Method and apparatus for analysing cerebellar atrophy

Abstract

A method for analyzing cerebellar atrophy includes the steps of: an analysis device receiving a medical image of a subject; the analysis device segmenting a cerebellar region and a cerebrospinal fluid (CSF) region within the cerebellar region in the medical image; the analysis device calculating the ratio of the cerebrospinal fluid region to the ratio of the cerebellar region; and the analysis device analyzing the state of cerebellar atrophy of the subject based on the ratio.

Description

Technology Field

[0001] The technology described below relates to a method and device for analyzing cerebellar atrophy through medical imaging. Background Technology

[0002] Cerebellar atrophy is one of the multiple system atrophies (MSA), which are subtypes of Parkinson's syndrome. Cerebellar atrophy is also known as MSA-C.

[0003] Cerebellar atrophy can be diagnosed through neurological examinations, genetic testing, and imaging diagnostics. In other words, cerebellar atrophy can be evaluated using medical imaging techniques such as MRI (Magnetic Resonance Imaging). Prior art literature

[0004] Korean Registered Patent Publication 10-2313656 The problem to be solved

[0005] Conventional methods for analyzing cerebellar atrophy utilized the sizes of the entire brain region and the cerebellum region in medical images. In other words, conventional methods used the relative size of the cerebellum compared to the entire brain region.

[0006] The technology described below, unlike conventional methods, discovers how the size of the cerebellar region changes upon the occurrence of cerebellar atrophy and aims to provide a method for diagnosing cerebellar atrophy based solely on the cerebellar region by utilizing this finding. means of solving the problem

[0007] A method for analyzing cerebellar atrophy includes the steps of: an analysis device receiving a medical image of a subject; the analysis device segmenting a cerebellar region and a cerebrospinal fluid (CSF) region within the cerebellar region in the medical image; the analysis device calculating the ratio of the cerebrospinal fluid region to the ratio of the cerebellar region; and the analysis device analyzing the state of cerebellar atrophy of the subject based on the ratio. Effects of the invention

[0008] Using the technology described below, it is possible to quantitatively evaluate cerebellar atrophy through medical imaging.

[0009] By using the technique described below, it is possible to diagnose cerebellar atrophy with higher sensitivity and specificity compared to conventional methods.

[0010] By using the technique described below, the complexity of the technique can be reduced by limiting the region of interest in medical images to the cerebellum. Brief explanation of the drawing

[0011] Figure 1 shows the results of comparing the cerebellar regions of a patient with cerebellar atrophy and a patient without cerebellar atrophy. Figure 2 shows the overall process of the analysis device analyzing cerebellar atrophy. Figure 3 is an example of dividing each region of a medical image. Figure 4 shows the configuration of the analysis device. Specific details for implementing the invention

[0012] The technology described below may be subject to various modifications and may have various embodiments. Specific embodiments of the technology described below may be described in the drawings of the specification. However, this is for the purpose of explaining the technology described below and is not intended to limit the technology described below to specific embodiments. Accordingly, it should be understood that all modifications, equivalents, and substitutions that fall within the spirit and scope of the technology described below are included in the technology described below.

[0013] Terms such as First, Second A, B, etc., may be used to describe various components. However, these terms are used merely to distinguish one component from others and are not intended to limit the components to such terms.

[0014] Terms such as first, second, A, B, etc., may be used to describe various components, but such components are not limited by the said terms and are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of rights of the technology described below, the first component may be named the second component, and similarly, the second component may be named the first component. The term “and / or” includes a combination of multiple related described items or any of the multiple related described items.

[0015] In terms used in this specification, singular expressions should be understood to include plural expressions unless the context clearly indicates otherwise, and terms such as "includes" should be understood to mean that the described features, number, step, operation, component, part, or combination thereof exist, and not to exclude the existence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0016] Before providing a detailed description of the drawings, it is to clarify that the classification of components in this specification is merely based on the primary function each component is responsible for. That is, two or more components described below may be combined into a single component, or a single component may be divided into two or more components based on more subdivided functions. Furthermore, each component described below may additionally perform some or all of the functions of other components in addition to its own primary function, and it is obvious that some of the primary functions of each component may be exclusively performed by other components.

[0017] Furthermore, in performing the method or operation method, each process constituting the method may occur differently from the specified order unless a specific order is clearly indicated in the context. That is, each process may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

[0019] We will examine the terms used to explain the technology below.

[0020] The cerebellar volume may be a space that includes the space occupied by the human cerebellum and the cerebrospinal fluid (CSF) surrounding the cerebellum.

[0021] The cerebellar region is the empty space surrounding the cerebellum that is occupied by cerebrospinal fluid.

[0023] Figure 1 shows the results of comparing the cerebellar regions of a patient with cerebellar atrophy and a patient without cerebellar atrophy. Patients without cerebellar atrophy may include normal individuals or patients with idiopathic Parkinson's disease (IPD).

[0024] In Fig. 1, the cerebellar region within the cerebellum is indicated in yellow. In Fig. 1, the white matter and grey matter regions of the cerebellum are indicated in red. In Fig. 1, the cerebellar region may be the combined area of ​​yellow and red. In Fig. 1, other brain regions excluding the cerebellum are indicated in blue.

[0025] Figures 1 (A) and (B) are images of the cerebellar region of a patient with cerebellar atrophy. Figure 1 (A) is an image taken in the sagittal plane. Figure 1 (B) is an image taken in the coronal plane.

[0026] Figures 1 (C) and (D) are images of the cerebellar region of a non-cerebellar atrophy patient. Figure 1 (C) is an image taken in the sagittal plane. Figure 1 (D) is an image taken in the coronal plane. When comparing Figures 1 (A) and (B) with Figures 1 (C) and (D), it can be seen that the size of the cerebellar region is similar.

[0027] Comparing Figures 1(A) and (B) with Figures 1(C) and (D), it can be confirmed that the cerebellum of patients with cerebellar atrophy is smaller than that of non-patients. Additionally, it can be confirmed that the cerebellar folds of patients with cerebellar atrophy are more numerous than those of non-patients.

[0028] Comparing Figures 1(A) and (B) with Figures 1(C) and (D), it can be observed that the space surrounding the cerebellum is larger in patients with cerebellar atrophy compared to those without cerebellar atrophy. Additionally, it can be observed that the cerebellar region within the cerebellar area is larger in patients with cerebellar atrophy compared to those without cerebellar atrophy.

[0029] Figure 1 illustrates that in cerebellar atrophy, only the cerebellum atrophies, and the cerebellar region does not shrink or atrophy. In other words, cerebellar atrophy shows a phenomenon in which the size (or volume) of the cerebellum in the cerebellar region decreases, and the cerebrospinal fluid region expands by the same amount as the shrinking region. Therefore, cerebellar atrophy can be diagnosed or evaluated based on the size (or volume) of the cerebellar region, the size (or volume) of the cerebellum, and the size (or volume) of the cerebrospinal fluid region.

[0031] Below, we examine the overall process by which the analysis device analyzes cerebellar atrophy.

[0032] Figure 2 shows the process of an analysis device (100) analyzing cerebellar atrophy.

[0033] The analysis device of Fig. 2 is a device that analyzes cerebellar atrophy using the aforementioned principle. In other words, the analysis device may be a device that analyzes cerebellar atrophy based on the size of the cerebellar region and the size of the cerebrospinal fluid region within the cerebellar region.

[0034] The analysis device can receive medical images.

[0035] Medical imaging can be MRI (Magnetic Resonance Imaging). Medical imaging can be an image of the human brain. Medical imaging can be an image of the human brain taken in the axial, sagittal, and coronal planes. Medical imaging can be an image of the cerebellum. Medical imaging can be a three-dimensional image.

[0036] The analysis device can segment the cerebellar region in medical images. The analysis device can segment the cerebellar region within the cerebellar region in medical images.

[0037] The analysis device can segment a desired region of interest using software for brain region segmentation. The analysis device can segment the cerebellar region using commercial software. The analysis device can segment the cerebellar region into the cerebellar region.

[0038] Meanwhile, the analysis device can segment the cerebellar region in medical images using a segmentation model. The analysis device can segment the cerebellar region into the cerebellar region using a segmentation model.

[0039] The analysis device can calculate the size (or volume) of the cerebellar region. The analysis device can calculate the size (or volume) of the cerebellar region and the cerebellar region. The analysis device can calculate the ratio of the size (or volume) of the cerebellar region to the cerebellar region and the cerebellar region. In other words, the analysis device can calculate the ratio of the size (or volume) occupied by the cerebellar region and the cerebellar region. It is assumed below that the analysis device performs the analysis based on the volume of the region of interest using MRI.

[0041] Mathematical formula 1 is the ratio of volumes ( Ratio It is a formula for calculating ).

[0042]

[0043] The analysis device can analyze the state of cerebellar atrophy in the subject based on the calculated ratio. In other words, the analysis device can analyze the presence of cerebellar atrophy by utilizing the aforementioned principle.

[0045] Below, we examine the results of the analysis device dividing each region of the medical image.

[0046] Figure 3 is an example of segmenting each region of a medical image. Figures 3 (A) and (B) show the results of segmenting the brain region, the cerebellar region, and the cerebellar fluid region within the cerebellar region in a brain MRI image. Figure 3 (A) is an image taken in the sagittal plane. Figure 3 (B) is an image taken in the coronal plane.

[0047] In Fig. 3, the analysis device used a conventional brain image analysis method to segment each region of the brain MRI image. In Fig. 3, the analysis device used a computational anatomy toolbox (CAT) to segment each region of the brain. Based on statistical parametric mapping (SPM), CAT can analyze VBM (Voxel-based morphometry), SBM (Surface-based morphometry), and ROI (Region of Interest) of brain images. SPM may be software designed for the analysis of brain imaging data sequences. CAT can perform spatial registration and noise removal operations using reference data. CAT can perform standardized and normalized image analysis. The analysis device can use Tissues_cat12 and lpba40 among the atlases output as analysis results of CAT. Tissues_cat12 is a segmentation of white matter, gray matter, cerebrospinal fluid (CSF), etc., within the brain image based on VBM. Lpba40 may divide 56 regions, including the cerebellum.

[0048] In Figures 3 (A) and (B), the cerebellum is shown in blue. In Figures 3 (A) and (B), the cerebrospinal fluid region within the cerebellar region is shown in green. In Figures 3 (A) and (B), the remaining region excluding the cerebellar region is shown in red.

[0049] In Figures 3 (A) and (B), the cerebellar region may include a green area and a blue area.

[0050] Figure 3 shows that the analysis device effectively divides each region of the medical image.

[0052] Below, we examine the results of comparing the performance of the proposed cerebellar atrophy analysis method with conventional analysis methods.

[0053] Tables 1 to 4 show the results of comparing the proposed cerebellar atrophy analysis method with the conventional analysis method.

[0054] The data used for verification consisted of a total of 159 subjects. The normal control group consisted of 106 subjects. There were 23 subjects with MSA-C (Medium Spontaneous Atrophy of Cerebellar Atrophy). There were 30 subjects with idiopathic Parkinson's disease. The medical images were t1-weighted brain MR images.

[0055] The conventional method used for comparison is a method in which an analysis device analyzes cerebellar atrophy through the ratio of the total brain volume to the volume of the cerebellum. The conventional method analyzed cerebellar atrophy using the ratio calculated through the following mathematical formula 2.

[0056]

[0057] Table 1 shows the results of a statistical t-test comparing the proposed method with the conventional method.

[0058] MSA-C Non-MSA-C (Normal, IPD) method Average (Deviation) Average (Deviation) p-value Conventional 0.077(0.1) 0.092(0.005) 1.1 e-06 proposal 0.256(0.082) 0.127(0.25) 2.1 e-07

[0059] Table 1 shows that when the Ratio is calculated using the conventional method, the difference between MSA-C and Non-MSA-C is not significant (0.077 vs 0.092). Table 1 shows that when the Ratio is calculated using the proposed method, the difference between MSAC-C and Non-MSAC-C is significant (0.256 vs 0.127).

[0060] Table 1 shows that when the Ratio is calculated using the proposed method, there is a distinct statistical difference between groups compared to the conventional method.

[0062] Table 2 shows the results of classifying medical images into normal individuals or patients with MSA-C.

[0063] Conventional proposal Metrics Average (Deviation) Average (Deviation) ROC-AUC 0.898 (0.07) 0.911 (0.07) Sensitivity 0.95 (0.09) 0.95 (0.09) Specificity 0.99 (0.01) 0.99 (0.01) F1-score 0.7 (0.04) 0.765 (0.09) Accuracy 0.914 (0.01) 0.93 (0.03)

[0064] The ROC-AUC of the conventional method is 0.898. The ROC-AUC of the proposed method is 0.911. The Sensitivity of the conventional method is 0.95. The Sensitivity of the proposed method is 0.95. The Specificity of the conventional method is 0.99. The Specificity of the proposed method is 0.99. The F1-score of the conventional method is 0.7. The F1-score of the proposed method is 0.765. The Accuracy of the conventional method is 0.914. The Accuracy of the proposed method is 0.93. Looking at the results in Table 2, it can be confirmed that the analysis device classifies medical images as normal individuals or patients with cerebellar atrophy better than the conventional technology.

[0066] Table 3 shows the results of the analysis device classifying patients with idiopathic Parkinson's disease (IPD) and cerebellar atrophy (MSA-C).

[0067] Conventional proposal Metrics Average (Deviation) Average (Deviation) ROC-AUC 0.85 (0.14) 0.878 (0.1) Sensitivity 0.79 (0.11) 0.82 (0.09) Specificity 0.65 (0.2) 0.7 (0.17) F1-score 0.85 (0.07) 0.898 (0.05) Accuracy 0.809 (0.1) 0.869 (0.07)

[0068] The ROCAUC of the conventional method is 0.85. The ROCAUC of the proposed method is 0.878. The Sensitivity of the conventional method is 0.79. The Sensitivity of the proposed method is 0.82. The Specificity of the conventional method is 0.65. The Specificity of the proposed method is 0.7. The F1-socre of the conventional method is 0.85. The F1-socre of the proposed method is 0.898. The Accuracy of the conventional method is 0.809. The Accuracy of the proposed method is 0.869. Looking at the results in Table 3, it can be confirmed that the analysis device classifies medical images into patients with idiopathic Parkinson's disease or cerebellar atrophy better than the conventional technology.

[0070] Table 4 shows the results of distinguishing between patients with cerebellar atrophy (Normal, IPD) and patients with cerebellar atrophy.

[0071] Conventional proposal Metrics Average (Deviation) Average (Deviation) ROC-AUC 0.89 (0.07) 0.926 (0.09) Sensitivity 0.96 (0.09) 1 (0) Specificity 0.99 (0.01) 1 (0) F1-score 0.68 (0.04) 0.815 (0.24) Accuracy 0.93 (0.01) 0.962 (0.04)

[0072] The ROCAUC of the conventional method is 0.89. The ROCAUC of the proposed method is 0.926. The Sensitivity of the conventional method is 0.96. The Sensitivity of the proposed method is 1. The Specificity of the conventional method is 0.99. The Specificity of the proposed method is 1. The F1-score of the conventional method is 0.68. The F1-score of the proposed method is 0.815. The Accuracy of the conventional method is 0.93. The Accuracy of the proposed method is 0.962. Looking at the results in Table 4, it can be confirmed that the analysis device classifies medical images as non-patients with cerebellar atrophy or as patients with cerebellar atrophy better than the conventional technology.

[0075] The configuration of the analysis device is described below.

[0076] Figure 4 is an example of an analysis device.

[0077] The analysis device (400) may correspond to the analysis device (100) described in FIG. 1.

[0078] The analysis device (400) may be physically implemented in various forms, such as a PC, laptop, smart device, server, or a chipset dedicated to data processing. The analysis device (400) may include an input device (410), a storage device (420), a computation device (430), an output device (440), an interface device (450), and a communication device (460).

[0079] The input device (410) may include an interface device (keyboard, mouse, touchscreen, etc.) for receiving certain commands or data. The input device (410) may include a configuration for receiving information through a separate storage device (USB, CD, hard disk, etc.). The input device (410) may receive input data through a separate measuring device or through a separate DB. The input device (410) may receive data via wired or wireless communication. The input device (410) may receive medical images. The medical images received by the input device (410) may be MRI images.

[0080] The storage device (420) can store information received through the input device (410). The storage device (420) can store information generated during the process of computation by the computation device (430). That is, the storage device (420) can include memory. The storage device (420) can store the results calculated by the computation device (430). The storage device (420) can store medical images. The storage device (410) can store an object segmentation model.

[0081] The computing device (430) can segment the cerebellar region and the cerebrospinal fluid region within the cerebellar region in the medical image. The computing device (430) can calculate the volume of the cerebellar region and the volume of the cerebrospinal fluid region. The computing device (430) can calculate the ratio between the calculated volumes. Based on the calculated ratio, the computing device (430) can analyze whether the subject has cerebellar atrophy. The computing device (430) can segment regions within the medical image using an object segmentation model.

[0082] The output device (440) may be a device that outputs certain information. The output device (440) may output an interface required for a data process, input data, analysis results, etc. The output device (440) may be implemented in various physical forms, such as a display, a document output device, etc.

[0083] The interface device (450) may be a device that receives certain commands and data from the outside. The interface device (450) may receive conversation information from a physically connected input device or an external storage device. The interface device (450) may receive control signals for controlling the analysis device (400). The interface device (450) may output the results analyzed by the analysis device (400).

[0084] The communication device (460) may refer to a configuration that receives and transmits certain information via a wired or wireless network. The communication device (460) may receive conversation information. The communication device (460) may receive control signals necessary to control the analysis device. The communication device (460) may transmit the results analyzed by the analysis device.

[0086] The aforementioned method for analyzing cerebellar atrophy may be implemented as a program (or application) comprising an executable algorithm that can be executed on a computer. The program may be provided by being stored on a transitory or non-transitory computer-readable medium.

[0087] A non-transient readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short moment, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on a non-transient readable medium such as a CD, DVD, hard disk, Blu-ray disc, USB, memory card, ROM (read-only memory), PROM (programmable read-only memory), EPROM (Erasable PROM, EPROM), EEPROM (Electrically EPROM), or flash memory.

[0088] Transient readable media refers to various types of RAM such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synclink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).

[0089] The embodiments and drawings attached to this specification merely clearly illustrate a part of the technical ideas included in the aforementioned technology, and it is self-evident that variations and specific embodiments that can be easily inferred by a person skilled in the art within the scope of the technical ideas included in the specification and drawings of the aforementioned technology are all included within the scope of the rights of the aforementioned technology.

Claims

Claim 1 A step in which an analysis device receives a medical image of a subject; a step in which the analysis device segments the cerebellar region and the cerebrospinal fluid (CSF) region within the cerebellar region in the medical image; a step in which the analysis device calculates the ratio occupied by the cerebrospinal fluid region within the cerebellar region; and a step in which the analysis device compares the calculated ratio with the ratio occupied by the cerebrospinal fluid region within the cerebellar region in a non-MSA-C state; and a step in which the analysis device classifies the subject as MSA-C or non-MSA-C based on the comparison result. The step of classifying the subject into a state of cerebellar atrophy (MSA-C) or a state of non-cerebellar atrophy (Non-MSA-C) comprises: a step of classifying the subject into a state of non-cerebellar atrophy (Non-MSA-C) when, based on the comparison result, it is determined that the ratio of the cerebellar region occupied by the subject is smaller than the ratio of the cerebellar region occupied by the subject in the state of non-cerebellar atrophy (Non-MSA-C); and the step of classifying the subject into a state of cerebellar atrophy (MSA-C) or a state of non-cerebellar atrophy (Non-MSA-C) comprises: a step of classifying the subject into a state of cerebellar atrophy (MSA-C) when, based on the comparison result, it is determined that the ratio of the cerebellar region occupied by the subject is larger than the ratio of the cerebellar region occupied by the subject in the state of non-cerebellar atrophy (Non-MSA-C). A method for analyzing cerebellar atrophy, comprising: a medical image being an image of a human brain; wherein the non-MSA-C state includes a normal state and idiopathic Parkinson's disease (IPD). Claim 2 A method for analyzing cerebellar atrophy in which the medical image in paragraph 1 is MRI (Magnetic Resonance Imaging). Claim 3 Input device for receiving medical images of a subject; and a computing device that segments the cerebellar region and the cerebrospinal fluid (CSF) region within the cerebellar region in the medical image, calculates the ratio of the CSF region to the cerebellar region, compares the calculated ratio with the ratio of the CSF region to the cerebellar region in a non-MSA-C state, and classifies the subject as MSA or non-MSA-C based on the comparison result; wherein classifying the subject as MSA or non-MSA-C is based on the comparison result being that if it is determined that the ratio of the CSF region to the cerebellar region of the subject is smaller than the ratio of the CSF region to the cerebellar region in a non-MSA-C state, the subject is classified as non-MSA A cerebellar atrophy analysis device comprising classifying the subject into a state (Non-MSA-C), wherein classifying the subject into a cerebellar atrophy state (MSA-C) or a non-MSA-C state includes classifying the subject into a cerebellar atrophy state (MSA-C) when, based on the comparison results, it is determined that the ratio of the cerebellar region to the cerebellar region of the subject is greater than the ratio of the cerebellar region to the cerebellar region in the non-MSA-C state, wherein the medical image is an image of a human brain, and the non-MSA-C state includes a normal state and idiopathic Parkinson's disease (IPD). Claim 4 In paragraph 3, the medical image is a cerebellar atrophy analysis device that is an MRI (Magnetic Resonance Imaging).

Images