Radiation dose management device, radiation dose management method, and storage medium

Abstract

A radiation dose management apparatus manages a radiation dose received by a subject, and includes an acquisition unit that acquires image data of the subject, an organ segmentation unit that segments the image data by organs of the subject with respect to the acquired image data, a voxel dose distribution map generation unit that generates a voxel dose distribution map indicating a radiation dose of each voxel based on a specific radiation dose model and the acquired image data of the subject, and an organ dose calculation unit that calculates a radiation dose of each organ based on the segmented image data of the organs and the radiation dose of each voxel in the generated voxel dose distribution map. The radiation dose management apparatus further includes a voxel dose correction unit that corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor.

Description

Technical Field

[0001] The embodiments of the present invention relate to a radiation dose management device, a radiation dose management method, and a storage medium. Background Technology

[0002] X-ray-based examinations have become increasingly common in order to understand the disease status of patients (sometimes referred to as examinees). On the one hand, X-ray-based examinations can help understand the condition of the disease; on the other hand, they can expose patients to harmful radiation exposure. Therefore, managing the radiation dose to patients is important from the perspective of minimizing their radiation exposure.

[0003] During a CT scan, the scan console provides either CTDIvol (CT Dose Index volume) or DLP (dose-length product) values ​​according to the selected protocol. If the dose value exceeds the threshold, a further scan is not permitted. The radiologist will review the scan protocol and, if deemed safe to continue, will enter the reason for proceeding.

[0004] like Figure 10 As shown, the existing technology uses a method to calculate the radiation dose received within a cylindrical acquisition area. This method is not targeted at the radiation dose of a specific organ. Furthermore, when the scanning ranges overlap, the average radiation dose values ​​within the overlapping scanning ranges are added together to calculate the cumulative dose. Therefore, this method cannot accurately determine the radiation dose received by radiation-sensitive organs.

[0005] Furthermore, in the existing technology, the self-repair capacity of organs to radiation damage is not considered when calculating the cumulative dose after multiple scans. Therefore, the final cumulative dose obtained is an uncorrected cumulative dose and is not an accurate cumulative dose. Summary of the Invention

[0006] The present invention was made in view of the above circumstances, and its object is to provide a radiation dose management device, a radiation dose management method and a storage medium capable of accurately obtaining the radiation dose of an organ.

[0007] Technical solution 1 relates to a radiation dose management device for managing the radiation dose received by a subject, comprising: an acquisition unit for acquiring image data of the subject; an organ segmentation unit for segmenting the acquired image data according to the organs of the subject; a voxel dose distribution map generation unit for generating a voxel dose distribution map representing the radiation dose of each voxel based on a specific radiation dose model and the acquired image data of the subject; and an organ dose calculation unit for calculating the radiation dose of each organ based on the segmented organ image data and the radiation dose of each voxel in the generated voxel dose distribution map.

[0008] Based on the above structure, the radiation dose can be obtained on an organ-by-organ basis, thereby enabling more accurate and targeted management of radiation dose to organs, especially radiation-sensitive organs.

[0009] The radiation dose management device described above may also include a voxel dose correction unit, which corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor, wherein the specific correction factor is the time elapsed after radiation and the inherent repair constant of each organ, and the organ dose calculation unit calculates the corrected radiation dose of each organ based on the segmented organ image data and the corrected radiation dose of each voxel.

[0010] Based on the above configuration, the radiation dose received by each organ can be corrected according to the time elapsed after radiation and the inherent repair constant of each organ. Thus, the residual radiation dose of each organ at the current moment can be obtained on an organ-by-organ basis, taking into account the time elapsed after radiation and the inherent repair constant of each organ.

[0011] The radiation dose management device described above may also include a voxel dose accumulation unit. When there are multiple image data for the subject, the voxel dose accumulation unit adds up the corrected voxel doses of the corresponding voxels in each of the acquired image data to calculate the corrected voxel dose accumulation value for each voxel. The organ dose calculation unit calculates the corrected radiation dose for each organ based on the segmented organ image data and the calculated corrected voxel dose accumulation value for each voxel.

[0012] Based on the above configuration, it is possible to correct the radiation dose of voxels in image data obtained after multiple scans and obtain the residual cumulative radiation dose of each organ at the current moment.

[0013] The aforementioned radiation dose management device may also include a radiation dose risk warning unit, which, based on the corrected radiation dose of each organ, indicates the deterministic and random risks of each organ.

[0014] Based on the above structure, it can provide doctors with a wealth of reference information and provide early warnings of risks.

[0015] Technical solution 2 relates to a radiation dose management method for managing the radiation dose received by a subject, comprising the following steps: an acquisition step, acquiring image data of the subject; an organ segmentation step, segmenting the acquired image data according to the organs of the subject; a voxel dose distribution map generation step, generating a voxel dose distribution map representing the radiation dose of each voxel based on a specific radiation dose model and the acquired image data of the subject; and an organ dose calculation step, calculating the radiation dose of each organ based on the segmented organ image data and the radiation dose of each voxel in the generated voxel dose distribution map.

[0016] Technical solution 3 relates to a storage medium storing a program that enables a computer to execute the radiation dose management method of technical solution 2.

[0017] Invention Effects

[0018] According to the above technical solution, the present invention can obtain the radiation dose of each organ, especially radiation-sensitive organs, and can make dose corrections taking into account the self-repair ability of each organ over time, thereby accurately obtaining the residual radiation dose of the organ.

[0019] Furthermore, the present invention can also provide alerts for deterministic and stochastic risks based on the radiation dose received by the obtained organs. Attached Figure Description

[0020] Figure 1 This is a schematic diagram illustrating the radiation dose management device according to the first embodiment.

[0021] Figure 2A This is a schematic diagram illustrating the segmentation process of the organ segmentation unit in the first embodiment. Figure 2B This is a schematic diagram representing the voxel dose distribution map generated by the voxel dose distribution map generation unit. Figure 2C This is a schematic diagram showing the radiation dose received by each voxel according to its organ.

[0022] Figure 3 This is a flowchart illustrating the management method of the radiation dose management device according to the first embodiment.

[0023] Figure 4 This is a flowchart illustrating a modified example of the radiation dose management device of the first embodiment, showing a management method.

[0024] Figure 5 A schematic diagram illustrating the radiation dose management device according to the second embodiment.

[0025] Figure 6 It is a graph showing the change of residual radiation dose over time.

[0026] Figure 7 This is a flowchart illustrating the management method of the radiation dose management device according to the second embodiment.

[0027] Figure 8 This is a diagram used to illustrate certain risks.

[0028] Figure 9A This is a schematic diagram illustrating the training process of the radiation dose model. Figure 9B This is a schematic diagram illustrating the application of a trained radiation dose model.

[0029] Figure 10 This indicates the method of accumulating radiation dose used in existing technologies. Detailed Implementation

[0030] Hereinafter, embodiments will be described with reference to the accompanying drawings. In the following embodiments, parts labeled with the same reference numerals will be considered as parts that perform the same operations, and repeated descriptions will be omitted as appropriate.

[0031] Given the potential dangers of ionizing radiation, dose assessment is a crucial aspect of CT research. Monte Carlo (MC) estimation is a commonly used algorithm for estimating patient-specific doses, but its slow speed limits its routine application. Faster alternatives include applying patient-specific conversion factors to CTDIvol, utilizing pre-acquired phantom data, and employing analytical approximations to replace MC estimation; however, these methods come at the cost of reduced accuracy.

[0032] To achieve both high accuracy and high performance, this implementation employs a preferred estimation method. It utilizes a deep learning approach, employing a deep convolutional neural network (DCNN) to generate a Monte Carlo (MC) radiation dose model. The basic principle is to use dual-channel data consisting of CT images and first-order dose estimates as input to a three-dimensional U-network, training the U-network to reproduce the corresponding Monte Carlo (MC) dose estimate. For a detailed implementation method, please refer to "Real-Time Patient-Specific CT Dose Estimation using a Deep Convolutional Neural Network" by Joscha Maier, Elias Eulig, Sabrina Dorn, Stefan Sawall, and Marc Kachelrieβ, Senior Member, IEEE.

[0033] By utilizing this trained radiation dose model, the voxel dose distribution map of the CT image input into the radiation dose model can be calculated in real time.

[0034] Figure 9A and Figure 9B The training process and application process of the radiation dose model are shown respectively. Figure 9A As shown, during the training process of the radiation dose model, a deep learning method is used, employing CT images and first-order dose estimation results to train the radiation dose model. After training is completed, as shown... Figure 9B As shown, the CT image used for analysis is input into a trained radiation dose model to obtain a distribution map of the radiation dose received by each voxel in the CT image (also known as a voxel dose distribution map). The voxel dose distribution map represents the estimated value of the radiation dose received by each voxel in the CT image, and the magnitude of the radiation dose is represented by color distribution or grayscale distribution.

[0035] Alternatively, deep learning methods can be used during the training of the voxel radiation dose model. This involves training the voxel radiation dose model using CT and PET image data. After training, the CT and PET image data used for analysis are input into the trained radiation dose model to obtain a voxel dose distribution map. For details on the method of generating voxel dose distribution maps based on PET-CT images, please refer to "Deep-dose: a voxel dose estimation method using deep convolutional neural network for personalized internal dosimetry" (Min Sun Lee1, Donghwi Hwang1,2, Joong Hyun Kim3 & Jae Sun Lee, see www.nature.com / scientificreports).

[0036] Furthermore, the method described in "Deep Dose: A Voxel Dose Estimation Method Based on Deep Convolutional Neural Networks for Individual Internal Radiation Dose Measurement" is also applicable to SPECT-CT images. After training the radiation dose model, the CT image data and SPECT image data used for analysis are input into the trained radiation dose model, and voxel dose distribution maps can also be obtained.

[0037] The method for generating the radiation dose model described above is the preferred method used in this invention, but other methods for generating radiation dose models can also be used.

[0038] The radiation dose model described above is used in the radiation dose management devices of various embodiments, which can be any one of X-ray CT devices, PET-CT devices, and SPECT-CT devices. Each embodiment is described below. Furthermore, only three modalities of radiation dose models are illustrated here, but the invention is not limited to these.

[0039] (First Implementation)

[0040] Figure 1 This is a schematic diagram illustrating the radiation dose management device according to the first embodiment.

[0041] The radiation dose management device of this embodiment can calculate the radiation dose to an organ from a scan obtained through retrieval. Here and in the preceding and following text, "scan" refers to radiographic imaging that produces ionizing radiation, such as a CT scan.

[0042] Different organs have different sensitivities to radiation. Examples of radiation-sensitive organs include the lens, thyroid gland, and reproductive organs (such as the uterus and ovaries). When receiving the same radiation dose, radiation-sensitive organs will suffer greater damage. Therefore, it is desirable to be able to differentiate between organs and accurately determine the radiation dose received by each organ.

[0043] The radiation dose management device 10 includes an acquisition unit 11, an organ segmentation unit 12, a voxel dose distribution map generation unit 13, and an organ dose calculation unit 14.

[0044] The acquisition unit 11 acquires image data of interest based on input from the scanning console. This image data is generated after the subject has undergone a scan that causes ionizing radiation, such as a CT scan. After the doctor inputs the subject's user information into the scanning console, they can use this information to retrieve the image data of a specific scan of interest. For example, if the doctor wants to know the radiation dose received by the patient in the previous scan, the acquisition unit 11 acquires the image data from the previous scan.

[0045] The organ segmentation unit 12 segments the acquired image data according to the organs of the examined body. For example... Figure 2A As shown in the figure, a schematic diagram of organ segmentation using existing organ segmentation methods is presented. The examination area in the figure is the skull region. The grid lines represent the segmented cerebellum, the horizontal stripes represent the segmented skull, and the vertical stripes represent the ear structures. This embodiment does not limit the organ segmentation method or the segmented examination area; any method and any examination area can be segmented. Furthermore, this embodiment focuses on the radiation dose to the organs, particularly the radiation dose to radiation-sensitive organs. Therefore, it is also advisable to emphasize and display the segmented organs that are sensitive to radiation.

[0046] The voxel dose distribution map generation unit 13 generates a voxel dose distribution map representing the radiation dose of each voxel using the trained radiation dose model described above and the acquired image data of the subject. As previously explained, after inputting the acquired image data of the subject into the trained radiation dose model, the voxel dose distribution map generation unit 13 outputs a voxel dose distribution map representing the distribution of each voxel in the image data. In this voxel dose distribution map, the estimated radiation dose received by each voxel is represented by color or grayscale intensity.

[0047] As an example, Figure 2B The text shows the use of... Figure 2AThe same image data is used to generate the voxel dose distribution map. Figure 2B The image uses grayscale values ​​from dark to light to represent the radiation dose received from high to low. The grayscale value of each pixel in the image represents the magnitude of the radiation dose received by that voxel. Furthermore, for ease of observation, it is preferable to... Figure 2B The grayscale level is represented by color, and different colors are used to represent the magnitude of the radiation dose received. Additionally, Figure 2B The diagram illustrates the changes in radiation dose for convenience, and may not reflect actual values.

[0048] The organ dose calculation unit 14 calculates the radiation dose received by each organ based on the segmented organ image data and the radiation dose received by each voxel in the generated voxel dose distribution map. Figure 2C This is an example of how radiation dose is received by individual voxels within an organ. For ease of observation, the boundaries of each organ can be highlighted. Additionally, the boundaries of radiation-sensitive organs can also be highlighted.

[0049] The organ dose calculation unit 14 is input with the voxel dose distribution map generated by the voxel dose distribution map generation unit 13 and the segmentation result after segmentation by the organ segmentation unit 12. After calculation using the following formula (1), the radiation dose value of each organ can be obtained according to the organ.

[0050]

[0051] Among them, D T ρ is the radiation dose received by the organ. V is the volume of the organ region. D(X,Y,Z) is the radiation dose received by each voxel (X,Y,Z). ρ(X,Y,Z) is the mass density of each voxel.

[0052] The radiation dose received by each organ region is estimated as the average dose within that organ region by calculation based on the above equation (1).

[0053] Due to the dangers of ionizing radiation, doctors sometimes want to know the radiation dose a patient has received before starting a scan to ensure the health and safety of the patient. For example, this could involve retrieving and obtaining image data from the patient's historical scans before beginning the current scan. Knowing the previous radiation dose allows for adjustments to the scanning protocol or special protection for certain radiation-sensitive organs.

[0054] Figure 3 Steps 100 (S100) to 500 (S500) show a flowchart of the management method of the radiation dose management device of this embodiment.

[0055] In step 100, the doctor inputs the patient information, i.e. the information of the subject, into the scanning console.

[0056] In step 200, the acquisition unit retrieves image data of interest from the database based on the input subject information. In step 200, the doctor can obtain information such as whether this is the subject's first scan, when previous scans were performed, and the scanned images obtained after the scan. Regarding the image data of interest, for example, to understand previous radiation exposure, the acquisition unit retrieves image data from historical scans that are associated with the subject information.

[0057] In step 300, the voxel dose distribution map generation unit generates a voxel dose distribution map representing the distribution of each voxel in the image data based on the acquired image data and the trained model.

[0058] In step 400, the organ segmentation unit uses the image data obtained in step 200 to segment each organ in the image data according to the organ and determine the boundaries of each organ.

[0059] In step 500, the radiation dose received by each organ is calculated using the voxel dose distribution map calculated in step 300 and the segmented organs in step 400.

[0060] Furthermore, although not explicitly stated above, the results of each step can be displayed through a scanning console or similar means. Additionally, the execution order of steps 300 and 400 is not limited to... Figure 3 The order shown can be reversed, or steps 300 and 400 can be executed simultaneously.

[0061] In addition, the above example illustrates the case in step 200 where the radiation dose to the organ is calculated using the results of a certain scan in the historical scan. However, sometimes it is also desirable to know the radiation dose to the organ from the current scan (the actual scan at the current moment), so the radiation dose to the organ from the current scan can also be calculated.

[0062] In this case, in step 200, after performing a scanning or imaging procedure such as a CT scan that generates a certain amount of ionizing radiation, the acquisition unit acquires the image data of the current scan and uses it as the object of analysis to analyze the radiation dose received by each organ.

[0063] According to the configuration of the first embodiment, it is possible to obtain the radiation dose received by each organ, especially radiation-sensitive organs.

[0064] (Modified Example)

[0065] In the first embodiment, a case is described in which the radiation dose to an organ is calculated using image data obtained from historical scans and the current scan.

[0066] However, a pre-operative positioning scan can also be performed before the formal scan. The positioning scan is the primary procedure in image examination, generating a positioning image used to determine the scan area and scan slices. Subsequently, based on the scan area and scan slices determined by the positioning image, and the scan parameters of the selected scan procedure, an image generated by the scan procedure can be simulated. This positioning scan is a pre-scan before the formal scan begins, and the radiation dose received is extremely small compared to the formal scan. By performing this positioning scan, the radiation dose to the subject during the formal scan can be estimated.

[0067] Figure 4 A flowchart illustrating a modified example of the radiation dose management device according to the first embodiment is shown. Figure 4 In step 200', a positioning scan is performed and a simulation of the actual scan is conducted. Then, the acquisition unit acquires the simulated image data and uses it as the analysis object to calculate the radiation dose received by each organ.

[0068] Based on the above configuration, it is also possible to obtain the radiation dose received by each organ, especially radiation-sensitive organs. Furthermore, when using localization images to simulate the radiation dose from the actual scan, it is possible to estimate the radiation dose from the actual scan with almost no radiation exposure.

[0069] (Second Implementation)

[0070] The difference between the second embodiment and the first embodiment is that the radiation dose management device 10' in the second embodiment also includes a correction unit 15, which corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor.

[0071] The differences between the first and second embodiments will be explained below, omitting the descriptions of the same parts.

[0072] The researchers of this invention discovered that while the body's tissue cells are damaged when exposed to radiation, they repair themselves over time, thus reducing the number of damaged cells. Furthermore, the number of damaged cells is directly proportional to the radiation dose, and the residual radiation dose in the body decreases as the body repairs itself.

[0073] Therefore, if the self-repair function of each organ is not considered and the radiation dose distribution map at the time of scanning is used to calculate the radiation dose received by each organ, the calculated radiation dose will be inaccurate.

[0074] Based on the above understanding, a correction unit 15 was added in the second embodiment.

[0075] The correction unit 15 receives a voxel dose distribution map from the voxel dose distribution map generation unit 13, and corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor, then outputs the corrected result to the organ dose calculation unit 14. The correction factor is the time elapsed after radiation and the inherent repair constant of each organ.

[0076] The corrected radiation dose of each voxel (also known as the residual radiation dose or residual dose) can be expressed by the following formula (2).

[0077] D t =D*exp(-μt) (2)

[0078] Where D is the radiation dose received by the voxel immediately after irradiation. t represents the time elapsed after irradiation. t This is the residual dose of the voxel after taking into account time factors and the repair constants of each organ; the residual dose D is... t The curve showing the change over time is as follows: Figure 6 As shown, the residual dose in the body of the tested individual decreases rapidly over time. μ is the repair constant, which is related to the self-repair capacity of cells. Different cells have different self-repair capabilities, thus this repair constant varies from organ to organ. For different organs, the residual dose D... t The curves that change over time also differ.

[0079] At t=0, the image data obtained is the image data just scanned at the current moment, and the corrected radiation dose of this scan is the same as that without correction.

[0080] By performing the above calculation, the corrected radiation dose (residual dose) for each voxel can be obtained.

[0081] The result corrected by the correction unit 15 is then output to the organ dose calculation unit 14, which calculates the residual dose of the organ.

[0082] As described above, the organ dose calculation unit 14 calculates the corrected radiation dose (residual dose) of each organ based on the segmented organ image data and the calculated corrected radiation dose (residual dose) of each voxel.

[0083] In addition, since multiple scans may be performed on the same subject, it is necessary to accumulate the correction results of multiple scans to calculate the cumulative dose of each voxel after correction, and further calculate the cumulative dose of each organ after correction.

[0084] To achieve the above objective, the correction unit 15 also includes a voxel dose accumulation unit 16. When the acquisition unit 11 acquires image data from multiple scans and the voxel dose distribution map generation unit 13 generates a voxel dose distribution map based on each image data, the voxel dose accumulation unit 16 adds up the corrected voxel doses of the corresponding voxels in each acquired image data, thereby calculating the corrected voxel dose accumulation value for each voxel.

[0085] The cumulative dose of voxels can be expressed by the following formula (3).

[0086] D A =∑Di*exp(-μ*ti) (3)

[0087] Where i represents the number of scans. ti represents the time elapsed after each radiation session. For example, we can define i = 0 as the current scan, i = 1 as the previous scan, i = 2 as the scan before that, and so on. When i = 0, t0 is 0, indicating no time has elapsed since the current radiation session; when i = 1, t1 represents the time elapsed since the previous scan; and when i = 2, t2 represents the time elapsed since the scan before that. Di is the radiation dose received by the voxel after each radiation session. A The cumulative dose value is the corrected dose (residual dose) of each voxel after taking into account the time factor and the repair constant of each organ.

[0088] By performing the above calculation, the cumulative dose value can be obtained by correcting and accumulating the voxels in each scan.

[0089] The output of the voxel dose accumulator 16 is then input to the organ dose calculation unit 14, which calculates the cumulative residual dose of the organ.

[0090] According to the configuration of the second embodiment, it is possible to correct the radiation dose of voxels in image data obtained from multiple scans and obtain the cumulative residual dose of each organ at the current moment.

[0091] Figure 7 A flowchart illustrating the management method of the radiation dose management device according to the second embodiment is shown.

[0092] In step 100, the doctor inputs the patient information, i.e. the information of the subject, into the scanning console.

[0093] In step 200, the acquisition unit acquires the image data of interest. As described in the first embodiment, the image data of interest may be at least one of the following: image data from historical scans, image data after the current formal scan, and image data simulated based on the determined scan range and scan plane and the scan parameters of the selected scan program (i.e., pre-scanned image data) by performing a positioning scan to determine the scan range and scan plane. The image data of interest may be one or more. In the case of multiple images, the image data of interest may be multiple image data composed of multiple historical scan image data, or multiple image data composed of one or more historical scan image data and formal scan image data.

[0094] In step 300, the voxel dose distribution map generation unit generates a voxel dose distribution map representing the distribution of each voxel in the image data based on the acquired image data and the trained model.

[0095] In step 400, the organ segmentation unit uses the image data obtained in step 200 to segment each organ in the image data according to the organ and determine the boundaries of each organ.

[0096] Next, unlike the first embodiment, in the second embodiment, steps 600 and 700 are performed instead of step 500.

[0097] In step 600, when the acquisition unit acquires image data of interest and the voxel dose distribution map generation unit generates a voxel dose distribution map based on the acquired image data of interest, the correction unit corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor, and outputs the corrected result to the organ dose calculation unit. When there are multiple image data of interest, the correction unit adds the corrected voxel doses of the corresponding voxels in each acquired image data by the voxel dose accumulation unit, obtains the corrected cumulative radiation dose of the voxels, and outputs it to the organ dose calculation unit.

[0098] In step 700, the organ dose calculation unit calculates the radiation dose of the organ based on the corrected voxel radiation dose. When there are multiple image data of interest, the organ dose calculation unit calculates the cumulative radiation dose of the organ based on the corrected voxel cumulative radiation dose.

[0099] Furthermore, although not explicitly stated above, the results of each step can be displayed through a scanning console or similar means. Additionally, the execution order of steps 300 and 400 is not limited to... Figure 7The order shown can be reversed, or steps 300 and 400 can be executed simultaneously.

[0100] According to the configuration of the second embodiment, the radiation dose of voxels in the scanned image data can be corrected, and the residual radiation dose of each organ at the current moment can be obtained. Furthermore, the radiation dose of voxels in image data obtained from multiple scans can be corrected, and the residual cumulative radiation dose of each organ at the current moment can be obtained.

[0101] (Third Implementation)

[0102] The third embodiment relates to the application of the calculated radiation dose to an organ.

[0103] In the third embodiment, radiation risk can be warned based on the calculated radiation dose to the organ.

[0104] [Confirmed Risk]

[0105] Deterministic risks are used to describe the causal relationship between ionizing radiation and certain side effects. Deterministic risks are also referred to as non-random effects to distinguish them from random, accidental risks (such as inducing cancer).

[0106] Certainty risks depend on factors such as dose, dose rate, irradiated volume, and type of radiation. There is a threshold for certainty risks; below this threshold, the risk will not occur. This threshold can be low and varies from person to person. However, once the threshold is exceeded, the severity of the risk increases with the amount of radiation received.

[0107] Examples of certain risks include the following situations.

[0108] Skin erythema: 2-5 Gy

[0109] Irreversible skin damage: 20-40 Gy

[0110] Hair loss: 2-5 Gy

[0111] Infertility: 2-3 Gy

[0112] Cataract: 0.5 Gy

[0113] Lethality (whole body): 3-5 Gy

[0114] Fetal malformations: 0.1-0.5 Gy

[0115] Here, Gy is the unit of absorbed radiation dose, which is the energy that ionizing radiation imparts to a unit mass of matter.

[0116] The following explanation uses the lens as an example. As a sensitive organ, the lens will rapidly die when the radiation dose it receives exceeds a certain threshold.

[0117] In procedures such as head CT scans, sinus CT scans, orbital CT scans, and neurointerventional surgeries, the eyes are likely to be within X-ray range. Without proper eye protection, this can result in a high radiation dose. Patients with recurrent or chronic diseases often require frequent CT scans; for example, 26% of patients with hydrocephalus receive a radiation dose exceeding 150 millisieverts (msv) within three years. In pediatric patients, repeated head CT scans can lead to an average cumulative dose to the lens of the eye of 26 mGy over several years, potentially reaching 1.3 Gy. Therefore, it is essential to pay close attention to the radiation dose received.

[0118] In this embodiment, the radiation dose management device can display the radiation dose of an organ at the current moment, the deterministic risk threshold of that organ, and warning information according to the organ.

[0119] For example, when using the radiation dose management device of the first and second embodiments to manage the radiation dose to the lens, the scanning control console displays prompts such as "The radiation dose received by this organ is AAA," "The threshold for certain risk to this organ is XXX," and "If the threshold of XXX is exceeded, there will be a very high risk." Thus, doctors can decide on subsequent treatment plans based on the obtained prompts.

[0120] [Random Risk]

[0121] If cells mutate, leading to gene mutations or malignant metastasis, it can result in random cancer or genetic risks. Random risks are often late-onset, with cancer being a typical example. Random risks can also be referred to as tissue-specific risks. Based on the radiation dose or cumulative radiation dose of each organ obtained from the first and second embodiments, the probability of tissue-specific changes occurring in each organ can be calculated. The probability P of tissue-specific risk can be expressed by the following formula.

[0122] P = r T *D T *W R

[0123] Where, r T The risk factor is related to the target area and the patient's age, as shown in Table 1 below. The numbers in the table represent the number of cancer cases per 100 msv of radiation dose, with a sample size of 100,000 people. T Cumulative organ dose. W R It is the weighting factor for radiation; for X-rays and gamma rays, this value is 1.

[0124] Table 1

[0125] Stomach 66 30 37 lung 373 166 208 colon 203 96 118 liver 32 14 18 bladder 153 75 91 Uterus 37 14 19 ovaries 76 28 37 prostate 67 34 41 breast 865 160 299 thyroid 200 18 54 leukemia 133 68 82

[0126] In this embodiment, the radiation dose management device displays the current radiation dose received by an organ, its tissue-specific risk threshold, and warning information, tailored to the organ. For example, the scanning control console can display in real-time warnings such as "The radiation dose received by this organ is AAA," "The tissue-specific risk threshold for this organ is XXX," and "If the XXX threshold is exceeded, there will be a very high risk." This allows doctors to determine subsequent treatment plans based on the obtained warning information.

[0127] The warnings regarding the aforementioned deterministic and random risks can be displayed on the scanning console as described above, or they can be provided via voice or other means.

[0128] According to the configuration of the third embodiment, it is possible to provide doctors with a wealth of reference information and provide early warnings of risks.

[0129] While several embodiments of the present invention have been described above, these embodiments are provided as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other ways, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included within the scope and spirit of the invention, and also within the scope of the invention as described in the claims and its equivalents.

Claims

1. A radiation dose management device for managing the radiation dose received by a subject, comprising: The acquisition department acquires image data of the subject. The organ segmentation unit segments the acquired image data according to the organs of the subject. The voxel dose distribution map generation unit generates a voxel dose distribution map representing the radiation dose of each voxel based on a specific radiation dose model and the obtained image data of the subject. The organ dose calculation unit calculates the radiation dose of each organ based on the segmented organ image data and the radiation dose of each voxel in the generated voxel dose distribution map; and The voxel dose correction unit corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor, which is the time elapsed after radiation and the inherent repair constant of each organ.

2. The radiation dose management device as described in claim 1, The organ dose calculation unit calculates the corrected radiation dose of each organ based on the segmented organ image data and the corrected radiation dose of each voxel.

3. The radiation dose management device as described in claim 1, It also has a voxel dose accumulation unit. When multiple image data exist regarding the subject being examined. The voxel dose accumulator adds the corrected voxel doses of the corresponding voxels in each of the acquired image data to calculate the corrected voxel dose accumulator value for each voxel. The organ dose calculation unit calculates the corrected radiation dose to each organ based on the segmented organ image data and the calculated corrected cumulative voxel dose value for each voxel.

4. The radiation dose management device as described in claim 1, The acquisition unit acquires at least one of the following: historical scan image data of the subject, current scan image data, and pre-scan image data, wherein the pre-scan is a positioning scan performed prior to the current scan. When the image data of the subject acquired by the acquisition unit is the image data of the current scan, the voxel dose correction unit does not correct the image data.

5. The radiation dose management device as described in claim 1, The specific radiation dose model is generated based on deep learning methods.

6. The radiation dose management device as described in claim 1, It also has a display control unit that displays the radiation dose of each organ on the display unit.

7. The radiation dose management device as described in claim 2 or 3, It also has a radiation dose risk warning unit that, based on the corrected radiation dose of each organ, indicates the deterministic and random risks of each organ.

8. The radiation dose management device as described in claim 1, The radiation dose management device is an X-ray CT device, a PET-CT device, or a SPECT-CT device.

9. A method for managing radiation dose, comprising the following steps: The acquisition step involves obtaining image data of the subject. The organ segmentation step involves segmenting the acquired image data according to the organs of the subject. The voxel dose distribution map generation step generates a voxel dose distribution map representing the radiation dose of each voxel based on a specific radiation dose model and the obtained image data of the subject. The organ dose calculation step involves calculating the radiation dose of each organ based on the segmented organ image data and the radiation dose of each voxel in the generated voxel dose distribution map; and The voxel dose correction step corrects the radiation dose of each voxel in the voxel dose distribution map based on a specific correction factor, which is the time elapsed after radiation and the inherent repair constant of each organ.

10. A storage medium storing a program that causes a computer to execute the radiation dose management method of claim 9.

Images