Cochlear implant wearing effect prediction device and cochlear implant wearing effect prediction method

Abstract

To provide a wearing effect prediction system and wearing effect prediction method for use in preoperative evaluation of cochlear implant surgery.SOLUTION: A cochlear implant wearing effect prediction device includes a computer, a U / I, and an image display device. The computer comprises: an image data acquisition unit that acquires brain images of a subject; a cortical thickness calculation unit that calculates the cerebral cortical thickness in a plurality of regions of interest specified on a brain map and set based on clinical test results by means of a surface-based morphometry method from the acquired brain images; a wearing effect prediction unit that inputs the calculated cortical thickness to a wearing effect prediction model formula statistically derived from cortical thicknesses and wearing outcomes of patients who have undergone cochlear implant surgery, and calculates a probability value of hearing improvement success if the subject wears the cochlear implant; a storage unit; a data output unit that outputs the calculated success probability value as a prediction result; and a control unit that controls the overall device. The wearing effect prediction model formula is derived using logistic regression analysis, with the cortical thicknesses in the plurality of regions of interest as explanatory variables and the hearing improvement success odds as objective variable.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a cochlear implant wearing effect prediction device and a cochlear implant wearing effect prediction method, and more particularly to a cochlear implant wearing effect prediction device and a cochlear implant wearing effect prediction method that non-invasively predict the cochlear implant wearing effect before surgery using brain image data of a subject. [Background technology]

[0002] A cochlear implant (CI) is a neuroprosthesis used to improve sound perception in patients with sensorineural hearing loss. Cochlear implants, when used in conjunction with speech therapy, can improve speech understanding in both noisy and quiet environments. However, clinical outcomes of cochlear implant surgery vary widely between individuals.

[0003] Although there are intraoperative tests such as electrically evoked auditory brainstem response and electrocochleography to predict clinical outcomes, these tests cannot be performed preoperatively. Therefore, there is an important clinical need to obtain preoperative indicators of the success of cochlear implant surgery.

[0004] In recent years, methods have been proposed that involve obtaining brain images using MRI (Magnetic Resonance Imaging) and processing the obtained brain images to diagnose signs of dementia such as cerebral infarction and blood vessel-related diseases, as well as atrophy of specific parts of the brain.

[0005] For example, Patent Document 1 discloses a method for diagnosing a specific dementia disease by creating a gray matter image of a brain cross section from a captured MRI image, comparing the gray matter image with a gray matter image of a healthy subject on a voxel basis, and performing analysis using ROIs (regions of interest) defined by a set of voxels set for each disease. However, the idea of ​​predicting the effect of wearing a cochlear implant from an MRI image has not yet been conceived.

[0006] Subsequently, a method for predicting the effect of cochlear implantation from the results of MRI image analysis of the brain was published (Non-Patent Document 1). In Non-Patent Document 1, a method for predicting the outcome of cochlear implantation using MRI images before cochlear implantation surgery for people with acquired hearing loss is proposed based on the correlation between gray matter probabilities of specific parts of the brain cortex by voxel-based morphometry (VBM) and word recognition score (WRS). However, in the analysis by VBM, information such as the thickness and surface area of ​​the cortex related to the gyri and sulci formed in the human brain, and the degree to which the cortex is folded (gyral index) is lost. For this reason, a new method for improving the accuracy of preoperative evaluation of cochlear implant candidates is required. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] JP 2005-230456 A [Non-patent literature]

[0008] [Non-Patent Document 1] Zhe Sun, et al. 11 others, Cortical reorganization following auditory deprivation predicts cochlear implant performance in postlingually deaf adults, Hum Brain Mapp.2021 Jan;42(1):233-244. Summary of the Invention [Problem to be solved by the invention]

[0009] The present invention has been made in consideration of the problems of the conventional technology described above, and an object of the present invention is to provide a highly sensitive cochlear implant wearing effect prediction device and cochlear implant wearing effect prediction method for preoperative evaluation of cochlear implant surgery candidates. [Means for solving the problem]

[0010] In order to achieve the above object, one aspect of the present invention provides a cochlear implant wearing effect prediction device that predicts the cochlear implant wearing effect using brain image data of a subject, the cochlear implant wearing effect prediction device comprising a host computer, a user interface connected to the host computer, and an image display device connected to the host computer, the host computer including an image data acquisition unit that acquires brain image data of the subject, a cortical thickness calculation unit that calculates cerebral cortical thickness in a plurality of regions of interest (ROIs) that are set based on the results of a preliminary clinical test and positioned on a predetermined brain map by surface morphometry from the acquired brain image data, and a cortical thickness calculation unit that calculates each of the calculated cortical thicknesses in the plurality of ROIs based on the cortical thickness and wearing results of a patient who has undergone cochlear implant surgery. a wearing effect prediction unit that inputs the above data into a cochlear implant wearing effect prediction model formula derived in advance using a statistical method, and calculates a probability value of successful hearing improvement when the subject wears the cochlear implant; a memory unit that stores brain image data of the subject, position information of the multiple regions of interest on the brain map, and the cochlear implant wearing effect prediction model formula; a data output unit that outputs the calculated probability value of successful hearing improvement to the image display device as a predicted result of the cochlear implant wearing effect for the subject; and a control unit that controls the image data acquisition unit, the cortical thickness calculation unit, the wearing effect prediction unit, the memory unit, the data output unit, and the device in general, wherein the cochlear implant wearing effect prediction model formula is derived by logistic regression analysis using the cortical thickness in the multiple regions of interest as an explanatory variable and the odds of successful hearing improvement as an objective variable.

[0011] The brain image data is preferably brain image data acquired by an MRI (Magnetic Resonance Imaging) device. The multiple regions of interest correspond to locations on the predetermined brain map identified as hearing-related regions in the preliminary clinical test, and the cortical thickness of the cerebral cortex in the multiple regions of interest can be calculated using a cortical thickness measurement software tool compatible with surface-based morphometry implemented in the host computer. Preferably, the predetermined brain atlas is a Desikan-Killiany cortical atlas, and the plurality of regions of interest include at least the right entorhinal cortex and the left medial orbitofrontal cortex.

[0012] In order to achieve the above object, one aspect of the present invention provides a method for predicting the effects of wearing a cochlear implant, which is executed by a cochlear implant wearing effect prediction device including a host computer, a user interface connected to the host computer, and an image display device connected to the host computer, and the host computer includes an image data acquisition unit, a memory unit in which brain image data, position information of a plurality of regions of interest positioned on a predetermined brain map, and a cochlear implant wearing effect prediction model formula are stored, and a control unit that performs predetermined calculation processing for predicting the effects of wearing the cochlear implant and controls the entire device, and the control unit performs a process of performing ... The method includes the steps of: acquiring image data; calculating cerebral cortical thickness in a plurality of regions of interest (ROIs) that are set based on the results of a preliminary clinical test and positioned on a predetermined brain map using a surface morphometry method from the acquired brain image data; inputting each of the calculated cortical thicknesses in the plurality of regions of interest into a cochlear implant wearing effect prediction model formula that has been derived in advance using a statistical method from the cortical thickness and wearing results of patients who have undergone cochlear implant surgery, to calculate a probability value of successful hearing improvement when the subject wears the cochlear implant; and outputting the calculated probability value of successful hearing improvement to the image display device as a predicted result of the effect of wearing the cochlear implant for the subject.

[0013] The brain image data is preferably brain image data acquired by an MRI (Magnetic Resonance Imaging) device. The multiple regions of interest correspond to locations on the predetermined brain map identified as hearing-related regions in the preliminary clinical test, and the cortical thickness of the cerebral cortex in the multiple regions of interest can be calculated using a cortical thickness measurement software tool compatible with surface-based morphometry implemented in the host computer. Preferably, the predetermined brain atlas is a Desikan-Killiany cortical atlas, and the plurality of regions of interest include at least the right entorhinal cortex and the left medial orbitofrontal cortex. Effect of the Invention

[0014] According to the present invention, the effect of wearing a cochlear implant on a subject who is a candidate for cochlear implant surgery can be predicted before surgery, so that an appropriate treatment plan can be proposed to the patient. In particular, for patients with acquired hearing loss, the present invention makes it possible to more accurately predict the effects of wearing a cochlear implant, thereby improving the success rate of treatment. Furthermore, according to the present invention, it becomes easier to select patients suitable for cochlear implantation, and effective treatment can be started early. [Brief description of the drawings]

[0015] [Figure 1] 1 is a block diagram showing the overall configuration of a cochlear implant wearing effect prediction device according to one embodiment of the present invention. [Diagram 2] 1 is a block diagram showing the functional configuration of a cochlear implant wearing effect prediction device according to an embodiment of the present invention. [Diagram 3] 1 is a flowchart showing a process for predicting the effect of wearing a cochlear implant executed by the cochlear implant wearing effect prediction device according to one embodiment of the present invention. [Figure 4] FIG. 2 is a diagram showing regions of interest (ROIs) that are significantly associated with the effect of wearing a cochlear implant, as found by rank logistic regression analysis in this embodiment. [Diagram 5]FIG. 13 is a diagram showing an example of the results of a Spearman correlation analysis between the probability value (probability of success) calculated by the cochlear implant wearing effect prediction device according to the present embodiment and monosyllable sound discrimination ability. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, specific examples of the embodiments of the present invention will be described in detail with reference to the drawings.

[0017] FIG. 1 is a block diagram showing the overall configuration of a cochlear implant effect prediction device according to one embodiment of the present invention.

[0018] A cochlear implant wearing effect prediction device 10 according to one embodiment of the present invention comprises a host computer 11, a user interface 12 connected to the host computer 11, and an image display device 13. The host computer 11 is connected to an MRI (magnetic resonance imaging) device 20 that takes scan images of the subject's brain via a specified communication line 5, and is configured to receive the brain scan images of the subject taken by the MRI device 20 (hereinafter referred to as MRI brain images).

[0019] In this embodiment, for example, a high-resolution three-dimensional T1-weighted whole brain MRI image is used as the MRI brain image. The cochlear implant wearing effect prediction device 10 may be directly connected to the MRI device 20, or may be configured to acquire the subject's MRI brain image data stored in a computer-readable storage medium via the user interface 12.

[0020] The cochlear implant wearing effect prediction device 10 may further be connected to a data server 30 via a predetermined communication network 6 including the Internet, a dedicated line, etc. The data server 30 is, for example, a server of a medical image management system, and manages an image database 31 in which a large amount of data on MRI brain images (image data and subject information associated therewith) is accumulated, and the cochlear implant wearing effect prediction device 10 may acquire necessary data from the data server 30. When the MRI brain image data of the subject is stored in the image database 31 of the data server 30, the cochlear implant wearing effect prediction device 10 may be configured to acquire the subject's MRI brain image data from the data server 30 via the predetermined communication network 6.

[0021] The host computer 11 of the cochlear implant wearing effect prediction device 10 is configured by installing a control program such as an operating system and application programs such as a client program or a server program on a computer hardware such as a personal computer, a server computer, or a workstation equipped with hardware such as a CPU, memory, a storage device, and various communication ports (not shown). In particular, it is preferable that a dedicated application program for performing analysis and display processing of MRI images, and a software program for performing statistical processing of various clinical test data are installed.

[0022] The user interface 12 includes input devices such as a keyboard and a mouse, and refers to a setting input screen, a selection screen, and the like displayed on the image display device 13 in response to various arithmetic processing executed by a predetermined application program, to receive commands given by the user of the cochlear implant wearing effect prediction device 10 by pressing various operation buttons on the input device or inputting characters, as well as input of information about the subject, and transmits the information to the host computer 11. The subject information includes the name, age, sex, etiology of hearing loss (congenital or acquired), and the like.

[0023] The image display device 13 is configured, for example, as a high-definition monitor made of a liquid crystal display (LCD), and displays various operation buttons and images processed by application programs on the display screen. The image display device 13 (monitor device) can also be configured to have a touch panel on the display screen to accept input from a finger, a touch pen, or the like.

[0024] The data server 30 stores MRI brain images of subjects with normal hearing (also called hearing controls) and subjects with severe to profound hearing loss in order to contribute to clinical applications of hearing. The MRI brain images of each subject taken by the MRI device 20 are digitally converted, unique identification information is added, and the image data of the MRI brain images are stored in the image database 31 of the data server 30 in a file format that complies with the DICOM (Digital Image and Communications in Medicine) standard.

[0025] FIG. 2 is a block diagram showing the functional configuration of a cochlear implant wearing effect prediction device according to one embodiment of the present invention.

[0026] The host computer 11 of the cochlear implant wearing effect prediction device 10 according to this embodiment is configured to include a CPU, memory, etc. (not shown), and includes a control unit 100 that controls the functional means and components included in the computer, a memory unit 200 that stores various data and computer programs, and a communication unit 300 that handles data communication between the host computer 11 and external devices such as the MRI device 20 and data server 30 via a communication line and / or communication network.

[0027] The control unit 100 includes an instruction information input unit 110 that accepts commands and subject information input via the user interface 12, an image data acquisition unit 120 that acquires image data of MRI brain images taken by the MRI device 20, and a data output unit 130 that outputs brain images, data processing results, etc. to the image display device 13.

[0028] Furthermore, the control unit 100 includes an image analysis unit 140, a wearing effect prediction unit 150, and a prediction result determination unit 160 as functional units realized by causing the CPU to execute a predetermined dedicated application program.

[0029] When subject information is input to the instruction information input unit 110 of the control unit 100, the image data acquisition unit 120 receives image data of the MRI brain image of the corresponding subject from the data server 30 directly from the MRI apparatus 20 or via a specified communication network 6 based on the input subject information, or reads and acquires image data of the subject's MRI brain image stored in a computer-readable storage medium via a user interface 12 having a specified reading means (not shown), and stores the acquired image data of the MRI brain image in the memory unit 200 in association with the subject information.

[0030] When image data of an MRI brain image (hereinafter referred to as brain image data) is acquired, the control unit 100 causes the image analysis unit 140 to function and execute image analysis processing of the acquired brain image data.

[0031] The image analysis unit 140 converts the data format of the acquired brain image data (e.g., DICOM format) into a predetermined data format for image analysis (e.g., NIfTI format), performs surface-based morphometry on the converted brain image data, and calculates the cortical thickness in each of a plurality of regions of interest (ROI) previously set for a predetermined brain map. In this embodiment, FreeSurfer, which is known as image analysis software used for surface morphometry, is used, and the Desikan-Killiany cortical atlas is used for the predetermined brain map. However, the present invention is not limited to this. The plurality of regions of interest set for the predetermined brain map are determined based on the results of a preliminary clinical test, and the setting method will be described in detail later.

[0032] The wearing effect prediction unit 150 uses the log odds reflecting the effectiveness (hearing improvement success) based on the degree of hearing improvement of monosyllables after surgery for hearing-impaired people who have undergone cochlear implant surgery as the objective variable, and the cortical thickness in the above-mentioned multiple regions of interest (ROI) as the explanatory variable, and inputs the cortical thickness in the subject's region of interest (ROI) calculated by the cortical thickness comparison unit 140 into a cochlear implant wearing effect prediction model formula using an odds ratio derived by rank logistic regression analysis, and calculates the probability value of successful hearing improvement from the value obtained. The cochlear implant wearing effect prediction model formula will be described in detail later.

[0033] The prediction result determination section 160 transmits the probability value calculated by the wearing effect prediction section 150 to the data output section 130 as a prediction result of the cochlear implant wearing effect on the subject.

[0034] The storage unit 200 stores the subject's brain image data, a cochlear implant wearing effect prediction model, brain map data, etc. Also, information such as the subject's age, sex, etiology of hearing loss (congenital or acquired), etc. associated with the subject's brain image data is stored.

[0035] Below, we will explain the processing steps taken by the cochlear implant wearing effect prediction device 10 in this embodiment to calculate the cortical thickness in multiple pre-set regions of interest (ROIs) from the subject's brain image data, and to calculate the probability value of successful hearing improvement when the subject wears a cochlear implant (representing the cochlear implant wearing effect) from the calculated cortical thickness.

[0036] FIG. 3 is a flowchart showing a process of predicting the effect of wearing a cochlear implant executed by the cochlear implant wearing effect prediction device according to one embodiment of the present invention.

[0037] First, when subject information is input to the instruction information input unit 110 of the host computer 11 via the user interface 12 of the cochlear implant wearing effect prediction device 10, the control unit 100 causes the image data acquisition unit 120 to function, and the image data acquisition unit 120 acquires image data of the MRI brain image (brain image data) of the relevant subject based on the input subject information, either directly from the MRI device 20 or from the data server 30 via a specified communication network 6 (step S10).

[0038] The image data acquisition unit 120 associates the acquired brain image data of the subject with the input information of the subject and stores the data in the storage unit 200. The image data acquisition unit 120 may acquire the brain image data of the subject from a predetermined storage device or computer-readable storage medium in which the brain image data of the subject captured by the MRI device 20 is stored.

[0039] When the subject's brain image data is acquired, the control unit 100 causes the image analysis unit 140 to function, and the image analysis unit 140 converts the data format of the acquired brain image data into the data format for the specified image analysis described above, and executes a program of FreeSurfer, which is pre-installed image analysis software, to cause a surface-based morphometry (SBM) means to function, perform surface morphometry on the converted brain image data, and calculate the cortical thickness in each of multiple regions of interest (ROI) that have been pre-set on a specified brain map (in other words, positioned in specific areas on the specified brain map) (step S11).

[0040] The image analysis unit 140 models the cerebral gray-white matter boundary and the pia surface by fitting them to a mesh structure consisting of fine triangles using the surface morphometry (SBM) method, and then performs processing to calculate the actual (measured) values ​​of the cerebral cortical surface area, cortical thickness, and cortical volume from the coordinates of the vertices of this mesh structure.

[0041] When the wearing effect prediction unit 150 receives the subject's cortical thickness in each of a plurality of preset regions of interest (ROI) from the image analysis unit 140, it reads out a cochlear implant wearing effect prediction model formula derived in advance and stored in the storage unit 200, inputs the subject's cortical thickness (measured value) in the plurality of regions of interest (ROI) calculated by the image analysis unit 140 into the explanatory variables of the cochlear implant wearing effect prediction model formula, and calculates logarithmic odds. The cochlear implant wearing effect prediction model formula is expressed by the following formula 1.

[0042]

number

[0043] Here, p is the probability of success (improvement in hearing for monosyllabic words, described below), β1 to β11 are regression coefficients calculated from the results of pre-clinical tests, described below, and X1 to X11 are the cortical thicknesses (measured values) of the subject in each of multiple regions of interest (ROIs) that have been set in advance.

[0044] For convenience of explanation, X1 to X11 are described as explanatory variables in Equation 1, but the present invention is not limited thereto, and the number of explanatory variables is determined according to the number of regions of interest (ROIs) whose significance has been verified by clinical results.

[0045] Thereafter, the wearing effect prediction unit 150 converts the logarithmic odds into a probability value by inputting them into the following formula 2, and determines the converted probability value p as the probability value of successful hearing improvement (i.e., the predicted result of the cochlear implant wearing effect) (step S12).

[0046]

number

[0047] The wearing effect prediction unit 150 transmits data on the determined probability value of successful hearing improvement to the prediction result determination unit 160.

[0048] When the prediction result determination unit 160 receives the data on the probability value of successful hearing improvement, it displays the data in a predetermined display format on the image display device via the data output unit 130 as a predicted result of the effect of wearing a cochlear implant, and also stores it in the memory unit 200 in association with the subject's information (step S13).

[0049] Hereinafter, a method for setting a plurality of regions of interest on a predetermined brain map and a method for deriving a cochlear implant effect prediction model equation in this embodiment will be described.

[0050] As a preliminary clinical study, the inventors used an MRI device to take high-resolution 3D T1-weighted whole-brain MRI images of hearing-impaired patients who had undergone cochlear implant surgery and subjects with healthy hearing, and analyzed the brain imaging data using surface-based morphometry (SBM).

[0051] The healthy hearing subjects (normal hearing controls) ranged in age from 20 to 85 years, and included 530 men and 1,388 women. For each gender, regression lines were constructed for each of the 34 regions of interest (ROIs) specified in the Desikan-Killiany cortical atlas (y-axis, cortical thickness; x-axis, age).

[0052] Cortical thickness was also measured for 34 regions of interest (ROIs) in patients with hearing loss who had undergone cochlear implant surgery using the Desikan-Killiany cortical atlas. A total of 64 patients with hearing loss were surveyed, consisting of 22 men and 42 women, of whom 19 had congenital hearing loss and 45 had acquired hearing loss. The mean age of the congenital hearing loss patients was 28.4 years, and the mean age of the acquired hearing loss patients was 63.0 years.

[0053] Cortical thickness in each region of interest (ROI) in the cortical atlas for hearing-impaired subjects was determined by predicted values ​​based on the age of the hearing-impaired subject calculated using a regression line derived from the brain imaging data of normal-hearing controls, and the residuals between the predicted values ​​and the measured values ​​were used for the following statistical analyses.

[0054] Six months after cochlear implant surgery, hearing-impaired patients were given a speech hearing test in quiet conditions at a volume that was easy for the subjects to hear, and the speech discrimination score (correct answer rate) from the test was used to evaluate the monosyllabic hearing (speech discrimination ability) of the subjects wearing the cochlear implant. The subjects' monosyllabic hearing was ranked as "poor" if the score was less than 30%, "good" if it was 30% to 70%, and "very good" if it was 70% or more (effectiveness of the cochlear implant).

[0055] Rank logistic regression analysis was performed with the effectiveness of cochlear implants based on the speech audiometry test described above as the dependent variable, and the residual cortical thickness in each region of interest (ROI), gender, and etiology of hearing loss (congenital or acquired) as covariates. Odds ratios and 95% confidence intervals were calculated for each ROI. The widely used statistical software SPSS was used for this analysis.

[0056] Using the rank logistic regression analysis described above, we found a significant association between cochlear implant efficacy and cortical thickness in specific regions of interest (ROIs) out of 34. Specifically, in the right hemisphere of the brain, a significant association was found in the entorhinal cortex (odds ratio = 8.62, 95% confidence interval = 1.94-38.29, p = 0.005), and in the left hemisphere, a significant association was found in the medial orbitofrontal cortex (odds ratio = 104.79, 95% confidence interval = 3.38-3249.49, p = 0.008).

[0057] 4 is a diagram showing regions of interest (ROIs) found to be significantly associated with the effect of wearing a cochlear implant by rank logistic regression analysis in this embodiment, in which A indicates the right entorhinal cortex and B indicates the left medial orbitofrontal cortex. Using the method described above, a region of interest (ROI) that is significantly related to the effect of wearing a cochlear implant is set as a region of interest (predetermined region of interest) for calculating cortical thickness by surface morphometry (SBM) using the image analysis unit 140.

[0058] Next, to derive a model equation for predicting the effectiveness of cochlear implant use using odds ratios, the right entorhinal cortex and left medial orbitofrontal cortex were selected as the main areas of interest, and significant (p<0.01) odds ratios were converted into coefficients for a logistic regression model by calculating the natural logarithm of each odds ratio.

[0059] The coefficients for the two selected regions of interest (ROIs) are: Right entorhinal cortex: β1=ln(8.62)=2.15, Left medial orbitofrontal cortex: β2=ln(104.79)=4.65.

[0060] Based on the above evaluation results, the following formula 3 is set as the cochlear implant wearing effect prediction model formula derived in this embodiment, and the following formula 4 is set as a formula for calculating the probability of successful hearing improvement, and these model formulas and calculation formulas are stored in the memory unit 200 of the cochlear implant wearing effect prediction device 10.

[0061]

number

[0062]

number

[0063] where X1 is the cortical thickness of the right entorhinal cortex and X2 is the cortical thickness of the left medial orbitofrontal cortex.

[0064] A Spearman correlation analysis between the probability value calculated by Equation 4 and monosyllabic hearing (used as a measure of the speech discrimination ability of cochlear implant patients) confirmed a significant correlation (p=0.004, correlation coefficient=0.362).

[0065] Furthermore, when comparing the correlation coefficients between the congenital hearing loss group and the acquired hearing loss group, it was found that the correlation coefficient was higher in the acquired hearing loss group (p<0.001, correlation coefficient=0.668) than in the congenital hearing loss group (p=0.029, correlation coefficient=0.501).

[0066] Fig. 5 is a diagram in which the probability value (probability of successful hearing improvement) calculated by Equation 4 is plotted for the monosyllable speech discrimination ability for the above-mentioned hearing-impaired patients, divided into congenital hearing-impaired group and acquired hearing-impaired group. As is clear from Fig. 5, the cochlear implant effect prediction model derived by this embodiment can more accurately predict the cochlear implant effect for patients with acquired hearing loss.

[0067] The cochlear implant wearing effect prediction model equation derived by the above method is stored in the storage unit 200 of the host computer 11 of the cochlear implant wearing effect prediction device 10 according to one embodiment of the present invention, and is used to calculate the probability value of successful hearing improvement, which represents the cochlear implant wearing effect of the subject, in the wearing effect prediction unit 150. Note that the regions of interest (ROI) set as explanatory variables of the cochlear implant wearing effect prediction model equation are not limited to the right entorhinal cortex and left medial orbitofrontal cortex, and other regions of interest for which significant associations have been detected may be added.

[0068] Therefore, in the cochlear implant wearing effect prediction device 10 of this embodiment, the image analysis unit 140 calculates the cortical thickness in a specific region of interest (right entorhinal cortex and left medial orbitofrontal cortex) in a specified brain map from the brain image data of the subject acquired by the image data acquisition unit 120 in the host computer 11, and the wearing effect prediction unit 150 inputs the calculated cortical thickness into a cochlear implant wearing effect prediction model formula derived in advance based on the results of a preliminary clinical test to calculate logarithmic odds, and converts this into a probability value, thereby predicting the probability of cochlear implant success (cochlear implant wearing effect).

[0069] As described above, the cochlear implant effect prediction device and the cochlear implant effect prediction method according to an embodiment of the present invention can predict the effect of cochlear implantation on a subject before surgery by surface-based morphometry using non-invasive MRI brain images, and therefore can provide important information for patient selection and clinical decision-making. Also, it can propose an appropriate treatment plan for hearing-impaired patients.

[0070] Although the embodiments of the present invention have been described in detail above with reference to the drawings, the present invention is not limited to the above-described embodiments and can be modified in various ways without departing from the technical scope of the present invention. [Explanation of symbols]

[0071] 5. Communication lines 6. Communication Network 10. Cochlear implant effect prediction device 11 Host Computer 12 User Interface 13 Image display devices 20 MRI machine 30 Data Server 31 Image Database 100 Control section 110 Instruction information input section 120 Image data acquisition unit 130 Data output section 140 Image Analysis Unit 150 Wearing Effect Prediction Department 160 Prediction result judgment unit 200 Storage section 300 Communications Department

Claims

1. A cochlear implant wearing effect prediction device that predicts the cochlear implant wearing effect by using brain image data of a subject, The cochlear implant wearing effect prediction device includes a host computer, a user interface connected to the host computer, and an image display device connected to the host computer, The host computer includes: An image data acquisition unit that acquires brain image data of the subject; a cortical thickness calculation unit that calculates the cortical thickness of the cerebral cortex in a plurality of regions of interest (ROIs) that are set based on the results of a preliminary clinical test and positioned on a predetermined brain map by surface morphometry from the acquired brain image data; A wearing effect prediction unit that inputs each of the calculated cortical thicknesses in the plurality of regions of interest into a cochlear implant wearing effect prediction model formula derived in advance using a statistical method from the cortical thickness and wearing results of a patient who has undergone cochlear implant surgery, and calculates a probability value of successful hearing improvement when the subject wears the cochlear implant; A memory unit for storing brain image data of the subject, position information of the plurality of regions of interest on the brain map, and a cochlear implant wearing effect prediction model formula; a data output unit that outputs the calculated probability value of successful hearing improvement to the image display device as a predicted effect of wearing the cochlear implant on the subject; The image data acquisition unit, the cortical thickness calculation unit, the wearing effect prediction unit, the storage unit, the data output unit, and a control unit that controls the entire device, The cochlear implant effect prediction device is characterized in that the cochlear implant effect prediction model formula is derived by logistic regression analysis using the cortical thickness in the multiple regions of interest as an explanatory variable and the odds of successful hearing improvement as a target variable.

2. 2. The cochlear implant effect prediction device according to claim 1, wherein the brain image data is brain image data taken by an MRI (Magnetic Resonance Imaging) device.

3. The device for predicting the effect of wearing a cochlear implant according to claim 1, characterized in that the multiple regions of interest correspond to positions on the predetermined brain map identified as hearing-related regions in the preliminary clinical test, and the cortical thickness of the cerebral cortex in the multiple regions of interest is calculated using a cortical thickness measurement software tool corresponding to surface-based morphometry implemented in the host computer.

4. The cochlear implant effect prediction device according to claim 1, characterized in that the predetermined brain map is a Desikan-Killiany cortical atlas, and the multiple regions of interest include at least the right entorhinal cortex and the left medial orbitofrontal cortex.

5. A method for predicting an effect of wearing a cochlear implant, which is executed by a cochlear implant wearing effect prediction device including a host computer, a user interface connected to the host computer, and an image display device connected to the host computer, The host computer includes an image data acquisition unit, a memory unit that stores brain image data, position information of a plurality of regions of interest that are located on a predetermined brain map, and a cochlear implant effect prediction model formula, and a control unit that performs predetermined calculation processing for predicting the cochlear implant effect and controls the entire device. The control unit: acquiring brain image data of a subject by the image data acquisition unit; calculating cerebral cortical thickness in a plurality of regions of interest (ROIs) that are set based on the results of a preliminary clinical test and positioned on a predetermined brain atlas by surface morphometry from the acquired brain image data; A step of inputting each of the calculated cortical thicknesses in the plurality of regions of interest into a cochlear implant effect prediction model formula derived in advance using a statistical method from the cortical thickness and wearing results of a patient who has undergone cochlear implant surgery, and calculating a probability value of successful hearing improvement when the subject wears a cochlear implant; and outputting the calculated probability value of successful hearing improvement to the image display device as a predicted result of the effect of wearing a cochlear implant for the subject.

6. 6. The method for predicting the effect of wearing a cochlear implant according to claim 5, wherein the brain image data is brain image data taken by an MRI (Magnetic Resonance Imaging) device.

7. The method for predicting the effect of wearing a cochlear implant according to claim 5, characterized in that the multiple regions of interest correspond to positions on the predetermined brain map identified as hearing-related regions in the preliminary clinical test, and the cortical thickness of the cerebral cortex in the multiple regions of interest is calculated using a cortical thickness measurement software tool corresponding to surface-based morphometry implemented in the host computer.

8. The cochlear implant effect prediction device according to claim 5, wherein the predetermined brain map is a Desikan-Killiany cortical atlas, and the plurality of regions of interest include at least the right entorhinal cortex and the left medial orbitofrontal cortex.

Images