Information processing device, information processing method, and program

Abstract

An image generation device of the present disclosure is provided with: a generation unit that generates, using a tomographic image of an eye to be examined, a first analysis result having two-dimensional thickness data of the retina layer of the eye to be examined, and generates a second analysis result different from the first analysis result, by using a machine learning model trained using a plurality of thickness maps and the first analysis result; and a display control unit that controls a display unit so as to display information relating to the second analysis result.

Description

Information processing device, information processing method, and program

[0001] The present disclosure relates to an information processing device, an information processing method, and a program.

[0002] As ophthalmic devices, devices for obtaining two-dimensional images of the fundus of a subject's eye (hereinafter referred to as fundus camera devices) and devices for obtaining tomographic images of a subject's eye using optical coherence tomography (OCT) with low-coherence light (hereinafter referred to as OCT devices) have been put to practical use.

[0003] If the thickness of each layer can be measured from retinal tomographic images taken with these devices, it will be possible to quantitatively diagnose the progression of diseases such as glaucoma and the degree of recovery after treatment. To quantitatively measure the thickness of these layers, the retinal layer thickness obtained from the test eye is compared with statistical values ​​of normal eye thickness (normal eye database) that store the thickness of retinal layers in normal eyes, and any deviations from the statistical values ​​of normal eye thickness are displayed. Note that as the axial length of the eye increases, the eyeball elongates, resulting in a thinner overall retina, which can result in a larger discrepancy when compared with statistical values ​​of thickness of a typical normal eye.

[0004] Here, Patent Document 1 discloses a technology that has a database (first database) that stores fundus layer thickness information for multiple eyes with long axial lengths, as well as a normal eye database (second database), and acquires analytical information by referring to either one of the databases.

[0005] JP 2015-84865 A

[0006] Here, the thickness statistical values ​​stored in the normal eye database are average values ​​for multiple eyes. However, normal thickness values ​​vary among individuals. For example, even if there is a difference between the thickness statistical values ​​stored in the normal eye database and the measured thickness, it may be normal for that individual. In such cases, if a diagnosis is made with reference to the normal eye database, the normal eye may be diagnosed as abnormal, even if the subject's eye is normal.

[0007] In addition to thickness, there are also individual differences in the shape of nerve fibers. For example, although it is considered normal for nerve fibers to have an arched shape, a non-arched shape may be considered normal for an individual. In such cases, for example, if a diagnosis is made with reference to a normal eye database, even a normal subject's eye may be diagnosed as abnormal.

[0008] Therefore, an object of the present disclosure is to improve the accuracy of diagnosis.

[0009] The image generating device of the present disclosure includes a generation unit that uses a tomographic image of the test eye to generate a first analysis result having two-dimensional thickness data of the retinal layer of the test eye, and generates a second analysis result that differs from the first analysis result using a machine learning model trained using multiple thickness maps and the first analysis result, and a display control unit that controls a display unit to display information related to the second analysis result.

[0010] According to the present disclosure, the accuracy of diagnosis can be improved.

[0011] 1 is a diagram showing a schematic configuration example of an image processing system (information processing system) according to embodiment 1. FIG. 1 is a diagram for explaining the structure of an eye, a tomographic image, and a fundus image. FIG. 2 is a diagram for explaining the structure of an eye, a tomographic image, and a fundus image. FIG. 3 is a diagram for explaining the structure of an eye, a tomographic image, and a fundus image. FIG. 4 is a flowchart showing a processing flow in an image processing system. FIG. 2 is a diagram for explaining a screen for displaying images in embodiment 1. FIG. 3 is a diagram for explaining the concept of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 1. FIG. 4 is a diagram for explaining the concept of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 1. FIG. 5 is a diagram for explaining learning of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 1. FIG. 6 is a diagram for explaining inference of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 1. FIG. 7 is a diagram showing a schematic configuration example of an ophthalmologic apparatus according to embodiment 2. FIG. 8 is a diagram for explaining a screen for displaying images in embodiment 2. FIG. 10 is a diagram for explaining an image generated by a difference calculation unit in an image processing device (information processing device) according to embodiment 2. FIG. 11 is a diagram for explaining an image generated by a difference calculation unit in an image processing device (information processing device) according to embodiment 2. FIG. 12 is a diagram showing an example of a schematic configuration of an ophthalmologic apparatus according to embodiment 3. FIG. 13 is a diagram for explaining the concept of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 3. FIG. 14 is a diagram for explaining the concept of an image generation model held by a generation unit in an image processing device (information processing device) according to embodiment 3. FIG. 15 is a diagram for explaining a screen for displaying an image in embodiment 4.

[0012] Hereinafter, exemplary embodiments for implementing the disclosed technology will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, and relative positions of components described in the following embodiments are arbitrary and can be changed depending on the configuration of an apparatus to which the disclosed technology is applied or various conditions. In addition, the same reference numerals are used between drawings to indicate identical or functionally similar elements.

[0013] [Embodiment 1] In this embodiment, an ophthalmic apparatus that performs both OCT imaging and visible light fundus imaging will be described as an example of an ophthalmic apparatus using the disclosed technology. The ophthalmic apparatus of this embodiment detects retinal layers from captured OCT data and generates two-dimensional thickness data of the retinal layers from information on the detected layers. Then, two-dimensional thickness data inferred by an image generation model is generated from the two-dimensional thickness data.

[0014] In addition, the image processing system (information processing system) in this embodiment may be configured so that an ophthalmic device (composite examination unit) using the disclosed technology is connected to a network, receives an order (examination instruction sheet) from a doctor's personal computer, performs an examination of the subject's eye in accordance with the order, and sends the results to the doctor's personal computer.

[0015] FIG. 1 is a diagram showing the configuration of an image processing system 100 (information processing system) including an image processing device 300 according to this embodiment. As shown in FIG. 1, the image processing system 100 is configured by the image processing device 300 being communicably connected to a tomographic imaging device (also called an OCT device) 200, a fundus imaging device 400, an external storage unit 500, a display unit 600, and an input unit 700 via interfaces. Here, "communicable" includes not only direct communication between devices but also communication via other devices. The image processing device 300 is an example of an information processing device.

[0016] The tomographic imaging device 200 is a device that captures tomographic images of the eye. The device used for the tomographic imaging device is, for example, an SD-OCT or SS-OCT. Note that since the tomographic imaging device 200 is a known device, a detailed description will be omitted, and the following description will focus on capturing tomographic images in response to instructions from the image processing device 300.

[0017] 1, a galvanometer mirror 201 is used to scan the fundus with measurement light and defines the fundus imaging range by OCT. A drive control unit 202 controls the driving range and speed of the galvanometer mirror 201 to define the fundus imaging range in the planar direction and the number of scanning lines (scanning speed in the planar direction). For simplicity, the galvanometer mirror is shown as a single unit, but in reality it is composed of two mirrors, one for X scanning and one for Y scanning, allowing the measurement light to scan a desired range on the fundus.

[0018] The focus 203 is for focusing the measurement light on the retinal layer of the fundus through the anterior segment of the eye, which is the subject's eye. The measurement light is focused on the retinal layer of the fundus through the anterior segment of the eye, which is the subject's eye, by a focus lens (not shown). The measurement light that irradiates the fundus is reflected and scattered by each retinal layer and returns.

[0019] The internal fixation lamp 204 is composed of a display unit 241 and a lens 242. The display unit 241 uses a plurality of light-emitting diodes (LDs) arranged in a matrix. The lighting positions of the light-emitting diodes are changed according to the area to be photographed under the control of the drive control unit 202. Light from the display unit 241 is guided to the subject's eye via the lens 242. The light emitted from the display unit 241 is 520 nm, and a desired pattern is displayed by the drive control unit 202.

[0020] The coherence gate stage 205 is controlled by the drive control unit 202 to accommodate differences in axial length of the subject's eye. The coherence gate refers to the position where the optical path of the measurement light and the reference light in OCT is equal. Furthermore, as an imaging method, controlling the position of the coherence gate controls imaging on the retinal layer side or deeper than the retinal layer. Here, the structure of the eye and images acquired by the image processing system will be described using Figures 2A to 2C.

[0021] Fig. 2A shows a schematic diagram of an eyeball. In Fig. 2A, C represents the cornea, CL represents the crystalline lens, V represents the vitreous body, M represents the macula (the center of the macula represents the fovea), and D represents the optic disc. The tomographic imaging apparatus 200 according to this embodiment will be described mainly for the case of imaging the posterior pole of the retina, including the vitreous body, macula, and optic disc. Although not described in this embodiment, the tomographic imaging apparatus 200 can also image the anterior segment of the eye, such as the cornea and crystalline lens.

[0022] FIG. 2B shows an example of a tomographic image of the retina captured by the tomography device 200. In FIG. 2B, AS represents an A-scan, a unit of image acquisition in OCT tomographic images. A group of multiple A-scans constitute a single B-scan. This B-scan is called a tomographic image (or tomogram). In FIG. 2B, V represents the vitreous body, M represents the macula, D represents the optic nerve head, and La represents the lamina cribrosa. Furthermore, L1 represents the boundary between the internal limiting membrane (ILM) and the nerve fiber layer (NFL), L2 represents the boundary between the nerve fiber layer and the ganglion cell layer (GCL), L3 represents the photoreceptor inner-outer segment junction (ISOS), L4 represents the retinal pigment epithelium (RPE), L5 represents Bruch's membrane (BM), and L6 represents the choroid. In the tomographic image, the horizontal axis (the main scanning direction of OCT) is the x-axis, and the vertical axis (depth direction) is the z-axis.

[0023] FIG. 2C shows an example of a fundus image acquired by the fundus image capturing device 400. The fundus image capturing device 400 is a device for capturing fundus images of the eye, and examples of such devices include a fundus camera and a scanning laser ophthalmoscope (SLO). In FIG. 2C, M represents the macula, D represents the optic disc, and the thick curve represents the retinal blood vessels. In the fundus image, the horizontal axis (the main scanning direction of the OCT) is the x-axis, and the vertical axis (the sub-scanning direction of the OCT) is the y-axis. The tomographic imaging device 200 and the fundus image capturing device 400 may be configured as an integrated device or as separate devices.

[0024] The image processing device 300 includes an image acquisition unit 301, a storage unit 302, an image processing unit 303, an instruction unit 304, and a display control unit 305. The image acquisition unit 301 includes a tomographic image generation unit 311, which acquires signal data of a tomographic image captured by the tomographic imaging device 200 and performs signal processing to generate a tomographic image. The image acquisition unit 301 also acquires fundus image data captured by the fundus imaging device 400. The generated tomographic image and fundus image are then stored in the storage unit 302. The storage unit 302 stores various types of information and data (including image data). The storage unit 302 also stores programs that the processing circuitry 250 reads and executes to realize various functions. For example, the storage unit 302 may be implemented by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, a hard disk, an optical disk, or the like. The storage unit 302 stores statistical values ​​of retinal layer thickness for multiple eyes (normal eye database).

[0025] The image processing unit 303 includes an analysis unit 331 and a generation unit 332. The analysis unit 331 detects retinal layers from captured OCT data and generates two-dimensional thickness data of the retinal layers (hereinafter also referred to as a thickness map) as a first analysis result using information on the detected layers. The generation unit 332 has a data generation engine (machine learning model) including a machine learning engine, and uses the thickness map generated by the analysis unit 331 as the first analysis result as input to generate a thickness map as a second analysis result.

[0026] The display control unit 305 controls the display unit 600. The display control unit 305 controls the display unit to display information related to the second analysis result. The information related to the second analysis result is, for example, the thickness map itself as the second analysis result. The information related to the second analysis result is, for example, a comparison result obtained by comparing the second analysis result with the first analysis result. Here, the comparison result includes, for example, difference information between the second analysis result and the first analysis result.

[0027] The analysis unit 331 and the generation unit 332 do not necessarily have to be separate units and may be realized as a single generation unit. Alternatively, the generation unit 332 may be provided in an external device different from the image processing device 3100. In this configuration, the image processing device 3100 outputs the first analysis result generated by the analysis unit 331 to the external device. The external device generates a second analysis result using the input first analysis result and outputs the generated second analysis result to the image processing device 3100. The image processing device 3100 then displays the input second analysis result on the display unit 600.

[0028] The image processing device 300 may be configured using, for example, a general-purpose computer. Alternatively, the image processing device 300 may be configured using a computer dedicated to the ophthalmologic device. The image processing device 300 includes a processor (not shown) and a storage medium including a memory such as an optical disk or a ROM (Read Only Memory). The processor may be a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). The processor is not limited to a CPU or an MPU, but may also be a GPU (Graphics Processing Unit) or an FPGA (Field-Programmable Gate Array). Each component of the image processing device 300 other than the storage unit 302 may be configured by a software module executed by a processor such as a CPU, MPU, or GPU. Each of the components may be configured as a circuit that performs a specific function, such as an ASIC, or an independent device. The processor or circuit may include a digital signal processor (DSP), a data flow processor (DFP), or a neural processing unit (NPU).

[0029] The external memory unit 500 stores information about the subject's eye (such as the patient's name, race, age, sex, and eye measurements such as axial length), as well as the captured image data, shooting parameters, image analysis parameters, and parameters set by the operator, all of which are associated with each other.

[0030] The input unit 700 is, for example, a mouse, a keyboard, or a touch screen, and the operator issues instructions to the image processing device 300 , the tomographic image capturing device 200 , and the fundus image capturing device 400 via the input unit 700 .

[0031] Next, the processing procedure of the image processing device 300 of this embodiment will be described with reference to Fig. 3. Fig. 3 is a flowchart showing the flow of operation processing including imaging and analysis of the entire system in this embodiment.

[0032] In step S301, the subject's eye information acquisition unit (not shown) externally acquires a subject's identification number as information for identifying the subject's eye. Based on the subject's identification number, the unit acquires information about the subject's eye stored in the external storage unit 500 and stores the information in the storage unit 302.

[0033] <Step S302> In step S302, the subject's eye is scanned and photographed. When the operator selects start of scanning (not shown), the tomographic imaging apparatus 200 controls the drive control unit 202 to operate the galvanometer mirror 201 and perform tomographic image scanning. The galvanometer mirror 201 is composed of an X scanner for the horizontal direction and a Y scanner for the vertical direction. Therefore, by changing the orientation of each of these scanners, scanning can be performed in both the horizontal (X) and vertical (Y) directions in the apparatus coordinate system. Furthermore, by simultaneously changing the orientation of these scanners, scanning can be performed in a combined horizontal and vertical direction, enabling scanning in any direction on the fundus plane.

[0034] When performing imaging, various imaging parameters are adjusted. Specifically, at least the position of the internal fixation light, the scan range, the scan pattern, the coherence gate position, and the focus are set. The drive control unit 202 controls the light-emitting diodes of the display unit 241 to control the position of the internal fixation light 204 so as to capture the center of the macula or the optic disc. The scan pattern is set to a raster scan, radial scan, cross scan, or other scan pattern for capturing a three-dimensional volume. In this embodiment, the scan pattern is described as a 3D scan for capturing a three-dimensional volume. After adjusting these imaging parameters, imaging is performed by the operator selecting start imaging (not shown). Note that, although not described in this embodiment, tracking of the subject's eye is performed to reduce the influence of fixational eye movement, thereby scanning the subject's eye.

[0035] <Step S303> In step S303, a tomographic image is generated. The tomographic image generating unit 311 generates a tomographic image by performing a general reconstruction process on each interference signal.

[0036] First, the tomographic image generating unit 311 performs fixed pattern noise reduction on the interference signal. Fixed pattern noise reduction is performed by extracting fixed pattern noise by averaging multiple detected A-scan signals and subtracting this from the input interference signal. Next, the tomographic image generating unit 311 performs desired window function processing to optimize depth resolution and dynamic range, which are in a trade-off relationship when Fourier transform is performed over a finite interval. Next, a tomographic signal is generated by performing FFT processing.

[0037] <Step S304> In step S304, the analysis unit 331 performs layer detection. Boundaries of retinal layers are detected in multiple tomographic images captured by the tomography apparatus 200. The retinal layer boundaries are detected using a machine learning model such as deep learning, or by image processing.

[0038] For example, when data is input into a machine learning model, data according to the design of the machine learning model is output. Output data that is highly likely to correspond to the input data according to the training trends from the training data is output. Furthermore, the likelihood of each type of output data trained from the training data is output as a numerical value. Specifically, for example, when a tomographic image acquired by OCT is input into a machine learning model trained with first training data, the model outputs the label of the imaging region captured in the image, or, depending on the design, the model outputs the probability for each imaging region label. Furthermore, when an OCT tomographic image is input into a machine learning model trained with second training data, the model outputs a labeled image. The input and output data formats and combinations of the paired groups constituting the training data may be an image and the other a numerical value, one consisting of multiple images and the other a string, or both images, and are implemented in a combination appropriate for the embodiment. When an image is divided into rectangular regions for training, retinal layers are detected in each rectangular region. That is, a group of rectangular region images is obtained in which each pixel is labeled. The analysis unit 331 arranges each of the labeled rectangular area images in the same positional relationship as the rectangular area images, combines them, and obtains a labeled image in which the retina is detected. Based on the output labeled image, the boundaries between each layer of the retina are detected as boundary lines.

[0039] In this embodiment, the boundary between the ILM and NFL, the boundary between the NFL and GCL, the ISOS, the RPE, and the BM are detected. Although not shown, other boundary lines, such as the boundary between the outer plexiform layer (OPL) and the outer nuclear layer (ONL), the boundary between the inner plexiform layer (IPL) and the inner nuclear layer (INL), the boundary between the INL and OPL, and the boundary between the GCL and IPL, may also be detected. Then, a process for smoothing the shape of the detected boundary line is performed. For example, the coordinate values ​​of the boundary line shape may be considered as time-series data based on a signal, and the shape of the boundary line may be smoothed using a Savitzky-Golay filter or smoothing processes such as a simple moving average, a weighted moving average, or an exponential moving average.

[0040] <Step S305> In step S305, the detected boundary line and tomographic image are displayed on the display unit 600. FIG. 4 shows an example of a screen displayed on the display unit 600. Reference numeral 410 denotes the entire screen, 401 denotes a patient tab, 402 denotes an imaging tab, 403 denotes a report tab, and 404 denotes a settings tab. The diagonal lines in the report tab 403 indicate the active state of the report screen. In this embodiment, an example of displaying the report screen 410 will be described. Reference numeral 411 denotes a tomographic image, and reference numerals 415 and 416 denote examples in which boundary lines are superimposed on the tomographic image. Reference numeral 412 denotes a retinal thickness graph determined from the boundary lines 415 and 416. Reference numeral 405 denotes an SLO image, and reference numeral 406 denotes a thickness map, which is a first analysis result obtained by analyzing using the boundary line results determined by the analysis unit 331. In this embodiment, an example of a thickness calculated from boundary lines 415 and 416 is shown, in which a color thickness map in which colors are assigned according to thickness values ​​is superimposed on the SLO image 405. Reference numeral 407 denotes a thickness map generated by the generation unit 332 as a second analysis result using the thickness map 406 as input by the analysis unit 331. Like the thickness map 406, the generated thickness map 407 is also a color thickness map in which colors are assigned according to thickness values. Reference numeral 408 denotes a color bar showing the relationship between retinal thickness and color when colors are assigned according to thickness values ​​when the thickness map is displayed on the display unit. The generation unit 332 is a model including an image processing system that outputs a thickness map using, for example, a rule base or machine learning (particularly, deep learning technology). Below, a machine learning algorithm including an image processing system using deep learning technology is described using FIGS. 5A, 5B, 6, 7A, and 7B. In this embodiment, the machine learning model is a trained machine learning model 3325 that has been trained using training data including, for example, a group of thickness maps of normal eyes related to medical images.

[0041] The training data consists of one or more groups of thickness maps of normal eyes. The machine learning models shown in FIGS. 5A and 5B are examples of autoencoders. An autoencoder consists of an encoder and a decoder. Here, the encoder converts input data into low-dimensional data, and the decoder restores data from the low-dimensional data. When training an autoencoder, training is performed using a group of thickness maps corresponding to normal eyes. When inference is performed using a model 3325 trained using a group of thickness maps of normal eyes, in FIG. 5A, a thickness map 501 of a normal eye is tensorized and input to the machine learning model 3325, and the generation unit 332 generates a still image of the tensor output by the machine learning model 3325 to output a thickness map 502 similar to a normal eye. In FIG. 5B, a thickness map 511 of an affected eye is tensorized and input to the machine learning model 3325, and the generation unit 332 generates a still image of the tensor output by the machine learning model 3325 to output a thickness map 512 similar to a normal eye. Here, the normal eye-like thickness map 512 can also be rephrased as a thickness map estimated for the affected eye if it were a normal eye. Note that the tensor in the description of this embodiment refers to a format in which a group of pixel values ​​of an image (thickness map) or the like is expressed as a multidimensional array, and is used as a data input / output format for the machine learning model 3325. Furthermore, images (thickness maps) and tensors can be converted to each other.

[0042] Types of thickness maps include, for example, a thickness map of only the NFL, a thickness map of all layers from the ILM to the RPE, a thickness map of the GCC (Ganglion Cell Complex) including the NFL, GCL, and IPL, etc. In this embodiment, an example of a thickness map of all layers from the ILM to the RPE will be described.

[0043] Next, FIG. 6 shows an example of thickness map learning. This figure is a diagram for explaining preprocessing for learning a thickness map. 611 shows a thickness map of all layers for the left eye. 602 shows a thickness map of all layers for the right eye, and the thickness map 612 is an example of the thickness map 602 flipped left and right. The thickness map 613 is an example of a thickness map of all layers in which the thickness map 603 is rotated using position information of the optic disc 604 and the macula 605 so that the positions of the optic disc 604 and the macula 605 are horizontal. Note that the positions of the optic disc 604 and the macula 605 are detected by the image processing unit 303 when the OCT data used for the learning data is captured. During learning, it is desirable to learn the position information of characteristic parts to a certain extent, as in the thickness maps 611 to 613 shown in FIG. 6. In particular, the left and right eyes are flipped left and right to match one of the orientations. Although not shown, thickness maps may be scaled so that the distance between the macula and the optic disc is the same between different thickness maps. Furthermore, the macular region may be shifted so that it is positioned at the center of the image. While Fig. 6 shows a thickness map obtained when the fixation light position during imaging is set at the center of the macula (or the center of the posterior pole), this is not limiting and a thickness map may be used in which the fixation light position during imaging is set at the center of the optic disc. In this case, in order to align the position information of the characteristic parts to a certain extent, as in the example shown in Fig. 6, the position information of the optic disc detected by the image processing unit 303 may be shifted so as to align it with the center of the image.

[0044] Here, in the example of thickness map 613, the thickness map rotated so that the positions of the optic disc 604 and the macula 605 are horizontal is shown as an example in which the thickness map is the same size as before rotation, but this is not limited to this. For example, even when the thickness map is rotated or shifted, as shown in thickness map 623, a region in which the thickness map can fit may be prepared and the rotation may be performed within that region. For example, if the thickness map size is 256 x 256, the region size is 300 x 300. The example of thickness map 623 shown in the figure shows an example in which the map is rotated so that the positions of the optic disc 604 and the macula 605 are horizontal, and the macula 605 is shifted to the center of the region. Expanding the region as shown in thickness map 623 allows learning to be performed without losing any data, even when the thickness map is rotated or shifted. When expanding the region of the thickness map in this way and performing learning, the thickness maps 611 and 612 are also preprocessed to expand their size in the same way as thickness map 623 for learning.

[0045] The accuracy evaluation and error (loss) calculation between the thickness map in the teacher data assigned for training or validation and the thickness map output by the machine learning model 3325 can be performed using the following calculation methods. Specifically, for example, methods such as MSE (Mean Squared Error) and SSIM (Structural Similarity) can be used to quantify the error or similarity. Furthermore, when calculating the accuracy evaluation and error (loss) in the training of the machine learning model 3325, a calculation target may be selected taking into account semantic regions, which are regions in the thickness map included in the training data that can be divided according to the appearance depicted in the thickness map or information related to the thickness map. Specifically, the semantic regions are retinal thickness values. Non-meaningful regions include marginal regions that appear when an image is rotated, such as in the thickness map 613, and expanded regions, such as in the thickness map 623. Since these margin areas and extension areas are areas in which thickness values ​​are not depicted (and do not affect diagnosis), when training the machine learning model 3325, the performance and characteristics of the machine learning model 3325 can be adjusted by using only the semantic areas that affect diagnosis as the target for accuracy evaluation and error (loss) calculation.

[0046] Next, FIGS. 7A and 7B show an example of inferring a thickness map. When the generation unit 332 generates a thickness map using the machine learning model 3325 that has been trained using a group of thickness maps such as those shown in FIG. 6, the generation unit 332 applies the same preprocessing to the thickness map as during training to perform inference. For example, FIG. 7A shows an example in which the machine learning model 3325 has been trained to fit the left eye during training. In FIG. 7A, when a thickness map 706 for the right eye is input to the generation unit 332, the thickness map for the right eye is flipped left and right and input to the generation unit 332 as thickness map 701, and the output thickness map 702 is flipped left and right to display thickness map 707. Note that if the machine learning model 3325 has been trained to fit the right eye during training, the thickness map for the left eye is flipped left and right and input to the generation unit 332 as thickness map 701.

[0047] Similarly, Figure 7B shows an example in which the tilt is adjusted using position information of the optic disc and macula during learning, and learning is performed using an expanded region, as shown in thickness map 623. In Figure 7B, when thickness map 716 is input to the generation unit 332, thickness map 711 is generated by rotating the thickness map so that the positions of the optic disc and macula are horizontal within the same region size as during learning. This is then input to the generation unit 332, and thickness map 712 output from the generation unit 332 is rotated so that the positions of the optic disc and macula are restored, and the expanded region is clipped, and thickness map 717 returned to the original size is displayed.

[0048] In the present embodiment, the thickness map is learned or inferred using the thickness values ​​themselves, and colors are assigned according to the thickness values ​​when displayed on the display unit 600. However, this is not limiting. Learning and inference may be performed using a color thickness value map obtained by converting thickness values ​​into colors. In this case, the inferred thickness map may be displayed on the display unit 600 without assigning color values ​​according to the thickness values.

[0049] <Step S306> In step S306, an instruction acquisition unit (not shown) acquires an instruction from outside as to whether or not to terminate the capturing of tomographic images by the image processing system 100. This instruction is input by the operator using the input unit 700. If an instruction to terminate the processing is acquired, the image processing system 100 terminates the processing. On the other hand, if the processing is not to be terminated and the capturing is to continue, the process returns to step S302 and the capturing is continued. In this manner, the processing of the image processing system 100 is performed.

[0050] According to the above-described configuration, in this embodiment, a normal-looking thickness map can be inferred from an input thickness map. Because the thickness map is based on the input, it is possible to display a thickness map that shows only the normal thickness while maintaining the structure of the subject's eye.

[0051] [Variation 1 of the First Embodiment] In this embodiment, an autoencoder has been described as an example of a machine learning model, but this is not limiting. For example, a variational autoencoder, a convolutional autoencoder, etc. may also be used. Furthermore, generative adversarial networks and their derivatives such as AnoGAN, Efficient-GAN, and GANomally may also be used.

[0052] [Embodiment 2] In the first embodiment, an example was shown in which a thickness map inferred from a thickness map of an input image was displayed. In this embodiment, difference information between the thickness map of the input image and the inferred thickness map is generated. Furthermore, an object is to generate difference information between the thickness map of the input image and statistical values ​​of thicknesses of normal eyes (normal eye database) and display this difference information to the operator.

[0053] FIG. 8 is a diagram illustrating an example of the schematic configuration of an image processing system 1000 including an image processing device 3103 according to the second embodiment. In FIG. 8, components similar to those shown in FIG. 1 are denoted by the same reference numerals, and detailed description thereof will be omitted. Here, the image processing unit 3103 in the image processing system 1000 of FIG. 8, which differs from that of the first embodiment, will be described. The image processing unit 3103 includes a first difference calculation unit 333 and a second difference calculation unit 334. The first difference calculation unit 333 calculates difference information (hereinafter also referred to as a difference thickness map) between the retinal layer thickness map, which is the first analysis result, and the thickness map generated by the generation unit 332, which is the second analysis result. The second difference calculation unit 334 calculates difference information between the retinal layer thickness map, which is the first analysis result, and retinal layer thickness statistics for multiple eyes stored in the storage unit 302. Then, the display unit 600 displays at least two of the following: the retinal layer thickness map which is the first analysis result; the retinal layer thickness map which is the second analysis result; the difference information generated by the first difference calculation unit 333; and the difference information generated by the second difference calculation unit 334.

[0054] This will be described with reference to FIG. 9 . FIG. 9 shows a report screen 410 displayed on the display unit 600 in this embodiment, and is an example of a display with a different layout than that shown in FIG. 4 . FIG. 9 shows an example in which a retinal layer thickness map 406, a thickness map 407 generated by the generation unit 332, a differential thickness map 927 generated by the first difference calculation unit 333, and a differential thickness map 926 generated by the second difference calculation unit 334 are displayed together with a tomographic image 411 and a vertical plane tomographic image 912. Here, the thickness map and differential thickness map will be described with reference to FIGS. 10A and 10B . FIG. 10A shows a thickness map 406, which is a first analysis result generated from the retinal layers analyzed by the analysis unit, in which the course of nerve fibers 1001 is typical and the thickness is normal, but in which a bulge is visible around the macular region 1002 due to disease. The thickness map 407 is a thickness map generated by the generation unit 332 using the thickness map 406 as input, and is an example in which the thickness around the macula is output as a normal-looking thickness. The differential thickness map 927 is difference information generated by the first difference calculation unit 333 and is a differential thickness map between the thickness maps 406 and 407. The differential thickness map 926 is difference information generated by the second difference calculation unit 334 and is a differential thickness map between the thickness map 406 and the statistical values ​​of retinal layer thickness for multiple eyes stored in the memory unit 302. Note that the differences calculated by the first difference calculation unit 333 and the second difference calculation unit 334 represent the differences between different thickness maps, such as differences or ratios, in multiple thickness maps. As shown in FIG. 10A , in the case of a typical eye, the difference information displays differences 1003 and 1008 due to bulging information around the macula.

[0055] In FIG. 10B , thickness map 1406 is a first analysis result generated from the retinal layer analyzed by the analysis unit, showing an example in which the course of nerve fibers 1011 is atypical but the thickness is normal, and a bulge is visible around the macular region 1012 due to disease. Thickness map 1407 is a thickness map generated by the generation unit 332 using thickness map 1406 as input, showing an example in which the thickness around the macular region appears normal. Difference thickness map 1927 is difference information generated by the first difference calculation unit 333, and is a difference thickness map between thickness maps 1406 and 1407. Difference thickness map 1926 is difference information generated by the second difference calculation unit 334, and is a difference thickness map between thickness map 1406 and statistical values ​​of retinal layer thickness for multiple eyes stored in the memory unit 302. As shown in difference thickness map 1926, difference information 1014 is displayed in a location where the thickness is within the normal range but deviates from the statistical values. As described here, the generation unit 332 generates a thickness map based on the input thickness map, and therefore generates a thickness map based on learned information that is less affected by thickness due to individual differences in shape, thickness, and axial length.

[0056] 10A and 10B are displayed on a report screen 410. Furthermore, on the report screen 410, grids 916 and 917 are superimposed on the retinal layer thickness maps 406 and 407, respectively, and the numerical values ​​within each grid are displayed in grids 9160 and 9170. Although not shown, the numerical values ​​within the grids may be displayed not only in the grids but also in a table format or a graph format.

[0057] In the report screen 410 of this embodiment, the thickness map and the differential thickness map are displayed side by side as described above, but the present invention is not limited to this. For example, it is possible to switch between the thickness map generated by the analysis unit 331 and the thickness map generated by the generation unit 332 and display them, and to switch between the differential thickness maps accordingly.

[0058] According to the configuration described above, in this embodiment, the operator can make an integrated judgment by referring to any of the thickness map obtained by the analysis unit analyzing the retinal layers, the thickness map generated by the generation unit, the difference information generated by the first difference calculation unit, and the second difference calculation unit.

[0059] Third Embodiment In the first and second embodiments, examples of generating a thickness map and generating difference information of the thickness map were described. In this embodiment, an example will be described in which a condition acquisition unit is provided that acquires conditions at the time of imaging and conditions for the thickness map, and a thickness map is generated based on these conditions.

[0060] FIG. 11 is a diagram illustrating an example of the schematic configuration of an image processing system 2000 including an image processing device 3203 according to the third embodiment. In FIG. 11, components similar to those shown in FIGS. 1 and 8 are denoted by the same reference numerals, and detailed description thereof will be omitted. Here, the image processing unit 3203 in the image processing system 2000 of FIG. 11, which differs from those in the first and second embodiments, will be described. The image processing unit 3203 includes a condition acquisition unit 335. The generation unit 3320 generates a thickness map based on the conditions acquired by the condition acquisition unit 335. In this regard, a machine learning algorithm including an image processing system using deep learning technology will be described with reference to FIGS. 12A and 12B. FIGS. 12A and 12B are diagrams illustrating an image generation model 33250 included in the generation unit 3320 according to the third embodiment.

[0061] 12A and 12B inputs the thickness map 1201 converted into a tensor and the thickness map tensor assigned with a scalar value indicating the condition 335 to the machine learning model 33250. The generation unit 3320 then converts the tensor output by the machine learning model 33250 into a still image and outputs it as the thickness map 1202.

[0062] For example, conditions acquired by the condition acquisition unit 335 include imaging conditions, conditions of the subject's eye, and analysis conditions. Imaging conditions include, for example, imaging date and time, name of the imaging region, imaging area, imaging angle of view, imaging method, image resolution and gradation, image pixel size, image filter, and information on the image data format. Information on the subject's eye includes, for example, age (date of birth), gender, and race. Analysis conditions include, for example, information on retinal layers contained in a thickness map. Information on retinal layers is information on layers contained in a thickness map generated from a tomographic image (type of thickness map). Types of thickness maps include, for example, a thickness map having the thickness of each layer, a thickness map having the thickness of all layers from the ILM to the RPE, and a thickness map of the GCC including the NFL, GCL, and IPL.

[0063] FIG. 12A illustrates an example in which a condition indicating a thickness map of all layers is input as an analysis condition along with a thickness map, and a thickness map corresponding to the condition is output as an output. FIG. 12B illustrates an example in which a condition indicating a thickness map of GCC is input as an analysis condition along with a thickness map, and a thickness map corresponding to the condition is output as an output. For example, the scalar value of the analysis condition is 001001 for a thickness map of all layers, and 001002 for a GCC thickness map. In addition to the cases of FIGS. 12A and 12B, other conditions may be input, such as the center of the macula as the imaging site, 10×10 mm as the imaging area, and age, sex, race, and eye-related measurement values ​​(axial length, refractive value, etc.) as information about the subject's eye. Note that the thickness map does not necessarily need to be assigned a single condition; multiple conditions may be assigned simultaneously. Alternatively, multiple conditions may be expressed as a single scalar value.

[0064] According to the configuration described above, in this embodiment, by learning conditions together with the thickness map when learning the thickness map, various types of thickness maps can be learned simultaneously. Then, by inputting conditions together with the thickness map during inference, a thickness map based on the input conditions can be output during output.

[0065] [Embodiment 4] In this embodiment, an example of displaying a time-series thickness map will be described. The time-series thickness map is a thickness map generated using each of a plurality of tomographic images taken at different times. Descriptions of components having the same functions as those in the first to third embodiments will be omitted here.

[0066] FIG. 13 is a display example of a time-series thickness map in this embodiment. In FIG. 13, the time-series thickness maps are displayed arranged horizontally. Below each thickness map, a difference thickness map is displayed arranged vertically. Reference numeral 1301 denotes a thickness map (Thickness Map) generated by the analysis unit. Reference numeral 1302 denotes a difference thickness map (Deviation) between the retinal layer thickness map 1301 generated by the second difference calculation unit 334 and the statistical values ​​of retinal layer thickness stored in the storage unit 302. Reference numeral 1303 denotes a difference thickness map (Estimate Deviation) between the retinal layer thickness map, which is the first analysis result generated by the first difference calculation unit 333, and the thickness map, which is the second analysis result generated by the generation unit 332.

[0067] As shown in this embodiment, multiple thickness maps and multiple differential thickness maps can be displayed. The differential thickness map 1303 generated by the first difference calculation unit 333 may display the difference from a reference map, which is a thickness map (second analysis result) inferred from a thickness map (first analysis result) obtained at a certain reference time. Alternatively, the differential thickness map 1303 may display the difference between each thickness map (each first analysis result) inferred from each thickness map obtained at each time (each second analysis result). Alternatively, the image processing unit 3203 may have an integration unit (not shown) that uses multiple thickness maps (second analysis results) inferred from each thickness map (first analysis result) obtained at each time to generate a statistical thickness map (e.g., an average map obtained by averaging each second analysis result) by integrating the respective second analysis results, and display the difference from the statistical thickness map. The statistical thickness map integrating the respective second analysis results is an example of a third analysis result.

[0068] Furthermore, although not shown, thickness maps generated by the generation unit 332 at each time may be displayed. These thickness maps may be displayed side by side or may be displayed alternately. Furthermore, instead of using time-series data, thickness maps may be generated from the thickness maps of the left and right eyes, and a difference thickness map may be generated.

[0069] According to the configuration described above, in this embodiment, a normal-like thickness map can be inferred from thickness maps of multiple inputs, such as time series and left and right eye thickness maps. Then, differences from statistical values ​​of normal eye thickness and differences from the inferred thickness map can be simultaneously displayed.

[0070] (Variation 1) In the various embodiments described above, examples have been described in which only one type of thickness map is trained, and examples have been described in which multiple types of thickness maps are trained together with conditions (e.g., imaging conditions, conditions of the subject's eye, analysis conditions), thereby supporting multiple types of thickness maps. However, this is not limiting. For example, a configuration may be adopted in which a machine learning model trained on one or several types of thickness maps for each condition is trained, and the machine learning model used is switched depending on the conditions. For example, as an example of switching a machine learning model trained depending on conditions, conditions of the subject's eye, such as race, age, and gender, are learned together for each type of thickness map, such as a thickness map of each layer, a thickness map of all layers, or a GCC thickness map. Then, during inference, a machine learning model may be selected depending on the analysis conditions (type of thickness map used as input), and inference may be performed together with the thickness map and the conditions of the subject's eye. This example is not limiting, and a configuration may be adopted in which machine learning models are trained for each type of thickness map collectively and for each condition of the subject's eye (e.g., race, gender, age, etc.), and a machine learning model is selected depending on the conditions of the subject's eye during inference. Regarding age, rather than dividing by one year, it is better to group them into ranges such as 30s to 40s, 50s to 60s, etc.

[0071] (Variation 2) In the various embodiments described above, examples of learning using training data including a group of thickness maps of normal eyes have been described. However, this is not limiting. For example, learning may be performed using training data including a group of thickness maps of diseased eyes. In this case, when a normal thickness map is input, this machine learning model outputs a thickness map similar to that of a diseased eye. Then, when a thickness map of a diseased eye is input, it outputs a thickness map of the diseased eye. Therefore, the more normal the eye, the more the difference is calculated when difference information from the output thickness map is generated. Furthermore, machine learning models may be trained for each type of disease. For example, disease types include glaucoma, diabetic retinopathy, and age-related macular degeneration. By performing inference using a machine learning model tailored to each disease, it is possible to predict a disease by determining that the result with the smallest difference between the input and output thickness maps is most likely to be the disease.

[0072] (Other Embodiments) The disclosed technology can also be realized by executing the following process. That is, the disclosed technology can also be realized by providing software (programs) that realize one or more functions of the various embodiments described above to a system or device via a network or a storage medium, and having a computer (or a CPU, MPU, etc.) of the system or device read and execute the programs. The computer has one or more processors or circuits, and may include multiple separate computers or a network of multiple separate processors or circuits to read and execute computer-executable instructions. In this case, the processor or circuit may include a central processing unit (CPU), a microprocessing unit (MPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), or a field-programmable gateway (FPGA). The processor or circuit may also include a digital signal processor (DSP), a data flow processor (DFP), or a neural processing unit (NPU).

[0073] (Configuration 1) An information processing device comprising: a generation unit that generates a first analysis result having two-dimensional thickness data of the retinal layer of a test eye using a tomographic image of the test eye, and generates a second analysis result different from the first analysis result using a machine learning model trained using multiple thickness maps and the first analysis result; and a display control unit that controls a display unit to display information related to the second analysis result.

[0074] (Configuration 2) The information processing device according to Configuration 1, wherein the machine learning model is a machine learning model trained using the plurality of thickness maps corresponding to a plurality of normal eyes.

[0075] (Configuration 3) The information processing device according to either of Configurations 1 or 2, wherein the display control unit controls the display unit to display a comparison result between the first analysis result and the second analysis result as information about the second analysis result.

[0076] (Configuration 4) The information processing device according to any one of configurations 1 to 3, wherein the display control unit controls the display unit to display at least one of a comparison result between the first analysis result and the second analysis result, and a comparison result between the first analysis result and a thickness map obtained from a normal eye database.

[0077] (Configuration 5) The information processing device according to any one of Configurations 1 to 4, wherein the generation unit generates a second analysis result that is different from the first analysis result by using the machine learning model, the first analysis result, and information about the subject's eye.

[0078] (Configuration 6) The information processing device according to Configuration 5, wherein the information relating to the subject's eye is at least one of the subject's race, the subject's age, the subject's sex, and the axial length of the subject's eye.

[0079] (Configuration 7) The information processing device according to any one of Configurations 1 to 6, wherein the generation unit generates a second analysis result that is different from the first analysis result by using the machine learning model, the first analysis result, and information about retinal layers contained in the first analysis result.

[0080] (Configuration 8) The information processing device according to Configuration 7, wherein the information on the retinal layers is information on layers included in the first analysis result.

[0081] (Configuration 9) The information processing device according to any one of configurations 1 to 8, wherein the display control unit controls the display unit to display the first analysis result and the second analysis result side by side.

[0082] (Configuration 10) The information processing device according to any one of configurations 1 to 9, wherein the display control unit controls the display unit to display side by side a comparison result between the first analysis result and the second analysis result, a comparison result between the first analysis result and a thickness map acquired from a normal eye database, the first analysis result, and the second analysis result.

[0083] (Configuration 11) An information processing device described in any of configurations 1 to 10, wherein the display control unit controls the display unit to display side by side: a plurality of first analysis results generated using each of a plurality of tomographic images obtained by photographing the subject's eye at different times; each of the comparison results between each of the plurality of first analysis results and a thickness map obtained from a normal eye database; and each of the comparison results between each of the plurality of first analysis results and a second analysis result generated using the machine learning model and at least one of the plurality of first analysis results.

[0084] (Configuration 12) An information processing device described in any of configurations 1 to 11, wherein the display control unit controls the display unit to display side by side: a plurality of first analysis results generated using each of a plurality of tomographic images obtained by photographing the subject's eye at different times; a comparison result between each of the plurality of first analysis results and a thickness map obtained from a normal eye database; and a comparison result between each of the plurality of first analysis results and a third analysis result obtained by averaging each of the plurality of second analysis results generated using the machine learning model and the plurality of first analysis results.

[0085] (Configuration 13) The information processing device described in any of configurations 1 to 12, wherein the generation unit has the machine learning model that receives as input a first analysis result of one of the right eye and the left eye of the subject eye, and the generation unit, when receiving the first analysis result of one of the right eye and the left eye of the subject eye, generates the second analysis result by inputting the first analysis result of the one of the right eye and the left eye of the subject eye into the machine learning model, and when receiving the first analysis result of the other of the right eye and the left eye of the subject eye, generates the second analysis result by flipping the other first analysis result left and right and inputting it into the machine learning model.

[0086] (Configuration 14) An information processing device comprising: an analysis unit that generates a first analysis result having two-dimensional thickness data of the retinal layer of a test eye using a tomographic image of the test eye; and a display control unit that controls a display unit to display information regarding a second analysis result that is different from the first analysis result and that is generated using a machine learning model trained using multiple thickness maps and the first analysis result.

[0087] (Configuration 15) An information processing system including: an OCT device that captures a tomographic image of a subject's eye; and the image generating device according to any one of Configurations 1 to 14 that is communicably connected to the OCT device.

[0088] (Method 1) An information processing method comprising: a generation step of generating a first analysis result having two-dimensional thickness data of the retinal layer of a test eye using a tomographic image of the test eye, and generating a second analysis result different from the first analysis result using a machine learning model trained using a plurality of thickness maps and the first analysis result; and a display control step of controlling a display unit to display information related to the second analysis result.

[0089] (Program 1) A program that causes a computer to execute the information processing method described in Method 1.

[0090] The present invention is not limited to the above-described embodiments, and various modifications and variations can be made without departing from the spirit and scope of the present invention. Therefore, the following claims are appended to apprise the public of the scope of the present invention.

[0091] This application claims priority based on Japanese Patent Application No. 2023-202721, filed November 30, 2023, the entire contents of which are incorporated herein by reference.

[0092] REFERENCE SIGNS LIST 100 Image processing system 200 Tomographic imaging device (OCT device) 300 Image processing device 400 Fundus image capturing device 500 External storage unit 600 Display unit 700 Input unit

Claims

1. An information processing device comprising: a generation unit that uses a tomographic image of a test eye to generate a first analysis result having two-dimensional thickness data of the retinal layer of the test eye, and generates a second analysis result different from the first analysis result using a machine learning model trained using multiple thickness maps and the first analysis result; and a display control unit that controls a display unit to display information regarding the second analysis result.

2. The information processing device according to claim 1, wherein the machine learning model is a machine learning model trained using the multiple thickness maps corresponding to multiple normal eyes.

3. An information processing device according to claim 1, wherein the display control unit controls the display unit to display a comparison result between the first analysis result and the second analysis result as information relating to the second analysis result.

4. The information processing device of claim 1, wherein the display control unit controls the display unit to display at least one of a comparison result between the first analysis result and the second analysis result, and a comparison result between the first analysis result and a thickness map obtained from a normal eye database.

5. The information processing device according to claim 1, wherein the generation unit uses the machine learning model, the first analysis result, and information regarding the subject's eye to generate a second analysis result that is different from the first analysis result.

6. The information processing device according to claim 5, wherein the information relating to the subject's eye is at least one of the subject's race, the subject's age, the subject's sex, and the axial length of the subject's eye.

7. The information processing device of claim 1, wherein the generation unit generates a second analysis result different from the first analysis result using the machine learning model, the first analysis result, and information regarding retinal layers contained in the first analysis result.

8. An information processing device according to claim 7, wherein the information relating to the retinal layers is information about the layers included in the first analysis result.

9. The information processing device according to claim 1, wherein the display control unit controls the display unit to display the first analysis result and the second analysis result side by side.

10. The information processing device of claim 1, wherein the display control unit controls the display unit to display side by side a comparison result between the first analysis result and the second analysis result, a comparison result between the first analysis result and a thickness map obtained from a normal eye database, and the first analysis result and the second analysis result.

11. The information processing device of claim 1, wherein the display control unit controls the display unit to display side by side: a plurality of first analysis results generated using each of a plurality of tomographic images obtained by photographing the test eye at different times; a comparison result between each of the plurality of first analysis results and a thickness map obtained from a normal eye database; and a comparison result between each of the plurality of first analysis results and a second analysis result generated using the machine learning model and at least one of the plurality of first analysis results.

12. The information processing device according to claim 1, wherein the display control unit controls the display unit to display side by side: a plurality of first analysis results generated using each of a plurality of tomographic images obtained by photographing the test eye at different times; a comparison result between each of the plurality of first analysis results and a thickness map obtained from a normal eye database; and a comparison result between each of the plurality of first analysis results and a third analysis result obtained by averaging each of the plurality of second analysis results generated using the machine learning model and the plurality of first analysis results.

13. The information processing device according to claim 1, wherein the generation unit has a machine learning model that receives as input a first analysis result of one of the right eye and the left eye of the subject eye, and when the first analysis result of one of the right eye and the left eye of the subject eye is input, the generation unit generates the second analysis result by inputting the first analysis result of the one of the right eye and the left eye of the subject eye into the machine learning model, and when the first analysis result of the other of the right eye and the left eye of the subject eye is input, the generation unit generates the second analysis result by flipping the other first analysis result left and right and inputting it into the machine learning model.

14. An information processing device comprising: an analysis unit that generates a first analysis result having two-dimensional thickness data of the retinal layer of a test eye using a tomographic image of the test eye; and a display control unit that controls the display unit to display information regarding a second analysis result different from the first analysis result that is generated using a machine learning model trained using multiple thickness maps and the first analysis result.

15. An information processing system comprising: an OCT device that captures a tomographic image of a test eye; and an image generating device according to any one of claims 1 to 14 that is communicatively connected to the OCT device.

16. An information processing method comprising: a generation step of generating a first analysis result having two-dimensional thickness data of the retinal layer of a test eye using a tomographic image of the test eye, and generating a second analysis result different from the first analysis result using a machine learning model trained using multiple thickness maps and the first analysis result; and a display control step of controlling a display unit to display information related to the second analysis result.

17. A program for causing a computer to execute the information processing method according to claim 16.

Images