Apparatus for predicting compression fracture and method thereof

Abstract

The present invention discloses a device and method for predicting the occurrence of compression fractures. In other words, the present invention trains an AI-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density level, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, habitual smoking amount, habitual alcohol intake, blood levels, steroid use, history of steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI. By predicting the probability of compression fracture occurrence by preset periods, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of preventive measures for osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures, the invention [prescribes] when to [prescribe] to users who have undergone bone density testing, users diagnosed with osteopenia or osteoporosis based on bone density testing, etc. By providing information on whether compression fractures may occur, the user can exercise greater caution and be informed of when X-ray examinations, preventive measures, and treatments are necessary.

Description

Technology Field

[0001] The present invention relates to a device and method for predicting the occurrence of a compression fracture, and more specifically, to a device and method for predicting the occurrence of a compression fracture by performing training on an artificial intelligence-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density value, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the spinal segments where they occurred, muscle mass measured in MRI, presence of underlying diseases, usual smoking amount, usual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured in spinal X-ray, CT, and MRI, thereby predicting the probability of compression fracture occurrence by a preset period, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examination, the necessity of measures for the prevention of osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc. Background Technology

[0002] Compression fractures are a condition in which the spine is crushed and broken. While they can occur due to injury, they often occur naturally in patients with osteopenia or osteoporosis.

[0003] Furthermore, most patients do not detect it at the time of occurrence and endure back pain until they visit a hospital later and are diagnosed with a compression fracture. Prior art literature

[0004] Korean Published Patent No. 10-2020-0095504 [Title: 3D Medical Image Analysis Method and System for Identifying Vertebral Fractures] The problem to be solved

[0005] The objective of the present invention is to provide a device and method for predicting the occurrence of a compression fracture and predicting the probability of a compression fracture occurring at a predetermined period, the estimated time remaining until a compression fracture occurs, the necessity of spinal imaging examinations, the necessity of measures to prevent osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures, by performing training on an artificial intelligence-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density value, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the spinal segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount, usual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, angle and length values ​​measured by spinal X-ray, CT, and MRI.

[0006] Another objective of the present invention is to provide a compression fracture occurrence prediction device and method that predicts the probability of compression fracture occurrence by a preset period, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of measures to prevent osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc., by performing training on an artificial intelligence-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density value, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the spinal segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount, usual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, angle and length values ​​measured by spinal X-ray, CT, and MRI. means of solving the problem

[0007] A compression fracture occurrence prediction device according to an embodiment of the present invention may include: a collection unit that collects essential input information and optional input information; and a control unit that performs artificial intelligence-based machine learning based on the collected essential input information and optional input information, and predicts, based on the machine learning results, the probability of compression fracture occurrence by period set in relation to the user, the estimated time remaining until compression fracture occurrence by period, the ROC curve (Receiver Operating Characteristic curve) for compression fracture by period, the AUC value (Area Under Curve value) by period, whether spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture.

[0008] As an example related to the present invention, the necessity of the spinal imaging examination may include information for recommending one of the following examinations based on the probability of the occurrence of the compression fracture: an X-ray examination, an ultrasound examination, a computed tomography examination, and a magnetic resonance imaging examination.

[0009] As an example related to the present invention, the necessity of measures for preventing at least one of osteopenia, osteoporosis, and compression fracture, and the necessity of administering a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture may include information for recommending any one of bisphosphonates, female hormone preparations, tissue-selective estrogen complexes (TSECs), selective estrogen receptor modulators (SERMs), RANKL inhibitors, parathyroid hormone preparations, PTHrP analogs, sclerostin inhibitors, activated vitamin D preparations, calcitonin preparations, calcium preparations, vitamin D preparations, vitamin K2 preparations, thoracic spine braces, thoracolumbar spine braces, lumbar spine braces, and lumbosacral braces.

[0010] A method for predicting the occurrence of a compression fracture according to an embodiment of the present invention may include: a step of collecting essential input information and optional input information by a collection unit; and a step of performing artificial intelligence-based machine learning based on the collected essential input information and optional input information by a control unit, and predicting, based on the machine learning results, the probability of occurrence of a compression fracture by period, the estimated time remaining until the occurrence of a compression fracture by period, the ROC curve for the compression fracture by period, the AUC value by period, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture.

[0011] As an example related to the present invention, the prediction step may perform machine learning using the essential input information and the optional input information as input values ​​for a preset artificial intelligence-based compression fracture prediction model, and based on the machine learning results, predict the probability of compression fracture occurrence by period, the estimated time remaining until compression fracture occurrence by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture, in relation to the user.

[0012] As an example related to the present invention, the control unit performs learning on the essential input information and the optional input information on a pre-learned deep learning algorithm model or machine learning algorithm model to predict the importance of the result derivation of each element, the odds ratio, the probability of occurrence of compression fracture by a preset period, the estimated time remaining until occurrence of compression fracture by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture; The method may further include a step of outputting, through at least one of a display unit and a voice output unit, the predicted importance, odds ratio, preset probability of occurrence of compression fracture by period, estimated time remaining until occurrence of compression fracture by period, ROC curve for compression fracture by period, AUC value by period, whether spinal imaging examination is necessary, whether measures for prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether medication for at least one of osteopenia, osteoporosis, and compression fracture is necessary.

[0013] As an example related to the present invention, the control unit acquires real-time video information related to a user through a camera unit; the control unit analyzes the acquired real-time video information to determine whether a preset event has occurred; when the event occurs, the control unit performs other machine learning based on artificial intelligence based on the real-time video information and the essential input information, and predicts, based on the other machine learning results, the probability of a compression fracture occurring due to the event, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture. And the method may further include a step of controlling the display unit to display, by the control unit, the probability of a compression fracture occurring due to an event occurrence in relation to the predicted user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture.

[0014] As an example related to the present invention, the event may be any one of the following: when the user's body's tilt change per unit of time or tilt change exceeds a preset threshold, when the user falls, and when the user falls.

[0015] As an example related to the present invention, when at least one of the cases is met, the probability of a compression fracture occurring due to the predicted event occurrence exceeds a preset threshold value and the estimated period remaining until the occurrence of a compression fracture due to the predicted event occurrence is less than or equal to another preset threshold value, the control unit determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user and generates emergency rescue alarm information; and the control unit may further include the step of transmitting the generated emergency rescue alarm information to an emergency rescue server through a communication unit.

[0016] As an example related to the present invention, when at least one of the cases is met—where the probability of a compression fracture occurring due to the predicted event exceeds a preset threshold value and the estimated period remaining until the occurrence of a compression fracture due to the predicted event is less than or equal to another preset threshold value—the control unit determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user and selects medical information capable of providing real-time treatment from among a plurality of medical information preset in the storage unit; and the control unit may further include the step of performing a remote medical treatment function through a video call function with a terminal possessed by the medical information capable of providing real-time treatment. Effects of the invention

[0017] The present invention trains an AI-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density level, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, habitual smoking amount, habitual alcohol intake, blood levels, steroid use, history of steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI. By predicting the probability of compression fracture occurrence by preset periods, the estimated time remaining until a compression fracture occurs, the necessity of spinal imaging examinations, the necessity of preventive measures for osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures, the invention [informs] users who have undergone bone density testing, users diagnosed with osteopenia or osteoporosis based on bone density testing, etc., regarding when By providing information on whether compression fractures may occur, it is effective in enabling the user to exercise greater caution and to indicate when X-ray examinations, measures to prevent osteopenia and osteoporosis, and the administration of medication are necessary.

[0018] Furthermore, the present invention trains an AI-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density, diet, weekly exercise volume, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, habitual smoking amount, habitual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI. By predicting the probability of compression fracture occurrence by preset periods, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of preventive measures for osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures, the invention provides information related to compression fractures in the case of high-risk exercises for users engaging in various types of exercise, thereby enabling the user to recognize the risk of compression fractures. It has the effect of enabling users to perform exercises suitable for them, and in the case of patients with osteopenia and osteoporosis, it enables them to supplement their diet suitable for them. Brief explanation of the drawing

[0019] FIG. 1 is a block diagram showing the configuration of a compression fracture occurrence prediction device according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating a method for predicting the occurrence of a compression fracture according to an embodiment of the present invention. Specific details for implementing the invention

[0020] It should be noted that the technical terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. Furthermore, unless specifically defined otherwise in this invention, the technical terms used in this invention should be interpreted in the sense generally understood by those skilled in the art to which this invention pertains, and should not be interpreted in an overly broad or overly narrow sense. Additionally, if a technical term used in this invention is an incorrect technical term that fails to accurately express the concept of the invention, it should be replaced with a technical term that can be correctly understood by those skilled in the art. Moreover, general terms used in this invention should be interpreted according to their prior definitions or the context, and should not be interpreted in an overly narrow sense.

[0021] Furthermore, singular expressions used in the present invention include plural expressions unless the context clearly indicates otherwise. Terms such as "composed of" or "comprising" in the present invention should not be interpreted as necessarily including all of the various components or steps described in the invention, and should be interpreted as meaning that some of the components or steps may not be included, or that additional components or steps may be included.

[0022] Additionally, terms including ordinal numbers, such as first, second, etc., used in the present invention may be used to describe components, but the components should not be limited by the terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component.

[0023] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. Identical or similar components are given the same reference number regardless of the drawing symbols, and redundant descriptions thereof will be omitted.

[0024] Furthermore, in describing the present invention, detailed descriptions of related prior art are omitted if it is determined that such descriptions could obscure the essence of the invention. Additionally, it should be noted that the attached drawings are intended only to facilitate an understanding of the concept of the present invention and should not be interpreted as limiting the concept of the present invention.

[0025] FIG. 1 is a block diagram showing the configuration of a compression fracture occurrence prediction device (100) according to an embodiment of the present invention.

[0026] As illustrated in FIG. 1, the compression fracture occurrence prediction device (100) is composed of a collection unit (110), a communication unit (120), a storage unit (130), a display unit (140), a voice output unit (150), and a control unit (160). Not all components of the compression fracture occurrence prediction device (100) illustrated in FIG. 1 are essential components, and the compression fracture occurrence prediction device (100) may be implemented with more components than those illustrated in FIG. 1, or with fewer components.

[0027] The above compression fracture occurrence prediction device (100) can be applied to various terminals such as a smartphone, portable terminal, mobile terminal, foldable terminal, personal digital assistant (PDA), portable multimedia player (PMP) terminal, telematics terminal, navigation terminal, personal computer, laptop computer, slate PC, tablet PC, ultrabook, wearable device (e.g., smartwatch, smart glass, head-mounted display, etc.), Wibro terminal, IPTV terminal, smart TV, digital broadcasting terminal, AVN terminal, A / V system, flexible terminal, digital signage device, artificial intelligence speaker, etc.

[0028] The above-mentioned collection unit (110) collects (or receives) pre-set essential input information, optional input information, etc., according to user input (or user / patient selection / touch / control). Here, the essential input information includes gender, height, weight, Body Mass Index (BMI), age, bone mineral density values, etc. At this time, the bone mineral density (BMD) value (or T-score / Z-score) indicates normal if it is -1.0 or higher, osteopenia if it is -1.0 to -2.5, osteoporosis if it is -2.5 or lower, and severe osteoporosis if it is -2.5 or lower accompanied by a fracture. In addition, the above-mentioned optional input information includes diet, weekly exercise amount (or daily / hourly exercise amount), presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the level (or vertebral segment) at which they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount (or smoking status), usual alcohol intake (or drinking status), blood levels, steroid use, history of steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, angle and length values ​​measured by spinal imaging diagnostic devices (e.g., spinal X-ray, CT, MRI, etc.). At this time, the above-mentioned diet includes information regarding the intake of milk, anchovies, protein, omega-3 fatty acids, vitamin D, vitamin K2, health foods, and medications, as well as the presence and amount of such intake. Furthermore, the above-mentioned presence of underlying diseases includes information regarding the presence of diseases such as rheumatoid arthritis, endocrine diseases, and dietary diseases including mental illnesses. In addition, the above blood values ​​include calcium levels, parathyroid hormone levels, thyroid hormone levels, vitamin D levels, vitamin K2 levels, etc.In addition, the values ​​measured by the aforementioned spinal imaging diagnostic device are values ​​measured from spinal X-ray, CT, and MRI, including pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), C7 vertical tilt, L1-S1 lordosis angle, L4-S1 lordosis angle, sagittal vertical axis (including, for example, distance: C7PL to post corner of S1), thoracic kyphosis angle, segmental kyphosis angle, coronal balance, global tilt, relative pelvic version, lordosis distribution index, relative spinopelvic alignment, and relative lumbar lordosis angle. It includes lordosis angle), global alignment and proportion score (Global Alignment and Proportion (GAP) score), etc.

[0029] In addition, the above-mentioned collection unit (110) may receive and collect data transmitted from OCS, EMR, PACS systems, and bone density testing devices used in hospitals, clinics, etc.

[0030] At this time, the collection unit (110) can collect personal information, medical information, required input information, optional input information, etc. related to the user, in accordance with a pre-set electronic document-type consent form for the use of personal information, consent form for the use of medical information, etc., and can utilize and provide the same to other terminals.

[0031] The communication unit (120) communicates with any internal component or any at least one external terminal through a wired / wireless communication network. At this time, the any external terminal may include a server (not shown), a terminal (not shown), etc. Here, wireless internet technologies include Wireless LAN (WLAN), DLNA (Digital Living Network Alliance), Wibro (Wireless Broadband), Wimax (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), IEEE 802.16, Long Term Evolution (LTE), LTE-A (Long Term Evolution-Advanced), and Wireless Mobile Broadband Service (WMBS), and the communication unit (120) transmits and receives data according to at least one wireless internet technology within a range that includes internet technologies not listed above. In addition, short-range communication technologies may include Bluetooth, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, Near Field Communication (NFC), Ultra Sound Communication (USC), Visible Light Communication (VLC), Wi-Fi, and Wi-Fi Direct.In addition, wired communication technologies may include Power Line Communication (PLC), USB communication, Ethernet, serial communication, and optical / coaxial cables.

[0032] In addition, the communication unit (120) can mutually transmit information with any terminal via a Universal Serial Bus (USB).

[0033] In addition, the communication unit (120) transmits and receives wireless signals to and from a base station, the server, the terminal, etc. on a mobile communication network built according to technical standards or communication methods for mobile communication (e.g., GSM (Global System for Mobile communication), CDMA (Code Division Multi Access), CDMA2000 (Code Division Multi Access 2000), EV-DO (Enhanced Voice-Data Optimized or Enhanced Voice-Data Only), WCDMA (Wideband CDMA), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), etc.).

[0034] In addition, the communication unit (120) transmits the collected essential input information, optional input information, etc., to the server, the terminal, etc., under the control of the control unit (160).

[0035] The above storage unit (130) stores various user interfaces (UI), graphic user interfaces (GUI), etc.

[0036] In addition, the storage unit (130) stores data and programs, etc., necessary for the operation of the compression fracture occurrence prediction device (100).

[0037] That is, the storage unit (130) can store a number of applications (or applications) running on the compression fracture occurrence prediction device (100), data for the operation of the compression fracture occurrence prediction device (100), and commands. At least some of these applications may be downloaded from an external server via wireless communication. Additionally, at least some of these applications may exist on the compression fracture occurrence prediction device (100) from the time of shipment for the basic functions of the compression fracture occurrence prediction device (100). Meanwhile, the applications may be stored in the storage unit (130), installed on the compression fracture occurrence prediction device (100), and driven by the control unit (160) to perform the operation (or function) of the compression fracture occurrence prediction device (100).

[0038] Additionally, the storage unit (130) may include at least one storage medium among Flash Memory Type, Hard Disk Type, Multimedia Card Micro Type, Card Type Memory (e.g., SD or XD memory, etc.), Magnetic Memory, Magnetic Disk, Optical Disk, RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), and PROM (Programmable Read-Only Memory). Additionally, the compression fracture occurrence prediction device (100) may operate a web storage that performs the storage function of the storage unit (130) on the internet, or may operate in relation to said web storage.

[0039] In addition, the storage unit (130) stores the collected essential input information, optional input information, etc. under the control of the control unit (160).

[0040] The above display unit (or display unit) (140) can display various content, such as various menu screens, using a user interface and / or a graphic user interface stored in the storage unit (130) under the control of the control unit (160). Here, the content displayed on the display unit (140) includes various text or image data (including various information data) and menu screens, etc., including data such as icons, list menus, and combo boxes. Additionally, the above display unit (140) may be a touch screen.

[0041] Additionally, the display unit (140) may include at least one of a Liquid Crystal Display (LCD), a Thin Film Transistor-Liquid Crystal Display (TFT LCD), an Organic Light-Emitting Diode (OLED), a Flexible Display, a 3D Display, an e-ink Display, and a Light Emitting Diode (LED).

[0042] In addition, the display unit (140) displays the collected essential input information, optional input information, etc. under the control of the control unit (160).

[0043] The voice output unit (150) outputs voice information included in a signal processed by the control unit (160). Here, the voice output unit (140) may include a receiver, a speaker, a buzzer, etc.

[0044] In addition, the voice output unit (150) outputs guidance voice generated by the control unit (160).

[0045] Additionally, the voice output unit (150) outputs voice information (or sound information) corresponding to the collected essential input information, optional input information, etc., under the control of the control unit (160).

[0046] The above-mentioned control unit (controller, or MCU (microcontroller unit)) (160) performs the overall control function of the compression fracture occurrence prediction device (100).

[0047] Additionally, the control unit (160) executes the overall control function of the compression fracture occurrence prediction device (100) using the program and data stored in the storage unit (130). The control unit (160) may include RAM, ROM, CPU, GPU, and a bus, and the RAM, ROM, CPU, GPU, etc. may be connected to each other through the bus. The CPU can access the storage unit (130) and perform booting using the O / S stored in the storage unit (130), and can perform various operations using various programs, content, data, etc. stored in the storage unit (130).

[0048] In addition, the control unit (160) utilizes the previously collected essential input information and optional input information as data for continuous machine learning (or deep learning). Here, the input dataset for machine learning can perform training and testing functions by dividing the essential input information and optional input information into a training set and a test set at a preset ratio (e.g., including 7:3, 8:2, etc.). Additionally, the input dataset for machine learning includes essential input information, optional input information, etc., that are collected later. In addition, the output dataset for the machine learning described above is the part to be predicted, which learns from collected essential and optional input information and subsequently predicts the following: the probability of compression fracture occurrence by period, the estimated time remaining until the occurrence of a compression fracture by period, the ROC curve (Receiver Operating Characteristic curve) for compression fractures by period, the AUC value (Area Under Curve value) by period, whether spinal imaging examination is necessary, whether measures for the prevention of osteopenia, osteoporosis, compression fractures, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fractures, etc. is necessary (or whether medication for osteopenia, osteoporosis, compression fractures, etc. is necessary).

[0049] That is, the control unit (160) performs a learning function to classify (or predict / confirm / judge) the following regarding the compression fracture prediction model through pre-set learning data: the probability of compression fracture occurrence by period, the estimated time remaining until compression fracture occurrence by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for prevention of osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary. At this time, the server (200) can store raw data such as essential input information and optional input information in parallel and distributed, refine unstructured data, structured data, and semi-structured data included in the stored raw data (or training data, etc.), perform preprocessing including classification into metadata, perform analysis including data mining on the preprocessed data, and build big data by conducting learning, training, and testing based on at least one type of machine learning. At this time, at least one type of machine learning may be any one of supervised learning, semi-supervised learning, unsupervised learning, reinforcement learning, and deep reinforcement learning, or a combination of at least one of these. And data mining may include classification, which predicts the class of new data by learning a training dataset with known classes by exploring inherent relationships between preprocessed data, or clustering, which groups data based on similarity without class information.

[0050] In addition, the control unit (160) performs a learning function to predict (or generate / classify / confirm) the following for a compression fracture prediction model in relation to a specific raw data and a user: the probability of a compression fracture occurring for each period, the estimated time remaining until a compression fracture occurs for each period, the ROC curve for a compression fracture for each period, the AUC value for each period, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0051] In this way, the control unit (160) performs a learning function on the compression fracture prediction model in the form of a neural network (or a feedforward neural network) through the learning data, etc.

[0052] Additionally, the control unit (160) controls the collection unit (110) to collect the essential input information, the optional input information, etc., based on user input. At this time, the control unit (160) may collect the essential input information, the optional input information, etc., to provide prediction information on when a compression fracture may occur from a general user, a user (or patient) who has undergone a bone density test, a user (or patient) who has been diagnosed with osteopenia, osteoporosis, etc., in a bone density test.

[0053] In addition, the control unit (160) performs artificial intelligence-based machine learning based on the received essential input information, the optional input information, etc., and based on the machine learning results, predicts (or generates / classifies / confirms) the probability of compression fracture occurrence by period, the estimated time remaining until compression fracture occurrence by period, the ROC curve (Receiver Operating Characteristic curve) for compression fracture by period, the AUC value (Area Under Curve value) by period, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary. Here, the periods include 3 months, 6 months, 1 year, 2 years, 3 years, 4 years, etc. In addition, the necessity of the above-mentioned spinal imaging examination, the necessity of measures to prevent osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc. (or information regarding the necessity of the relevant spinal imaging examination, the necessity of measures to prevent osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc.) may be set (or determined) based on the probability of compression fracture occurrence and / or the estimated time remaining until the occurrence of a compression fracture. That is, the necessity of the above-mentioned spinal imaging examination includes information for recommending one of the following examinations: X-ray examination, ultrasound examination, tomography examination (or Computerized Tomography (CT) examination), or Magnetic Resonance Imaging (MRI) (or Nuclear Magnetic Resonance (NMR) examination).In addition, it includes information on whether there is a need for measures to prevent the above-mentioned osteopenia, osteoporosis, compression fractures, etc., and / or whether there is a need to administer a treatment for the above-mentioned osteopenia, osteoporosis, compression fractures, etc., and information for recommending any one of the following: bisphosphonates, female hormone preparations, tissue-selective estrogen complexes (TSECs), selective estrogen receptor modulators (SERMs), RANKL inhibitors, parathyroid hormone preparations, PTHrP analogs, sclerostin inhibitors, activated vitamin D preparations, calcitonin preparations, calcium preparations, vitamin D preparations, vitamin K2 preparations, thoracic spine braces, thoracolumbar spine braces, lumbar spine braces, and lumbosacral braces, and / or a treatment.

[0054] That is, the control unit (160) performs machine learning (or artificial intelligence / deep learning) using the essential input information, the optional input information, etc., as input values ​​for a preset artificial intelligence-based compression fracture prediction model, and based on the machine learning results (or artificial intelligence results / deep learning results), predicts (or generates / classifies / confirms) the probability of compression fracture occurrence by period, the estimated remaining time until compression fracture occurrence by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc., are necessary in relation to the user. At this time, the control unit (160) can calculate (or predict) the probability of compression fracture occurrence by period based on the predicted AUC value (or AUC) for each period.

[0055] Additionally, the control unit (160) outputs (or displays) through the display unit (140) and / or the voice output unit (150) a preset period-by-period probability of compression fracture occurrence related to the predicted (or generated / classified / identified) user, an estimated period remaining until the occurrence of a compression fracture, an ROC curve for compression fractures by period, an AUC value by period, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fractures, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fractures, etc. is necessary.

[0056] In addition, the control unit (160) performs learning on a pre-learned (or set) deep learning algorithm model or machine learning algorithm model using the essential input information, the optional input information, etc. as input values, and predicts (or generates / classifies / confirms) the importance of the result derivation of each element, the odds ratio, the probability of compression fracture occurrence per set period, the estimated time remaining until compression fracture occurrence per period, the ROC curve for compression fracture per period, the AUC value per period, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary. Here, the deep learning algorithm model or machine learning algorithm model includes a Convolutional Neural Network, a Recurrent Neural Network, a Long Short Term Memory Network, a Gated Recurrent Unit, a Feedforward neural network, a linear regression model, a polynomial regression model, a support vector regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, etc.

[0057] Additionally, the control unit (160) outputs (or displays) the importance, odds ratio, preset period-by-period probability of compression fracture occurrence, estimated time remaining until period-by-period compression fracture occurrence, period-by-period ROC curve for period-by-period compression fracture, period-by-period AUC value, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary through the display unit (140) and / or the voice output unit (150).

[0058] In addition, the control unit (160) acquires (or captures / collects) real-time video information related to the user through a camera unit (not shown).

[0059] At this time, the control unit (160) can detect (or measure / collect) the rate of change in tilt (or rate of change in tilt) of the user's body (or physical body), and falls / falls in the up / down direction (or vertical direction) through a sensor unit (not shown) or a smart device worn by the user. Here, the smart device includes a smart watch, a smart band, etc., which includes a sensor unit (not shown) capable of measuring location information, altitude information, speed information, tilt information, and changes in movement.

[0060] Additionally, the control unit (160) analyzes the acquired (or captured / collected) real-time video information to check (or determine) whether a preset event has occurred. Here, the event includes cases where the preset change in tilt (or tilt change) per hour of the user's body exceeds a preset threshold, where the user falls, or where the user falls.

[0061] As a result of the above verification, if the above event occurs, the control unit (160) performs other machine learning based on artificial intelligence based on the acquired (or captured / collected) real-time video information, the above essential input information, etc., and based on the results of other machine learning, predicts (or generates / classifies / confirms) the probability of a compression fracture occurring due to the occurrence of the event in relation to the user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0062] That is, the control unit (160) performs other machine learning (or other artificial intelligence / other deep learning) using the real-time video information, the essential input information, etc. as input values ​​for the pre-set artificial intelligence-based compression fracture prediction model, and based on the other machine learning results (or other artificial intelligence results / other deep learning results), predicts (or generates / classifies / confirms) the probability of a compression fracture occurring according to the event in relation to the user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0063] Additionally, the control unit (160) outputs (or displays) through the display unit (140) and / or the voice output unit (150) the probability of a compression fracture occurring due to the occurrence of the corresponding event related to the predicted (or generated / classified / identified) user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0064] Additionally, if the probability of a compression fracture occurring due to the predicted event exceeds a preset threshold value and / or the remaining estimated period until the occurrence of a compression fracture due to the predicted event is less than or equal to another preset threshold value, the control unit (160) determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user, generates emergency rescue alarm information, and transmits (or provides) the generated emergency rescue alarm information to an emergency rescue server (not shown) (e.g., an emergency rescue 119 server, a nearby hospital server where the user is located, etc.) via the communication unit (120). Here, the emergency rescue alarm information includes address information where the user is located, the real-time video information, the required input information, event details (e.g., falling, tripping, etc.), and the date and time information of the event occurrence.

[0065] Accordingly, the above user may be transported to a hospital, etc., in accordance with the emergency rescue function.

[0066] Additionally, if the probability of a compression fracture occurring due to the predicted event exceeds a preset threshold value and / or the remaining estimated period until the occurrence of a compression fracture due to the predicted event is less than or equal to another preset threshold value, the control unit (160) determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user, selects (or chooses) information of a doctor capable of providing real-time treatment from among a plurality of doctor information pre-registered in the storage unit (130), and may perform a remote medical treatment function through a video call function with a terminal (not shown) possessed by the doctor corresponding to the selected (or chosen) information capable of providing real-time treatment.

[0067] In addition, if the above check result indicates that the above event does not occur, the control unit (160) repeatedly performs the process of acquiring real-time video information and analyzing the real-time video information acquired in real-time to check whether the above preset event has occurred.

[0068] In an embodiment of the present invention, the compression fracture occurrence prediction device (100) can perform various functions in the form of a dedicated app or website (e.g., information collection function, importance of deriving results of each element based on artificial intelligence / analysis, odds ratio, probability of compression fracture occurrence by preset period, estimated time remaining until compression fracture occurrence by period, ROC curve for compression fracture by period, AUC value by period, necessity of spinal imaging examination and osteopenia and osteoporosis, prediction function regarding measures for prevention of compression fracture and necessity of administering therapeutic agents, etc.).

[0069] In this way, based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount, usual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI, training is performed on an artificial intelligence-based compression fracture prediction model to predict the probability of compression fracture occurrence by a preset period, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of measures to prevent osteopenia, osteoporosis, and compression fractures, and the necessity of administering medication for osteopenia, osteoporosis, and compression fractures.

[0070] In addition, based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount, usual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI, training is performed on an artificial intelligence-based compression fracture prediction model to predict the probability of compression fracture occurrence by a preset period, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examination, the necessity of measures to prevent osteopenia, osteoporosis, and compression fractures, and the necessity of administering treatment drugs for osteopenia, osteoporosis, and compression fractures.

[0071] Hereinafter, a method for predicting the occurrence of a compression fracture according to the present invention will be described in detail with reference to FIGS. 1 and 2.

[0072] FIG. 2 is a flowchart illustrating a method for predicting the occurrence of a compression fracture according to an embodiment of the present invention.

[0073] First, the collection unit (110) collects (or receives) pre-set essential input information, optional input information, etc., according to user input (or user / patient selection / touch / control). Here, the essential input information includes gender, height, weight, Body Mass Index (BMI), age, bone density values, etc. Additionally, the optional input information includes diet, weekly exercise amount (or daily / hourly exercise amount), whether there are flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the level (or vertebral segment) where they occurred, muscle mass measured by MRI, presence of underlying diseases, usual smoking amount (or smoking status), usual alcohol intake (or drinking status), blood levels, whether steroids are taken, history of receiving steroid injections, history of cancer, type of cancer, whether cancer has metastasized, degree of cancer healing, angle and length values ​​measured by a spinal imaging diagnostic device (e.g., spinal X-ray, CT, MRI, etc.). At this time, the above diet includes information regarding the presence and amount of intake of milk, anchovies, protein, omega-3 fatty acids, vitamin D, vitamin K2, health foods, and drugs. In addition, the presence of the above underlying disease includes information regarding the presence of diseases such as rheumatoid arthritis, endocrine diseases, and dietary diseases including mental illness. Furthermore, the above blood values ​​include calcium levels, parathyroid hormone levels, thyroid hormone levels, vitamin D levels, vitamin K2 levels, etc.In addition, the values ​​measured by the aforementioned spinal imaging diagnostic device are values ​​measured from spinal X-ray, CT, and MRI, including pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), C7 vertical tilt, L1-S1 lordosis angle, L4-S1 lordosis angle, sagittal vertical axis (including, for example, distance: C7PL to post corner of S1), thoracic kyphosis angle, segmental kyphosis angle, coronal balance, global tilt, relative pelvic version, lordosis distribution index, relative spinopelvic alignment, and relative lumbar lordosis angle. It includes lordosis angle), global alignment and proportion score (Global Alignment and Proportion (GAP) score), etc.

[0074] In addition, the above-mentioned collection unit (110) may receive and collect data transmitted from OCS, EMR, PACS systems, and bone density testing devices used in hospitals, clinics, etc.

[0075] For example, the first collection unit (110) collects first required input information, first optional input information, etc. according to the input of user Hong Gil-dong (S210).

[0076] Subsequently, the control unit (160) performs artificial intelligence-based machine learning based on the received essential input information, the optional input information, etc., and based on the machine learning results, predicts (or generates / classifies / confirms) the probability of compression fracture occurrence by period, the estimated time remaining until compression fracture occurrence by period, the ROC curve (Receiver Operating Characteristic curve) for compression fracture by period, the AUC value (Area Under Curve value) by period, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary. Here, the periods include 3 months, 6 months, 1 year, 2 years, 3 years, 4 years, etc. In addition, the necessity of the above-mentioned spinal imaging examination, the necessity of measures to prevent osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc. (or information regarding the necessity of the relevant spinal imaging examination, the necessity of measures to prevent osteopenia, osteoporosis, compression fractures, etc., and the necessity of administering therapeutic agents for osteopenia, osteoporosis, compression fractures, etc.) may be set (or determined) based on the probability of compression fracture occurrence and / or the estimated time remaining until the occurrence of a compression fracture. That is, the necessity of the above-mentioned spinal imaging examination includes information for recommending one of the following examinations: X-ray examination, ultrasound examination, tomography examination (or Computerized Tomography (CT) examination), or Magnetic Resonance Imaging (MRI) (or Nuclear Magnetic Resonance (NMR) examination).In addition, it includes information on whether there is a need for measures to prevent the above-mentioned osteopenia, osteoporosis, compression fractures, etc., and / or whether there is a need to administer a treatment for the above-mentioned osteopenia, osteoporosis, compression fractures, etc., and information for recommending any one of the following: bisphosphonates, female hormone preparations, tissue-selective estrogen complexes (TSECs), selective estrogen receptor modulators (SERMs), RANKL inhibitors, parathyroid hormone preparations, PTHrP analogs, sclerostin inhibitors, activated vitamin D preparations, calcitonin preparations, calcium preparations, vitamin D preparations, vitamin K2 preparations, thoracic spine braces, thoracolumbar spine braces, lumbar spine braces, and lumbosacral braces, and / or a treatment.

[0077] That is, the control unit (160) performs machine learning (or artificial intelligence / deep learning) using the essential input information, the optional input information, etc., as input values ​​for a preset artificial intelligence-based compression fracture prediction model, and based on the machine learning results (or artificial intelligence results / deep learning results), predicts (or generates / classifies / confirms) the probability of compression fracture occurrence by period, the estimated remaining time until compression fracture occurrence by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc., are necessary in relation to the user. At this time, the control unit (160) can calculate (or predict) the probability of compression fracture occurrence by period based on the predicted AUC value (or AUC) for each period.

[0078] Additionally, the control unit (160) outputs (or displays) through the display unit (140) and / or voice output unit (150) the probability of a compression fracture occurring by a preset period, the estimated time remaining until the occurrence of a compression fracture by a period, the ROC curve for the compression fracture by a period, the AUC value by a period, whether a spinal imaging examination is necessary, whether measures are necessary to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0079] For example, the first control unit (160) performs machine learning using the collected first essential input information, first optional input information, etc. as input values ​​for the artificial intelligence-based compression fracture prediction model, and based on the machine learning results, predicts the probability of a first compression fracture occurring by period related to Hong Gil-dong, the estimated time remaining until the occurrence of a first compression fracture by period, the ROC curve for the first compression fracture by period, the first AUC value by period, whether a first spinal imaging examination is necessary, whether measures to prevent the first osteopenia, osteoporosis, compression fracture, etc., and whether a treatment drug for the first osteopenia, osteoporosis, compression fracture, etc. is necessary, respectively.

[0080] In addition, the first control unit outputs the probability of a compression fracture occurring in the first period for the predicted Hong Gil-dong, the estimated time remaining until the occurrence of a compression fracture in the first period, the ROC curve for the compression fracture in the first period, the AUC value in the first period, whether a first spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary through the first display unit (140) and the first voice output unit (150) (S220).

[0081] In addition, the control unit (160) performs learning on a pre-learned (or set) deep learning algorithm model or machine learning algorithm model using the essential input information, the optional input information, etc. as input values, and predicts (or generates / classifies / confirms) the importance of the result derivation of each element, the odds ratio, the probability of compression fracture occurrence per set period, the estimated time remaining until compression fracture occurrence per period, the ROC curve for compression fracture per period, the AUC value per period, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary. Here, the deep learning algorithm model or machine learning algorithm model includes a Convolutional Neural Network, a Recurrent Neural Network, a Long Short Term Memory Network, a Gated Recurrent Unit, a Feedforward neural network, a linear regression model, a polynomial regression model, a support vector regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, etc.

[0082] Additionally, the control unit (160) outputs (or displays) the importance, odds ratio, preset period-by-period probability of compression fracture occurrence, estimated time remaining until period-by-period compression fracture occurrence, period-by-period ROC curve for period-by-period compression fracture, period-by-period AUC value, whether spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary through the display unit (140) and / or the voice output unit (150).

[0083] For example, the first control unit performs machine learning using the collected first essential input information, first optional input information, etc. as input values ​​for the support vector machine model, and based on the machine learning results, predicts the second importance, second odds ratio, probability of occurrence of the second compression fracture by period, estimated time remaining until occurrence of the second compression fracture by period, ROC curve for the second compression fracture by period, second AUC value by period, whether there is a need for a second spinal imaging examination, whether there is a need for measures to prevent the second osteopenia, osteoporosis, compression fracture, etc., and whether there is a need for medication for the second osteopenia, osteoporosis, compression fracture, etc.

[0084] In addition, the first control unit outputs the second importance, second odds ratio, probability of occurrence of the second compression fracture by period, estimated time remaining until occurrence of the second compression fracture by period, ROC curve for the second compression fracture by period, second AUC value by period, whether a second spinal imaging examination is necessary, whether measures to prevent the second osteopenia, osteoporosis, compression fracture, etc., and whether medication for the second osteopenia, osteoporosis, compression fracture, etc. is necessary through the first display unit and the first voice output unit (S230).

[0085] In addition, the control unit (160) acquires (or captures / collects) real-time video information related to the user through a camera unit (not shown).

[0086] At this time, the control unit (160) can detect (or measure / collect) the rate of change in tilt (or rate of change in tilt) of the user's body (or physical body), and falls / falls in the up / down direction (or vertical direction) through a sensor unit (not shown) or a smart device worn by the user. Here, the smart device includes a smart watch, a smart band, etc., which includes a sensor unit (not shown) capable of measuring location information, altitude information, speed information, tilt information, and changes in movement.

[0087] For example, the first control unit is linked with a first camera unit (not shown) configured in the living room of the house where Hong Gil-dong is located, and obtains first real-time video information within the living room (S240).

[0088] Subsequently, the control unit (160) analyzes the acquired (or captured / collected) real-time video information to check (or determine) whether a preset event has occurred. Here, the event includes cases where the preset change in tilt per hour of the user's body (or body) exceeds a preset threshold, where the user falls, or where the user falls.

[0089] For example, the first control unit analyzes the acquired first real-time video information to check whether the event has occurred (S250).

[0090] As a result of the above verification, if the above event occurs, the control unit (160) performs other machine learning based on artificial intelligence based on the acquired (or captured / collected) real-time video information, the above essential input information, etc., and based on the results of other machine learning, predicts (or generates / classifies / confirms) the probability of a compression fracture occurring due to the occurrence of the event in relation to the user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0091] That is, the control unit (160) performs other machine learning (or other artificial intelligence / other deep learning) using the real-time video information, the essential input information, etc. as input values ​​for the pre-set artificial intelligence-based compression fracture prediction model, and based on the other machine learning results (or other artificial intelligence results / other deep learning results), predicts (or generates / classifies / confirms) the probability of a compression fracture occurring according to the event in relation to the user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0092] Additionally, the control unit (160) outputs (or displays) through the display unit (140) and / or the voice output unit (150) the probability of a compression fracture occurring due to the occurrence of the corresponding event related to the predicted (or generated / classified / identified) user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary.

[0093] Additionally, if the probability of a compression fracture occurring due to the predicted event exceeds a preset threshold value and / or the remaining estimated period until the occurrence of a compression fracture due to the predicted event is less than or equal to another preset threshold value, the control unit (160) determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user, generates emergency rescue alarm information, and transmits (or provides) the generated emergency rescue alarm information to an emergency rescue server (not shown) (e.g., an emergency rescue 119 server, a nearby hospital server where the user is located, etc.) via the communication unit (120). Here, the emergency rescue alarm information includes address information where the user is located, the real-time video information, the required input information, event details (e.g., falling, tripping, etc.), and the date and time information of the event occurrence.

[0094] Accordingly, the above user may be transported to a hospital, etc., in accordance with the emergency rescue function.

[0095] Additionally, if the probability of a compression fracture occurring due to the predicted event exceeds a preset threshold value and / or the remaining estimated period until the occurrence of a compression fracture due to the predicted event is less than or equal to another preset threshold value, the control unit (160) determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user, selects (or chooses) information of a doctor capable of providing real-time treatment from among a plurality of doctor information pre-registered in the storage unit (130), and may perform a remote medical treatment function through a video call function with a terminal (not shown) possessed by the doctor corresponding to the selected (or chosen) information capable of providing real-time treatment.

[0096] For example, when an event occurs in which Hong Gil-dong falls from the living room sofa to the floor through the analysis of the first real-time video information, the first control unit performs machine learning using the first real-time video information, the first essential input information, etc. as input values ​​for the artificial intelligence-based compression fracture prediction model, and based on the machine learning results, predicts the probability of a third compression fracture occurring due to the fall event in relation to Hong Gil-dong, the estimated time remaining until the occurrence of a third compression fracture, the ROC curve for the third compression fracture, the third AUC value, whether a third spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc., and whether there is a need to administer a treatment for osteopenia, osteoporosis, compression fracture, etc.

[0097] In addition, the first control unit outputs, through the first display unit and the first voice output unit, the probability of a third compression fracture occurring due to a fall event in relation to the predicted Hong Gil-dong, the estimated time remaining until the occurrence of the third compression fracture, the ROC curve for the third compression fracture, the third AUC value, whether a third spinal imaging examination is necessary, whether measures to prevent osteopenia, osteoporosis, compression fracture, etc. are necessary, and whether medication for osteopenia, osteoporosis, compression fracture, etc. is necessary (S260).

[0098] As described above, embodiments of the present invention train an AI-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density level, diet, weekly exercise amount, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, habitual smoking amount, habitual alcohol intake, blood levels, steroid use, history of receiving steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI. By doing so, the model predicts the probability of compression fracture occurrence by preset periods, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of preventive measures for osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures. This applies to users who have undergone bone density testing and users diagnosed with osteopenia or osteoporosis based on bone density testing. By providing information on when compression fractures may occur, the user can be encouraged to exercise greater caution and be informed of the timing for X-ray examinations, measures to prevent osteopenia and osteoporosis, and the administration of medication.

[0099] Furthermore, as previously described, the embodiments of the present invention train an AI-based compression fracture prediction model based on input values ​​such as the user's gender, height, weight, body mass index, age, bone density, diet, weekly exercise volume, presence of flat feet, type of shoes frequently worn, history of previous spinal and joint surgery, presence of previous compression fractures and the vertebral segments where they occurred, muscle mass measured by MRI, presence of underlying diseases, habitual smoking amount, habitual alcohol intake, blood levels, steroid use, history of steroid injections, history of cancer, type of cancer, presence of cancer metastasis, degree of cancer healing, and angle and length values ​​measured by spinal X-ray, CT, and MRI. By predicting the probability of compression fracture occurrence by preset periods, the estimated time remaining until the occurrence of a compression fracture, the necessity of spinal imaging examinations, the necessity of preventive measures for osteopenia, osteoporosis, and compression fractures, and the necessity of administering therapeutic agents for osteopenia, osteoporosis, and compression fractures, the invention provides information related to compression fractures in the case of high-risk exercises for users engaging in various types of exercise, thereby enabling the user It is possible to recognize the risk of compression fractures and enable the user to perform exercises suitable for them, and in the case of patients with osteopenia and osteoporosis, to supplement their diet suitable for them.

[0100] A person skilled in the art to which the present invention pertains will be able to make modifications and variations to the foregoing without departing from the essential characteristics of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical concept of the present invention, and the scope of the technical concept of the present invention is not limited by these embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0101] 100: Compression fracture occurrence prediction device 110: Collection unit 120: Communication unit 130: Storage unit 140: Display unit 150: Voice output unit 160: Control unit

Claims

Claim 1 A collection unit that collects at least one of essential input information and optional input information, and collects data received from an OCS, EMR, PACS system and a bone density testing device;The system includes a control unit that performs AI-based machine learning based on at least one piece of information collected above, and predicts, based on the machine learning results, the probability of a compression fracture occurring by a preset period in relation to the user, the estimated time remaining until a compression fracture occurs by a preset period, the ROC curve (Receiver Operating Characteristic curve) for the compression fracture by a preset period, the AUC value (Area Under Curve value) for the compression fracture by a preset period, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture, and whether a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture is necessary. The control unit acquires real-time video information related to the user through a camera unit, analyzes the acquired real-time video information to confirm whether a preset event has occurred, and when the event occurs, performs another AI-based machine learning based on the real-time video information and the essential input information, and based on the other machine learning results, predicts, in relation to the user, the probability of a compression fracture occurring due to the event, the estimated time remaining until a compression fracture occurs, the ROC curve, the AUC value, and whether a spinal imaging examination is necessary. A compression fracture occurrence prediction device characterized by predicting whether there is a need for preventive measures for at least one of osteopenia, osteoporosis, and compression fracture, and whether there is a need for administration of a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture, and controlling the display unit to display, in relation to the predicted user, the probability of compression fracture occurrence due to event occurrence, the estimated time remaining until compression fracture occurrence, the ROC curve for compression fracture, the AUC value, whether spinal imaging examination is necessary, whether there is a need for preventive measures for at least one of osteopenia, osteoporosis, and compression fracture, and whether there is a need for administration of a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture. Claim 2 A compression fracture occurrence prediction device according to claim 1, characterized in that the determination of whether a spinal imaging examination is necessary includes information for recommending one of an X-ray examination, an ultrasound examination, a computed tomography examination, and a magnetic resonance imaging examination according to the probability of the occurrence of the compression fracture. Claim 3 A compression fracture occurrence prediction device according to claim 1, wherein the necessity of measures for preventing at least one of osteopenia, osteoporosis, and compression fracture, and the necessity of administering a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture include information for recommending any one of bisphosphonates, female hormone preparations, tissue-selective estrogen complexes (TSECs), selective estrogen receptor modulators (SERMs), RANKL inhibitors, parathyroid hormone preparations, PTHrP analogs, Sclerostin inhibitors, activated vitamin D preparations, calcitonin preparations, calcium preparations, vitamin D preparations, vitamin K2 preparations, thoracic spine braces, thoracolumbar spine braces, lumbar spine braces, and lumbosacral braces. Claim 4 A step of collecting at least one piece of information among essential input information and optional input information by a collection unit; a step of performing artificial intelligence-based machine learning based on the collected at least one piece of information by a control unit, and predicting, based on the machine learning results, the probability of occurrence of compression fracture by period, the estimated time remaining until occurrence of compression fracture by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether medication for at least one of osteopenia, osteoporosis, and compression fracture is necessary in relation to the user. A step of acquiring real-time video information related to a user through a camera unit by the control unit; a step of analyzing the acquired real-time video information by the control unit to determine whether a preset event has occurred; and when the event occurs, a step of performing other machine learning based on artificial intelligence by the control unit based on the real-time video information and the essential input information, and predicting, based on the other machine learning results, the probability of a compression fracture occurring due to the event in relation to the user, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture. A method for predicting the occurrence of a compression fracture, comprising the step of controlling a display unit by the above-described control unit to display, in relation to the predicted user, the probability of a compression fracture occurring due to the occurrence of an event, the estimated time remaining until the occurrence of a compression fracture, the ROC curve for the compression fracture, the AUC value, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture. Claim 5 A method for predicting the occurrence of a compression fracture according to claim 4, wherein the predicting step comprises performing machine learning using the essential input information and the optional input information as input values ​​for a preset artificial intelligence-based compression fracture prediction model, and, based on the machine learning results, predicting, in relation to the user, the probability of a compression fracture occurring by a preset period, the estimated period remaining until the occurrence of a compression fracture by a period, the ROC curve for the compression fracture by a period, the AUC value by a period, whether a spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether there is a need to administer a therapeutic agent for at least one of osteopenia, osteoporosis, and compression fracture. Claim 6 In claim 4, the control unit performs learning on the essential input information and the optional input information on a pre-learned deep learning algorithm model or machine learning algorithm model to predict the importance of the result derivation of each element, the odds ratio, the preset probability of compression fracture occurrence by period, the estimated time remaining until compression fracture occurrence by period, the ROC curve for compression fracture by period, the AUC value by period, whether spinal imaging examination is necessary, whether measures for the prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether medication for at least one of osteopenia, osteoporosis, and compression fracture is necessary; A method for predicting the occurrence of a compression fracture, further comprising the step of outputting, through at least one of a display unit and a voice output unit, the predicted importance, odds ratio, preset probability of occurrence of a compression fracture by period, estimated time remaining until occurrence of a compression fracture by period, ROC curve for compression fracture by period, AUC value by period, whether spinal imaging examination is necessary, whether measures for prevention of at least one of osteopenia, osteoporosis, and compression fracture are necessary, and whether medication for at least one of osteopenia, osteoporosis, and compression fracture is necessary. Claim 7 A method for predicting the occurrence of compression fractures according to claim 4, wherein the determination of whether the spinal imaging examination is necessary includes information for recommending any one of X-ray examination, ultrasound examination, tomography examination, computerized tomography (CT) examination, magnetic resonance imaging (MRI) and nuclear magnetic resonance (NMR) examination. Claim 8 A method for predicting the occurrence of a compression fracture according to claim 4, wherein the event is any one of the following: when the user's body's tilt change per hour or tilt change exceeds a preset threshold, when the user falls, and when the user falls. Claim 9 A method for predicting the occurrence of a compression fracture according to claim 4, wherein, when at least one of the following applies: the probability of a compression fracture occurring due to the occurrence of the predicted event exceeds a preset threshold value, and the estimated period remaining until the occurrence of a compression fracture due to the occurrence of the predicted event is less than or equal to another preset threshold value, the control unit determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user and generates emergency rescue alarm information; and the control unit further comprises the step of transmitting the generated emergency rescue alarm information to an emergency rescue server through a communication unit. Claim 10 A method for predicting the occurrence of a compression fracture according to claim 4, wherein, when at least one of the cases is met, the probability of a compression fracture occurring due to the occurrence of the predicted event exceeds a preset threshold value and the estimated period remaining until the occurrence of a compression fracture due to the occurrence of the predicted event is less than or equal to another preset threshold value, the control unit determines that emergency rescue or real-time non-face-to-face medical treatment is required for the user and selects medical information capable of real-time treatment from among a plurality of medical information preset in the storage unit; and further comprises the step of performing a remote medical treatment function through a video call function with a terminal possessed by the medical information capable of real-time treatment, by the control unit.

Images