Prediction system, prediction device, prediction method, control program, and recording medium

Abstract

To highly accurately predict effects of a drug administered to a subject.SOLUTION: A prediction system includes an acquisition unit and a prediction unit. The acquisition unit acquires subject information including a medical image showing bones of a subject. The prediction unit outputs drug prediction information regarding effects of drugs acting on at least one of bone formation and bone absorption, using a drug effect prediction model for predicting effects of drugs on the basis of input information including the subject information and drug information on drugs.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to a prediction system, a prediction device, a prediction method, and a control program for predicting the effect of a drug acting on bone.

Background Art

[0002] Patent Document 1 discloses a technique for predicting the effect of an anticancer agent.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[0004] A prediction system according to one aspect of the present disclosure includes an acquisition unit that acquires target information including a medical image of a target bone, and a prediction unit that outputs pharmacodynamic prediction information regarding the effect of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the target information. The prediction unit outputs the pharmacodynamic prediction information using a pharmacodynamic prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0005] A prediction device according to one aspect of the present disclosure includes an acquisition unit that acquires target information including a medical image of a target bone, and a prediction unit that outputs pharmacodynamic prediction information regarding the effect of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the target information. The prediction unit outputs the pharmacodynamic prediction information using a pharmacodynamic prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0006] Further, a prediction method according to an aspect of the present disclosure includes an acquisition step in which a computer acquires target information including a medical image in which a target bone is imaged, and a prediction step in which the computer outputs drug efficacy prediction information regarding the effect of a drug having at least one of an effect on bone formation and an effect on bone resorption from input information including the target information. In the prediction step, the drug efficacy prediction information is output using a drug efficacy prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0007] Further, a prediction system according to each aspect of the present disclosure may be implemented by one or more computers. In this case, a control program for a prediction system that realizes the prediction system by operating a computer as each part (software element) included in the prediction system, and a computer-readable recording medium on which the control program is recorded also fall within the scope of the present disclosure.

Brief Description of Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0009] Drugs for improving the state of a target bone include drugs that act on bone formation, drugs that act on bone resorption, and drugs that adjust the balance of bone remodeling, etc. These drugs are administered to the target alone or in combination of multiple types.

[0010] According to one aspect of the present disclosure, the effect of a drug when administered to a target can be accurately predicted.

[0011] Hereinafter, the present disclosure will be described in detail.

[0012] The state of a target bone can be affected by the balance between bone formation and bone resorption, lifestyle habits such as diet and exercise, and genetics, etc. For example, osteoporosis is a disease characterized by low bone mass and abnormal microstructure of bone tissue, with increased bone fragility and an increased risk of fracture. Osteoporosis is often seen in the elderly and postmenopausal women. Osteoporosis is not limited to these, and can also occur, for example, due to local circulatory disorders, calcium metabolism disorders, etc.

[0013] If a fracture occurs at a site important for motor function due to deterioration of the state of a target bone, surgery and hospitalization may be required. As a result, the QOL (quality of life) of the target can be significantly reduced. In order to reduce the decline in the QOL of the target, it is important to detect the tendency of deterioration of the state of the target bone as early as possible and start an intervention by an appropriate method including administration of an appropriate drug.

[0014] Hereinafter, each embodiment according to the present disclosure will be described. In the following description, when the subject is a human (i.e., "subject"), it will be described by way of example, but the subject is not limited to humans. The subject may be, for example, a non-human mammal such as a horse family, a cat family, a dog family, a bovine family, or a pig family. And the present disclosure includes embodiments in which the "subject" is replaced with "animal" among the following embodiments, as long as they are embodiments applicable to these animals.

[0015] [Embodiment 1] (Overview of Prediction Device 1) A prediction device 1 according to an aspect of the present disclosure outputs drug efficacy prediction information regarding the effect of a drug when administered to a subject (hereinafter sometimes referred to as "subject") from subject information that is information regarding the subject and drug information regarding the drug.

[0016] The subject information includes at least any one of fracture risk factor information regarding the subject's fracture risk factors, bone strength information regarding the subject's bone strength, information regarding the subject's health condition, bone density information indicating the bone density of the subject's bone, bone metabolism information indicating the bone metabolism state of the subject's bone, and a medical image in which the subject's bone appears. In the present disclosure, the subject's bone may be a plurality of bones included in each part of the subject's body such as the chest, waist, neck, hip joint, head, leg, arm, finger, and foot. Alternatively, in the present disclosure, the subject's bone may be a specific bone such as the subject's spine, thoracic vertebra, lumbar vertebra, pelvis, cervical vertebra, skull, femur, fibula, tibia, radius, middle phalanx, phalanx, calcaneus, middle metatarsal bone, and phalanx.

[0017] The subject information may include fracture risk factor information regarding the subject's fracture risk factors. The fracture risk factor information can use, for example, information or its results used by a fracture risk assessment tool (FRAX (registered trademark); Fracture Risk Assessment Tool). Also, the fracture risk factor information may include at least one piece of information from the following information group A.

[0018] (Information group A): Type (e.g., race), age, gender, weight, height, presence or absence of fractures, fracture location, fracture history, fracture history of family (e.g., parents), glucocorticoid, rheumatoid arthritis, secondary osteoporosis, underlying diseases (e.g., food and / or drug allergies, diseases related to the onset of osteoporosis and diseases not related, etc.), smoking history, drinking habits (e.g., drinking frequency and amount of alcohol, etc.), occupational history, exercise history, medical history (e.g., medical history of bone diseases), menstruation (e.g., cycle and presence or absence, etc.), menopause (e.g., possibility and presence or absence, etc.), estimated time of menopause (e.g., estimated year, month, or date, etc.), artificial joints (e.g., type, presence or absence, and time of replacement surgery such as spinal implants or knee joints), results of blood tests, results of urine tests, drugs being taken, gene sequences.

[0019] Information group A is information that is useful for determining the bone density, bone metabolism, fracture possibility, and balance between bone formation and bone resorption of the subject. Therefore, by using target information including at least one piece of information from information group A, the prediction device 1 can more accurately predict the effect of the drug to be administered to the subject.

[0020] Information regarding the presence or absence of fractures may be detected based on a medical image of the subject. The fracture location may be information regarding the fracture location of the subject estimated based on a medical image of the subject. Alternatively, the information regarding the presence or absence of fractures may be information regarding locations where the subject is predicted to have a high possibility of future fractures.

[0021] Information regarding the results of blood tests may be, for example, information regarding the results of at least any one of biochemical tests, sugar metabolism tests, and endocrine tests.

[0022] Biochemical tests may include, for example, information regarding the results of at least one test from the following blood test group B.

[0023] (Blood test group B): Total protein (TP), albumin (ALB), aspartate aminotransferase (AST), alanine aminotransferase (ALT), lactate dehydrogenase (LDH), bone ALP, creatinine (Cre), calcium (Ca), inorganic phosphorus (IP), total cholesterol (T-Cho), high-density lipoprotein cholesterol (HDL-C), triglyceride (TG), low-density lipoprotein cholesterol (calculated LDL-C), C-reactive protein (CRP).

[0024] The carbohydrate metabolism test may include information on at least one test result among the following carbohydrate metabolism test group C, for example.

[0025] (Carbohydrate metabolism test group C): Glucose (Glu), hemoglobin A1c (HbA1c).

[0026] The endocrine test may include information on at least one test result among the following endocrine test group D, for example.

[0027] (Endocrine test group D): Intact parathyroid hormone (intact PTH), 25-hydroxyvitamin D (25(OH)D), 1,25-dihydroxyvitamin D (1,25(OH)2D), thyroid-stimulating hormone (YSH), CS, estradiol (E2), free testosterone (free-TST).

[0028] The information on the blood test result may include information on at least one test result among the following urine test group E, for example.

[0029] (Urine test group E): Calcium (Ca), phosphorus (P), creatinine (Cre).

[0030] The information on the drugs being taken may include information such as drug name, amount being taken, duration of taking the drugs, etc. The information on the drugs being taken may also include information on the steroid drugs being used.

[0031] The bone density information may be, for example, a measured value obtained by measuring the bone density of a predetermined site of a subject. The bone density can be measured by methods such as DXA (dual - energy X - ray absorptiometry), ultrasonic method, and MD (micro densitometry) method. Also, the bone density can be estimated from a simple X - ray image, an MRI (magnetic resonance imaging) image, a CT (computed tomography) image, an ultrasonic image, and a PET image in which the subject's bone appears. The DXA method is a method of measuring bone density using X - rays of two different energies. The DXA method can generally be measured for the lumbar spine, proximal femur, and distal radius. In a DXA device for measuring bone density using the DXA method, when the bone density of the lumbar spine is measured, X - rays are irradiated from the front of the subject's lumbar spine. Also, in the DXA device, when the bone density of the proximal femur is measured, X - rays are irradiated from the front of the subject's proximal femur. The ultrasonic method is a method of measuring bone density by applying ultrasonic waves to bones such as the lumbar spine, femur, heel, and tibia. Here, "the front of the lumbar spine" and "the front of the proximal femur" are intended to be the directions facing the imaging sites such as the lumbar spine and the proximal femur correctly, and may be the ventral side of the subject's body or the dorsal side of the subject's body. Note that the proximal femur includes, for example, at least one of the neck, trochanter, diaphysis, and the entire proximal femur (neck, trochanter, and diaphysis). The MD method is a method of measuring bone density by simultaneously imaging the bones of the hand and an aluminum plate using X - rays and comparing the shades of the imaged images. When measuring bone density by the MD method, X - rays are irradiated from the back of both hands. The location of the bone measured by the MD method is, for example, the second metacarpal bone of the hand. By using the bone density information indicating the bone density measured by the DXA method with high measurement accuracy, the accuracy of the drug efficacy prediction information output from the prediction device 1 can be improved. The bone density information may be information regarding bone quality described later instead of bone density, or may include information regarding bone quality in addition to bone density.

[0032] The bone density information may be the bone density (estimated value) of the subject's bone estimated by the prediction device 1 from a medical image in which the subject's bone is shown. Alternatively, the bone density information may be the future bone density (predicted value) of the subject's future bone predicted by the prediction device 1 from a medical image in which the subject's bone is shown. Thus, the prediction device 1 may be configured to estimate and / or predict the future bone density of the subject's bone from a medical image in which the subject's bone is shown. In this case, the prediction device 1 may include a learned bone density estimation / future prediction model using the medical image in which the subject's bone is shown as an explanatory variable and the measured value of the bone density of the subject's bone in the medical image as an objective variable. The prediction device 1 may include a learned future bone density prediction model using the medical image in which the subject's bone is shown as an explanatory variable and the measured value of the bone density of the subject's bone after a predetermined period has elapsed from the time when the medical image was taken as an objective variable.

[0033] The prediction device 1 may calculate at least one of the estimated value and the predicted value of the bone density of the bone from the entire bone or a part of the medical image. When the prediction device 1 calculates the estimated value and the predicted value of the bone density from a part of the medical image, it may calculate at least one of the estimated value and the predicted value of the bone density from a plurality of parts. When the prediction device 1 partially calculates the estimated value and the predicted value of the bone density, the prediction device 1 may partially calculate a region mainly including cancellous bone or a region mainly including cortical bone. By being able to partially calculate at least one of the estimated value and the predicted value of the bone density from the medical image in this way, it becomes easier to predict fractures of the vertebra or the proximal part of the femur. Not limited to bone density, the bone quality described later can also be calculated from the entire bone or a part of the medical image. As a method of focusing on bone density or bone quality from a part of the medical image, for example, a learned segmentation method or the like can be used.

[0034] Bone density can use values used, for example, in osteoporosis guidelines (not limited to this, such as, for example, the 2015 edition of the Prevention and Treatment Guidelines of the Japanese Osteoporosis Society, hereinafter the same). Also, bone density only needs to be a value related to the density of bone, and an independent index can be used. Specifically, bone density is, for example, bone mineral density per unit area (g / cm 2 ), bone mineral density per unit volume (g / cm 3 ), YAM (%), T-score, and Z-score, and may be represented by at least one of them. YAM is an abbreviation of "Young Adult Mean" and may be called young adult mean percentage or young mean percentage. Bone density information may include information regarding its location (for example, at least one of the proximal femur and vertebrae) in addition to at least one of the measured value, estimated value, and predicted value of bone density.

[0035] Here, the prediction device 1 may be configured to use, as an alternative to a medical image, the bone density of the subject's bone measured by a DXA device, MD method, ultrasonic method, etc. as bone density information. In this case, the subject can use, for example, the bone density measured by a DXA device within one year as bone density information. Or, the bone density information may be the future bone density of the subject's bone predicted by the prediction device 1 from at least one of the bone density measured by a DXA device, fracture risk factor information, and bone metabolism information.

[0036] Bone metabolism information may include at least any one of osteoblast-related bone formation marker information, osteoclast-related bone resorption marker information, and bone matrix-related marker information regarding bone mass of the subject. Bone metabolism information can be analyzed, for example, from a blood or urine sample of the subject. As bone metabolism information, for example, data analyzed before drug treatment can be used. Bone metabolism information may be, for example, data analyzed once, or may be multiple data obtained by analyzing samples collected multiple times. When there are multiple data, they can be used as a change rate, an average value, or a baseline value as bone metabolism information. Also, as bone metabolism information, for example, in addition to that analyzed before drug treatment, that analyzed after drug treatment (for example, after 3 to 6 months have passed since the start of drug administration, etc.) can be used.

[0037] Here, the bone formation marker information may be information regarding at least any one of alkaline phosphatase and type I procollagen-N-propeptide. The bone resorption marker information may be information regarding at least any one of deoxypyridinoline, type I collagen cross-linked N-telopeptide, type I collagen cross-linked C-telopeptide, tartrate-resistant acid phosphatase-5b. The bone matrix-related marker information may be information regarding at least any one of undercarboxylated osteocalcin, pentosidine, and homocysteine.

[0038] Bone quality can be determined based on, for example, at least one of the statistical properties of bone, the morphological properties of bone, the mechanical properties of bone, and the chemical properties of bone. Bone quality can be determined based on, for example, at least one of bone metabolism markers, gender, race, presence or absence of menopause, age, cortical bone state, cancellous bone state, cancellous bone trabecular state, disease information, bone evaluation information, drug information, and presence or absence of fractures. More specifically, bone quality can be determined using, for example, but not limited to, at least one of bone formation markers, bone resorption markers, bone quality markers (e.g., vitamin K value), cortical bone thickness, trabecular bone density, trabecular bone orientation, and trabecular bone score. The trabecular bone score is an index obtained from raw data (density grayscale image) of a lumbar spine DXA image in the anterior-posterior direction (front-back direction of the subject).

[0039] Disease information may include, for example, at least one of osteoporosis, rheumatism, osteonecrosis (e.g., femoral head osteonecrosis), systemic sclerosis, kidney disease, and marble bone disease. Bone evaluation information may include information evaluated by a fracture risk assessment tool. Drug information may include, for example, at least one of the trade name, generic name, dosage, administration period, and administration method (e.g., oral, intravenous injection, intramuscular injection, subcutaneous injection, etc.) of a drug containing at least one of a drug that suppresses bone resorption, a drug that promotes bone formation, and other drugs (e.g., calcium preparations, vitamin preparations, female hormone preparations, etc.).

[0040] Bone quality may also include, for example, the type of medullary cavity shape. The medullary cavity shape can be classified, for example, using the Dorr classification. The medullary cavity shape can be classified as follows using at least one of the cortical bone thickness and the medullary cavity shape. Type A: A type with thick cortical bone and a narrow and thin medullary cavity. Type B: An intermediate between Type A and Type C, with a medullary cavity that is neither narrow nor wide. Type C: A type with thin cortical bone and a wide medullary cavity.

[0041] The prediction device 1 can use any medical image that can analyze the bone mineral density, bone shape, and the state of tissues around the bone (such as tendons, etc.) of the subject to be predicted. Therefore, the medical image may include, for example, at least one of a simple X-ray image, a CT (Computed Tomography) image, an MRI (Magnetic Resonance Imaging) image, a PET image, and an ultrasonic image. When using a simple X-ray image as the medical image, the imaging site of the simple X-ray image is not particularly limited. The simple X-ray image used as the medical image may show at least any one of the subject's head, neck, chest, waist, hip joint, knee joint, ankle joint, foot, toes, shoulder joint, elbow joint, wrist joint, hand, fingers, or temporomandibular joint. Also, the estimation simple X-ray image used to estimate the bone mineral density and fracture location, etc. of the subject's bone may be a frontal image in which the target site is imaged from the front or a lateral image in which the target site is imaged from the side. When using a CT image as the medical image, for example, at least one of a three-dimensional image, a cross-sectional image in a direction perpendicular to the body axis connecting the head and legs (for example, a horizontal section), and a cross-sectional image in a direction parallel to the body axis (for example, a sagittal section or a coronal section, etc.) can be used.

[0042] When the site to be analyzed is the subject's hip joint (such as the femoral neck and the femoral head), the prediction device 1 may use a simple X-ray image taken from a direction different from the front and the side, such as an axial image and a Rauenstein image, as the medical image.

[0043] The target information may include time information corresponding to each piece of information included in the target information. The time information may be the time or point in time when each piece of information included in the target information was acquired, the examination date on which the subject was examined, the diagnosis date on which the subject was diagnosed, and the date on which each piece of information was input into an electronic medical record, etc.

[0044] The target information may be information including the time information as described above, or may be information substantially regarding the target person at the current time. "Substantially at the current time" is intended to regard a predetermined period as the same as the current time. Here, the predetermined period may be set arbitrarily. For example, within the past three months, within the past six months, or within the past year can be set as the predetermined period. More specifically, even if the time point when one piece of target information is acquired and the time point when other target information is acquired deviate within the predetermined period, they can be regarded as the same time point. The target information may be, for example, information regarding the target person at the time when the bone density information and bone metabolism information of the target person are acquired, or at the time when a medical image of the target person's bone is taken.

[0045] The target information may be information regarding a past or future target person instead of the current time. For example, the target information may be information regarding the target person at the time when the bone density information and bone metabolism information of the target person were acquired in the past, or at the time when a medical image of the target person's bone was taken in the past.

[0046] The drug information may include information regarding a drug having at least any one of an action on bone formation, an action on bone resorption, and an action of adjusting the balance of bone remodeling. The drug information may include at least information regarding the name of the drug and / or its mechanism of action. By using such drug information, the prediction device 1 can more accurately predict the effect of the drug to be administered to the target person. The drug information may include information regarding the chemical structural formula and physical properties of each drug.

[0047] For the name of the drug, for example, at least one of the chemical name, generic name, trade name, and common name of the active ingredient can be used. As the name of the drug, a development code of a drug manufacturer, a clinical trial institution, and / or a medical institution, etc. may be used. These names of the drugs are not limited to only one, and a plurality of names may be retained. In this case, the usability of the data can be improved. As information on the mechanism of action, for example, biochemical information on the mechanism by which the medicinal ingredient contained in the drug acts on the target molecule, physiological and pharmacological information on the conditions important for the medicinal ingredient to act on the target molecule, etc. may be used. Alternatively, the information on the mechanism of action may include, for example, information on at least one of the action on bone formation, the action on bone resorption, and the action of adjusting the balance of bone remodeling.

[0048] In addition, the drug information may include auxiliary information regarding the dosage form, manufacturer, lot number, production number, administration method, administration conditions, administration period, administration interval, contraindication information, side effect information, drug price, price, and approved country of each drug. The administration method may include, for example, oral administration or injection, etc. Also, the injection in the administration method may include information regarding the injection site such as intravenous, intramuscular, intrathecal, or subcutaneous. When using the drug information including these auxiliary information, the prediction device 1 may output the drug efficacy prediction information together with the auxiliary information regarding each drug and present the drug efficacy prediction information and the auxiliary information to medical personnel. Alternatively, the prediction device 1 may output the drug efficacy prediction information regarding the effect of the drug when the drug is administered, for each auxiliary information. Thereby, the prediction device 1 can assist medical personnel who determine a treatment plan including drug administration to the subject. Here, the effect of the drug is the influence exerted on the subject when the drug is administered to the subject, including the degree and / or presence or absence of the influence.

[0049] The drug efficacy prediction information may be fracture risk information indicating how the risk of the subject's bone fracture changes due to the administration of a drug to the subject. The fracture risk information in the drug efficacy prediction information includes, for example, information indicating any of an increase in fracture risk, the same level, and a decrease. Alternatively, the drug efficacy prediction information may be a prediction result of the future bone density of the subject's bone. Alternatively, the drug efficacy prediction information may be a prediction result of the current bone density of the subject's bone. The drug efficacy prediction information may be information regarding whether the drug under consideration for administration to the subject can improve the state of the subject's bone. Alternatively, the drug efficacy prediction information may be information regarding to what extent the drug under consideration for administration to the subject can improve the state of the subject's bone. The drug efficacy prediction information may be information indicating the state of the subject's bone after continuing the administration of the drug to the subject for a predetermined period, or information indicating the change in the subject's bone around a predetermined period. Some drugs acting on the state of the bone do not have immediate efficacy and exhibit effects after administration for a predetermined period. According to the above configuration, the prediction device 1 can output drug efficacy prediction information indicating the effect of a drug even if the drug does not have immediate efficacy.

[0050] The drug efficacy prediction information may be prediction information regarding a drug for which administration to a subject should be considered, and which can change the state of the subject's bone when the administration to the subject is continued for a predetermined period. In this case, the drug efficacy prediction information may include at least any of information regarding a drug having an effect on bone formation, information regarding a drug having an effect on bone resorption, and information regarding a drug that adjusts the balance of bone remodeling. When the drug efficacy prediction information includes information regarding both of the above drugs, action information indicating whether the drug included in the drug efficacy prediction information acts on bone formation or acts on bone formation may be associated with each drug.

[0051] Agents having an effect on bone formation include, but are not limited to, for example, teriparatide acetate, and teriparatide (recombinant), etc. Agents having an effect on bone resorption include, but are not limited to, for example, calcitonin preparations, bisphosphonate preparations, and anti-RANKL monoclonal antibodies, etc. Agents having an effect on adjusting the balance of bone remodeling include, for example, active vitamin D3 preparations (such as calcitriol, eldecalcitol, or alfacalcidol, etc.).

[0052] More specifically, the drug efficacy prediction information may be information indicating the degree of improvement of a predetermined drug for a subject in percentage, or may be information indicating the improvement ratio of bone density in percentage.

[0053] In addition, the drug efficacy prediction information may include, for example, a plurality of pieces of prediction information regarding the drug efficacy according to the administration period for administering the drug to the subject, etc. Specifically, the drug efficacy prediction information may include two or more pieces of prediction information regarding the drug efficacy when the administration period for administering the drug to the subject is, for example, 3 months, 6 months, 1 year, and 3 years later, etc. Alternatively, when the drug efficacy prediction information includes prediction information for a plurality of drugs, the appropriate administration periods for each drug may be different. Or, the drug efficacy prediction information may be prediction information regarding the drug efficacy at the time when the classification defined in the guidelines, etc. changes (for example, the time when the class changes from the "osteoporosis" class to the "osteopenia" class).

[0054] In addition, the drug efficacy prediction information may include a prediction image showing a prediction result regarding how the bones of the subject will change depending on the drug to be administered. Such drug efficacy prediction information utilizes subject information including a medical image in which the subject's bones are depicted. The prediction image may show, in addition to the prediction result indicating the change in the bones of the subject, a prediction result indicating the influence on tissues around the bones and the like. Further, a plurality of prediction images may be displayed as the administration period elapses, or the prediction image may be changed like a video in one image area. The display of a plurality of prediction images may be, for example, a parallel display of prediction images corresponding to a plurality of administration periods and the like.

[0055] Also, the drug efficacy prediction information may include side effect prediction information regarding side effects predicted when a drug is administered to the subject. The side effect prediction information may be, for example, information regarding at least one side effect among the following side effect groups.

[0056] (Side effect group) Fatigue, allergic reaction, anaphylactic shock, headache, vomiting, gastrointestinal disorder, constipation, vascular disorder, heartburn, kidney disorder, liver disorder, jaundice, hypercalcemia, hypocalcemia, osteonecrosis of the jaw, osteomyelitis of the jaw, osteonecrosis of the external auditory canal, urinary tract stone, spasm, palpitation, loss of appetite, and hyperuricemia.

[0057] The side effect prediction information may include prediction information regarding specific side effects that may occur in the subject administered with the drug, or may include information (such as occurrence probability) indicating the possibility of one or more side effects occurring in the subject. The side effect prediction information may include a plurality of prediction information regarding the occurrence of side effects according to the administration period and the like for administering the drug to the subject. Further, the side effect prediction information may include prediction information predicting how the probability of side effects occurring in the subject changes due to the intake of other drugs, diet, or daily activities different from the drug to be predicted. For example, the side effect prediction information may include prediction information indicating how the probability of side effects occurring in the subject changes when taking a drug different from the drug to be predicted to what extent (for example, intake period and intake amount).

[0058] When a drug is administered to a subject, the prediction device 1 outputs drug efficacy prediction information that accurately predicts the effect of the drug. Thereby, the prediction device 1 can present to the subject and medical personnel in charge of the subject, etc., at a time when the drug has not been administered, how the state of the subject's bone can change due to the administration of the drug.

[0059] (Configuration of Prediction System 100a) First, the configuration of the prediction system 100a according to one aspect of the present disclosure will be described with reference to FIG. 1. FIG. 1 is a diagram showing a configuration example of the prediction system 100a in the medical facility 8 in which the prediction device 1 is introduced. Hereinafter, as an example, the prediction device 1 that acquires a medical image in which the bone of the subject is imaged will be described as the target information.

[0060] The prediction system 100a includes the prediction device 1 and one or more terminal devices 7 communicably connected to the prediction device 1. The prediction device 1 is a computer that outputs drug efficacy prediction information from a medical image showing the bone of the subject and transmits the drug efficacy prediction information to the terminal device 7. The prediction device 1 may be a cloud-type device or an on-premises type device provided within the medical facility 8 or a company that provides an analysis service.

[0061] The terminal device 7 (reception unit) functions as a reception unit that receives drug efficacy prediction information in the prediction system 100a. That is, the terminal device 7 receives drug efficacy prediction information from the prediction device 1 and presents the drug efficacy prediction information. The terminal device 7 is a computer used by medical personnel such as doctors belonging to the medical facility 8. The terminal device 7 is, for example, a personal computer, a tablet terminal, a smartphone, or the like. The terminal device 7 has a communication unit that performs data transmission and reception with other devices, an input unit such as a keyboard and a microphone, a display unit that can display information included in the drug efficacy prediction information, an output unit such as a speaker, and the like. Although an example is shown in which a LAN (local area network) is arranged in the medical facility 8 shown in FIG. 1 and the prediction device 1 and the terminal device 7 are connected to the LAN, the present invention is not limited thereto. For example, the network in the medical facility 8 may be the Internet, a telephone communication line network, an optical fiber communication network, a cable communication network, a satellite communication network, or the like.

[0062] In the LAN in the medical facility 8, in addition to the prediction device 1 and the terminal device 7, a medical image management device 6 (storage unit) that can function as a storage unit for storing target information in the prediction system 100a may be communicably connected. The medical image management device 6 is a computer that functions as a server for managing medical images captured at the medical facility 8. In this case, the prediction device 1 may acquire a medical image in which the bones of the target person are shown from the medical image management device 6.

[0063] In the LAN in the medical facility 8, in the prediction system 100a, an electronic medical record management device 5 (storage unit) that can function as a storage unit for storing target information may be communicably connected. The electronic medical record management device 5 is a computer that functions as a server for managing various target information related to the target person. In this case, the prediction device 1 may acquire various target information related to the target person from the electronic medical record management device 5.

[0064] The LAN within the medical facility 8 may be communicably connected to an external communication network. In the medical facility 8, the prediction device 1 and the terminal device 7 may be directly connected without going through the LAN.

[0065] (Configuration of the Prediction Device 1) Subsequently, the configuration of the prediction device 1 applied to the prediction system 100a shown in FIG. 1 will be described with reference to FIG. 3. FIG. 3 is a block diagram showing an example of the configuration example of the prediction device 1.

[0066] The prediction device 1 includes a control unit 2 that comprehensively controls each part of the prediction device 1, and a storage unit 3 that stores various data used by the control unit 2. The control unit 2 includes an acquisition unit 21, a drug information application unit 22, a prediction unit 23, an output control unit 24, and a learning unit 25. In addition to the control program 31, which is a program for performing various controls of the prediction device 1, the storage unit 3 stores learning data 32, a drug database 33, and a learned drug efficacy prediction model 34 (drug efficacy prediction model). In one example, target information may be stored in the storage unit 3.

[0067] <Acquisition Unit 21> The acquisition unit 21 acquires target information including at least any one of risk factor information regarding the fracture risk factors of the subject, bone density information indicating the bone density of the subject, bone metabolism information indicating the bone metabolism state of the subject, and a medical image in which the bone of the subject appears. For example, the acquisition unit 21 shown in FIG. 3 may be able to acquire risk factor information regarding the fracture risk factors of the subject, bone density information indicating the bone density of the subject, and bone metabolism information indicating the bone metabolism state from the electronic medical record management device 5. The acquisition unit 21 shown in FIG. 3 may be able to acquire a medical image, which is an image of the subject's bone, from the medical image management device 6. The target information is input data that is input to the prediction unit 23.

[0068] <Drug Information Application Unit 22> The drug information application unit 22 acquires drug information regarding a drug having at least any one of information on an action on bone formation, an action on bone resorption, and a drug for adjusting the balance of bone remodeling. For example, as shown in FIG. 3, the drug information application unit 22 may read the drug database 33 of the storage unit 3 to acquire drug information. Alternatively, the drug information application unit 22 may acquire drug information from an external drug information management device via a network such as the Internet. The drug information can be input data input to the prediction unit 23.

[0069] <Prediction unit 23> The prediction unit 23 generates and outputs pharmacodynamic prediction information regarding the effect of a drug when the drug is administered to a subject by inputting the subject information into a learned pharmacodynamic prediction model 34 that predicts the effect of the drug based on the drug information regarding the drug having an action on bone. The prediction unit 23 may output at least any one of a prediction result indicating the difference between the future bone density and the current bone density of the subject's bone, and a prediction result indicating the change of the subject's bone density from the current time as the pharmacodynamic prediction information.

[0070] For example, when medical images and bone metabolism information of a subject are input as the subject information, the prediction unit 23 may be configured to output a list of drugs to be considered for administration to the subject and pharmacodynamic information (described later) regarding the effect of each drug included in the list even if the drug information regarding the drug to be administered to the subject is not input. That is, the input of drug information is not essential for the prediction unit 23 to output the pharmacodynamic information.

[0071] The prediction unit 23 may output a prediction result regarding the fracture site of the subject. Alternatively, the prediction unit 23 may output a prediction result indicating the possibility that the subject will fracture in the future. Specific examples of the information output from the prediction unit 23 will be described later (see FIG. 7).

[0072] The learned drug efficacy prediction model 34 may be learned using learning data 32 that uses subject information regarding each of a plurality of subjects to which a drug has been administered and drug information regarding the drug administered to each of the subjects as explanatory variables, and the effect of the drug in each of the subjects as the objective variable. The learned drug efficacy prediction model 34 is, for example, a model obtained as a result of the learning unit 25 described later performing learning processing on an unlearned neural network. The subject may be of the same species as the target, and hereinafter, the case where the subject is a human will be described as an example.

[0073] The prediction unit 23 may output a prediction result when a drug is administered and a prediction result when a drug is not administered. In this case, the learned drug efficacy prediction model 34 may be learned using learning data 32 that uses subject information regarding each of a plurality of subjects to which a drug has been administered and subject information regarding each of a plurality of subjects to which a drug has not been administered as explanatory variables, and the effect of the drug in each of the subjects as the objective variable.

[0074] The prediction unit 23 may output a prediction result when different types of drugs are administered. In this case, the learned drug efficacy prediction model 34 may be learned using learning data 32 that uses subject information regarding each of a plurality of subjects to which drug A has been administered and subject information regarding each of a plurality of subjects to which a drug B of a different type from drug A has been administered as explanatory variables, and the effect of the drug in each of the subjects as the objective variable.

[0075] The prediction unit 23 may output a prediction result when different combinations of drugs are administered. In this case, the learned drug efficacy prediction model 34 may be learned using learning data 32 that uses subject information regarding each of a plurality of subjects with different drug combinations as explanatory variables, and the effect of the drug in each of the subjects as the objective variable.

[0076] The prediction unit 23 performs calculations based on the learned drug efficacy prediction model in response to the input of target information (e.g., medical images) and drug efficacy information to the input layer 231 (see FIG. 4), and outputs drug efficacy prediction information from the output layer 232 (see FIG. 4). As an example, the prediction unit 23 may be configured to extract feature quantities from the target information and the drug efficacy information and use them as input data. Known algorithms such as those listed below can be applied to the extraction of the feature quantities. · Convolutional Neural Network (CNN). · Auto encoder. · Recurrent Neural Network (RNN). · LSTM (Long Short-Term Memory). · ConvLSTM (Convolutional Long Short-Term Memory).

[0077] The learned drug efficacy prediction model 34 is a calculation model used when the prediction unit 23 performs calculations based on the input data. The learned drug efficacy prediction model 34 is generated by performing machine learning on an unlearned neural network using the learning data 32 described later. The learning data 32, the configuration of the neural network, and specific examples of the learning process will be described later.

[0078] <Output control unit 24> The output control unit 24 transmits the drug efficacy prediction information output from the prediction unit 23 to the terminal device 7. The prediction device 1 may be configured to include a display unit (not shown). In that case, the output control unit 24 causes the display unit to display the drug efficacy prediction information.

[0079] <Learning unit 25> The learning unit 25 controls the learning process for an unlearned neural network. By performing the learning process on the unlearned neural network, the learning unit 25 creates a neural network that functions as the prediction unit 23 from the neural network. This learning uses learning data 32 (described later). A specific example of the learning performed by the learning unit 25 will be described later.

[0080] (Learning data 32) The learning data 32 is data including explanatory variables and objective variables used for machine learning to generate a learned drug efficacy prediction model 34 from an unlearned neural network. In the learning data 32, subject information regarding each of a plurality of subjects to whom a drug has been administered, and drug information regarding the drug administered to each of the subjects are explanatory variables, and the effect of the drug in each of the subjects is the objective variable.

[0081] Here, the subject information is information including at least any one of risk factor information regarding the fracture risk factor of the subject, bone density information indicating the bone density of the subject, bone metabolism information indicating the bone metabolism state of the subject, and a medical image in which the bone of the subject is shown. The subject information may include risk factor information regarding the fracture risk factor of the subject. The fracture risk factor information may include at least one piece of information from the above-mentioned information group A. Here, the bone density information may be a measured value obtained by measuring the bone density of a predetermined site of the subject. The bone density may be measured by a method such as DXA (dual-energy X-ray absorptiometry), an ultrasonic method, and an MD (micro densitometry) method. Alternatively, the bone density may be estimated from a simple X-ray image, an MRI (magnetic resonance imaging) image, a PET (Positron Emission Tomography) image, and a CT (computed tomography) image in which the bone of the subject is shown.

[0082] The effect of the drug in each subject may be the degree to which the drug administered to each subject improves the bone condition of each subject. Alternatively, the effect of the drug in each subject may be information indicating the bone condition of each subject after the drug has been continuously administered for a predetermined period, or the change in the bone condition before and after the period during which the drug has been continuously administered for a predetermined period.

[0083] (Configuration of Prediction Unit 23) Hereinafter, the configuration of the prediction unit 23 will be described with reference to FIG. 4. The configuration shown in FIG. 4 is an example, and the configuration of the prediction unit 23 is not limited thereto.

[0084] As shown in FIG. 4, the prediction unit 23 performs an operation on the input data input to the input layer 231 based on the learned drug effect prediction model 34, and outputs drug effect prediction information from the output layer 232.

[0085] The prediction unit 23 in FIG. 4 includes a neural network having an input layer 231 and an output layer 232. The neural network may be any neural network suitable for handling time-series information. For example, it may be an LSTM or the like. The neural network may be any neural network suitable for handling a combination of time-series information and position information. For example, the neural network may be a ConvLSTM network that combines a CNN and an LSTM. The input layer 231 can extract the feature amount of the temporal change of the input data. The output layer 232 can calculate a new feature amount based on the feature amount extracted by the input layer 231, the temporal change of the input data, and the initial value. The input layer 231 and the output layer 232 have a plurality of LSTM layers. Each of the input layer 231 and the output layer 232 may have three or more LSTM layers.

[0086] (Learning Process by Learning Unit 25) Hereinafter, the learning process for generating the learned drug effect prediction model 34 will be described with reference to FIG. 5. FIG. 5 is a flowchart showing an example of the flow of the learning process by the learning unit 25.

[0087] The learning unit 25 acquires learning data 32 from the memory unit 3 (step S1). The learning data 32 includes explanatory variables to be input to the input layer 231. Here, the explanatory variables are subject information regarding each of a plurality of subjects to which a drug has been administered, and drug information regarding the drug administered to each of the subjects.

[0088] Subsequently, the learning unit 25 inputs subject information regarding a certain subject A and drug information regarding the drug administered to subject A to the input layer 231 (step S2).

[0089] Next, the learning unit 25 acquires output data regarding the effect of the drug administered to subject A from the output layer 232 (step S3). This output data contains the same content as the target variable of the learning data 32. In FIG. 5, the order of step S2 and step S3 may be reversed. Alternatively, in FIG. 5, step S2 and step S3 may be executed simultaneously.

[0090] Subsequently, the learning unit 25 acquires the target variable regarding subject A included in the learning data 32. Then, the learning unit 25 compares the output data acquired in step S3 with the target variable regarding subject A, calculates an error (step S4), and adjusts the drug effect prediction model during learning so that the error is reduced (step S5).

[0091] For the adjustment of the drug effect prediction model during learning, any known method can be applied. For example, the error backpropagation method may be adopted as a method for adjusting the drug effect prediction model. The adjusted drug effect prediction model becomes a new drug effect prediction model, and in subsequent calculations, the prediction unit 23 uses the new drug effect prediction model. At the adjustment stage of the drug effect prediction model, the parameters (for example, filter coefficients, weighting coefficients, etc.) used in the prediction unit 23 can be adjusted.

[0092] When the error does not fall within a predetermined range and when the explanatory variables for all the subjects included in the learning data 32 have not been input (NO in step S6), the learning unit 25 returns to step S2 and repeats the learning process. When the error falls within a predetermined range and when the explanatory variables for all the subjects included in the learning data 32 have been input (YES in step S6), the learning unit 25 ends the learning process.

[0093] When the learning process as described above is adopted, the prediction unit 23 can output drug efficacy prediction information regarding the effect of the drug to be administered to the subject from the target information and the drug information. Also, when the learning process as described above is adopted, the prediction unit 23 can output drug efficacy prediction information regarding one or more drugs that are candidates for the drug to be administered to the subject from the target information.

[0094] 〔Another Embodiment〕 The learning data 32 may include, as explanatory variables, information regarding each of a plurality of combinations of drugs administered to the subjects, and may include, as objective variables, information regarding the effects of the plurality of drugs in each of the specimens. When the learning process using such learning data 32 is adopted, the prediction unit 23 can output drug efficacy prediction information for each of the combinations of a plurality of drugs to be administered to the subject.

[0095] (Processing Performed by the Prediction Device 1) Hereinafter, the flow of the process performed by the prediction device 1 will be described with reference to FIG. 6. FIG. 6 is a flowchart showing an example of the flow of the process performed by the prediction device 1. FIG. 6 shows an example of the process of outputting drug efficacy prediction information from the target information.

[0096] The target information regarding the target bone is stored in a medical image management device 6, an electronic medical record management device 5, etc. (storage step). First, the acquisition unit 21 acquires the target information (step S11: acquisition step) and inputs it to the prediction unit 23.

[0097] The prediction unit 23 generates and outputs drug efficacy prediction information regarding the effect of a drug when administered to the subject by inputting the subject information into the learned drug efficacy prediction model 34 (step S12: generation step and prediction step). The drug efficacy prediction information output from the prediction unit 23 may be transmitted to the terminal device 7, and in this case, the terminal device 7 receives the drug efficacy prediction information (reception step).

[0098] In step S12, drug information regarding a drug having at least one of an action on bone formation and an action on bone resorption may be input into the learned drug efficacy prediction model 34 together with the subject information.

[0099] (Medical service realized by the prediction device 1) FIG. 7 is a diagram showing an example of a medical service realized by the prediction device 1.

[0100] When the subject is examined at the medical facility 8 based on the results of a regular health check or subjective symptoms such as suspicion of a fracture, subject information regarding the subject can be collected by the user (e.g., a medical professional) during the examination. The subject information is information including at least one of bone density information indicating the bone density of the acquired bone, bone metabolism information indicating the bone metabolism state of the subject's bone, and a medical image in which the subject's bone appears.

[0101] When the user inputs the acquired subject information into the prediction device 1, the prediction device 1 outputs drug efficacy prediction information regarding the drug (and / or its candidate) to be administered to the subject. Alternatively, it may be configured such that the user inputs the acquired subject information and drug information regarding the drug (and / or its candidate) to be administered to the subject into the prediction device 1.

[0102] The prediction device 1 may output, as drug efficacy prediction information, the future bone density when the subject has not received drug administration and the future bone density when the subject has received drug administration. Here, the output bone density is the bone mineral density per unit area (g / cm 2 ), the bone mineral density per unit volume (g / cm 3) may be represented by at least one of YAM (%), T-score, and Z-score. YAM is an abbreviation of "Young Adult Mean" and may be called young adult average percentage or young average percentage. For example, from the output layer 230, bone density represented by bone mineral density per unit area (g / cm 2 ) and bone density represented by YAM may be output, or bone density represented by YAM, bone density represented by T-score, and bone density represented by Z-score may be output. Also, the output bone density may be categorized such that a T-score of -1.0 or higher is displayed as normal bone density, a T-score from -1.0 to -2.5 is displayed as low bone density or osteopenia, and a T-score of -2.5 or lower is categorized as osteoporosis. Alternatively, it may be a YAM ratio. The YAM ratio is a value indicating a diagnostic criterion, and the region where the YAM ratio is from 100% to 80% may be categorized as "normal", the region where the YAM ratio is from 80% to 70% may be categorized as "osteopenia", and the region where the YAM ratio is lower than 70% may be categorized as "osteoporosis".

[0103] FIG. 7 shows drug efficacy prediction information such as "Bone density % after X years if left untreated", "Bone density % after X years if drug A is used", and "Bone density % after X years if drug B is used". A user who has obtained such drug efficacy prediction information can appropriately consider and determine whether to administer a drug to a subject based on the drug efficacy prediction information, and which of drug A and drug B is more preferably administered. The prediction device 1 may present the probability of future fractures instead of bone density, or present the probability of future fractures in addition to bone density, as the drug efficacy prediction information in FIG. 7.

[0104] FIG. 7 merely shows an example. For example, as the drug prediction information, there may be included efficacy prediction information such as "when drug A and drug B are used in combination, bone density after X years is ○%", "when drug A, drug B, drug C, ··· are used in combination, bone density after X years is ○%", etc. Further, as the drug prediction information, there may be included efficacy prediction information such as "when drug A is administered at P mg / day, bone density after X years is ○%", "when drug A is administered at Q mg / day, bone density after X years is ○%", etc. Since there are also drugs with a limited length of administration period, as the drug prediction information, there may be included efficacy prediction information such as "when drug A is administered for the first K years and then drug B is administered, bone density after X years is ○%", etc. Further, as the drug prediction information, there may be included information regarding the fracture risk such as "the possibility of developing a fracture within M years is ○%", instead of the bone density.

[0105] The prediction device 1 may be capable of outputting the efficacy prediction information shown in (1) to (6) shown in FIG. 7. In FIG. 7, the "target drug" is intended to mean one or more drugs to be administered to the subject and / or drugs that are candidates therefor.

[0106] (1) The degree of the effect of the drug when the drug is administered to the subject.

[0107] (2) Information regarding whether the subject is a person in whom the target drug may act effectively.

[0108] (3) The type in which the target drug may act effectively.

[0109] (4) The degree of the drug effect by combining the target drugs.

[0110] (5) The cause of the subject's osteoporosis.

[0111] (6) The relationship with the concomitant drugs. For example, the efficacy prediction information shown in (5) above may be information indicating, for example, whether the cause is bone formation alone, bone resorption alone, or both bone formation and bone resorption.

[0112] When various additional information about the subject is input, the prediction device 1 may output the type of drug suitable for the subject.

[0113] For a subject with extremely low bone mass, a subject who has already suffered a fracture, and a subject with a high fracture risk, the prediction device 1 may output the type of drug with a high effect of increasing bone density. Examples of drugs with a high effect of increasing bone density include parathyroid hormone drugs and the like. Here, the information on bone mass may be information measured by a bone density measuring device such as DXA, or information obtained by estimating bone density using learned parameters from X-ray images. Regarding the fracture risk, information on fracture risk factors by FRAX (registered trademark) may be used, information predicted using learned parameters from X-ray images may be used, or a combination thereof may be used. Bone mass is the sum of bone salts and bone matrix proteins. In the present disclosure, bone mass is an index related to bone density, and the bone mass is the amount of bone tissue within the skeleton.

[0114] The prediction device 1 may output drug efficacy prediction information corresponding to the type of drug effective in suppressing fractures. Such drug efficacy prediction is useful, for example, for a subject with a high fracture risk. Examples of drugs effective in suppressing fractures include bisphosphonate preparations and denosumab. The prediction device 1 may, for example, extract the type of drug effective in fractures of a specific site for a subject with a high fracture risk at the specific site, and output drug efficacy prediction information corresponding to the drug. More specifically, for example, when the specific site is the proximal femur, the prediction device 1 may extract the type of drug effective in healing fractures of the proximal femur and output drug efficacy prediction information corresponding to the drug. The specific site can include at least one of the vertebral body and the proximal femur, but is not limited thereto. The prediction of fractures at a specific site may be based on, for example, partially estimating or predicting at least one of bone density and bone quality in a region mainly including at least one of cancellous bone and cortical bone.

[0115] More specifically, when the prediction device 1 predicts at least one of the bone density of the vertebrae and fractures as drug efficacy prediction information, it may partially focus on a region mainly including the cancellous bone of the vertebrae. Further, when the prediction device 1 predicts at least one of the bone density of the proximal femur and fractures as drug efficacy prediction information, it may partially focus on a region mainly including the cortical bone of the femur.

[0116] Based on the measurement of the bone metabolism marker, for a subject in whom bone resorption is considered to be enhanced, the prediction device 1 may output the type of drug corresponding to a bone resorption inhibitor. Examples of drugs corresponding to bone resorption inhibitors include, for example, calcitonin drugs, bisphosphonate preparations, anti-RANKL antibodies, selective estrogen receptor modulators (SERM preparations), female hormone preparations, and the like.

[0117] For a subject in whom bone formation is considered to be decreased, the prediction device 1 may output the type of drug corresponding to a bone formation promoting drug. Examples of drugs corresponding to bone formation promoting drugs include, for example, active vitamin D3, parathyroid hormone drugs, and vitamin K2 preparations.

[0118] The information regarding the drug included in the drug efficacy prediction information output by the prediction device 1 may be selected based on information regarding the administration period, swallowing, and cost. Here, the administration period may include at least one of information regarding the age of the subject and menopause. For a subject who needs to take the drug for a long term, the prediction device 1 may output the type of drug corresponding to a drug suitable for long-term administration. Examples of drugs corresponding to drugs suitable for long-term administration include, for example, SERM preparations.

[0119] For subjects who are thought to have no problem taking medicines orally, the prediction device 1 may output types of medicines that correspond to oral agents. On the other hand, for subjects who are thought to have difficulty taking medicines orally, the prediction device 1 may output types of medicines that correspond to intravenous formulations and subcutaneous injection formulations. For subjects at risk of fracture and with progressing symptoms of osteoporosis, the prediction device 1 may output types of medicines that correspond to high-cost medicines. Examples of high-cost medicines include parathyroid hormone medicines.

[0120] The prediction device 1 may output a type of drug suitable for a subject based on information on a disease, symptoms, etc., for which a drug is contraindicated or inappropriate for the subject. For example, for a subject suffering from renal failure, the prediction device 1 may be configured not to output information on drugs such as bisphosphonates and SERMs. For example, for a subject with reduced renal function, the prediction device 1 may be configured not to output information on active vitamin D3 preparations. Without being limited thereto, the prediction device 1 may present efficacy prediction information even if a drug is contraindicated or inappropriate for the subject. In this case, it may be clearly indicated that the drug is contraindicated or inappropriate, and this may be displayed in a manner that is easy for the user to understand. The prediction device 1 may also display the reason why the drug is contraindicated or inappropriate for the subject so that the user can understand.

[0121] The prediction device 1 may be configured to output a virtual image corresponding to the future bone density predicted for the bones of the subject, and the virtual image. For example, when a simple chest X-ray image of the subject is used as the target information, it may be an image that reflects the way the bones shown in the simple chest X-ray image would appear corresponding to the future bone density. The prediction device 1 may be capable of outputting an image and / or a video showing the difference (i.e., the change in bone density) between the simple chest X-ray image of the target information and the virtual image. For example, by allowing the subject to view such an image and / or video, it is possible to clearly show the subject the risk of fracture occurring in their own bones and the risk of developing osteoporosis in their own bones, etc. Therefore, even for a subject without any subjective symptoms, it is possible to effectively convey the necessity of taking measures against the risk of fracture and the risk of developing osteoporosis, etc., and lead to early intervention.

[0122] By outputting at least any one of the information (1) to (6) above as drug efficacy prediction information, the prediction device 1 can inform the user whether a drug that is expected to have an effect when administered to the subject. A user who has obtained such drug efficacy prediction information can appropriately consider and select the drug to be administered to the subject.

[0123] 〔Embodiment 2〕 (Configuration of the prediction system 100b) The prediction device 1 may not be a computer installed in a predetermined medical facility 8, but may be communicably connected to a LAN disposed in each of a plurality of medical facilities 8 via a communication network 9. FIG. 2 is a diagram showing a configuration example of a prediction system 100b according to another aspect of the present disclosure.

[0124] In the LAN within the medical facility 8a, in addition to one or more terminal devices 7a (reception units), an electronic medical record management device 5a (storage unit) and a medical image management device 6a (storage unit) may be communicably connected. Also, in the LAN within the medical facility 8b, in addition to the terminal device 7b (reception unit), an electronic medical record management device 5b (storage unit) and a medical image management device 6b (storage unit) may be communicably connected. In the following, when the medical facilities 8a and 8b are not particularly distinguished, they are referred to as "medical facility 8". Also, for the terminal devices 7a and 7b, and the medical image management devices 6a and 6b, when not particularly distinguished, they are respectively referred to as "terminal device 7" and "medical image management device 6".

[0125] In FIG. 2, an example is shown in which the LANs of the medical facilities 8a and 8b are connected to the communication network 9. The prediction device 1 only needs to be communicably connected to the devices within each medical facility via the communication network 9, and is not limited to the configuration shown in FIG. 2. For example, the prediction device 1 may be installed within the medical facility 8a or within the medical facility 8b.

[0126] In the prediction system 100b adopting such a configuration, the prediction device 1 can acquire the medical image of the subject Pa who has been examined at the medical facility 8a from the medical image management device 6a of the medical facility 8a. Then, the prediction device 1 transmits drug efficacy prediction information regarding the effect of the drug being considered for administration to the subject Pa to the terminal device 7a installed at the medical facility 8a. Similarly, the prediction device 1 can acquire the medical image of the subject Pb who has been examined at the medical facility 8b and transmit drug efficacy prediction information regarding the effect of the drug being considered for administration to the subject P to the terminal device 7b installed at the medical facility 8b.

[0127] In this case, the medical images of each subject may include identification information unique to each medical facility 8 (for example, facility ID) assigned to each medical facility 8 that examines each subject, and identification information unique to each subject (for example, subject ID, patient ID) assigned to each subject. Based on this identification information, the prediction device 1 can correctly transmit the drug efficacy prediction information output from the medical image related to the subject to the terminal device 7 of each medical facility 8 where the subject has been examined.

[0128] 〔Embodiment 3〕 (Outline of the prediction device 1a) Hereinafter, other embodiments of the present disclosure will be described below. For the sake of convenience of explanation, members having the same functions as those described in the above embodiment are denoted by the same reference numerals, and their descriptions will not be repeated.

[0129] FIG. 8 is a block diagram showing the configuration of the prediction device 1a according to the present embodiment. The prediction device 1a is different from the prediction device 1 shown in FIG. 3 in that it further includes an image analysis function and a function of generating processed image data 351 obtained by changing a medical image so as to include region-of-interest information 352 regarding the region of interest focused on in the process of image analysis.

[0130] The prediction device 1a includes a control unit 2a that comprehensively controls each part of the prediction device 1a, and a storage unit 3a that stores various data used by the control unit 2a. The control unit 2a includes an acquisition unit 21, a drug information application unit 22, a prediction unit 23, an output control unit 24a, a learning unit 25, and an image analysis unit 26. In addition to a control program 31 that is a program for performing various controls of the prediction device 1a, the storage unit 3a stores learning data 32, a drug database 33, a learned drug efficacy prediction model 34 (drug efficacy prediction model), and analysis result data 35. The analysis result data 35 may store the region-of-interest information 352 output from the image analysis unit 26. Further, the analysis result data 35 stores the processed image data 351 output from the image analysis unit 26. The processed image data 351 is image data obtained by the image analysis unit 26 adding the analysis result and the region-of-interest information to the medical image to be analyzed.

[0131] The processed image data 351 may be image data obtained by adding information indicating the region that serves as the main basis to the input image data. The information indicating the region is information that allows the range of the region to be understood, such as coloring or framing. In image analysis, since there are many cases where a certain region of an image serves as the main basis for estimation, the user can confirm the region that serves as the basis for estimation based on such information indicating the region. In this case, the display image displayed on the terminal device 7 used by the user includes a processed image with attention region information added to the medical image input to the first analysis model 261. The output control unit 24a may generate a display image in which the medical image and the processed image are arranged in parallel. Such a display image makes it easy for the user to compare the original analyzed image and the processed image. Alternatively, the terminal device 7 may be able to switch between and display the medical image and the processed image according to a user operation. For example, the terminal device 7 may alternately display the medical image and the processed image when the user clicks a switch button displayed on the display screen.

[0132] <Image analysis unit 26> The image analysis unit 26 includes a first analysis model 261. The first analysis model 261 is, for example, a trained machine model that analyzes medical images and outputs an estimation result regarding the state change of bones from the input medical image. Further, in addition to the estimation result, the first analysis model 261 may output attention region information 352 regarding the attention region that is the region of interest within the medical image and that is focused on during the process of outputting the estimation result. The image analysis unit 26 may store the attention region information 352 in the storage unit 3a. Here, the region within the medical image may be, for example, a region inside the outer periphery of the medical image. The attention region information 352 may be located, for example, in all regions within the medical image, or only a part of it may overlap. Or, when the attention region is not in the medical image (or, for example, when it is the entire medical image), it may not overlap with the region within the medical image. In this case, for example, a message indicating that the attention region does not exist (or, for example, that the entire medical image is the attention region) may be displayed outside the medical image.

[0133] Further, the region of interest information 352 may be a heatmap indicating the region that is the basis for the estimation result. In this case, for example, the outer edge of the heatmap will indicate the segmentation region. A heatmap is a method of representing the magnitude of bone density with the concentration of an arbitrary color. For example, the region of interest information 352 may be a heatmap indicating the degree of attention, or a heatmap indicating the numerical value of bone density. Further, the region of interest information 352 may be a heatmap indicating the possibility (probability) of fracture. In this case, the processed image may be an image in which a heatmap indicating the region of interest information is superimposed on the medical image. The image used for the heatmap may be a still image or a moving image. By showing it as a moving image, for example, it becomes easy to visually recognize the relationship between each heatmap by fading various heatmaps in order. Also, in the case of a heatmap of bone density including areas other than the segmentation region, a part of the segmentation region may be framed.

[0134] The medical image input to the first analysis model 261 is an image including an image of the bone of the subject, and outputs an estimation result regarding the state of the bone. For example, the first analysis model 261 may be a learned model that is learned to output an estimation result or a calculation result regarding the state of the bone, such as bone density, relative comparison of bone density, and possibility of fracture, from an X-ray image of the bone. The bone density may be the calculated bone density of the bone site included in the medical image, or the average bone density of the whole body estimated from the image data. Alternatively, the bone density may use the bone density of the lowest part. As a method for calculating the bone density from the medical image, a known method can be used. The relative comparison of bone density is the ratio of the estimated bone density to the YAM. The possibility of fracture is the possibility that the bone at a specific site (for example, the femoral neck, etc.) fractures. The estimation result regarding the state of the bone may be the estimation result at the time when the medical image was taken, or may be a prediction at a time point after a predetermined period has elapsed from that time point.

[0135] The first analysis model 261 may output at least any one of the following information as an estimation result. · The bone density estimated at the time when a medical image is taken. · The bone density predicted at the time when a predetermined period has elapsed since the time when a medical image is taken. · The fracture site estimated at the time when a medical image is taken and the possibility thereof. · The bone density at a past time point retroactively by a predetermined period from the time when a medical image is taken. · The fracture site predicted at the time when a predetermined period has elapsed since the time when a medical image is taken and the possibility thereof.

[0136] When the bone density of the subject has not been measured, the acquisition unit 21 cannot acquire the bone density information of the subject from the electronic medical record management device 6. In such a case, the acquisition unit 21 may acquire, from the image analysis unit 26, an estimation result including the bone density of the bone estimated by the first analysis model 261 from a medical image including an image of the bone of the subject. In this case, the prediction unit 23 may input the estimation result regarding the bone density acquired from the first analysis model 261 into the learned drug efficacy prediction model 34 as target information.

[0137] Next, a learning method of the first analysis model 261 will be described. The learning of the first analysis model 261, for example, the learning for estimating the bone density, can be performed using learning data including an X-ray image of a bone whose bone density has been specified as an explanatory variable and the bone density of the subject shown in the X-ray image as an objective variable. The learning for estimating the possibility of fracture can be performed using learning data including an X-ray image of the subject's bone as an explanatory variable and data regarding whether or not the subject has fractured within a predetermined period thereafter as an objective variable. The relative comparison of the bone density does not need to be learned and is obtained by dividing the estimated bone density by YAM.

[0138] Furthermore, the learning of future predictions may be performed using, as teacher data, X-ray images of bones for which bone density has been specified and data on the bone density or whether a fracture has occurred in the subject after a predetermined period of time. As data on whether a fracture has occurred, for example, information linking fracture information for an X-ray image showing a fractured bone and non-fracture information for an X-ray image showing a non-fractured bone can be used. By including information such as the age, gender, weight, etc. of the subject who has become the learning data, and information on lifestyle habits such as exercise volume, diet content, smoking, drinking, etc. in the learning data, a first analysis model 261 that enables more accurate estimation or prediction can be constructed.

[0139] The first analysis model 261 may be a cell pathological analysis model. In this case, the medical image to be input is a microscopic image of cells of the subject, and the estimation result may be information regarding pathological mutations of the cells.

[0140] In medical images, not only changes in the images associated with changes in bone density and the like estimated by the first analysis model but also changes in the images due to other reasons may appear. Examples of other reasons include certain diseases, treatment scars, various intentionally entered information, and items worn by the subject. Examples of certain diseases include arterial calcification, bone sclerosis, fractures, organ tumors, etc. Examples of treatment scars include implants, bone cement, etc. Examples of intentionally entered information include characters such as "L" and "R" indicating directions. Examples of items worn by the subject include necklaces, etc.

[0141] Due to such reasons, the pixel values (e.g., luminance) of medical images often change compared to the case without other reasons. In addition, implants, necklaces, etc. show unique shapes, which may hide the information contained in the medical images used when the first analysis model 261 estimates bone density. Changes due to such reasons can affect the evaluation of bone density based on the estimation results of the image analysis unit 26. Usually, a doctor who views a medical image notices such changes and interprets the evaluation results of bone density after taking into account the information that caused such changes. However, in some cases, the doctor may not notice such changes in the medical image. Alternatively, in the estimation results output from the image analysis unit 26, doctors may not be able to determine how much the changes in the medical image due to other reasons are considered. In such cases, the doctor may be confused about how to interpret the output estimation results.

[0142] Therefore, the image analysis unit 26 may further include a second analysis model 262. The second analysis model 262 is a trained machine model that detects changes in medical images due to other reasons, such as certain diseases, treatment marks, various intentionally entered information, items worn by the subject, etc. The image analysis unit 26 may correct the medical image according to the detection results. Here, the correction may be at least one of image position adjustment (left - right, up - down direction or rotation, etc. with respect to the original position), luminance adjustment, and masking process. Alternatively, the image analysis unit 26 may analyze the medical image subjected to these corrections using the first analysis model 261.

[0143] The second analysis model 262 can be trained using images in which a doctor has previously annotated the positions of images due to certain diseases, treatment marks, various intentionally entered information, items worn by the subject, etc. The second analysis model 262 can be constructed by a known learning method for constructing an image analysis model.

[0144] The image analysis unit 26 may generate the processed image data 351 by using the medical image input to the first analysis model 261, the analysis result output from the first analysis model 261, and the region of interest information. Alternatively, the image analysis unit 26 may correct the image by using the result output from the second analysis model 262, and input the image to the first analysis model 261 to create the processed image data. Alternatively, the image analysis unit 26 may add information (derivation basis data) regarding the derivation basis of the change in the image due to other reasons to the image data generated by using the medical image input to the first analysis model 261, the analysis result output from the first analysis model 261, and the region of interest information, and generate the added image data as the processed image data 351.

[0145] For example, when the image analysis unit 26 detects a change due to other reasons in the medical image by the second analysis model 262, the image analysis unit 26 may generate the processed image data 351 by excluding the region corresponding to the detected change from the original medical image. Then, the image analysis unit 26 may output an estimation result (including bone density) regarding the state of the bone shown in the processed image data 251 by inputting the generated processed image data 351 to the first analysis model 261. According to this configuration, the image analysis unit 26 can output the estimation result from the medical image including the region corresponding to the change detected by the second analysis model 262 and the estimation result from the medical image excluding the region corresponding to the change detected by the second analysis model 262. By presenting both of these estimation results to the user, the user can be made to evaluate the magnitude of the influence of the region corresponding to the change detected by the second analysis model 262 on the estimation result.

[0146] The output control unit 24a transmits the drug effect prediction information output from the prediction unit 23 and the processed image data 351 generated by the image analysis unit 26 to the terminal device 7. The output control unit 24a may generate a web page by using the drug effect prediction information output from the prediction unit 23 and the processed image data 351 generated by the image analysis unit 26. In this case, the output control unit 24a may transmit access information for accessing the generated web page to the terminal device 7.

[0147] FIG. 9 is an example of a processed image including information (derivation basis data) regarding the basis for deriving a change in an image due to other reasons, which is generated by the image analysis unit 26 based on the analysis results analyzed by the first analysis model 261 and the second analysis model 262. The area indicated by the frame 1502 in FIG. 9 is the area where the implant 1502A for fixing the spine is shown. Also, the area indicated by the frame 1503 is the area where the calcified abdominal aorta 1503A is shown. In the figure, each area is indicated by a dotted line, but may be indicated by different colors, such as a red frame and a yellow frame. The area indicated by the frame 1501 indicates the area that is the basis for deriving the analysis result analyzed by the first analysis model 261. Thus, the range for detecting a change in the image due to other reasons may be outside the area (frame 1501) that is the basis for the analysis when performing the analysis regarding the state of the object.

[0148] FIG. 10 is an example of a browser image 1600 that displays the analysis results in both cases, including and excluding a change in the image due to other reasons. In this browser image 1600, for example, an analysis image 1602 and an analysis result 1601 including drug efficacy prediction information are displayed. The analysis image 1602 may be the same as the image shown in FIG. 9.

[0149] In the analysis result 1601, there are displays such as "When including the dotted-line area, your bone density is XX g / cm 2 " and "When excluding the dotted-line area, your bone density is YY g / cm 2 ". Also, in the analysis result 1601 shown in FIG. 10, drug efficacy prediction information such as "Probability of the drug 〇〇 being effective 〇%" is displayed.

[0150] In the analysis image 1602, it may be possible for a medical professional such as a doctor to select one or both of the areas 1604 and 1605 indicated by the dotted line. For example, when an operation on the selection button 1606 is received after the area 1604 is selected, a configuration may be such that detailed analysis results for the area 1604 are displayed in the analysis result 1601. The back button 1607 and the end button 1608 are the same as before.

[0151] 〔Embodiment 4〕 Hereinafter, other embodiments of the present disclosure will be described below. For convenience of explanation, members having the same functions as the members described in the above embodiments are denoted by the same reference numerals, and the description thereof will not be repeated.

[0152] Each function of the prediction devices 1 and 1a, including the drug efficacy prediction model that generates drug efficacy prediction information based on the target information and drug information regarding a drug having an effect on bone, may be realized by cloud computing. The prediction devices 1 and 1a may be, for example, a cloud server (not shown). In this case, in the prediction systems 100 and 100a, the cloud server may receive the target information from the terminal devices 7, 7a, 7b and generate drug efficacy prediction information based on the target information and drug information regarding a drug having an effect on bone. Then, the cloud server may transmit the generated drug efficacy prediction information to the terminal device 7 that is the transmission source of the target information. In the present embodiment, the terminal devices 7, 7a, 7b function as a storage unit that stores the target information to be transmitted to the cloud server, and also function as a reception unit that receives the drug efficacy prediction information from the cloud server.

[0153] 〔Example of Realization by Software〕 The control blocks of the prediction device 1 (particularly, the acquisition unit 21, the drug information application unit 22, the prediction unit 23, the output control unit 24, and the learning unit 25) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

[0154] In the latter case, the prediction device 1 includes a computer that executes instructions of a program, which is software for realizing each function. This computer includes, for example, one or more processors and a computer-readable recording medium that stores the above program. Then, in the above computer, the object of the present disclosure is achieved by the above processor reading and executing the above program from the above recording medium. As the above processor, for example, a CPU (Central Processing Unit) can be used. As the above recording medium, a "non-transitory tangible medium", for example, in addition to a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, etc. can be used. Further, it may further include a RAM (Random Access Memory) or the like for expanding the above program. Further, the above program may be supplied to the above computer via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program. One aspect of the present disclosure can also be realized in the form of a data signal embedded in a carrier wave, in which the above program is embodied by electronic transmission.

[0155] As described above, the invention according to the present disclosure has been described based on the drawings and examples. However, the invention according to the present disclosure is not limited to the above-described embodiments. That is, the invention according to the present disclosure can be variously modified within the scope shown in the present disclosure, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the invention according to the present disclosure. That is, it should be noted that those skilled in the art can easily make various deformations or modifications based on the present disclosure. Also, note that these deformations or modifications are included in the scope of the present disclosure.

[0156] 〔Summary 1〕 The prediction device according to Aspect 1A of the present disclosure includes an acquisition unit that acquires target information including at least any one of fracture risk factor information regarding a target's fracture risk factors, bone density information indicating the bone density of the target's bone, bone metabolism information indicating the bone metabolism state of the target's bone, and a medical image in which the target's bone appears, and a prediction unit that outputs efficacy prediction information regarding the effect of the drug when the drug is administered to the target by inputting the target information into a drug efficacy prediction model that predicts the effect of the drug based on drug information regarding a drug having an action on bone.

[0157] The prediction device according to Aspect 2A of the present disclosure may, in the above Aspect 1A, include information regarding a drug having at least any one of an action on bone formation and an action on bone resorption.

[0158] The prediction device according to Aspect 3A of the present disclosure may, in the above Aspect 1A or 2A, be learned using teacher data in which subject information regarding each of a plurality of subjects to whom the drug has been administered and the drug information regarding the drug administered to each of the subjects are explanatory variables, and the effect of the drug in each of the subjects is a target variable.

[0159] The prediction device according to Aspect 4A of the present disclosure may, in any one of the above Aspects 1A to 3A, include at least information regarding the name and mechanism of action of the drug.

[0160] The prediction device according to Aspect 5A of the present disclosure may, in any one of the above Aspects 1A to 4A, include information regarding each combination of a plurality of types of drugs that can change the state of the target's bone, and the prediction unit may output the efficacy prediction information for each combination of the plurality of types of drugs.

[0161] In the prediction device according to Aspect 6A of the present disclosure, in any one of Aspects 1A to 5A above, the prediction unit may output, as the drug efficacy prediction information, information indicating the state of the bone of the subject after administering the drug to the subject continuously for a predetermined period, or information indicating a change in the bone of the subject around the predetermined period.

[0162] In the prediction device according to Aspect 7A of the present disclosure, in any one of Aspects 1A to 6A above, the prediction unit may output, as the drug efficacy prediction information, information regarding a drug that can change the state of the bone of the subject after administering the drug to the subject continuously for a predetermined period.

[0163] In the prediction device according to Aspect 8A of the present disclosure, in any one of Aspects 1A to 7A above, the prediction unit may output, as the drug efficacy prediction information, a prediction result of the future bone density of the bone of the subject.

[0164] In the prediction device according to Aspect 9A of the present disclosure, in any one of Aspects 1A to 8A above, the prediction unit may output, as the drug efficacy prediction information, a prediction result of the current bone density of the bone of the subject.

[0165] In the prediction device according to Aspect 10A of the present disclosure, in any one of Aspects 1A to 9A above, the prediction unit may output, as the drug efficacy prediction information, at least one of a prediction result indicating the difference between the future bone density and the current bone density of the bone of the subject, and a prediction estimation result indicating the change in the bone density of the bone of the subject from the current state.

[0166] In the prediction device according to Aspect 11A of the present disclosure, in any one of Aspects 1A to 10A above, the prediction unit may output, as the drug efficacy prediction information, a prediction result regarding the fracture site of the subject.

[0167] In the prediction device according to Aspect 12A of the present disclosure, in any one of Aspects 1A to 11A above, the prediction unit may output, as the drug efficacy prediction information, a prediction result indicating the possibility that the subject will fracture in the future.

[0168] In the prediction device according to Aspect 13A of the present disclosure, in any of Aspects 1A to 12A above, the target information may include at least one of information regarding the type, age, gender, weight, height, presence or absence of fracture, fracture site, fracture history, family fracture history, glucocorticoid, rheumatoid arthritis, secondary osteoporosis, underlying disease, smoking history, drinking habit, occupational history, exercise history, medical history, blood test results, urine test results, drugs being taken, gene sequence, of the target.

[0169] In the prediction device according to Aspect 14A of the present disclosure, in any of Aspects 1A to 13A above, the bone density information may be a measured value obtained by measuring the bone density of the bone at a predetermined site of the target.

[0170] In the prediction device according to Aspect 15A of the present disclosure, in any of Aspects 1A to 14A above, the bone metabolism information may include at least any one of bone formation marker information regarding osteoblasts, bone resorption marker information regarding osteoclasts, and bone matrix-related marker information regarding bone mass, of the target.

[0171] In the prediction device according to Aspect 16A of the present disclosure, in the above Aspect 15A, the bone formation marker information may be information regarding at least any one of alkaline phosphatase and type I procollagen-N-propeptide, the bone resorption marker information may be information regarding at least any one of deoxypyridinoline, type I collagen cross-linked N-telopeptide, type I collagen cross-linked C-telopeptide, tartrate-resistant acid phosphatase-5b fraction (TRACP-5b), and the bone matrix-related marker information may be information regarding at least any one of undercarboxylated osteocalcin, pentosidine, and homocysteine.

[0172] The prediction device according to Aspect 17A of the present disclosure, in any of Aspects 1A to 16A above, the medical image may include at least one of an X-ray image, a CT (Computed Tomography) image, an MRI (Magnetic Resonance Imaging) image, a PET (Positron Emission Tomography) image, and an ultrasonic image of the body of the subject.

[0173] The prediction system according to Aspect 18A of the present disclosure includes the prediction device described in any of Aspects 1A to 17A above, and a terminal device communicably connected to the prediction device, the terminal device presenting the drug effect prediction information.

[0174] The prediction method according to Aspect 19A of the present disclosure includes an acquisition step of acquiring target information including at least any one of fracture risk factor information regarding a subject's fracture risk factors, bone density information indicating the bone density of the subject's bone, bone metabolism information indicating the bone metabolism state of the subject's bone, and a medical image showing the subject's bone, and a prediction step of outputting drug effect prediction information regarding the effect of the drug when the drug is administered to the subject by inputting the target information into a drug effect prediction model that predicts the effect of the drug based on drug information regarding a drug having an action on the bone.

[0175] The control program according to Aspect 20A of the present disclosure is a control program for causing a computer to function as the prediction device described in any of Aspects 1A to 17A above, and is a control program for causing a computer to function as the acquisition unit and the prediction unit.

[0176] The recording medium according to Aspect 21A of the present disclosure is a computer-readable recording medium recording the control program described in Aspect 20A above.

[0177] 〔Summary 2〕 The prediction system according to Aspect 1B of the present disclosure includes a storage unit that stores target information regarding a target bone, and a reception unit that receives drug effect prediction information regarding the effect of the drug when the drug is administered to the target. The target information includes at least a medical image in which the target bone is imaged, and the drug effect prediction information is generated based on a drug effect prediction model that predicts the effect of the drug based on the target information and drug information regarding a drug having an effect on the bone.

[0178] The prediction system according to Aspect 2B of the present disclosure is, in the above Aspect 1B, the drug effect prediction model may be learned using learning data having, as explanatory variables, subject information regarding each of a plurality of subjects to whom the drug has been administered and the drug information regarding the drug administered to each of the subjects, and having, as an objective variable, the effect of the drug in each of the subjects.

[0179] The prediction system according to Aspect 3B of the present disclosure is, in the above Aspect 1B or 2B, the drug effect prediction information may include at least any one of a prediction result indicating a difference between the future bone density and the current bone density of the target bone, a prediction result indicating a change in the bone density of the target bone from the present, a prediction result when medication is administered and a prediction result when no medication is administered, and a prediction result when different types of drugs are administered.

[0180] The prediction system according to Aspect 4B of the present disclosure is, in any one of the above Aspects 1B to 3B, the target information may include at least any one of fracture risk factor information regarding a fracture risk factor of the target, bone density information indicating the bone density of the bone of the target, and bone metabolism information indicating the bone metabolism state of the bone of the target.

[0181] The prediction system according to Aspect 5B of the present disclosure is, in any one of the above Aspects 1B to 4B, the drug information may include information regarding a drug having at least any one of an action on bone formation and an action on bone resorption.

[0182] In the prediction system according to Aspect 6B of the present disclosure, in any of Aspects 1B to 5B above, the drug information may include at least information regarding the name and / or mechanism of action of the drug.

[0183] In the prediction system according to Aspect 7B of the present disclosure, in any of Aspects 1B to 6B above, the drug information includes information regarding each of combinations of a plurality of drugs that can change the state of the bone of the subject, and the drug efficacy prediction information may include information regarding each of the combinations of the plurality of drugs.

[0184] In the prediction system according to Aspect 8B of the present disclosure, in any of Aspects 1B to 7B above, the drug efficacy prediction information may include information indicating the state of the bone of the subject after continuously administering the drug to the subject for a predetermined period, or information indicating a change in the bone of the subject before and after the predetermined period.

[0185] In the prediction system according to Aspect 9B of the present disclosure, in any of Aspects 1B to 8B above, the drug efficacy prediction information may include information regarding a drug that can change the state of the bone of the subject after continuously administering the drug to the subject for a predetermined period.

[0186] In the prediction system according to Aspect 10B of the present disclosure, in any of Aspects 1B to 9B above, the drug efficacy prediction information may include a predicted result of the bone density of the bone of the subject in the future.

[0187] In the prediction system according to Aspect 11B of the present disclosure, in any of Aspects 1B to 10B above, the drug efficacy prediction information may include a predicted result of the current bone density of the bone of the subject.

[0188] In the prediction system according to Aspect 12B of the present disclosure, in any of Aspects 1B to 11B above, the drug efficacy prediction information may include a predicted result regarding the fracture site of the subject.

[0189] In the prediction system according to Aspect 13B of the present disclosure, in any of Aspects 1B to 12B above, the drug effect prediction information may include a prediction result indicating the possibility that the subject will fracture in the future.

[0190] In the prediction system according to Aspect 14B of the present disclosure, in any of Aspects 1B to 13B above, the subject information includes at least any one of fracture risk factor information regarding the fracture risk factors of the subject, bone density information indicating the bone density of the subject's bone, and bone metabolism information indicating the bone metabolism state of the subject's bone, and the bone density information may be a measured value obtained by measuring the bone density of the bone at a predetermined site of the subject.

[0191] In the prediction system according to Aspect 15B of the present disclosure, in any of Aspects 1B to 14B above, the subject information includes at least any one of fracture risk factor information regarding the fracture risk factors of the subject, bone density information indicating the bone density of the subject's bone, and bone metabolism information indicating the bone metabolism state of the subject's bone, the medical image is a simple X-ray image of the subject's body, and the bone density information may be an estimated value estimated based on a bone density prediction model that estimates the bone density from the simple X-ray image in which the subject appears.

[0192] In the prediction system according to Aspect 16B of the present disclosure, in the above Aspect 15B, the bone density prediction model may be learned using teacher data having a medical image in which the bone of the subject appears as an explanatory variable and a measured value of the bone density of the bone of the subject in which the medical image was taken as an objective variable.

[0193] The prediction device according to Aspect 17B of the present disclosure includes an acquisition unit that acquires subject information regarding the subject's bone, and a prediction unit that outputs drug effect prediction information regarding the effect of a drug when the drug is administered to the subject, the subject information includes at least a medical image in which the subject's bone appears, and the prediction unit has a drug effect prediction model that predicts the effect of the drug based on the subject information and drug information regarding a drug having an action on the bone.

[0194] The prediction method according to Aspect 18B of the present disclosure includes a storage step of storing target information regarding a target bone, and a drug efficacy prediction information generation step of generating drug efficacy prediction information regarding the effect of the drug when the drug is administered to the target based on a drug efficacy prediction model that predicts the effect of the drug based on the target information and drug information regarding a drug having an action on the bone, and a reception step of receiving the drug efficacy prediction information, wherein the target information includes at least a medical image showing the target bone.

[0195] The control program according to Aspect 19B of the present disclosure is a control program for causing a computer to function as any one of the prediction systems according to Aspects 1B to 16B, and is a control program for causing a computer to function as the storage unit and the reception unit.

[0196] The recording medium according to Aspect 20B of the present disclosure is a computer-readable recording medium on which the control program according to Aspect 19B is recorded.

[0197] 〔Summary 3〕 The prediction system according to Aspect 1C of the present disclosure includes an acquisition unit that acquires target information including a medical image showing a target bone, and a prediction unit that outputs drug efficacy prediction information regarding the effect of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the target information, wherein the prediction unit outputs the drug efficacy prediction information using a drug efficacy prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0198] In the prediction system according to Aspect 2C of the present disclosure, in the above Aspect 1C, the prediction unit may output drug efficacy prediction information regarding both the effect of a drug having an action on bone formation and the effect of a drug having an action on bone resorption, respectively.

[0199] In the prediction system according to Aspect 3C of the present disclosure, in the above Aspect 1C or 2C, the input information may include the drug information.

[0200] In the prediction system according to Aspect 4C of the present disclosure, in any of the above Aspects 1C to 3C, the drug efficacy prediction model may be trained using learning data in which subject information regarding each of a plurality of subjects to whom the drug has been administered and drug information regarding the drug administered to each of the subjects are explanatory variables, and the effect of the drug in each of the subjects is the objective variable.

[0201] In the prediction system according to Aspect 5C of the present disclosure, in any of the above Aspects 1C to 4C, the drug efficacy prediction information may include at least any one of a prediction result indicating the difference between the future bone density and the current bone density of the target bone, a prediction result indicating the change in the bone density of the target bone from the present, a prediction result when medication is administered and a prediction result when no medication is administered, and a prediction result when different types of drugs are administered.

[0202] In the prediction system according to Aspect 6C of the present disclosure, in any of the above Aspects 1C to 5C, the target information may further include at least any one of fracture risk factor information regarding the fracture risk factors of the target, bone density information indicating the bone density of the target bone, and bone metabolism information indicating the bone metabolism state of the target bone.

[0203] In the prediction system according to Aspect 7C of the present disclosure, in any of the above Aspects 1C to 6C, the drug information may include at least information regarding the name and / or mechanism of action of the drug.

[0204] In the prediction system according to Aspect 8C of the present disclosure, in any of the above Aspects 1C to 7C, the drug information may include information regarding each combination of a plurality of types of drugs that can change the state of the target bone, and the drug efficacy prediction information may include information regarding each of the combinations of the plurality of types of drugs.

[0205] In the prediction system according to aspect 9C of the present disclosure, in any of aspects 1C to 8C above, the drug efficacy prediction information may include information indicating the state of the bone of the subject after continuous administration of the drug to the subject for a predetermined period, or information indicating changes in the bone of the subject before and after the predetermined period.

[0206] In the prediction system according to aspect 10C of the present disclosure, in any of aspects 1C to 9C above, the drug efficacy prediction information may include information regarding a drug that can change the state of the bone of the subject after continuous administration of the drug to the subject for a predetermined period.

[0207] In the prediction system according to aspect 11C of the present disclosure, in any of aspects 1C to 10C above, the drug efficacy prediction information may include a prediction result of the future bone density of the bone of the subject.

[0208] In the prediction system according to aspect 12C of the present disclosure, in any of aspects 1C to 11C above, the drug efficacy prediction information may include a prediction result of the current bone density of the bone of the subject.

[0209] In the prediction system according to aspect 13C of the present disclosure, in any of aspects 1C to 12C above, the drug efficacy prediction information may include a prediction result regarding the fracture site of the subject.

[0210] In the prediction system according to aspect 14C of the present disclosure, in any of aspects 1C to 13C above, the drug efficacy prediction information may include a prediction result indicating the possibility that the subject will fracture in the future.

[0211] In the prediction system according to aspect 15C of the present disclosure, in any of aspects 1C to 14C above, the subject information may further include bone density information indicating the bone density of the bone of the subject, and the bone density information may be a measured value obtained by measuring the bone density of a predetermined site of the bone of the subject.

[0212] In the prediction system according to Aspect 16C of the present disclosure, in any of Aspects 1C to 15C above, the target information may further include bone density information indicating the bone density of the target bone, the medical image may be a simple X-ray image of the target's body, and the bone density information may be an estimated value estimated based on a bone density prediction model that estimates the bone density from the simple X-ray image in which the target appears.

[0213] In the prediction system according to Aspect 17C of the present disclosure, in Aspect 16C above, the bone density prediction model may be learned using teacher data having, as an explanatory variable, a medical image in which a subject's bone appears, and, as an objective variable, a measured value of the bone density of the subject's bone in the medical image taken.

[0214] The prediction device according to Aspect 18C of the present disclosure includes an acquisition unit that acquires target information including a medical image in which a target bone appears, and a prediction unit that outputs drug effect prediction information regarding the effect of a drug having at least either an action on bone formation or an action on bone resorption from input information including the target information. The prediction unit outputs the drug effect prediction information using a drug effect prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0215] The prediction method according to Aspect 19C of the present disclosure includes an acquisition step in which a computer acquires target information including a medical image in which a target bone appears, and a prediction step in which the computer outputs drug effect prediction information regarding the effect of a drug having at least either an action on bone formation or an action on bone resorption from input information including the target information. In the prediction step, the drug effect prediction information is output using a drug effect prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug.

[0216] The control program according to Aspect 20C of the present disclosure is a control program for causing a computer to function as the prediction system according to any of Aspects 1C to 17C above, and is a control program for causing a computer to function as the acquisition unit and the prediction unit.

[0217] The recording medium according to aspect 21C of the present disclosure is a computer-readable recording medium on which the control program described in the above aspect 20 is recorded.

Explanation of Signs

[0218] 1, 1a Prediction device 5 Medical image management device (storage unit) 6 Electronic medical record management device (storage unit) 7 Terminal device (storage unit, receiving unit) 21 Acquisition unit 22 Drug information application unit 23 Prediction unit 24 Output control unit 25 Learning unit 32 Teacher data 34 Learned drug efficacy prediction model 100a, 100b Prediction system S11 Acquisition step S12 Generation step

Claims

1. An acquisition unit that acquires target information including a medical image showing a target bone; A prediction unit that outputs drug efficacy prediction information regarding the efficacy of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the target information; and is provided with, The prediction unit outputs the drug efficacy prediction information using a drug efficacy prediction model that predicts the efficacy of the drug based on the input information and drug information regarding the drug. Prediction system.

2. The prediction unit according to claim 1, wherein the prediction unit outputs drug efficacy prediction information regarding both the efficacy of a drug having an action on bone formation and the efficacy of a drug having an action on bone resorption, respectively.

3. The input information includes the drug information, The prediction system according to claim 1.

4. The drug efficacy prediction model uses, as explanatory variables, subject information regarding each of a plurality of subjects to whom the drug has been administered and the drug information regarding the drug administered to each of the subjects, and is learned using learning data having the efficacy of the drug in each of the subjects as an objective variable. The prediction system according to claim 1.

5. The drug efficacy prediction information includes at least one of a prediction result indicating the difference between the future bone density and the current bone density of the target bone, a prediction result indicating the change in the bone density of the target bone from the present, a prediction result when the drug is administered and a prediction result when the drug is not administered, and a prediction result when different types of drugs are administered. The prediction system according to claim 1.

6. The target information further includes at least one of fracture risk factor information regarding the fracture risk factors of the target, bone density information indicating the bone density of the target bone, and bone metabolism information indicating the bone metabolism state of the target bone. The prediction system according to claim 1.

7. The drug information includes at least information regarding the name and / or mechanism of action of the drug. The prediction system according to claim 1.

8. The drug information includes information regarding each combination of a plurality of types of drugs capable of changing the state of the target bone, The drug efficacy prediction information includes information regarding each of the combinations of the plurality of types of drugs. The prediction system according to claim 1.

9. The drug efficacy prediction information includes information indicating the state of the target bone after continuing the administration of the drug to the target for a predetermined period, or information indicating the change in the target bone before and after the predetermined period. The prediction system according to claim 1.

10. The drug efficacy prediction information includes information regarding a drug that can change the state of the bone of the subject after administration to the subject has been continued for a predetermined period. The prediction system according to claim 1.

11. The drug efficacy prediction information includes a prediction result of the future bone density of the bone of the subject. The prediction system according to claim 1.

12. The drug efficacy prediction information includes a prediction result of the current bone density of the bone of the subject. The prediction system according to claim 1.

13. The drug efficacy prediction information includes a prediction result regarding the fracture site of the subject. The prediction system according to claim 1.

14. The drug efficacy prediction information includes a prediction result indicating the possibility that the subject will fracture in the future. The prediction system according to claim 1.

15. The subject information further includes bone density information indicating the bone density of the bone of the subject, The bone density information is a measured value obtained by measuring the bone density of the bone at a predetermined site of the subject. The prediction system according to claim 1.

16. The subject information further includes bone density information indicating the bone density of the bone of the subject, The medical image is a simple X-ray image obtained by imaging the body of the subject, The bone density information is an estimated value estimated based on a bone density prediction model that estimates the bone density from the simple X-ray image in which the subject appears. The prediction system according to claim 1.

17. The bone density prediction model is learned using teacher data having, as an explanatory variable, a medical image in which the bone of a subject appears, and, as an objective variable, a measured value of the bone density of the bone of the subject for whom the medical image was taken. The prediction system according to claim 16.

18. An acquisition unit that acquires subject information including a medical image in which the bone of the subject appears; A prediction unit that outputs drug efficacy prediction information regarding the effect of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the subject information, The prediction unit outputs the drug efficacy prediction information using a drug efficacy prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug. Prediction device.

19. An acquisition step in which a computer acquires subject information including a medical image in which the bone of the subject appears; A prediction step in which a computer outputs drug efficacy prediction information regarding the effect of a drug having at least one of an action on bone formation and an action on bone resorption from input information including the subject information, In the prediction step, the drug efficacy prediction information is output using a drug efficacy prediction model that predicts the effect of the drug based on the input information and drug information regarding the drug. Prediction method.

20. A control program for causing a computer to function as the prediction system according to any one of claims 1 to 17, the control program for causing a computer to function as the acquisition unit and the prediction unit.

21. A computer-readable recording medium on which the control program according to claim 20 is recorded.

Images