number

[0124] JPEG2025080243000022.jpg51151

[0125] JPEG2025080243000023.jpg77151

[0126] JPEG2025080243000024.jpg51151

[0127] In one example, when light of six different wavelengths is used, each photodiode can generate six trend lines associated with the light of each wavelength. In this case, a total of six proportional coefficients of the calibration parameters of the six trend lines can be generated, one for each trend line. In addition, the proportional coefficients and intercept coefficients of the calibration parameters can be generated separately for each photodiode. That is, when light of six different wavelengths and 20 photodiodes are used, 120 proportional coefficients of the calibration parameters and 20 intercept coefficients of the calibration parameters can be generated.

[0128] JPEG2025080243000025.jpg80151

[0129] With such a configuration, system parameters for a plurality of photodiodes included in a medical device can be corrected identically. Also, as shown in FIG. 5, a normalized diffuse reflectance can be calculated based on the identically corrected system parameters, and physiological information can be estimated based on the normalized diffuse reflectance. Thus, the medical device according to the invention of the present disclosure may become unnecessary for additional calibration after performing the generation of calibration parameters once. That is, the invention according to the present disclosure can enhance the convenience for the user by eliminating the need for calibration using a phantom.

[0130] JPEG2025080243000026.jpg64151

[0131] In one embodiment, the normalized theoretical diffuse reflectance calculation unit 920 can calculate a first normalized theoretical diffuse reflectance 922_1 and a second normalized theoretical diffuse reflectance 922_2 based on an arbitrary light scattering coefficient 912 and an arbitrary light absorption coefficient 914. Specifically, the normalized theoretical diffuse reflectance calculation unit 920 can calculate a normalized theoretical diffuse reflectance from an arbitrary light scattering coefficient and an arbitrary light absorption coefficient using a diffuse reflectance theoretical formula. The diffuse reflectance theoretical formula can be expressed as the formula of Equation 5 above.

[0132] The normalized theoretical diffuse reflectance calculation unit 920 can calculate a pair of normalized theoretical diffuse reflectances based on a pair of optical characteristic values. For example, the pair of optical characteristic values can include an arbitrary light scattering coefficient 912 and an arbitrary light absorption coefficient 914. Also, the pair of normalized theoretical diffuse reflectances can include a first normalized theoretical diffuse reflectance 922_1 and a second normalized theoretical diffuse reflectance 922_2. At this time, one piece of learning data can include a pair of optical characteristic values and a pair of normalized theoretical diffuse reflectances.

[0133] In one embodiment, the normalized theoretical diffuse reflectance calculation process may be the same as the following process. As shown in FIG. 5, the theoretical formula of diffuse reflectance may be an equation related to the light scattering coefficient, the light absorption coefficient, and the distance between the light source and the photodiode. That is, the first theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the first distance between the light source and the first photodiode. Similarly, the second theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the second distance between the light source and the second photodiode. Also, the third theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the third distance between the light source and the third photodiode. Here, the first normalized theoretical diffuse reflectance 922_1 may be a value obtained by dividing the second theoretical diffuse reflectance by the first theoretical diffuse reflectance. Furthermore, the second normalized theoretical diffuse reflectance 922_2 may be a value obtained by dividing the third theoretical diffuse reflectance by the first theoretical diffuse reflectance. At this time, information on the first distance, the second distance, and the third distance may be input in advance to the normalized theoretical diffuse reflectance calculation unit 920.

[0134] A plurality of arbitrary light scattering coefficients 912 and arbitrary light absorption coefficients 914 may be generated. A plurality of normalized theoretical diffuse reflectance pairs may be generated based on each of the plurality of pairs of light characterization coefficients. A first set of training data may be generated based on the plurality of light characterization coefficients and the plurality of pairs of normalized theoretical diffuse reflectances corresponding to the light characterization coefficient pairs. As an example, the first set of training data may include 20,000,000 light characterization coefficient pairs and normalized theoretical diffuse reflectance pairs.

[0135] In one embodiment, the first initial light characteristic value estimation model 930_1 may be a deep learning-based model or a machine learning-based model trained using the first set of training data. Here, the machine learning-based model may be one of KNN (K-Nearest Neighbors), GB (Gradient Boost), and ANN (Artificial Neural Network). The first initial light characteristic value estimation model 930_1 trained based on the first set of training data may estimate initial light characteristic values ​​(light scattering coefficient and light absorption coefficient) based on the normalized diffuse reflectance of the second photodiode and the normalized diffuse reflectance of the third photodiode.

[0136] In one embodiment, the fourth to sixth distances may be distances between the light source and the fourth to sixth photodiodes, respectively. Information on the fourth distance, the fifth distance, and the sixth distance may be input in advance to the normalized theoretical diffuse reflectance calculation unit 920. At this time, the fourth distance, the fifth distance, and the sixth distance may be different from the first distance, the second distance, and the third distance, respectively. At this time, the fourth theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the fourth distance between the light source and the fourth photodiode. Also, the fifth theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the fifth distance between the light source and the fifth photodiode. Additionally, the sixth theoretical diffuse reflectance may be calculated based on the arbitrary light scattering coefficient 912, the arbitrary light absorption coefficient 914, and the sixth distance between the light source and the sixth photodiode. Here, the third normalized theoretical diffuse reflectance may be a value obtained by dividing the fifth theoretical diffuse reflectance by the fourth theoretical diffuse reflectance. Moreover, the fourth normalized theoretical diffuse reflectance 922_2 may be a value obtained by dividing the sixth theoretical diffuse reflectance by the fourth theoretical diffuse reflectance. The plurality of training data generated by repeating such a process may be a second set of training data. The second initial light characteristic value estimation model 930_2 may be trained based on the second set of training data.

[0137] For the seventh to ninth distances, by repeating the above process, a third set of training data can be generated. At this time, the third initial light characteristic value estimation model 930_3 can be trained based on the third set of training data. For the tenth to twelfth distances, by repeating the above process, a fourth set of training data can be generated. At this time, the fourth initial light characteristic value estimation model 930_4 can be trained based on the fourth set of training data. In FIG. 9, four initial light characteristic value estimation models 930_1 to 930_4 are shown being generated, but it is not limited thereto, and any number of initial light characteristic value estimation models can be generated depending on the number of photodiodes and the arrangement of the photodiodes and the light source.

[0138] When estimating the initial light characteristic values based on the normalized diffuse reflectance related to the first, second, and third distances, the first initial light characteristic value estimation model 930_1 is used. Similarly, when estimating the initial light characteristic values based on the normalized diffuse reflectance related to the fourth, fifth, and sixth distances, the second initial light characteristic value estimation model 930_2 is used. When estimating the initial light characteristic values based on the normalized diffuse reflectance related to the seventh, eighth, and ninth distances, the third initial light characteristic value estimation model 930_3 is used. Similarly, when estimating the initial light characteristic values based on the normalized diffuse reflectance related to the tenth, eleventh, and twelfth distances, the fourth initial light characteristic value estimation model 930_4 is used.

[0139] When the above-described numerical solver inputs arbitrary light scattering coefficients 912 and arbitrary light absorption coefficients 914 as initial values and estimates the final light scattering coefficient and the final light absorption coefficient, there are drawbacks such as long calculation time and low accuracy. When estimating the initial light scattering coefficient and the initial light absorption coefficient using the initial light characteristic value estimation model 930 of the invention according to the present disclosure and then estimating the final light scattering coefficient and the final light absorption coefficient using the numerical solver, the calculation time is short and the accuracy can also be improved. Thus, the invention according to the present disclosure has the advantage of being able to enhance the convenience of the user by providing rapid calculation and high accuracy.

[0140] FIG. 10 is a diagram showing an example of a medical device according to an embodiment of the present disclosure. As shown in the figure, a plurality of photodiodes and a plurality of light source groups 1012, 1014, 1022, 1024 can be arranged on one surface of the medical device. An example shown in the figure may be the same as the medical device shown in FIG. 1. The first light source group 1012 can include six light sources. The six light sources can irradiate light with different wavelengths from each other. The second to fourth light source groups 1014, 1022, 1024 can also include six light sources configured to irradiate light with different wavelengths from each other.

[0141] In one embodiment, the medical device can be a device on which calibration has been performed. For example, calibration for the medical device can be performed during the manufacturing process. Also, the medical device is used by being attached to the body. The process in which the medical device attached to the body detects optical data will be described in detail later.

[0142] In one embodiment, the first state 1010 can represent the detection relationship between the first light source group 1012 and the third light source group 1014 and the first set of photodiodes 1016. In the first measurement process, each of the first set of photodiodes 1016 can detect diffused light in a state where the first light source included in the first light source group 1012 irradiates light. At this time, the light intensity of the first light source can be constant during the first measurement process. After the first measurement process is performed, the first set of photodiodes 1016 can detect optical data (12 measurement voltage values) related to the first light source. Each of the second to sixth measurement processes is the same as the first measurement process, except that the second to sixth light sources included in the first light source group 1012 are used instead of the first light source in the first measurement process. Similarly, the thirteenth to eighteenth measurement processes related to the thirteenth to eighteenth light sources included in the third light source group 1014 can be performed.

[0143] In one embodiment, the second state 1020 may represent a detection relationship between the second light source group 1022 and the fourth light source group 1024 and the second set of photodiodes 1026. In the seventh measurement process, in a state in which the seventh light source included in the second light source group 1022 irradiates light, each of the second set of photodiodes 1026 may detect diffuse light. At this time, the light intensity of the seventh light source may be constant during the seventh measurement process. After the seventh measurement process is performed, the second set of photodiodes 1026 may detect optical data (12 measured voltage values) associated with the seventh light source. Each of the eighth to twelfth measurement processes is similar to the seventh measurement process, except that the eighth to twelfth light sources included in the second light source group 1022 are used instead of the seventh light source in the seventh measurement process. Similarly, the nineteenth to twenty-fourth measurement processes associated with the nineteenth to twenty-fourth light sources included in the fourth light source group 1024 may be performed.

[0144] The process of estimating the physiological information based on the optical data acquired by the first to twenty-fourth measurement processes will be described in detail later with reference to Figures 11 and 12. The physiological information estimated based on the optical data can include oxygenated hemoglobin content information, deoxygenated hemoglobin content information, water content information, fat content information, and bladder urine volume.

[0145] FIG. 11 is a diagram showing an example of estimating physiological information based on a plurality of optical data according to an embodiment of the present disclosure. In FIG. 11, the optical data (measured voltage value) detected by the first measurement process, the thirteenth measurement process, the seventh measurement process, and the nineteenth measurement process associated with the first wavelength (i.e., the first light source, the seventh light source, the thirteenth light source, and the nineteenth light source) among the first to twenty-fourth measurement processes described above will be mainly described. Here, the measured voltage value may mean a voltage value corrected using a calibration parameter. The optical data associated with the second to sixth wavelengths may be processed in the same manner as the optical data associated with the first wavelength.

[0146] JPEG2025080243000027.jpg54151

[0147] As described above in FIG. 4, the normalized diffuse reflectance can be calculated by dividing the voltage value of each photodiode by the reference photodiode voltage value. As shown, in the light data associated with the first light source and the thirteenth light source, the fifth and fifteenth photodiodes closest to the first and thirteenth light sources can be selected as the reference photodiodes. For example, the fifth photodiode in the first row of the plurality of photodiodes associated with the first light source and the fifteenth photodiode in the second row can be selected as the reference photodiode. As shown, in the light data associated with the seventh and nineteenth light sources, the sixth and sixteenth photodiodes closest to the seventh and nineteenth light sources can be selected as the reference photodiodes. For example, the sixth photodiode in the first row of the plurality of photodiodes associated with the seventh light source and the sixteenth photodiode in the second row can be selected as the reference photodiode.

[0148] JPEG2025080243000028.jpg45151

[0149] As shown, a normalized diffuse reflectance 1122 of the first photodiode for the seventh light source may be generated based on a measured voltage value 1112 of the first photodiode for the seventh light source and a measured voltage value 1114 of the sixth photodiode for the seventh light source, where the sixth photodiode may be a reference photodiode. Also, a normalized diffuse reflectance 1128 of the seventh photodiode for the thirteenth light source may be generated based on a measured voltage value 1116 of the fifth photodiode for the thirteenth light source and a measured voltage value 1118 of the seventh photodiode for the thirteenth light source, where the fifth photodiode may be a reference photodiode.

[0150] In one embodiment, the normalized diffuse reflectance data map 1120 may have a smaller number of data than the measured voltage data map 1110. Specifically, the normalized diffuse reflectance for a specific reference photodiode is not calculated. For example, the normalized diffuse reflectance for the measured voltage value 1116 of the fifth photodiode for the thirteenth light source is not calculated. Here, the fifth photodiode may be a reference photodiode. In the illustrated example, if the number of data in the measured voltage data map 1110 is 48 (4×12) and there are 8 measured voltage values ​​corresponding to the reference photodiode, the number of normalized diffuse reflectance data may be 40.

[0151] The light scattering coefficient can be associated with a particular region, where the light scattering coefficient can be a final light scattering coefficient. As an example, the first region can be a body part associated with the first photodiode and the second photodiode. Similarly, the nth region can be a body part associated with the nth photodiode and the n+1th photodiode.

[0152] JPEG2025080243000029.jpg42151

[0153] As an example, the light scattering coefficient of the nth region for the yth light source may be estimated based on the normalized diffuse reflectance of the nth photodiode for the yth light source and the normalized diffuse reflectance of the n+1th photodiode for the yth light source. For example, the light scattering coefficient 1132 of the first region for the seventh light source may be estimated based on the normalized diffuse reflectance 1122 of the first photodiode for the seventh light source and the normalized diffuse reflectance 1124 of the second photodiode for the seventh light source. As yet another example, the light scattering coefficient 1134 of the sixth region for the thirteenth light source may be estimated based on the normalized diffuse reflectance 1126 of the sixth photodiode for the thirteenth light source and the normalized diffuse reflectance 1128 of the seventh photodiode for the thirteenth light source.

[0154] Similarly, an optical absorption coefficient data map may be calculated based on the normalized diffuse reflectance data map 1120, where the optical absorption coefficient may be the final optical absorption coefficient. Since the optical absorption coefficient is part of the physiological information, the optical absorption coefficient data map is used as the physiological information.

[0155] In one embodiment, the light scattering coefficient data map 1130 may have a smaller number of data items than the normalized diffuse reflectance data map 1120. Specifically, one light scattering coefficient may be estimated based on two normalized diffuse reflectances. For example, if the number of normalized diffuse reflectance data map 1120 is 40 (4×10), the number of light scattering coefficient data map 1130 may be 32 (4×8). Similarly, the light absorption coefficient data map may have a smaller number of data items than the normalized diffuse reflectance data map 1120.

[0156] After the first to twenty-fourth measurement processes are performed based on the above, a light scattering coefficient data map and a light absorption coefficient data map can be calculated for the plurality of detected light data. In one embodiment, the light scattering coefficient data map and the light absorption coefficient data map can be generated for each wavelength. For example, after performing the second measurement process, the eighth measurement process, the fourteenth measurement process, and the twentieth measurement process related to the second wavelength, a light scattering coefficient data map and a light absorption coefficient data map for the second wavelength can be calculated for the plurality of detected light data. Similarly, after performing a plurality of measurement processes related to the nth wavelength, a light scattering coefficient data map and a light absorption coefficient data map for the nth wavelength can be calculated for the plurality of detected light data. The process of calculating a physiological information data map based on the plurality of light scattering coefficient data maps will be described in detail below with reference to FIG. 12.

[0157] FIG. 12 is a diagram illustrating an example of estimating physiological information based on a plurality of light scattering coefficient data maps according to an embodiment of the present disclosure. The plurality of light scattering coefficient data maps 1210 may include light scattering coefficient data maps for each wavelength. In one embodiment, the light scattering coefficient data map 1212_1 for the first wavelength may include all light scattering coefficients estimated based on a plurality of light data detected by performing a plurality of measurement processes associated with the first wavelength. Similarly, each of the light scattering coefficient data maps 1212_2 to 1212_6 for the second to sixth wavelengths is similar to the light scattering coefficient data map 1212_1 for the first wavelength, except that the light scattering coefficient data map 1212_1 for the first wavelength is based on a plurality of light data detected by performing a plurality of measurement processes associated with each of the second to sixth wavelengths instead of the first wavelength. Here, the light scattering coefficient may mean a final light scattering coefficient. In the illustrated example, the light scattering coefficient of the xth region for the yth light source may be represented by x, y.

[0158] In one embodiment, the physiological information data map can be calculated based on a plurality of light scattering coefficient data maps 1210. Specifically, the physiological information can be estimated based on a plurality of light scattering coefficients in a particular region for light sources associated with different wavelengths. For a method of estimating physiological information based on the light scattering coefficients, see the description of FIG. 5.

[0159] For example, physiological information on the xth region can be estimated based on the light scattering coefficients of [(x, y), (x, y+1), (x, y+2), (x, y+3), (x, y+4), (x, y+5)]. Here, (x, y) may be the light scattering coefficient of the xth region for the yth light source included in the light scattering coefficient data map 1212_1 for the first wavelength. Specifically, physiological information on the eighth region (e.g., oxygenated hemoglobin content information, deoxygenated hemoglobin content information, water content information, fat content information, etc.) can be estimated based on the light scattering coefficient of the eighth region for the first light source, the light scattering coefficient of the eighth region for the second light source, the light scattering coefficient of the eighth region for the third light source, the light scattering coefficient of the eighth region for the fourth light source, the light scattering coefficient of the eighth region for the fifth light source, and the light scattering coefficient of the eighth region for the sixth light source. Here, each of the first to sixth light sources can emit light of the first to sixth wavelengths, respectively.

[0160] In one embodiment, the plurality of physiological information data maps may include a first physiological information data map 1220, a second physiological information data map 1230, a third physiological information data map 1240, and a fourth physiological information data map 1250. Here, each of the first to fourth physiological information data maps 1220, 1230, 1240, and 1250 includes a first physiological information data map 1220, a second physiological information data map 1230, a third physiological information data map 1240, and a fourth physiological information data map 1250. 2 ) data map, deoxygenated hemoglobin (HHb) data map, water (H 2 O) data map, and Fat data map.

[0161] In one embodiment, the plurality of physiological information data maps 1220, 1230, 1240, and 1250 may have a smaller number of data than the plurality of light scattering coefficient data maps 1210. Specifically, four pieces of physiological information may be estimated based on six pieces of data included in the plurality of physiological information data maps. For example, each of the plurality of light scattering coefficient data maps 1210 may include 32 (4×8) pieces of data, and thus the plurality of light scattering coefficient data maps 1210 may include 192 (4×8×6) pieces of data. In this case, the plurality of physiological information data maps 1220, 1230, 1240, and 1250 may include 128 (4×8×4) pieces of data.

[0162] Using multiple light sources and multiple photodiodes, physiological information on multiple regions can be provided. The invention according to the present disclosure can provide physiological information on not only localized regions of the body but also wide areas of the body. Furthermore, by providing physiological information on multiple regions, the state of the organs contained in the body (e.g., the amount of urine stored in the bladder, the position of the bladder, etc.) can be specifically understood. In the case of a patient who does not feel the need to urinate, the invention according to the present disclosure can be used to receive physiological information on his / her bladder in real time or periodically. The patient can monitor the amount of urine stored in his / her bladder using the provided information and urinate at the appropriate time.

[0163] Fig. 13 is a block diagram showing an example of a method for estimating a urine volume in a bladder according to an embodiment of the present disclosure. In the contents described below with reference to Figs. 13 to 16, a "measurement cycle" can refer to a series of steps for detecting an optical data set using a medical device (e.g., the medical device 100 described with reference to Fig. 1) placed on the skin located on the bladder of a specific user. That is, the measurement cycle can include the first to twenty-fourth measurement steps shown in Fig. 10. The detailed steps of the measurement cycle can be understood from the contents described above with reference to Figs. 1 and 10.

[0164] In one embodiment, the processor can receive a light data set 1302 associated with a particular user by performing a measurement cycle. For example, the light data set 1302 can include the measured voltage data map 1110 shown in Figure 11. For example, the light data set 1302 can include 48 light data for each of six wavelengths.

[0165] In one embodiment, the light characteristic value set estimator 1310 can estimate a light characteristic value set for at least a part of a body of a specific user based on the light data set 1302. Here, the processor is the light characteristic value set estimator 1310. For example, the light characteristic value set estimator 1310 can calculate a normalized diffuse reflectance set for a plurality of photodiodes based on the light data set 1302. The light characteristic value set estimator 1310 can also estimate a light characteristic value set 1312 associated with at least a part of the body based on the normalized diffuse reflectance set. A series of processes performed by the light characteristic value set estimator 1310 can be understood from the contents described with reference to FIGS. 4 to 12.

[0166] As an example, the light characteristic value set 1312 may include the final light scattering coefficient 554 and the final light absorption coefficient 556 described with reference to Figure 5. For example, the light characteristic value set 1312 may include the light scattering coefficient data map 1130 and the light absorption coefficient data map described with reference to Figure 11. For example, the light characteristic value set 1312 may include 32 pairs of light scattering coefficients and light absorption coefficients for each of six wavelengths.

[0167] In one embodiment, the urine volume estimation model 1320 can estimate the bladder urine volume 1322 of a specific user based on the light characteristic value set 1312. In this case, the urine volume estimation model 1320 can be a deep learning-based model or a machine learning-based model that has learned a plurality of learning data sets. For example, the machine learning-based model can be an artificial neural network (ANN), a K-Nearest Neighbors (KNN), a gradient boost (GB), a linear regression model, a random forest model, or an Ada Boost model. In addition, the learning data set can include a pair of an actual urine volume and a light characteristic value set associated with the actual urine volume. The process of acquiring the learning data set and the process of the urine volume estimation model 1320 learning a plurality of learning data sets will be described in detail below with reference to FIG. 14 and FIG. 15.

[0168] In another embodiment, the urine volume estimation model 1320 can estimate the user's bladder urine volume 1322 based on the light characteristic value set 1312 and the obesity degree information 1314. Here, the urine volume estimation model 1320 can be a deep learning-based model or a machine learning-based model that has learned a plurality of learning data sets and learning obesity degree information. In this case, the obesity degree information can include information on fat in the body around the location of the bladder. For example, the obesity degree information can include a body mass index (BMI; Body Max Index), an obesity degree measured by an abdominal obesity measurement method, an obesity degree measured by a standard body weight method, a body fat index, an abdominal fat thickness measured using ultrasound, and the like. As another example, the obesity degree information can include the above-mentioned light absorption coefficient. With this configuration, the urine volume estimation model 1320 can accurately estimate the urine volume even in the case of an obese user by learning the obesity degree information 1314.

[0169] In one embodiment, the processor may calculate a physiological information set based on the light characteristic value set 1312. The process of calculating the physiological information set can be understood from the contents described with reference to FIG. 12. Here, the physiological information set may correspond to the light characteristic value set 1312. For example, the physiological information set may include light absorption coefficient data included in a light absorption coefficient data map. Also, the physiological information set may include a plurality of physiological information data maps 1220, 1230, 1240, 1250 calculated based on a plurality of light scattering coefficient data maps 1210 described above with reference to FIG. 12. In this case, the processor may estimate the urine volume 1322 using the light characteristic value set 1312 and a physiological information set corresponding to the light characteristic value set 1312. However, since the physiological information set may be calculated based on the light characteristic value set 1312, the description will be based on the light characteristic value set 1312.

[0170] For a user with a lot of fat around the skin where the bladder is located, the light data included in the light data set 1302 may have little change even though the amount of urine in the bladder increases / decreases. In one embodiment, if the amount of change in the light data included in the light data set 1302 is minimal even though the amount of urine in the bladder increases / decreases, the processor may output a result associated with the urine volume being impossible to estimate. For example, the processor may output a result associated with the urine volume being impossible to estimate if the obesity information 1314 is equal to or greater than a predetermined obesity standard value.

[0171] In one embodiment, if the estimated urine volume 1322 is equal to or greater than a predefined reference value, the processor may output a message associated with a urination recommendation. For example, the reference value may correspond to an amount of urine in the bladder at which a person feels the need to urinate on average. Specifically, the processor may output visual, auditory, or tactile information as a message associated with a urination recommendation via the user terminal / medical device. For example, the user terminal / medical device may output a pop-up window or a vibration / sound notification advising the patient to urinate. With this configuration, a patient wearing the medical device may urinate at an appropriate time by receiving a message associated with a urination recommendation.

[0172] The measurement cycle can be performed in real time or periodically. The disclosed invention can provide the patient with an estimated urine volume 1322 based on the light data set 1302 detected by the measurement cycle. The patient can be provided with the bladder volume in real time or periodically, allowing the patient to monitor the amount of urine stored in his / her bladder and void at the appropriate time.

[0173] FIG. 14 is a graph showing an example of training data according to an embodiment of the present disclosure. In an embodiment, the processor may receive an optical data set associated with the nth measurement by an nth measurement cycle (where n is 1, 2, 3, 4 or more). The processor may estimate an optical characteristic value of the nth measurement based on the optical data set of the nth measurement. In this case, the nth actual urine volume may be a value obtained by directly measuring the amount of urine in the bladder of a specific user at the time when the nth measurement cycle is performed. For example, the actual urine volume may be obtained by a bladder irrigation process, a urodynamic study (UDS) process, a clean intermittent catheterization (CIC) process, etc.

[0174] As an example, the actual urine volume can be obtained by a bladder irrigation process. Specifically, the bladder irrigation process can include draining urine from the bladder of a specific user using a Foley catheter. At this time, the amount of urine in the bladder of the specific user can be specifically confirmed using an ultrasonic bladder volume device (RU scanner, Residual Urine Scanner). Then, the bladder irrigation process can inject sterile saline into the bladder of the specific user. At this time, the actual urine volume can correspond to the volume of the injected sterile saline. For example, when all the urine in the bladder of the specific user is drained, the first actual urine volume can be about 0 ml. Then, when 100 ml of sterile saline is injected into the bladder of the specific user, the second actual urine volume can be 100 ml.

[0175] As another example, the actual urine volume may be obtained by a uro-rheology test process. Specifically, the uro-rheology test process may include a bladder irrigation process using a Foley catheter for UDS instead of a Foley catheter. The actual urine volume may be obtained by the bladder irrigation process included in the uro-rheology test process. In addition, various measurement data such as the internal pressure of the bladder, the activity of the bladder muscle, and the connection state between the urethra and the bladder may be obtained by the uro-rheology test process. As an example, the learning dataset may include the measurement data.

[0176] As another example, the actual urine volume can be obtained by a self-catheterization process. Specifically, the self-catheterization process can include draining urine from the bladder using a self-catheterization catheter. At this time, the amount of urine drained using the self-catheterization catheter can be measured (e.g., measuring the amount of urine drained using a catheter cup). At this time, the actual urine volume can be calculated using the amount of urine drained. For example, the self-catheterization process can drain all of the urine in the bladder in two times. 200 ml of urine can be drained from the bladder in the first time, and 150 ml of urine can be drained from the bladder in the second time. At this time, the first actual urine volume can be 350 ml, the second actual urine volume can be 150 ml, and the third actual urine volume can be 0 ml.

[0177] As an example, a plurality of actual urine volumes can be obtained. For example, a first actual urine volume corresponding to the minimum urine volume of the bladder of a specific user can be obtained. Also, an Xth actual urine volume corresponding to the minimum urine volume of the bladder of a specific user can be obtained (where X is 2, 3 or more). When X is 3 or more, the 2nd to X-1th actual urine volumes can be obtained as values ​​between the first actual urine volume and the Xth actual urine volume.

[0178] In the graph of FIG. 14, the x-axis may represent time, and the y-axis may represent the amount of urine in the bladder of a specific user. Referring to FIG. 14, the actual urine amount included in each of the first learning data 1410, the second learning data 1420, the third learning data 1430, and the fourth learning data 1440 may be displayed. Specifically, the first learning data 1410 may include a first actual urine amount at the time when the first measurement cycle is performed. Similarly, the nth learning data may include an nth actual urine amount at the time when the nth measurement cycle is performed. Additionally, the nth learning data may include a pair of the nth actual urine amount and the light characteristic value of the nth measurement. In FIG. 14, only the first learning data 1410 to the fourth learning data 1440 are shown, but the present invention is not limited thereto. For example, the multiple learning data may be obtained by more than four or less than four measurement cycles.

[0179] The minimum urine volume of the specific user's bladder may correspond to the amount of urine when all of the urine in the bladder is discharged. For example, the minimum urine volume of the specific user's bladder may be about 0 ml. Also, the maximum urine volume of the specific user's bladder may correspond to the maximum capacity of the bladder. For example, the maximum urine volume of the specific user's bladder may be about 400 ml to 500 ml. The minimum and maximum urine volumes of the specific user's bladder may differ for each individual user. Referring to FIG. 14, the first actual urine volume may correspond to the minimum urine volume of the specific user's bladder, and the third actual urine volume may correspond to the maximum urine volume of the specific user's bladder.

[0180] In one embodiment, the teacher model can be generated by learning a plurality of learning data sets. For example, the teacher model can be a linear regression model, a random forest model, etc. Also, the teacher model can input an additional learning urine volume to estimate an additional learning light characteristic value set.

[0181] As an example, the nth teacher model may be a teacher model that has learned the nth+1st learning data set and the nth learning data set. Here, the nth estimated additional learning urine volume may be selected as any value between the nth+1st actual urine volume and the nth actual urine volume. Here, the nth estimation may refer to a process of estimating a plurality of nth estimated additional learning light characteristic value sets using the nth teacher model based on a plurality of nth estimated additional learning urine volumes between the nth+1st actual urine volume and the nth actual urine volume. For example, the nth estimated multiple additional learning urine volumes may be selected as values ​​at a certain interval between the nth+1st actual urine volume and the nth actual urine volume. For example, when the nth+1st actual urine volume is 400 ml and the nth actual urine volume is 100 ml, the first additional learning urine volume of the nth estimation may be selected as 200 ml, and the second additional learning urine volume of the nth estimation may be selected as 300 ml.

[0182] For example, the first estimation may include a process of estimating a first additional learning light characteristic value set based on a first additional learning urine volume between the second actual urine volume and the first actual urine volume. Similarly, the first estimation may include a process of estimating a k-th additional learning light characteristic value set based on a k-th additional learning urine volume between the second actual urine volume and the first actual urine volume (k is 1, 2, 3 or more). In this case, the first additional learning data set of the first estimation may include the first additional learning urine volume and the first additional learning light characteristic value set. Similarly, the k-th additional learning data set of the first estimation may include the k-th additional learning urine volume and the k-th additional learning light characteristic value set.

[0183] 14, the additional learning urine volumes included in each of the first additional learning data set 1412_1 of the first estimation, the second additional learning data set 1412_2 of the first estimation, and the third additional learning data set 1412_3 of the first estimation may be displayed. Specifically, the first additional learning urine volume of the first estimation to the third additional learning urine volume of the first estimation may be displayed between the first actual urine volume and the second actual urine volume. As an example, each time point for the first additional learning data set 1412_1 of the first estimation to the third additional learning data set 1412_3 of the first estimation may correspond to a method in which the additional learning urine volume is selected between the second actual urine volume and the first actual urine volume. For example, when multiple additional learning urine volumes are selected as values ​​at regular intervals between the second actual urine volume and the first actual urine volume, each time point for the first additional learning data set 1412_1 of the first estimation to the third additional learning data set 1412_3 of the first estimation can be selected as a value at regular intervals between the second actual urine volume measurement time point and the first actual urine volume measurement time point. For example, if the second actual urine volume is 400 ml, the time point corresponding to the first learning data 1410 is 0 seconds, the first actual urine volume is 0 ml, the time point corresponding to the second learning data 1420 is 4000 seconds, the first additional learning urine volume of the first estimation is 100 ml, the second additional learning urine volume of the first estimation is 200 ml, and the third additional learning urine volume of the first estimation is 300 ml, then the time point corresponding to the first additional learning data set 1412_1 of the first estimation may be 1,000 seconds, the time point corresponding to the second additional learning data set 1412_2 of the first estimation may be 2,000 seconds, and the time point corresponding to the third additional learning data set 1412_3 of the first estimation may be 3,000 seconds.

[0184] Based on what has been described above about the first estimation between the second training data 1420 and the first training data 1410, the second estimation between the third training data 1430 and the second training data 1420, and the third estimation between the fourth training data 1440 and the third training data 1430 can be understood in a similar manner.

[0185] In FIG. 14, three additional learning data sets 1412_1 to 1412_3 of the first estimation are shown, but it is not limited thereto. For example, when the number of urine volumes for additional learning is selected to be more than 3 or less than 3, the number of additional learning data sets can also be generated to be more than 3 or less than 3. Further, in FIG. 14, it shows that four measurement cycles are performed, but it is not limited thereto. For example, the measurement can be performed a number of times more than 4 or less than 4. As another example, multiple measurements can be performed within 72 hours.

[0186] In one embodiment, when two measurement cycles are performed, two learning data sets can be obtained. For example, the first urine volume included in the first learning data set can correspond to the minimum urine volume of a specific user's bladder, and the second urine volume included in the second learning data set can correspond to the maximum urine volume of the specific user's bladder. At this time, by minimizing the data collection for the user, the inconvenience to the user can be minimized, and a urine volume prediction model for individuals can be generated.

[0187] In other embodiments, when multiple measurement cycles (for example, three or more times) are performed, multiple learning data sets can be obtained. At this time, multiple teacher models can be generated based on the multiple learning data sets, and multiple additional learning data sets can be generated based on the multiple teacher models. The urine volume prediction model can improve the accuracy of estimating the urine volume in the bladder by learning the multiple learning data sets and the multiple additional learning data sets.

[0188] 15 is a block diagram illustrating an example of a urine volume estimation model 1550 according to an embodiment of the present disclosure. In an embodiment, the urine volume estimation model 1550 can learn a plurality of training data sets 1512, 1514. Here, the nth training data set can include a pair of the nth actual urine volume and the nth light characteristic value set. That is, the plurality of training data sets 1512, 1514 can include a plurality of actual urine volumes 1512 and a plurality of light characteristic value sets 1514. The method of acquiring the plurality of training data sets can be understood from the plurality of training data 1410, 1420, 1430, 1440 described above with reference to FIG. 14.

[0189] In one embodiment, the urine volume estimation model 1550 can further learn a single or multiple additional learning data sets 1532, 1534. Here, the k-th additional learning data can include a pair of the k-th additional learning urine volume and the k-th additional learning light characteristic value set. That is, the multiple additional learning data sets 1532, 1534 can include multiple additional learning urine volumes 1532 and multiple additional learning light characteristic value sets 1534. A method for acquiring multiple additional learning data sets can be understood from the multiple additional learning data sets 1412_1 to 1412_3, 1422_1, 1422_2, 1432_1 to 1432_4 described above with reference to FIG. 14. Although FIG. 15 shows multiple pairs of the additional learning urine volume and the additional learning light characteristic value set, the present invention is not limited thereto, and the number of pairs of the additional learning urine volume and the additional learning light characteristic value set may be one.

[0190] In one embodiment, the urine volume estimation model 1550 can be trained by applying weights 1520 to the multiple training data sets 1512, 1514. Specifically, the weights 1520 can be information for the urine volume estimation model 1550 to adjust the learning weights of the multiple additional training data sets 1532, 1534 and the multiple training data sets 1512, 1514. For example, the weights 1520 can be determined in advance before data training of the urine volume estimation model 1550. Additionally or alternatively, the weights 1520 can be applied to the multiple additional training data sets and adjusted during the training process of the urine volume estimation model 1550.

[0191] In one embodiment, the urine volume estimation model 1550 can further learn the learning obesity level information 1540. In this case, the learning obesity level information 1540 can be obesity level information of a body that is the subject of measurements of multiple actual urine volumes 1512. Although a single learning obesity level information 1540 is shown in Fig. 15, the present invention is not limited to this. For example, when measurements of multiple actual urine volumes 1512 are performed on multiple bodies, the urine volume estimation model 1550 can learn multiple pieces of learning obesity level information.

[0192] The urine volume estimation model 1550 can learn by applying a weighting value 1520 to a plurality of learning data sets 1512 and 1514 and applying the weighting value 1520 to the plurality of learning data sets 1512 and 1514. This allows the urine volume estimation model 1550 to accurately estimate the urine volume. Furthermore, the urine volume estimation model 1550 can be provided for individual users by learning the learning obesity degree information 1540. Furthermore, the urine volume estimation model 1550 uses a machine learning model or a deep learning model for which app development support is relatively good, and the invention according to the present disclosure can easily develop a mobile app for a wearable device. Furthermore, the machine learning model or the deep learning model is easy to re-learn, and the invention according to the present disclosure can estimate the bladder urine volume for an individual. Furthermore, the urine volume estimation model 1550 is easy to maintain and improve, and has excellent model scalability and model versatility.

[0193] FIG. 16 is a graph showing a plurality of examples of a urine volume estimation model according to an embodiment of the present disclosure. Referring to FIG. 16, a urine volume estimation graph 1600 can display urine volume estimation results of a plurality of urine volume estimation models. Specifically, the urine volume estimation graph 1600 can display a first graph 1640, a second graph 1650, and a third graph 1660. In FIG. 16, the third graph 1660 can be displayed as a dotted line graph. In addition, the urine volume estimation graph 1600 can display a first actual urine volume included in a first learning data set 1610 and a second actual urine volume included in a second learning data set 1630. In addition, the urine volume estimation graph 1600 can display a first actual urine volume for comparison included in a first comparison data set 1622, a second actual urine volume for comparison included in a second comparison data set 1624, and a third actual urine volume for comparison included in a third comparison data set 1626. The urine volume estimation graph 1600 can display an index on the x-axis and the urine volume in the bladder (actual urine volume and / or estimated urine volume) on the y-axis. In this case, the index can be an indicator for expressing the flow of time. For example, the index of the time point when the first measurement cycle is performed can be indicated as 0, and the index of the time point when the second measurement cycle is performed can be indicated as 4.

[0194] In one embodiment, the first learning data set 1610 may include a pair of a first actual urine volume and a first set of light characteristic values. Also, the second learning data set 1630 may include a pair of a second actual urine volume and a second set of light characteristic values. In one embodiment, the first comparison data set 1622 may include a pair of a first comparison actual urine volume and a first comparison light characteristic value set. Similarly, the second comparison data set 1624 may include a pair of a second comparison actual urine volume and a second comparison light characteristic value set, and the third comparison data set 1626 may include a pair of a third comparison actual urine volume and a third comparison light characteristic value set. Here, the multiple learning data sets 1610, 1630 and the multiple comparison data sets 1622, 1624, 1626 may be obtained by performing a measurement cycle. For example, the multiple training data sets 1610, 1630 and the multiple comparison data sets 1622, 1624, 1626 may have been obtained from a particular user wearing a medical device (eg, medical device 100 shown in FIG. 1) on the skin positioned over the bladder.

[0195] For example, the first actual urine volume may be the actual urine volume in the bladder of the specific user at the time when the first measurement is performed. Also, the second actual urine volume may be the actual urine volume in the bladder of the specific user at the time when the second measurement of the bladder of the specific user is performed. With reference to Fig. 16, the first actual urine volume may be about 100 ml, and the second actual urine volume may be about 300 ml. The method of measuring the actual urine volume can be understood from the above content based on Fig. 14.

[0196] Similarly, the method of measuring the first to third actual urine volumes for comparison is the same as the method of measuring the first and second actual urine volumes. At this time, the first to third actual urine volumes for comparison may be actual urine volumes that are not used in learning the urine volume learning model. Also, the first to third light characteristic value sets for comparison may be light characteristic value sets that are not used in learning the urine volume learning model.

[0197] As an example, the first to third actual urine volumes for comparison may be selected between the first and second actual urine volumes. For example, the first to third actual urine volumes for comparison may be selected to have the same interval between the first and second actual urine volumes. With reference to FIG. 16, when the first actual urine volume is 100 ml and the second actual urine volume is 300 ml, the first actual urine volume for comparison may be selected to be 150 ml, the second actual urine volume to be 200 ml, and the third actual urine volume to be 250 ml. In FIG. 16, three actual urine volumes for comparison are selected, but the number of actual urine volumes for comparison may be more or less than three.

[0198] In Fig. 16, a "comparison measurement cycle" may refer to a measurement cycle for estimating a comparison light characteristic value set. In this case, an index of a time when a first comparison measurement cycle is performed may be indicated as 1, an index of a time when a second comparison measurement cycle is performed may be indicated as 2, and an index of a time when a third comparison measurement cycle is performed may be indicated as 3. As an example, an nth comparison measurement cycle may be performed corresponding to a time when an nth comparison actual urine volume is measured. Also, an nth comparison light characteristic value set may be estimated based on an nth comparison light data set detected by the nth comparison measurement cycle.

[0199] In one embodiment, the first urine volume estimation model may be a model trained with the first learning data set 1610 and the second learning data set 1630. For example, the first urine volume estimation model is an ANN model. The first graph 1640 can display a urine volume estimation result of the first urine volume estimation model. The teacher model may be a model trained with the first learning data set 1610 and the second learning data set 1630. For example, the teacher model is a Random Forest model or a Linear Regression model. The second graph 1650 can display a urine volume estimation result of the teacher model. The second urine volume estimation model may be a model trained with the first learning data set 1610, the second learning data set 1630, and a plurality of additional learning data sets. At this time, the plurality of additional learning data sets may be generated by the teacher model. For example, the second urine volume estimation model is an ANN model. The third graph 1660 can display a urine volume estimation result of the second urine volume estimation model.

[0200] A comparison of the urine volume estimation result of the first urine volume estimation model, the urine volume estimation result of the second urine volume estimation model, and the urine volume estimation result of the teacher model can be expressed as shown in Table 3 below. Referring to FIG. 16, a urine volume estimation graph 1600 may represent Table 3 below.

[0201] [Table 3]

[0202] In Table 3, the urine volume estimation model may include a first urine volume estimation model, a second urine volume estimation model, and a teacher model. The urine volume estimation model may estimate the urine volume by inputting a light characteristic value set (or a comparison light characteristic value set) corresponding to each index. For example, the first urine volume estimation model may generate a urine volume estimation result corresponding to about 39.1 ml by inputting a first comparison light characteristic value set. At this time, the estimation error of the first urine volume estimation model may be 110.9 ml obtained by subtracting 39.1 ml, which is the estimation result of the first urine volume estimation model, from 150 ml, which is the first comparison actual urine volume. Similarly, the second urine volume estimation model may generate a urine volume estimation result corresponding to about 115.6 ml by inputting a first comparison light characteristic value set. At this time, the estimation error of the second urine volume estimation model may be 34.4 ml obtained by subtracting 115.6 ml, which is the estimation result of the first urine volume estimation model, from 150 ml, which is the first comparison actual urine volume. It can be seen from Table 3 or urine volume estimation graph 1600 that the second urine volume estimation model trained on multiple learning datasets and multiple additional learning datasets has a smaller urine volume estimation error than the first urine volume estimation model trained on multiple learning datasets.

[0203] When the amount of the training data set is not large, multiple additional training data sets are generated by the teacher model to perform data augmentation. By allowing the urine volume prediction model to learn more data through data augmentation, an automated method for predicting urine volume in the bladder can be realized that is designed based on medical knowledge and diagnosis.

[0204] 17 is a flow chart illustrating a digital bladder volume prediction method 1700 according to an embodiment of the present disclosure. The method 1700 can be performed by a controller (or at least one processor) of a medical device, a user terminal, and / or at least one processor of an information processing system. The method 1700 begins with the processor receiving (S1710) a light data set associated with a particular user detected by a plurality of photodiodes. In one embodiment, the processor can estimate (S1720) a set of light characteristic values ​​for at least a portion of the particular user's body based on the light data set.

[0205] In one embodiment, the processor may estimate the bladder volume of the specific user using a urine volume estimation model based on the estimated light characteristic value set (S1730). Here, the urine volume estimation model may be a deep learning-based model or a machine learning-based model trained on a plurality of training data sets. Here, the plurality of training data sets may include pairs of the specific user's actual urine volume and the light characteristic value set associated with the actual urine volume.

[0206] In one embodiment, the processor may include a first learning data set and a second learning data set. Here, the first learning data set may include a pair of a first actual urine volume of a specific user and a first learning light characteristic value set associated with the first actual urine volume. Also, the second learning data set may include a pair of a second actual urine volume of a specific user and a second learning light characteristic value set associated with the second actual urine volume. At this time, the second actual urine volume may be greater than the first actual urine volume. For example, the first urine volume may correspond to a minimum urine volume of a bladder of the specific user, and the second urine volume may correspond to a maximum urine volume of a bladder of the specific user.

[0207] In one embodiment, a teacher model may be generated by learning a plurality of learning data sets. Also, the urine volume estimation model may further learn a single or a plurality of additional learning data sets. In this case, the additional learning data set may include a pair of an additional learning urine volume and an additional learning light characteristic value set estimated by inputting the additional learning urine volume into the teacher model. Here, the additional learning urine volume may be larger than the first urine volume and smaller than the second urine volume. Also, the urine volume estimation model may learn by applying a predetermined weight value to the plurality of learning data sets.

[0208] In one embodiment, the processor may receive obesity information associated with a particular user. In this case, the urine volume estimation model may further learn the training obesity information. The processor may then estimate a urine volume using the urine volume estimation model based on the received obesity information and the set of light characteristic values.

[0209] The above flow chart and description are merely illustrative and may be implemented in different ways in some embodiments, for example, the order of steps may be changed, some steps may be repeated, some steps may be omitted, or some steps may be added.

[0210] The above-mentioned method may be provided as a computer program stored in a computer-readable recording medium for execution by a computer. The medium may be a medium for continuously storing a computer-executable program or a medium for temporarily storing the program for execution or download. The medium may be various recording means or storage means in the form of a single or multiple hardware components combined, and may be distributed over a network without being limited to a medium directly connected to a computer system. Examples of the medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and ROMs, RAMs, flash memories, etc., configured to store program instructions. Other examples of the medium include recording media or storage media managed by app stores that distribute applications, or sites, servers, etc. that supply or distribute various other software.

[0211] The methods, operations, or techniques of the present disclosure can be embodied in a variety of ways. For example, such techniques can be embodied in hardware, firmware, software, or a combination thereof. Those of ordinary skill in the art will appreciate that the various exemplary logical blocks, modules, circuits, and algorithmic steps described in the present disclosure can be embodied in electronic hardware, computer software, or a combination of both. To clearly illustrate such interchangeability between hardware and software, the various exemplary components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is embodied as hardware or software will vary depending on the particular application and design requirements imposed on the overall system. Those of ordinary skill in the art may also embody the described functionality in various ways for each particular application, but such embodying should not be interpreted as departing from the scope of the present disclosure.

[0212] In a hardware implementation, the processing units used to perform the techniques may be embodied within one or more ASICs, DSPs, digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described in this disclosure, computers, or any combination thereof.

[0213] Thus, the various example logic blocks, modules and circuits described in this disclosure may be embodied or performed by any combination of general purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate and transistor logic, discrete hardware components, or any combination designed to perform the functions described herein. A general purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be embodied by a combination of computing devices, such as a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other configuration.

[0214] In a firmware and / or software implementation, the techniques may be embodied with instructions stored on a computer-readable medium, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, compact disc (CD), magnetic or optical data storage devices, etc. The instructions may be executable by one or more processors to cause the processors to perform certain aspects of the functions described in this disclosure.

[0215] If embodied in software, the techniques may be stored on or transferred via a computer-readable medium as one or more instructions or code. A computer-readable medium includes any medium that facilitates transfer of a computer program from one place to another, and includes both computer storage media and communication media. A storage medium may be any available medium that can be accessed by a computer. By way of non-limiting example, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to transport or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Additionally, any connection may be properly referred to as a computer-readable medium.

[0216] For example, if the software is transferred from a website, server, or other remote source using coaxial cable, fiber optic cable, lead wire, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, lead wire, digital subscriber line, or wireless technologies such as infrared, radio, and microwave are included within the definition of medium. As used herein, disk and disc include CDs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically while discs reproduce data optically using lasers. Combinations of the above should also be included within the scope of computer-readable media, etc.

[0217] A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any different forms of storage medium known in the art. An exemplary storage medium may be coupled to the processor such that the processor reads information from, and writes information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

[0218] Although the embodiments described above are described as utilizing aspects of the presently disclosed subject matter on one or more stand-alone computer systems, the present disclosure is not so limited and may be embodied in any computing environment, such as a network or distributed computing environment. Moreover, aspects of the subject matter in the present disclosure may be embodied in multiple processing chips or devices, and storage may be affected across multiple devices in the same manner. Such devices may include PCs, network servers, and handheld devices.

[0219] Although the present disclosure has been described in the present specification by way of some embodiments, various modifications and alterations can be made without departing from the present disclosure, which can be understood by those of ordinary skill in the art to which the present disclosure pertains. In addition, such modifications and alterations should be understood to fall within the scope of the claims appended hereto. [Explanation of symbols]

[0220] 100 Medical Devices 112_1 First photodiode 112_2 Second photodiode 112_20 20th photodiode 114_1 First light source group 114_2 Second light source group 114_3 Third light source group 114_4 Fourth light source group 120 User terminals

Claims

1. 1. A method for predicting bladder volume, performed by at least one processor, comprising: receiving a light data set associated with a particular user detected by a plurality of photodiodes, where the plurality of photodiodes are configured to detect light intensities associated with light projected onto skin located over the particular user's bladder; estimating a set of light characteristic values ​​for at least a portion of the particular user's body based on the light data set; and estimating a bladder urine volume of the specific user using a urine volume estimation model based on the estimated set of light characteristic values.

2. The urine volume estimation model is a deep learning-based model or a machine learning-based model that has learned a plurality of learning data sets, The urine volume prediction method according to claim 1 , wherein the plurality of learning data sets include a pair of an actual urine volume of the specific user and a set of light characteristic values ​​associated with the actual urine volume.

3. the plurality of training data sets include a first training data set and a second training data set; The first learning data set includes a pair of a first actual urine volume of the specific user and a first learning light characteristic value set associated with the first actual urine volume; The second learning data set includes a pair of a second actual urine volume of the specific user and a second learning light characteristic value set associated with the second actual urine volume; The method for predicting a urine volume according to claim 2 , wherein the second actual urine volume is greater than the first actual urine volume.

4. A teacher model is generated by training the plurality of training data sets; The urine volume estimation model is further trained on one or more additional training data sets; The additional learning data set includes a pair of an additional learning urine volume and an additional learning light characteristic value set estimated by inputting the additional learning urine volume into the teacher model, The urine volume prediction method according to claim 3 , wherein the additional learning urine volume is larger than the first urine volume and smaller than the second urine volume.

5. The urine volume prediction method according to claim 4 , wherein the urine volume estimation model is trained by applying predetermined weights to the multiple training data sets.

6. the first actual urine volume corresponds to a minimum urine volume of the particular user's bladder; The method of claim 3 , wherein the second actual urine volume corresponds to a maximum urine volume of the specific user's bladder.

7. The method of claim 1 , further comprising the step of outputting a message related to a urination recommendation when the estimated urine volume information is equal to or greater than a predetermined reference value.

8. The urine volume estimation model further learns learning obesity level information, The method further includes receiving obesity information associated with the particular user; The urine volume prediction method according to claim 2 , wherein the step of estimating the urine volume includes a step of estimating the urine volume using the urine volume estimation model based on the received obesity level information and the set of light characteristic values.

9. A computer program stored on a computer-readable recording medium for executing the method according to any one of claims 1 to 8 on a computer.

10. A user terminal, The Communications Department and Memory, at least one processor coupled to the memory and configured to execute at least one computer readable program contained in the memory; The at least one program comprises: receiving a light data set associated with a particular user detected by a plurality of photodiodes, where the plurality of photodiodes are configured to detect light intensities associated with light projected onto skin located over the particular user's bladder; estimating a set of light characteristic values ​​for at least a portion of the particular user's body based on the light data set; A user terminal including instructions for estimating the specific user's bladder urine volume using a urine volume estimation model based on the estimated light characteristic value set.

Images