WO2020180135A1 - 뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 - Google Patents
뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 Download PDFInfo
- Publication number
- WO2020180135A1 WO2020180135A1 PCT/KR2020/003131 KR2020003131W WO2020180135A1 WO 2020180135 A1 WO2020180135 A1 WO 2020180135A1 KR 2020003131 W KR2020003131 W KR 2020003131W WO 2020180135 A1 WO2020180135 A1 WO 2020180135A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- brain
- tissue image
- information
- disease
- learning
- Prior art date
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
- A61B5/004—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
- A61B5/0042—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part for the brain
Definitions
- the embodiment relates to an apparatus and method for predicting deep learning-based brain diseases, such as Alzheimer's disease.
- Alzheimer's disease is a neurodegenerative brain disease that affects many of the elderly and is clinically characterized by loss of cognitive function. As the life expectancy increases due to the development of medicine, Alzheimer's disease tends to increase, and as a result, the need for early diagnosis and predictive technology for appropriate measures such as slowing the progression of Alzheimer's disease is increasing. have.
- MRI which is a non-invasive method of brain imaging, does not have the effect of artificial shading by the skull, compared to Computed Tomography (CT), which is the same non-invasive method, so micro-diagnosis of cerebral perfusion is possible.
- CT Computed Tomography
- Alzheimer's disease can be a decisive factor for treatment before the point of time when a patient's cognitive function is rapidly deteriorated and daily life becomes difficult.
- Alzheimer's disease causes symptoms of atrophy of the patient's brain over time
- research is being conducted to detect and diagnose the patient's condition from brain images.
- diagnosing diseases using brain images since all brain structures between patients are different, in order to uniformly check their common degree of change, convert them to the same structure using a template, and check the degree of change for each anatomical area. Method can be used.
- MMSE Mini-Mental State Examination
- ADAS Alzheimer's Disease Assessment Score
- CDR Clinical Dementia Rating
- An object of the embodiment is to provide a brain disease prediction apparatus and method for more reliably predicting the progression of brain diseases such as Alzheimer's disease, and a learning device for predicting brain diseases.
- the brain disease prediction apparatus includes a data collection unit for collecting a plurality of brain images at different viewpoints, a data processing unit for extracting information on a brain tissue image of an ROI from among the brain images, and a plurality of the different viewpoints.
- the progression of the brain disease, the current progression rate of the brain disease, the future progression rate of the brain disease, and the brain disease by using information of a plurality of brain tissue images of the region of interest extracted from the brain image of It may include a diagnostic unit that predicts at least one or more information of clinical diagnostic evaluation scores according to the condition.
- the data processing unit includes a brain position alignment unit that processes the plurality of brain images so that the positions of the brain are aligned in the plurality of brain images, and a brain tissue image of the region of interest in the plurality of brain images in which the brain positions are aligned
- a brain tissue image extracting unit for extracting, and a removal unit for removing a region other than the brain tissue image of the region of interest from the plurality of brain images, and information on the brain tissue using the brain tissue image of the region of interest It may include an information extraction unit for extracting.
- the brain tissue image of the region of interest may include at least one of a middle temporal gyrus image, an enorhinal cortex image, a Fusiform gyrus, and a hippocampus image. have.
- the learning model may include a plurality of learning models each trained on a normal state, a mild cognitive impairment state, and an Alzheimer's disease state according to the information of the brain tissue image.
- the training data for pre-training the learning model in the diagnosis unit includes first training data including first brain tissue image information acquired at a first time point and a first diagnostic evaluation score for the first brain tissue image information, and , Second learning data including second brain tissue image information obtained at a second time point and a second diagnostic evaluation score for the second brain tissue image information, and third brain tissue image information obtained at a third time point It may include third learning data including a third diagnostic evaluation score for the third brain tissue image information.
- the training data for pre-training the learning model in the diagnosis unit may include data in which each of the first training data, the second training data, and the third training data, or a combination of two or more training data.
- the data collection unit may further collect clinical diagnosis evaluation score information according to the state of the brain tissue, and the diagnosis unit may use information of the brain tissue image and the clinical diagnosis evaluation score information as learning data of the learning model. have.
- the diagnosis unit may predict the brain tissue image information at a previous time point or the brain tissue image information at a later time point using the learning model.
- the diagnosis unit may reuse the brain tissue image information at a previous time point or the brain tissue image information at a later time point to be used as input data of the learning model.
- the learning model includes an encoding module unit and a decoating module unit, and the diagnosis unit receives the information of the brain tissue image as an input of the encoding module and provides information on the volume change amount, shape change amount, shape change amount, and position change amount of the brain tissue image. Extracting a feature that includes, and using the feature as an input of the decoding module unit, at least one or more information of a progression of a brain disease, a current progression of the brain disease, a later progression of the brain disease, and a clinical diagnosis evaluation score It is predictable.
- the brain disease prediction method includes the steps of collecting data of a plurality of brain images at different viewpoints, extracting information of brain tissue images of a region of interest among the brain images, and Brain disease progression, current progression rate of the brain disease, future progression rate of the brain disease, and the brain by using information of a plurality of brain tissue images of a region of interest extracted from a plurality of brain images as input to a pre-learned learning model. It may include predicting at least one or more information of the clinical diagnosis evaluation score according to the disease state.
- the learning device includes a learning data collection unit for collecting learning data, a progression of a brain disease, a current progression rate of the brain disease, a later progression rate of the brain disease, and a clinical trial according to the state of the brain disease. It may include a learning model that is trained using the training data to predict at least one or more of the diagnostic evaluation scores.
- the learning model may be any one of RNN, CNN, and DNN, or a combination thereof.
- the learning data collection unit includes first training data including first brain tissue image information acquired at a first time point and a first diagnostic evaluation score for the first brain tissue image information, and second brain tissue acquired at a second time point. Second learning data including tissue image information and a second diagnostic evaluation score for the second brain tissue image information, and third brain tissue image information acquired at a third time point and the third brain tissue image information 3 It may include third learning data including diagnostic evaluation scores.
- the learning data collection unit may collect data in which each of the first learning data, the second learning data, and the third learning data, or a combination of two or more learning data.
- the embodiment has an effect of diagnosing Alzheimer's disease more reliably using past brain image data, and providing significant information on the analysis and the basis for the diagnosis.
- the embodiment is a diagnostic method that can help the clinician's decision-making.
- the effect of a reliable diagnosis by synthesizing the predicted value of the progression probability of Alzheimer's disease and the clinical diagnostic evaluation score through clinical diagnostic evaluation scores. There is.
- FIG. 1 is a block diagram showing the configuration of a brain disease device according to an embodiment.
- FIG. 2 is a block diagram showing information collected by a data collection unit of a brain disease device according to an embodiment.
- FIG. 3 is a diagram illustrating a brain image collected by a data collection unit of a brain disease device according to an embodiment.
- FIG. 4 is a diagram showing diagnostic evaluation score information collected by a data collection unit of a brain disease device according to an embodiment.
- FIG. 5 is a block diagram showing the configuration of a data processing unit of a brain disease device according to an embodiment.
- FIG. 6 is a diagram illustrating a brain tissue extracted from a data processing unit of a brain disease device according to an embodiment.
- FIG. 7 is a diagram showing a learning model of a brain disease device according to an embodiment.
- FIG. 8 is a diagram showing a plurality of learning models of a brain disease device according to an embodiment.
- FIG. 9 is a diagram illustrating a mathematical modeling of a learning model of a brain disease device according to an embodiment.
- FIG. 10 is a diagram showing the performance results of the brain disease device according to the embodiment.
- FIG. 11 is a block diagram showing a brain disease method according to an embodiment.
- a component when connected to or is referred to as being connected to another component, it should be understood that it may be directly connected or connected to the other component, but other components may exist in the middle.
- FIG. 1 is a block diagram showing the configuration of a brain disease device according to an embodiment
- FIG. 2 is a block diagram showing information collected by a data collection unit of a brain disease device according to an embodiment
- FIG. 3 is a brain according to the embodiment
- FIG. 4 is a diagram showing diagnostic evaluation score information collected by a data collection unit of a brain disease device according to an embodiment
- FIG. 5 is a brain according to an embodiment.
- FIG. 6 is a diagram showing a brain tissue extracted from a data processing unit of a brain disease device according to an embodiment
- FIG. 7 is a learning model of a brain disease device according to an embodiment.
- FIG. 8 is a diagram illustrating a plurality of learning models of a brain disease device according to an embodiment
- FIG. 9 is a diagram illustrating a mathematical modeling of a learning model of a brain disease device according to an embodiment
- FIG. 10 Is a diagram showing the performance results of the brain disease device according to the embodiment.
- a brain disease prediction apparatus 1000 may include a data collection unit 100.
- the data collection unit 100 may collect brain images captured at different viewpoints of the same object. As shown in FIG. 2, the data collection unit 100 may include brain image information 110 and diagnostic evaluation score information 120. The data collection unit 100 may collect only the brain image information 110.
- the brain image may be an image taken from MRI.
- the brain image may include a brain image 10 taken at a first time point, a brain image 20 taken at a second time point, and a brain image 30 taken at a third time point, but the number is limited. It doesn't work.
- the first time point may be one year ago
- the second time point may be two years ago
- the third time point may be three years ago.
- the brain image of the current viewpoint may be further included.
- the diagnosis evaluation score information 120 may be a diagnosis evaluation score for a corresponding brain image.
- the diagnostic evaluation score may include pMCI-early, pMCI-late, and sMCI, and may further include MMSE, ADAS-cog 11, and ADAS-cog13, but is not limited thereto.
- the apparatus 1000 for predicting a brain disease may further include a data processing unit 200.
- the data processing unit 200 may perform image processing for obtaining brain volume information through a matching process and a brain tissue segmentation process.
- the data processing unit 200 may extract a sample having a process of obtaining an arbitrary sample set after performing a task having a voxel value normalized to the total volume of the brain for each region of interest by dividing an anatomical region from the brain volume information. .
- the data processing unit 200 corrects the symmetry of the input image and then removes a brain region that is not required for analysis.
- the brain tissue is separated. After converting all brain tissues into the same template space, a density map of the brain is obtained through a matching process. The obtained brain density map is divided into regions of interest for diagnosis by anatomical region, and then normalized to match the distribution of data values.
- the data processing unit 200 aligns the brain position, removes the skull and cerebellum, separates and matches the brain tissue, and separates a region of interest having volume information of the brain.
- the preprocessor extracts a sample by extracting volume information using a random sampling technique from a value obtained by normalizing the volume information of the brain separated for each region of interest to the total volume of the brain.
- the data processing unit 200 may separate a region of interest having volume information of the brain by using a distribution of several brain template information.
- a region of interest having brain volume information may be separated through distribution of brightness based on template information such as location information of brain tissues and brightness information of each brain tissue.
- the data processing unit 200 includes a brain position alignment unit 210, a brain tissue image extracting unit 220, a removal unit 230, and an information extracting unit 240.
- the brain position alignment unit 210 may process the brain image so that the brain position is aligned in the brain image.
- the brain position alignment unit 210 may align the brain position through a process of matching the brain image.
- the brain position alignment unit 210 may align the brain position by matching the symmetry of the brain image.
- the brain tissue image extractor 220 may extract a brain tissue image of an ROI from the brain image.
- Brain tissue may be a biomarker associated with Alzheimer's disease.
- an increase in Amyloid ⁇ and abnormal tau levels was observed in the renal cortical region, and the middle temporal gyrus tissue 42 of the brain 40 of FIG. 6A and the internal olfactory sense of FIG. 6B
- a decrease in volumes was observed in the cortical (entorhinal cortex) tissue 44 and the fusiform gyrus tissue 46, and in the hippocampus tissue 48 of FIG. 6C.
- the brain tissue image extraction unit 220 is one or more of a middle temporal gyrus tissue image, an endoorhinal cortex tissue image, a fusiform gyrus tissue image, and a hippocampus tissue image. Can be extracted.
- the removal unit 230 may remove a region of the brain image other than the brain tissue image of the region of interest. That is, the removal unit 230 may remove images of brain tissues such as the skull and cerebellum.
- the information extracting unit 240 may extract information on a brain tissue by using an image of a brain tissue of a region of interest.
- the information may include various pieces of information on brain tissue, and as an example, may include volume information. This makes it possible to acquire an image of the brain tissue of the specimen.
- the brain disease prediction apparatus 1000 may further include a diagnosis unit 400.
- the diagnosis unit 400 uses the learning model 300 learned in advance by inputting the information of the brain tissue image, the progression of the brain disease, the current progression speed of the brain disease, the future progression speed of the brain disease, and clinical At least one or more of the diagnostic evaluation scores may be predicted.
- the diagnosis unit 400 may diagnose Alzheimer's disease using this information.
- the learning model 300 may be disposed inside the diagnosis unit 400 or may be disposed in a space separate from the diagnosis unit 400.
- the learning model 300 is trained by learning data and labels using one or a combination of neural networks such as RNN (Recurrent Neural Network), CNN (Convolution Neural Network), DNN (Deep Neural Network), etc. Can be used by 400.
- RNN Recurrent Neural Network
- CNN Convolution Neural Network
- DNN Deep Neural Network
- the learning model 300 may include a learning data collection unit that collects learning data, but is not limited thereto.
- the learning model 300 may be a separate learning device configured with a learning data collection unit.
- the learning model 300 may include an encoding module 310 and a decoding module 320.
- the encoding module 310 may extract features including information on the brain tissue image by inputting information on the brain tissue image.
- features such as volume change, shape change, shape change, and position change of brain tissue may be extracted.
- the brain tissue image information acquired by time is used as the information of the brain tissue image, the amount of change therefor can be measured.
- the learning model 300 may be trained using a brain image of the same object in a predetermined cycle and a clinical diagnosis evaluation score obtained from the same object on the same day at the time of acquisition of each brain image.
- the learning model 300 includes first training data including a first brain tissue image acquired at a first time point and a first diagnostic evaluation score for the first brain tissue image, and first training data acquired at a second time point.
- Second learning data including a second brain tissue image and a second diagnostic evaluation score for the second brain tissue image, a third brain tissue image acquired at a third time point, and a third for the third brain tissue image It may be learned using third learning data including diagnostic evaluation scores.
- a plurality of learning models 300 may be provided and may be trained by respective training data.
- the second training data at the second time point, and the third learning data at the third time point are time series data.
- the value obtained from the third learning data can be learned by the second learning model as the second learning data, and the value obtained by the second learning data can be learned by the first learning model as the first learning data.
- the learning model 300 may include a plurality of learning models.
- the learning model 300 may include a first learning model 300a, a second learning model 300b, and a third learning model 300c.
- the first learning model 300a may be trained using information of a brain tissue image in a normal state.
- the second learning model 300b may be learned using information on a mild cognitive impairment state, which is an initial brain disease onset state.
- the third learning model 300c may be learned using information on the state of Alzheimer's disease, which is the state of brain disease invention. Of course, you can use the three data in combination to improve the accuracy of learning.
- the learning model 300 may be mathematically modeled.
- C 0 ,C 1 ,C 2 means cell state
- ht-1, ht, ht+1 means hidden state
- Xt means input data
- ⁇ t , ⁇ t+1 , ⁇ t+2 means the time difference information
- Mt, Mt+1, Mt+2 means the masking vector
- Z t means the variable It means the value reflecting the correlation information
- Means matrix product operator Can mean a concatenation operator.
- Equation 1 each value can be calculated by Equation 1.
- (1) process is an equation that reflects time difference information
- (2) is an equation that performs a matrix multiplication operation of time difference information and a masking vector in the input data
- (3) process is an equation that reflects correlation information between variables.
- the process is an equation that combines time difference information and correlation information between variables
- (6) the process uses the input value through the process (5) when there is a missing value, and input when there is no missing value It can be calculated using the data.
- the diagnosis unit 400 uses the brain tissue image and clinical diagnosis evaluation score information at the current time as input data.
- the brain image information at a previous viewpoint or the brain image information at a later viewpoint may be predicted.
- diagnosis unit 400 uses the learning model 300 using the brain image information of a previous time point or the brain image information of a later time point as the input data, the progression of the brain disease and the current progression of the brain disease It is possible to predict at least one or more information of a speed, a later progression speed of the brain disease, and a clinical diagnosis evaluation score.
- the diagnosis unit 400 may simultaneously predict the progression stage of a brain disease and a clinical diagnosis evaluation score from each temporal feature obtained from the features extracted by the learning model 300 or predict regardless of the sequence.
- a diagnosis method is finally performed through the diagnosis unit 400, and a method of distinguishing between classes and a prediction model suitable for the clinical diagnosis evaluation score is developed. Deep learning models can be used to construct.
- the diagnosis unit 400 evaluates the clinical diagnosis with the features extracted based on the brain image by the learning model 300. You can predict your score. In this case, the diagnosis unit 400 may diagnose a brain disease by integrating the probability of the progression step for each time and the predicted clinical diagnosis evaluation score.
- the data used for the performance result was the ADNI-based TADPOLE challenge dataset, and 655 subjects were used. Of these, 395 were used as training data, 131 were used as verification data, and 129 were used as test data.
- the performance result can be composed of CN (Cognitively Normal), MCI (Mild Cognitive Impairment), AD (Alzheimer's Disease) labels, and performance evaluation was evaluated using 5-fold cross validation.
- mAUC multi-class AUC
- MAE mean absolute error
- RMSE root mean square error
- Equation 2 The calculation of mAUC can be calculated by Equation 2.
- Equation 3 The calculation of MAE and RMSE can be calculated by Equation 3.
- a clinical diagnosis evaluation score is used as a case where the clinical diagnosis evaluation score is used (with Clinical scores) and when the clinical diagnosis evaluation score is not used (w/o Clinical scores). It can be seen that the performance of (with Clinical scores) is further improved. Therefore, it can be confirmed that the use of the clinical diagnostic evaluation score contributes to the improvement of classification performance.
- the brain tissue image information at the next time point and the actual brain tissue at the next time point estimated from the brain tissue image information at the current time point Represents image information and a value calculated by MAE corresponding to Equation 3.
- values calculated by MAE and RMSE corresponding to Equation 3 are shown. Therefore, it can be seen that the brain disease prediction of the examples is considerably improved compared to the prior art.
- FIG. 11 is a block diagram showing a brain disease method according to an embodiment.
- the brain disease method according to the embodiment may be performed by the brain disease device described above.
- Collecting a plurality of brain image data from different viewpoints (S100) may be performed by the data collection unit.
- the extracting (S200) of information on the brain tissue image of the region of interest among the brain images may be performed by the data processing unit.
- the progression of the brain disease, the current progression rate of the brain disease, and the future of the brain disease by using information of a plurality of brain tissue images of the region of interest extracted from the plurality of brain images at different points of time as input of a pre-learned learning model. Predicting at least one information of a progress rate and a clinical diagnosis evaluation score according to the brain disease state (S300) may be performed by the diagnosis unit.
- embodiments of the present invention described above can be implemented through various means.
- embodiments of the present invention may be implemented by hardware, firmware, software, or a combination thereof.
- the method according to the embodiments of the present invention includes one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing elements (DSPDs, Digital Signal Processing Devices), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, and the like.
- ASICs application specific integrated circuits
- DSPs digital signal processors
- DSPDs digital signal processing elements
- PLDs Programmable Logic Devices
- FPGAs Field Programmable Gate Arrays
- processors controllers, microcontrollers, microprocessors, and the like.
- the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above.
- a computer program in which software codes and the like are recorded may be stored in a computer-readable recording medium or a memory unit and driven by a processor.
- the memory unit may be located inside or outside the processor, and may exchange data with the processor through various known means.
- the present invention may be expressed as a block diagram including a plurality of blocks or a flowchart including a plurality of steps, and combinations of each block of the block diagram and each step of the flowchart may be performed by computer program instructions. Since these computer program instructions can be mounted on the encoding processor of a general-purpose computer, special purpose computer or other programmable data processing equipment, the instructions executed by the encoding processor of the computer or other programmable data processing equipment are each block of the block diagram or Each step of the flow chart will create a means to perform the functions described. These computer program instructions can also be stored in computer-usable or computer-readable memory that can be directed to a computer or other programmable data processing equipment to implement a function in a particular way.
- each block or each step may represent a module, segment, or part of code including one or more executable instructions for executing a specified logical function. It should also be noted that in some alternative embodiments, the functions mentioned in blocks or steps may occur out of order. For example, two blocks or steps shown in succession may in fact be performed substantially simultaneously, or the blocks or steps may sometimes be performed in the reverse order depending on the corresponding function.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Surgery (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Psychiatry (AREA)
- Neurology (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims (15)
- 서로 다른 시점의 복수의 뇌 영상을 수집하는 데이터 수집부;상기 뇌 영상 중 관심 영역의 뇌 조직 영상의 정보를 추출하는 데이터 처리부; 및상기 서로 다른 시점의 복수의 뇌 영상에서 추출한 관심 영역의 복수의 뇌 조직 영상의 정보를 미리 학습된 학습 모델의 입력으로 하여 뇌 질환의 진행 경과, 상기 뇌 질환의 현재 진행 속도, 상기 뇌 질환의 추후 진행 속도 및 상기 뇌 질환 상태에 따른 임상적 진단 평가 점수 중 적어도 하나 이상의 정보를 예측하는 진단부를 포함하는 뇌 질환 예측 장치.
- 제1항에 있어서,상기 데이터 처리부는,상기 복수의 뇌 영상에서 상기 뇌의 위치가 정렬되도록 상기 복수의 뇌 영상을 처리하는 뇌 위치 정렬부;상기 뇌 위치가 정렬된 상기 복수의 뇌 영상에서 상기 관심 영역의 뇌 조직 영상을 추출하는 뇌 조직 영상 추출부;상기 복수의 뇌 영상에서 상기 관심 영역의 상기 뇌 조직 영상을 제외한 영역을 제거하는 제거부;상기 관심 영역의 상기 뇌 조직 영상을 이용하여 상기 뇌 조직의 정보를 추출하는 정보 추출부;를 포함하는 뇌 질환 예측 장치.
- 제2항에 있어서,상기 데이터 처리부에서,상기 관심 영역의 뇌 조직 영상은 중측두회(middle temporal gyrus) 영상, 내 후각 피질(entorhinal cortex) 영상, 방추이랑(Fusiform gyrus) 및 해마 (Hippocampus) 영상 중 적어도 하나를 포함하는 뇌 질환 예측 장치.
- 제1항에 있어서,상기 학습 모델은 상기 뇌 조직 영상의 정보에 따른 정상 상태, 경도인지장애 상태 및 알츠하이머병 상태에 대해 각각 학습된 복수의 학습 모델을 포함하는 뇌 질환 예측 장치.
- 제1항에 있어서,상기 진단부에서,상기 학습 모델을 미리 학습시키기 위한 학습 데이터는 제1 시점에서 획득된 제1 뇌 조직 영상 정보와 상기 제1 뇌 조직 영상 정보에 대한 제1 진단평가점수를 포함하는 제1 학습 데이터와, 제2 시점에서 획득된 제2 뇌 조직 영상 정보와 상기 제2 뇌 조직 영상 정보에 대한 제2 진단평가점수를 포함하는 제2 학습 데이터와, 제3 시점에서 획득된 제3 뇌 조직 영상 정보와 상기 제3 뇌 조직 영상 정보에 대한 제3 진단평가점수를 포함하는 제3 학습 데이터를 포함하는 뇌 질환 예측 장치.
- 제5항에 있어서,상기 진단부에서,상기 학습 모델을 미리 학습시키기 위한 학습 데이터는 상기 제1 학습 데이터, 상기 제2 학습 데이터 및 상기 제3 학습 데이터의 각각 또는 2개 이상의 학습 데이터가 조합된 데이터를 포함하는 뇌 질환 예측 장치.
- 제1항에 있어서,상기 데이터 수집부는 상기 뇌 조직의 상태에 따른 임상적 진단평가점수 정보를 더 수집하고,상기 진단부는 상기 뇌 조직 영상의 정보와, 상기 임상적 진단평가점수 정보를 상기 학습 모델의 학습 데이터로 사용하는 뇌 질환 예측 장치.
- 제7항에 있어서,상기 진단부는,상기 학습 모델을 이용하여 이전 시점의 상기 뇌 조직 영상 정보 또는 이후 시점의 상기 뇌 조직 영상 정보를 예측하는 뇌 질환 예측 장치.
- 제8항에 있어서,상기 진단부는,이전 시점의 상기 뇌 조직 영상 정보 또는 상기 이후 시점의 상기 뇌 조직 영상 정보를 재이용하여 상기 학습 모델의 입력 데이터로 사용하는 뇌 질환 예측 장치.
- 제1항에 있어서,상기 학습 모델은,인코딩 모듈부와 디코팅 모듈부를 포함하고,상기 진단부는,상기 뇌 조직 영상의 정보를 상기 인코딩 모듈의 입력으로 하여 상기 뇌 조직 영상의 부피 변화량, 형태 변화량, 형상 변화량, 위치 변화량 정보를 포함하는 특징을 추출하고,상기 특징을 상기 디코딩 모듈부의 입력으로 하여 뇌 질환의 진행 경과, 상기 뇌 질환의 현재 진행 속도, 상기 뇌 질환의 추후 진행 속도, 임상적 진단 평가 점수 중 적어도 하나 이상의 정보를 예측하는 뇌 질환 예측 장치.
- 서로 다른 시점의 복수의 뇌 영상의 데이터를 수집하는 단계;상기 뇌 영상 중 관심 영역의 뇌 조직 영상의 정보를 추출하는 단계; 및상기 서로 다른 시점의 복수의 뇌 영상에서 추출한 관심 영역의 복수의 뇌 조직 영상의 정보를 미리 학습된 학습 모델의 입력으로 하여 뇌 질환의 진행 경과, 상기 뇌 질환의 현재 진행 속도, 상기 뇌 질환의 추후 진행 속도 및 상기 뇌 질환 상태에 따른 임상적 진단 평가 점수 중 적어도 하나 이상의 정보를 예측하는 단계를 포함하는 뇌 질환 예측 방법.
- 학습 데이터를 수집하는 학습 데이터 수집부; 및뇌 질환의 진행 경과, 상기 뇌 질환의 현재 진행 속도, 상기 뇌 질환의 추후 진행 속도 및 상기 뇌 질환 상태에 따른 임상적 진단 평가 점수 중 적어도 하나 이상의 정보를 예측하기 위해 상기 학습 데이터를 이용하여 학습되는 학습 모델을 포함하는 뇌 질환을 예측하기 위한 학습 장치.
- 제12항에 있어서,상기 학습 모델은 RNN, CNN 및 DNN 중 어느 하나 또는 이들의 조합인 뇌 질환을 예측하기 위한 학습 장치.
- 제12항에 있어서,상기 학습 데이터 수집부는 제1 시점에서 획득된 제1 시점에서 획득된 제1 뇌 조직 영상 정보와 상기 제1 뇌 조직 영상 정보에 대한 제1 진단평가점수를 포함하는 제1 학습 데이터와, 제2 시점에서 획득된 제2 뇌 조직 영상 정보와 상기 제2 뇌 조직 영상 정보에 대한 제2 진단평가점수를 포함하는 제2 학습 데이터와, 제3 시점에서 획득된 제3 뇌 조직 영상 정보와 상기 제3 뇌 조직 영상 정보에 대한 제3 진단평가점수를 포함하는 제3 학습 데이터를 포함하는 뇌 질환을 예측하기 위한 학습 장치.
- 제14항에 있어서,상기 학습 데이터 수집부는 상기 제1 학습 데이터, 상기 제2 학습 데이터 및 상기 제3 학습 데이터의 각각 또는 2개 이상의 학습 데이터가 조합된 데이터를 수집하는 뇌 질환을 예측하기 위한 학습 장치.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020217022211A KR102605720B1 (ko) | 2019-03-06 | 2020-03-05 | 뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/KR2019/002617 WO2020179950A1 (ko) | 2019-03-06 | 2019-03-06 | 딥러닝 기반 뇌 질환 진행 예측 방법 및 장치 |
KRPCT/KR2019/002617 | 2019-03-06 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020180135A1 true WO2020180135A1 (ko) | 2020-09-10 |
Family
ID=72337451
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2019/002617 WO2020179950A1 (ko) | 2019-03-06 | 2019-03-06 | 딥러닝 기반 뇌 질환 진행 예측 방법 및 장치 |
PCT/KR2020/003131 WO2020180135A1 (ko) | 2019-03-06 | 2020-03-05 | 뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2019/002617 WO2020179950A1 (ko) | 2019-03-06 | 2019-03-06 | 딥러닝 기반 뇌 질환 진행 예측 방법 및 장치 |
Country Status (2)
Country | Link |
---|---|
KR (1) | KR102605720B1 (ko) |
WO (2) | WO2020179950A1 (ko) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102383058B1 (ko) * | 2021-10-22 | 2022-04-08 | 주식회사 뉴로젠 | 시공간 기억 검사와 뇌 영상 정보를 활용한 인지 장애 예측 장치 및 방법 |
KR20230145724A (ko) * | 2022-04-11 | 2023-10-18 | 고려대학교 산학협력단 | 자기공명 영상의 딥러닝 판독 기술에 기반한 다양한 퇴행성 뇌질환 감별진단 방법 및 장치 |
KR20240000692A (ko) * | 2022-06-23 | 2024-01-03 | 고려대학교 산학협력단 | 멀티모달 뇌 영상 기반 뇌질환 바이오마커 추출 방법론 및 파이프라인 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014042684A (ja) * | 2012-08-28 | 2014-03-13 | Dainippon Printing Co Ltd | 医用画像処理装置、およびプログラム |
KR20170140757A (ko) * | 2016-06-10 | 2017-12-21 | 한국전자통신연구원 | 임상 의사결정 지원 앙상블 시스템 및 이를 이용한 임상 의사결정 지원 방법 |
KR20180040287A (ko) * | 2016-10-12 | 2018-04-20 | (주)헬스허브 | 기계학습을 통한 의료영상 판독 및 진단 통합 시스템 |
KR20180058466A (ko) * | 2016-11-24 | 2018-06-01 | 주식회사 셀바스에이아이 | 질환 발병 확률 예측 모델 학습 방법 및 장치 |
US20180310870A1 (en) * | 2017-05-01 | 2018-11-01 | The Charles Stark Draper Laboratory, Inc. | Deep learning architecture for cognitive examination subscore trajectory prediction in alzheimer's disease |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101796055B1 (ko) * | 2016-06-02 | 2017-11-10 | 고려대학교 산학협력단 | 다중 뇌 연결망 구축을 통한 뇌 상태 모니터링 방법 및 장치 |
KR20180002229A (ko) * | 2016-06-29 | 2018-01-08 | 원시스템주식회사 | 치매 정보 데이터베이스 구축을 위한 에이전트 장치 및 그 운영방법 |
KR101854071B1 (ko) * | 2017-01-13 | 2018-05-03 | 고려대학교 산학협력단 | 딥러닝을 사용하여 관심 부위 이미지를 생성하는 방법 및 장치 |
KR101754291B1 (ko) * | 2017-04-04 | 2017-07-06 | 이현섭 | 개인 맞춤형 뇌질병 진단 및 상태 판정을 위한 의료 영상 처리 시스템 및 방법 |
KR101881731B1 (ko) * | 2018-02-27 | 2018-07-25 | 한국과학기술정보연구원 | 기계 학습을 통한 치매 예측용 데이터 처리 장치 및 그 방법, 이를 수록한 기록 매체 |
KR102125127B1 (ko) | 2018-03-05 | 2020-06-19 | 고려대학교 산학협력단 | 딥러닝 기반 뇌 질환 진단 방법 |
-
2019
- 2019-03-06 WO PCT/KR2019/002617 patent/WO2020179950A1/ko active Application Filing
-
2020
- 2020-03-05 WO PCT/KR2020/003131 patent/WO2020180135A1/ko active Application Filing
- 2020-03-05 KR KR1020217022211A patent/KR102605720B1/ko active IP Right Grant
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014042684A (ja) * | 2012-08-28 | 2014-03-13 | Dainippon Printing Co Ltd | 医用画像処理装置、およびプログラム |
KR20170140757A (ko) * | 2016-06-10 | 2017-12-21 | 한국전자통신연구원 | 임상 의사결정 지원 앙상블 시스템 및 이를 이용한 임상 의사결정 지원 방법 |
KR20180040287A (ko) * | 2016-10-12 | 2018-04-20 | (주)헬스허브 | 기계학습을 통한 의료영상 판독 및 진단 통합 시스템 |
KR20180058466A (ko) * | 2016-11-24 | 2018-06-01 | 주식회사 셀바스에이아이 | 질환 발병 확률 예측 모델 학습 방법 및 장치 |
US20180310870A1 (en) * | 2017-05-01 | 2018-11-01 | The Charles Stark Draper Laboratory, Inc. | Deep learning architecture for cognitive examination subscore trajectory prediction in alzheimer's disease |
Also Published As
Publication number | Publication date |
---|---|
KR102605720B1 (ko) | 2023-11-24 |
WO2020179950A1 (ko) | 2020-09-10 |
KR20210096292A (ko) | 2021-08-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020180135A1 (ko) | 뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 | |
De Moura et al. | Deep convolutional approaches for the analysis of covid-19 using chest x-ray images from portable devices | |
WO2017051945A1 (ko) | 질환 모델 기반의 의료 정보 서비스 제공 방법 및 장치 | |
WO2023077603A1 (zh) | 一种异常脑连接预测系统、方法、装置及可读存储介质 | |
WO2020242239A1 (ko) | 앙상블 학습 알고리즘을 이용한 인공지능 기반 진단 보조 시스템 | |
Kamil | A deep learning framework to detect Covid-19 disease via chest X-ray and CT scan images. | |
WO2015072818A1 (ko) | 뇌질환 진단 서비스 장치 및 뇌질환 진단 서비스 방법 | |
WO2019083129A1 (ko) | 피검체에 대한 안저 영상의 판독을 지원하는 방법 및 이를 이용한 장치 | |
WO2019231104A1 (ko) | 심층 신경망을 이용하여 영상을 분류하는 방법 및 이를 이용한 장치 | |
WO2020231007A2 (ko) | 의료기계 학습 시스템 | |
WO2019098415A1 (ko) | 자궁경부암에 대한 피검체의 발병 여부를 판정하는 방법 및 이를 이용한 장치 | |
WO2023068619A1 (ko) | 시공간 기억 검사와 뇌 영상 정보를 활용한 인지 장애 예측 방법 및 이를 수행하는 장치 및 컴퓨터 판독 가능한 기록매체 | |
WO2020197233A2 (ko) | 다중시기 측부혈류 영상을 생성하기 위한 학습 방법 및 기계 학습을 이용한 다중시기 측부혈류 영상 생성 방법 | |
Serener et al. | Deep learning to distinguish COVID-19 from other lung infections, pleural diseases, and lung tumors | |
WO2023095989A1 (ko) | 뇌질환 진단을 위한 다중 모달리티 의료 영상 분석 방법 및 장치 | |
WO2022139068A1 (ko) | 딥러닝 기반의 폐질환 진단 보조 시스템 및 딥러닝 기반의 폐질환 진단 보조 방법 | |
WO2021201582A1 (ko) | 피부 병변의 원인 분석 방법 및 장치 | |
Pelaez et al. | A deep learning model to screen for Corona Virus Disease (COVID-19) from X-ray chest images | |
WO2021002669A1 (ko) | 병변 통합 학습 모델을 구축하는 장치와 방법, 및 상기 병변 통합 학습 모델을 사용하여 병변을 진단하는 장치와 방법 | |
WO2020246676A1 (ko) | 자궁경부암 자동 진단 시스템 | |
Bhan et al. | Diagnosing Parkinson’s disease in Early Stages using Image Enhancement, ROI Extraction and Deep Learning Algorithms | |
WO2020111557A1 (ko) | 혈관 지도 구축 장치, 방법 및 상기 방법을 실행하기 위한 컴퓨터 프로그램 | |
WO2022119347A1 (ko) | 초음파 영상 기반의 딥 러닝을 통한 관상동맥 경화반 조직 분석 방법, 장치 및 기록매체 | |
WO2017010612A1 (ko) | 의료 영상 분석 기반의 병리 진단 예측 시스템 및 방법 | |
WO2021177799A1 (ko) | 이미지 기반 코로나바이러스감염증 정량 분류 방법 및 시스템 |
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: 20766011 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 20217022211 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20766011 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20766011 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 21.04.2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20766011 Country of ref document: EP Kind code of ref document: A1 |