WO2022114819A1 - Dispositif et procédé d'évaluation d'ostéoporose par ia - Google Patents

Dispositif et procédé d'évaluation d'ostéoporose par ia Download PDF

Info

Publication number
WO2022114819A1
WO2022114819A1 PCT/KR2021/017548 KR2021017548W WO2022114819A1 WO 2022114819 A1 WO2022114819 A1 WO 2022114819A1 KR 2021017548 W KR2021017548 W KR 2021017548W WO 2022114819 A1 WO2022114819 A1 WO 2022114819A1
Authority
WO
WIPO (PCT)
Prior art keywords
osteoporosis
reading
user
fluoroscopic image
bone density
Prior art date
Application number
PCT/KR2021/017548
Other languages
English (en)
Korean (ko)
Inventor
원영준
Original Assignee
가톨릭관동대학교산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from KR1020200159472A external-priority patent/KR20220072164A/ko
Priority claimed from KR1020200159465A external-priority patent/KR20220072162A/ko
Priority claimed from KR1020200159462A external-priority patent/KR20220072159A/ko
Priority claimed from KR1020200159474A external-priority patent/KR20220072166A/ko
Application filed by 가톨릭관동대학교산학협력단 filed Critical 가톨릭관동대학교산학협력단
Publication of WO2022114819A1 publication Critical patent/WO2022114819A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the present invention relates to an AI osteoporosis reading apparatus and method, and more particularly, to an AI osteoporosis reading apparatus and method for determining whether a patient suspected of osteoporosis has osteoporosis through AI learning based on a fluoroscopic image of the osteoporotic patient.
  • bone mineral density BMD
  • osteoporosis in which a hole is formed in the bone. If you suffer from osteoporosis, even a small impact causes very serious consequences, such as broken bones and not easily joining broken bones, so periodic diagnosis is required to prevent osteoporosis or worsen symptoms.
  • Osteoporosis refers to a state in which bone mass is excessively reduced compared to that of a normal person, and is a clinical condition accompanied by fractures and deformation of bone morphology. That is, osteoporosis is a condition in which bone mass is abnormally reduced and is a pathological condition accompanied by fractures of the spine and femur, and bone deformation.
  • Conventional bone density information may be obtained by a medical professional reading a picture through a patient, such as an X-ray or ultrasound.
  • a medical professional who reads a fluoroscopic picture of a patient has low skill level, there is a risk of misdiagnosis.
  • AI artificial intelligence
  • the need to accurately determine whether a patient has osteoporosis by reading a fluoroscopic picture of the patient is emerging.
  • An object of the present invention is to provide an AI osteoporosis reading apparatus and method capable of determining whether a suspected patient has osteoporosis through AI learning based on a fluoroscopic image of the osteoporotic patient.
  • Another technical object of the present invention is to provide an AI osteoporosis reading apparatus and method capable of checking changes in bone density by storing fluoroscopic images for each user according to time series, and determining whether or not osteoporosis is based thereon.
  • a database in which fluoroscopic images of a plurality of osteoporosis patients are stored; Measuring unit for measuring the fluoroscopic image of the suspected patient; And based on the reading range, which is a specific region for measuring bone density in the fluoroscopic image stored in the database, through AI learning, the bone density reading information for the osteoporosis patient is generated, and the bone density reading information and the suspicious patient received from the measurement unit Comparing the bone density of the fluoroscopic images of the determining unit to determine whether the suspected patient has osteoporosis; Including, wherein the reading range includes a wrist portion of the fluoroscopic image, it provides an AI osteoporosis reading device.
  • the reading range includes: a) a range including the distal portion of the wrist, the inner area of the radius and steel frame, and the outer area of the radius and steel frame, b) a range including at least one of the middle portion of the clavicle, the distal portion and the lower portion of the humerus head; c) at least one of mandibular cortical bone thickness and mandibular medullary bone density; device can be provided.
  • the reading range may provide an AI osteoporosis reading device in which a preset area A is set from a preset tooth position of the mandible, but is set to an area excluding the teeth.
  • the read information in the fluoroscopic image of the osteoporosis patient, characterized in that at least one of the preset area A and oral length data is used, wherein the oral length data includes the mandible and the total length of the teeth, the tooth head and the gums. It is possible to provide an AI osteoporosis reader, which is any one of the length and the length of the eroded gum.
  • the database may include: a data unit for each user in which a perspective image for each user is stored; and a learning data unit for storing fluoroscopy images of a plurality of osteoporosis patients, wherein the data unit for each user stores fluoroscopic images for each user measured by the measurement unit for each user, providing an AI osteoporosis reading device can do.
  • the determination unit by comparing the fluoroscopic image for a specific user received from the measurement unit with the fluoroscopic image for the specific user pre-stored in the user-specific data unit, provides an AI osteoporosis reading device capable of confirming the state change of bone density can do.
  • storing a plurality of fluoroscopic images of osteoporosis patients in a database generating, by a determination unit, bone density reading information on bone density of an osteoporosis patient through AI learning based on a reading range that is a specific region for measuring bone density in the fluoroscopic image stored in the database; measuring, by the measuring unit, a fluoroscopic image of the suspected patient; storing the fluoroscopy image of the suspected patient measured by the measurement unit in the database; Comprising the step of determining, by the determination unit, the bone density reading information and the bone density of the fluoroscopic image of the suspected patient received from the measurement unit to determine whether the suspected patient has osteoporosis, wherein the reading range includes the wrist portion of the fluoroscopic image It provides a method for reading AI osteoporosis, characterized in that.
  • the reading range includes: a) a range including the distal portion of the wrist, the inner area of the radius and steel frame, and the outer area of the radius and steel frame, b) a range including at least one of the middle portion of the clavicle, the distal portion and the lower portion of the humerus head; c) at least one of mandibular cortical bone thickness and mandibular medullary bone density; method can be provided.
  • the reading range may provide an AI osteoporosis reading method in which a preset area A is set from a preset tooth position of the mandible, but is set to an area excluding the teeth.
  • the read information in the fluoroscopic image of the osteoporosis patient, characterized in that at least one of the preset area A and oral length data is used, wherein the oral length data includes the mandible and the total length of the teeth, the tooth head and the gums. It is possible to provide an AI osteoporosis reading method, which is any one of the length and the length of the eroded gum.
  • the database may include: a data unit for each user in which a perspective image for each user is stored; and a learning data unit for storing fluoroscopy images of a plurality of osteoporosis patients, wherein the storing of the fluoroscopic images of the suspected patient measured by the measurement unit in the database includes, in the data unit for each user, the fluoroscopy images measured by the measurement unit It is possible to provide an AI osteoporosis reading method, further comprising the step of storing the fluoroscopy image for each user for each user.
  • the determination unit After the step of storing the fluoroscopy image of the suspected patient measured by the measurement unit in the database, the determination unit, the fluoroscopy image for the specific user received from the measurement unit, is pre-stored in the data unit for each user. Comparing with the image, it is possible to provide an AI osteoporosis reading method further comprising the step of confirming a change in the state of bone density.
  • the change in bone density can be checked, and based on this, it is possible to determine whether osteoporosis is present.
  • FIG. 1 is a view showing an AI osteoporosis reading apparatus according to an embodiment of the present invention.
  • FIGS. 2 to 6 are diagrams illustrating a reading range in a perspective image according to various embodiments of the present invention.
  • FIG. 7 is a flowchart illustrating an AI osteoporosis reading method according to an embodiment of the present invention.
  • FIG. 8 is a flowchart of storing information for each user according to an embodiment of the present invention.
  • FIGS. 2 to 6 are diagrams showing a reading range in a fluoroscopic image according to various embodiments of the present invention.
  • the AI osteoporosis reading apparatus 10 may include a measurement unit 102 , a control unit 104 , a determination unit 106 , a communication unit 108 , and a database 110 .
  • the measurement unit 102 may generate a fluoroscopic image necessary for reading bone density by photographing the user.
  • the user is a person who takes a fluoroscopic image through the AI osteoporosis reading device 10 , and may mean an osteoporosis patient or a patient suspected of osteoporosis.
  • the measurement unit 102 may generate a fluoroscopic image necessary for reading a patient's bone density, and may provide it to the control unit 104 .
  • the fluoroscopic image is a photograph for reading bone density, and may include at least one of an X-ray image, a tomography image, an ultrasound image, and a magnetic resonance image (MRI).
  • the tomography may be, for example, CT.
  • the magnetic resonance may be, for example, nuclear magnetic resonance (NMR).
  • the measurement unit 102 may check a reading range before taking a fluoroscopic image, and may take a fluoroscopic image for the reading range.
  • the reading range is a specific area for measuring bone density in the fluoroscopic image, and may include, for example, at least one of a distal part of the wrist, an inner area of a radius and a steel frame, and an outer area of a radius and a steel frame.
  • the reading range is described as including at least one of the distal part of the wrist, the inner area of the radius and the steel frame, and the outer area of the radius and the steel frame, but is not limited thereto, and the reading range can be set to various parts according to the user's setting. of course there is
  • the reading range may belong to one of a range of 1.5 cm to 2.5 cm of the distal wrist and a range of 3 cm to 5 cm between the radius and the steel frame.
  • the reading range is in the range of 0.5 cm to 1.5 cm of the middle part of the clavicle, the range of 0.5 cm to 1.5 cm of the distal part by dividing the distal part into 3 parts, and the range of 0.5 cm to 1.5 cm of the part located far from the ribs, and 0.5 cm of the lower part of the humerus head to 1.5 cm.
  • the reading range may be set to an area excluding teeth while setting a preset area A from a preset tooth position of the mandible.
  • the reading range may be any one of L2, L3, and L4 of the thoracolumbar region.
  • the reading range may include at least one of the femoral neck and the femoral electron load among the thighs, or an angle formed by the femoral neck and the femoral electron load.
  • the controller 104 may control the AI osteoporosis measuring device 10 . Specifically, the control unit 104 may control the measurement unit 102 to take a fluoroscopic image of the user. Also, the control unit 104 may control the communication unit 106 to receive external information through communication with an external device and to transmit a see-through image.
  • the determination unit 106 may generate bone density reading information on the bone density of the osteoporosis patient through AI learning based on the reading range, which is a specific region for measuring bone density in the fluoroscopic image stored in the database 110 .
  • the determination unit 106 may be referred to as, for example, a “processor”.
  • the determination unit 106 may be implemented using at least one of a server, a computer, a PCB, a logic circuit, and a laptop.
  • the determination unit 106 may include an AI engine.
  • the AI engine may be referred to as an “artificial intelligence engine”.
  • the AI engine may perform AI learning using machine learning and/or deep learning.
  • AI learning may mean generating bone density reading information for the osteoporosis patient's bone density based on the reading range in the fluoroscopic image stored in the database 110 .
  • the bone density reading information may refer to information about a numerical value or an image about the bone density of an osteoporosis patient.
  • the determination unit 106 may determine whether the suspected patient has osteoporosis using the user medical history information, the bone density reading information, and the fluoroscopic image of the suspected patient.
  • the user medical history information may mean an individual's osteoporosis and fracture risk factors, disease history, drug history, personal genetic information, and the like.
  • the determination unit 106 compares and analyzes the bone density reading information generated through AI learning and the fluoroscopic image of the suspected patient measured by the measurement unit 102, but different reference points for each user through the user history information of the suspected patient established, and using this, it is possible to determine whether the suspected patient has osteoporosis.
  • the determination unit 106 may predict a change in the bone density state of the user by AI learning the change in the state of bone density for each user. Specifically, the determination unit 106 compares the perspective image of the specific user received from the measurement unit 102 with the perspective image of the specific user pre-stored in the database unit 110 for the specific user (ie, the measurement unit 102). A change in the state of bone density of a specific user of the fluoroscopy image received from In addition, the determination unit 106 AI learns the change in the fluoroscopic image for each user stored in the database 110, and compares the change in the bone density state of the specific user identified above with the AI learning result to predict the change in the bone density state of the user. can
  • the communication unit 108 may communicate with an external organization (not shown).
  • the external organization refers to an external device of an external organization, and may mean a server, system, or terminal of the external organization.
  • the communication unit 108 may transmit and receive the first signal S1 to and from an external organization.
  • the first signal S1 may include data.
  • the first signal S1 may include a fluoroscopic image necessary for reading osteoporosis.
  • the database 110 may store data necessary for the AI osteoporosis reading device 10 .
  • the database 110 may store fluoroscopic images of a plurality of osteoporosis patients.
  • the fluoroscopic images of the plurality of osteoporosis patients may include a fluoroscopic image taken through the measurement unit 102 as well as a fluoroscopic image received from the outside through the communication unit 108 .
  • the database 110 may include a data unit for each user and a learning data unit.
  • the data unit for each user stores the perspective images for each user in time series, and the perspective images for each user measured by the measurement unit 102 may be continuously updated for each user.
  • the learning data unit may store fluoroscopic images of a plurality of osteoporosis patients.
  • FIG. 7 is a flowchart illustrating an AI osteoporosis reading method according to an embodiment of the present invention.
  • the AI osteoporosis reading device 10 stores the fluoroscopic image of the osteoporosis patient ( S302 ). Specifically, the AI osteoporosis reading device 10 may store a fluoroscopic image of an osteoporosis patient. The AI osteoporosis reading device 10 may receive and store a fluoroscopic image of an osteoporosis patient from an external device, or may be captured and stored by the measurement unit 102 .
  • the AI osteoporosis reading device 10 may store the fluoroscopy image of the suspected patient captured through the measurement unit 102 in a pre-stored fluoroscopy image storage location of the user. Specifically, the AI osteoporosis reading apparatus 10 may store the fluoroscopic images for each user measured by the measurement unit 102 in the database 110 for each user.
  • the AI osteoporosis reading device 10 may check the bone density state change of the suspected patient.
  • the AI osteoporosis reading device 10 is a fluoroscopic image received from the measurement unit 102 when the previous fluoroscopy image of a specific user of the fluoroscopic image received from the measurement unit 102 is pre-stored in the database 110 .
  • a change in the state of bone density can be confirmed by comparing the fluoroscopic image of a specific user with the pre-stored fluoroscopy image.
  • the AI osteoporosis reading device 10 may confirm the reading range in the fluoroscopic image of the suspected patient (S308). Specifically, the AI osteoporosis reading device 10 may identify a position corresponding to the bone density reading information in the fluoroscopic image of the suspected patient.
  • the AI osteoporosis reading device 10 may determine whether the suspected patient has osteoporosis. Specifically, the AI osteoporosis reading device 10 may determine whether the suspected patient has osteoporosis by comparing the reading range confirmed in the fluoroscopic image of the suspected patient with the bone density reading information generated by AI learning.
  • the AI osteoporosis reading device 10 may receive basic user information ( S402 ).
  • the user basic information is basic information of a user who uses the AI osteoporosis reading device 10, and may mean basic information of the user, such as name, age, gender, blood type, and the like.
  • the AI osteoporosis reading apparatus 10 may determine whether the received user information is pre-stored information (S404). Specifically, the AI osteoporosis reading device 10 stores the user's basic information together with the fluoroscopic image of the user from the measurement unit 102 , and may check whether the corresponding user basic information is pre-stored information.
  • the AI osteoporosis reading apparatus 10 may store the user's fluoroscopy image measured by the measurement unit 102 together with the user's pre-stored fluoroscopic image (S406).
  • the AI osteoporosis reading device 10 stores fluoroscopy images for each user, so that changes in bone density of the patient can be checked in time series.

Abstract

La présente invention peut fournir un dispositif d'évaluation d'ostéoporose par IA comprenant : une base de données dans laquelle sont stockées des images fluoroscopiques d'une pluralité de patients atteints d'ostéoporose ; une unité de mesure servant à effectuer des mesures sur des images fluoroscopiques d'un patient présumé ; et une unité de détermination qui génère des informations d'évaluation de densité minérale osseuse concernant la densité minérale osseuse d'un patient atteint d'ostéoporose par apprentissage IA sur la base d'une plage d'évaluation qui est une partie spécifique dans laquelle la densité minérale osseuse est mesurée dans les images fluoroscopiques stockées dans la base de données, et qui compare les informations d'évaluation de densité minérale osseuse à la densité minérale osseuse des images fluoroscopiques du patient présumé reçues de l'unité de mesure, la plage d'évaluation comprenant une partie poignet des images fluoroscopiques.
PCT/KR2021/017548 2020-11-25 2021-11-25 Dispositif et procédé d'évaluation d'ostéoporose par ia WO2022114819A1 (fr)

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
KR1020200159472A KR20220072164A (ko) 2020-11-25 2020-11-25 Ai 골다공증 판독장치 및 방법
KR1020200159465A KR20220072162A (ko) 2020-11-25 2020-11-25 Ai 골다공증 판독장치 및 방법
KR10-2020-0159462 2020-11-25
KR1020200159462A KR20220072159A (ko) 2020-11-25 2020-11-25 Ai 골다공증 판독장치 및 방법
KR1020200159474A KR20220072166A (ko) 2020-11-25 2020-11-25 Ai 골다공증 판독장치 및 방법
KR10-2020-0159474 2020-11-25
KR10-2020-0159472 2020-11-25
KR10-2020-0159465 2020-11-25

Publications (1)

Publication Number Publication Date
WO2022114819A1 true WO2022114819A1 (fr) 2022-06-02

Family

ID=81754798

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2021/017548 WO2022114819A1 (fr) 2020-11-25 2021-11-25 Dispositif et procédé d'évaluation d'ostéoporose par ia

Country Status (1)

Country Link
WO (1) WO2022114819A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20170016778A (ko) * 2015-08-04 2017-02-14 재단법인 아산사회복지재단 심층신경망을 이용한 골 연령 산출방법 및 프로그램
US20180247020A1 (en) * 2017-02-24 2018-08-30 Siemens Healthcare Gmbh Personalized Assessment of Bone Health
US20190239843A1 (en) * 2014-07-21 2019-08-08 Zebra Medical Vision Ltd. Systems and methods for prediction of osteoporotic fracture risk
KR20200015379A (ko) * 2018-08-03 2020-02-12 고려대학교 산학협력단 인공지능 기반의 치과방사선사진을 이용한 골밀도 예측시스템 및 이에 의한 골밀도 예측 방법
KR20200085470A (ko) * 2019-01-07 2020-07-15 주식회사 씨아이메디칼 Ai 골밀도 판독 장치 및 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190239843A1 (en) * 2014-07-21 2019-08-08 Zebra Medical Vision Ltd. Systems and methods for prediction of osteoporotic fracture risk
KR20170016778A (ko) * 2015-08-04 2017-02-14 재단법인 아산사회복지재단 심층신경망을 이용한 골 연령 산출방법 및 프로그램
US20180247020A1 (en) * 2017-02-24 2018-08-30 Siemens Healthcare Gmbh Personalized Assessment of Bone Health
KR20200015379A (ko) * 2018-08-03 2020-02-12 고려대학교 산학협력단 인공지능 기반의 치과방사선사진을 이용한 골밀도 예측시스템 및 이에 의한 골밀도 예측 방법
KR20200085470A (ko) * 2019-01-07 2020-07-15 주식회사 씨아이메디칼 Ai 골밀도 판독 장치 및 방법

Similar Documents

Publication Publication Date Title
WO2018056544A1 (fr) Système de réalité augmentée destiné à la chirurgie dentaire et son procédé de mise en œuvre
McClelland et al. Fracture healing assessment comparing stiffness measurements using radiographs.
CN102985940B (zh) 读影委托系统、读影委托中介装置及读影结果评价方法
Chandhanayingyong et al. Teleconsultation for emergency orthopaedic patients using the multimedia messaging service via mobile phones
WO2021107688A2 (fr) Système de recommandation de service médical basé sur un indice quantitatif
CN103258111A (zh) 形成结构化的图像辅助的诊断
WO2010128818A2 (fr) Procédé et système de traitement d'images médicales
Larson et al. Accuracy of diagnosis of subtle chest disease and subtle fractures with a teleradiology system.
WO2019198981A1 (fr) Procédé d'analyse d'état de santé et de fourniture d'informations sur la base d'une image capturée, dispositif et support d'enregistrement associés
WO2022114819A1 (fr) Dispositif et procédé d'évaluation d'ostéoporose par ia
JP2006107134A (ja) 健康情報提供システム及びその方法
US20030229521A1 (en) Method for the administration of medical patient data
JP2005122237A (ja) 検査装置管理方法及びサーバ装置
JP2001128958A (ja) アイリス・アイリドロジ健康診断管理システム
EP1402811A1 (fr) Appareil de distribution d'image
EP4020493A2 (fr) Dispositif d'évaluation de la gravité et procédé de génération de modèle
Tachakra Level of diagnostic confidence, accuracy, and reasons for mistakes in teleradiology for minor injuries
WO2018147674A1 (fr) Appareil et procédé de diagnostic d'état médical sur la base d'une image médicale
US20040059606A1 (en) Method and apparatus for checking codings in the health service
KR20220072164A (ko) Ai 골다공증 판독장치 및 방법
JP2003256570A (ja) 画像診断支援システム
CN113409921A (zh) 医学成像装置、系统和方法
KR20220072162A (ko) Ai 골다공증 판독장치 및 방법
WO2016080736A1 (fr) Procédé pour fournir des informations d'aide au diagnostic par utilisation d'images médicales et son système
Franken Jr et al. Added value of radiologist consultation to family practitioners in the outpatient setting.

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21898645

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21898645

Country of ref document: EP

Kind code of ref document: A1