Method of providing information angiography for low contrast medium and device using the same

Abstract

The present specification provides a method for providing information on angiography for contrast reduction performed by a processor, comprising the steps of: receiving a computed tomography angiography (CTA) projection image of an individual and said CTA contrast agent concentration data; selecting one of a first model for each contrast agent concentration that is trained to generate a first extracted image of a blood vessel using the projection image of the CTA as input based on said contrast agent concentration data; generating a first extracted image using the received projection image as input to the selected first model; and generating a rendering image from said first extracted image.

Description

Technology Field

[0001] The present invention relates to a method for providing information on angiography for contrast agent reduction and a device using the same. Background Technology

[0002] Vascular disease is one of the issues threatening public health in a super-aging society, and cerebrovascular disease, in particular, is one of the top five causes of death. With the annual number of unruptured cerebral aneurysm patients in Korea increasing rapidly from 70,828 in 2016 to 143,808 in 2021, periodic follow-up monitoring is essential. In addition, the angiography imaging market is expected to grow significantly at an average annual rate of 8.8%, rising from approximately $46 million in 2022 to approximately $70 million in 2027.

[0003] Methods for examining vascular diseases can be divided into invasive and non-invasive methods. Invasive methods include Digital Subtraction Angiography (DSA), while non-invasive methods include Computed Tomography Angiography (CTA) and Magnetic Resonance Angiography (MRA). Non-invasive angiography devices generally acquire radiation-based projection images (CTA, MRA) after injecting a contrast agent, and then obtain images through post-processing. In particular, CTA is widely used because it offers shorter scanning times compared to MRA and has no imaging restrictions regarding metallic materials.

[0004] In this regard, iodinated contrast agents are generally used in CTA to increase the contrast of blood vessels. However, iodinated contrast agents used in CTA can cause various side effects, including local or systemic allergic reactions such as anaphylaxis, adverse reactions in the cardiovascular and renal systems, and symptoms that lead to fatal consequences such as extravasation.

[0005] Furthermore, recent studies have shown that irradiating with contrast agents increases the number of DNA damages compared to cases without contrast agents. In other words, as insufficient contrast agent washout during radiation therapy can have a negative impact, the use of low-concentration contrast agents is gaining traction.

[0006] Consequently, the use of non-ionic contrast agents with low osmotic pressure is being pursued as an alternative to iodinated contrast agents; however, this presents a limitation in that it is difficult to maintain maximum contrast enhancement. Furthermore, while high-molecular-weight iodine contrast agents are under development to overcome the short duration of contrast and toxicity of currently used low-molecular-weight contrast agents, there is a lack of verification regarding safety through clinical trials and the proven characteristics (quantity, density, injection rate, etc.) of the contrast agent that induce maximum angiography in the region of interest.

[0007] The background description of the invention is provided to facilitate a better understanding of the present invention. The matters described in the background description should not be construed as an acknowledgment that they exist as prior art. Prior art literature

[65535] Japanese Patent Publication No. JP 2023-502429 (January 24, 2023) The problem to be solved

[0008] Conventional contrast agent reduction methods in CTA include optimization methods for contrast agent injection speed and imaging conditions, dual-energy based contrast agent reduction methods, development of iodinated contrast agents based on polymer compounds, and deep learning-based contrast agent reduction methods.

[0009] More specifically, the method for optimizing the injection rate and imaging conditions of the contrast agent (hereinafter referred to as the contrast agent optimization method) is a method for optimizing the contrast effect according to the amount, concentration, injection rate, and tube voltage of the iodine contrast agent based on the HU value in each organ. Although the contrast agent can be reduced simply by changing the CTA protocol, it has the limitation that the reduction rate is negligible. In addition, as the contrast agent optimization method is performed in the same way as the conventional CTA method, there are limitations regarding double coating, artifact errors caused by movement, loss of procedural information regarding high-density materials, and information loss due to beam hardening phenomena.

[0010] Next, the dual-energy-based contrast agent reduction method is based on the characteristic that the linear attenuation coefficient varies according to the radiation energy of each material. It involves acquiring low and high-energy images and then separating the materials by reflecting the difference in linear attenuation coefficients between them at different energies. This dual-energy-based method can reduce the amount of contrast agent required by increasing the intravascular contrast agent signal through the maximum K-edge response to iodine based on low tube voltage settings. However, it has limitations, such as a very low contrast agent reduction rate and higher development costs compared to conventional CTA. Additionally, the dual-energy-based method still faces limitations regarding data loss caused by high-density materials and beam hardening phenomena.

[0011] Subsequently, in order to overcome the limitations of conventional iodine-containing low-molecular-weight contrast agents, the development of contrast agents based on iodine-containing radial polymer compounds is currently underway. More specifically, the development method for polymer-based iodinated contrast agents offers longer contrast duration, lower toxicity, fewer adverse reactions, and simpler manufacturing (purification) methods compared to conventional contrast agents; however, its clinical utility has not yet been sufficiently evaluated, and limitations regarding data loss due to high-density materials and beam hardening phenomena still remain.

[0012] Next, deep learning-based contrast agent reduction methods are being introduced, including deep learning-based methods trained on CTA images for normal contrast agent doses and CTA images for low contrast agent doses. However, because it is difficult to obtain a large amount of refined data for pre-training, machine learning is not sufficiently trained, which has limitations in that accurate prediction is difficult.

[0013] In other words, as mentioned above, conventional methods suffer from low accuracy and reliability of contrast imaging due to beam hardening. Currently, various methods are being attempted to correct for low contrast agent imaging and beam hardening; however, these methods all have limitations, such as the inclusion of virtual information, high development costs, loss of information regarding high-density procedures, limitations in reconstruction, and difficulties in securing training data.

[0014] Meanwhile, the inventors of the present invention recognized that by utilizing a calibration phantom reflecting various blood vessel thicknesses and a computational simulation system based thereon, various CTA projection images according to concentration can be obtained, and machine learning can be performed based on this. In other words, the inventors recognized that the limitations of training data related to conventional machine learning methods can be overcome through CTA system modeling using calibration phantoms and computational simulation, and discovered that medical images in various cases similar to reality can be generated through computational simulation modeling.

[0015] Furthermore, the inventors of the present invention have discovered that when data regarding blood vessels and data regarding procedures are extracted separately and then aligned to generate a rendering image, the blood vessel and procedure data can be extracted more precisely even from a single captured image, and errors caused by the conventional beam hardening phenomenon can be reduced.

[0016] Ultimately, the inventors of the present invention developed a method for providing information on angiography that can generate CTA projection images for different actual contrast agent concentrations through a calibration phantom reflecting various blood vessel thicknesses, acquire CTA projection images for various cases such as actual clinical images by reflecting them in a computer simulation system, train a machine learning model for different concentrations based on this, and generate a three-dimensional rendering image of a CTA projection image for a low-concentration contrast agent using the trained machine learning model for different concentrations.

[0017] Accordingly, the problem that the present invention aims to solve is to provide a machine learning-based method for providing information on angiography and a device based thereon, in which the limitations regarding the aforementioned contrast agent side effects, double coating, and beam hardening phenomena are overcome.

[0018] The problems of the present invention are not limited to those mentioned above, and other unmentioned problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0019] To solve the problem described above, the present invention provides a method for providing information on angiography for contrast reduction performed by a processor, comprising: receiving a computed tomography angiography (CTA) projection image and CTA contrast agent concentration data of an individual; selecting one of a first model for each contrast agent concentration that is trained to generate a first extracted image of a blood vessel using a projection image of CTA as input based on the contrast agent concentration data; generating a first extracted image using the received projection image as input to the selected first model; and generating a rendering image from the first extracted image.

[0020] According to a feature of the present invention, the step of receiving clinical data of an individual may be further included.

[0021] According to another feature of the present invention, the method may further include the step of predicting the probability of a vascular disease of an individual based on clinical data and rendering images.

[0022] According to another feature of the present invention, the step of predicting the probability may include the step of extracting features within a rendering image, and the step of predicting the probability of a vascular disease in an object using clinical data and features as input to a third model trained to predict the probability of a vascular disease in an object using clinical data and features as input.

[0023] According to another feature of the present invention, the CTA projection image may be a single projection image taken after the contrast agent is injected, but is not limited thereto.

[0024] According to another feature of the present invention, the concentration of the contrast agent may be 90% or less of the standard concentration of the contrast agent, but is not limited thereto.

[0025] According to another feature of the present invention, the first model according to the concentration of the contrast agent may be at least one of the following with respect to the reference contrast agent concentration: a first model with a concentration of 0%, a first model with a concentration of 10%, a first model with a concentration of 20%, a first model with a concentration of 30%, a first model with a concentration of 40%, a first model with a concentration of 50%, a first model with a concentration of 60%, a first model with a concentration of 70%, a first model with a concentration of 80%, a first model with a concentration of 90%, and a first model with a concentration of 100%, but is not limited thereto.

[0026] According to another feature of the present invention, prior to the selection step, the method may further include: a step of generating a second extracted image of a blood vessel from a received CTA projection image; a step of obtaining a blood vessel thickness from the second extracted image; a step of converting the image intensity of the second extracted image based on the blood vessel thickness to generate a third extracted image of a contrast agent concentration from the second extracted image; a step of applying the third extracted image of a contrast agent concentration to a computer simulation of a CTA system to generate at least one simulation image of a contrast agent concentration; a step of matching the received CTA projection image to each simulation image of a contrast agent concentration to generate a first data set of a contrast agent concentration; and a step of learning each first model of a contrast agent concentration based on the first data set of a contrast agent concentration.

[0027] According to another feature of the present invention, the step of generating a second extracted image may include removing data for bones within the received CTA projection image and then applying a threshold value.

[0028] According to another feature of the present invention, the step of acquiring blood vessel thickness may include measuring the blood vessel thickness in a second extracted image from one or more lateral directions, and calculating an average value for one or more measured blood vessel thicknesses.

[0029] According to another feature of the present invention, the step of generating a third extracted image may include: selecting at least one of a predetermined transformation formula for each blood vessel thickness based on blood vessel thickness; and applying the selected at least one transformation formula to the second extracted image to generate a third extracted image for each contrast agent concentration from the second extracted image.

[0030] According to another feature of the present invention, a predetermined conversion formula for each blood vessel thickness may be calculated based on a CTA projection image of a calibration phantom, but is not limited thereto.

[0031] According to another feature of the present invention, the third extracted image may be an image in which the received CTA projection image is aligned with each of the third extracted images according to concentration, but is not limited thereto.

[0032] According to another feature of the present invention, the step of generating a rendered image may be based on at least one of filtered back-projection (FBP), algebraic reconstruction technique (ART), maximum likelihood expectation maximization (ML-EM), and automap, but is not limited thereto.

[0033] According to another feature of the present invention, the step of generating a rendering image may include the step of generating a rendering image of a blood vessel using the first extracted image as input to a second model trained to generate a rendering image using the first extracted image as input.

[0034] According to another feature of the present invention, the step of generating a rendered image may further include a step of correcting the quality of the rendered image.

[0035] According to another feature of the present invention, the correction may include at least one of restoration, enhancement, registration, and linear correction, but is not limited thereto.

[0036] To solve other problems as described above, the present invention provides a device for providing information on angiography for contrast agent reduction, comprising a communication unit configured to receive a computed tomography angiography (CTA) projection image and CTA contrast agent concentration data of an individual, and a processor connected to communicate with the communication unit, wherein the processor is configured to select one of a first model for each contrast agent concentration that is trained to generate a first extracted image of a blood vessel by taking a projection image of CTA as input based on the contrast agent concentration data, generate a first extracted image by taking the received projection image as input to the selected first model, and generate a rendering image from the first extracted image.

[0037] According to a feature of the present invention, the communication unit may be further configured to receive clinical data of an individual.

[0038] According to another feature of the present invention, the processor may be further configured to predict the probability of a vascular disease of an object based on clinical data and rendering images.

[0039] According to another feature of the present invention, a processor may be configured to extract features within a rendered image and to predict the probability of a vascular disease in an individual using clinical data and features as input to a third model trained to predict the probability of a vascular disease in an individual using clinical data and features as input.

[0040] According to another feature of the present invention, the CTA projection image may be a single projection image taken after the contrast agent is injected, but is not limited thereto.

[0041] According to another feature of the present invention, the contrast agent concentration may be 90% or less of the standard contrast agent concentration, but is not limited thereto.

[0042] According to another feature of the present invention, the first model according to the concentration of the contrast agent may be at least one of the following with respect to the reference contrast agent concentration: a first model with a concentration of 0%, a first model with a concentration of 10%, a first model with a concentration of 20%, a first model with a concentration of 30%, a first model with a concentration of 40%, a first model with a concentration of 50%, a first model with a concentration of 60%, a first model with a concentration of 70%, a first model with a concentration of 80%, a first model with a concentration of 90%, and a first model with a concentration of 100%, but is not limited thereto.

[0043] According to another feature of the present invention, the processor may be further configured to generate a second extracted image of a blood vessel from a received CTA projection image prior to the selection of one of the first models for each contrast agent concentration, obtain a blood vessel thickness from the second extracted image, convert the image intensity of the second extracted image based on the blood vessel thickness, generate a third extracted image for each contrast agent concentration from the second extracted image, apply the third extracted image for each contrast agent concentration to a computer simulation of a CTA system to generate at least one simulation image for each contrast agent concentration, match the received CTA projection image to each simulation image for each contrast agent concentration to generate a first data set for each contrast agent concentration, and learn each first model for each contrast agent concentration based on the first data set for each contrast agent concentration.

[0044] According to another feature of the present invention, the processor may include the step of removing data for bones in a received CTA projection image and then applying a threshold value.

[0045] According to another feature of the present invention, the processor may be configured to measure the blood vessel thickness in a second extracted image from one or more lateral directions and to calculate an average value for the one or more measured blood vessel thicknesses.

[0046] According to another feature of the present invention, a processor may be configured to select at least one of a predetermined conversion formula for each blood vessel thickness based on blood vessel thickness, and apply the selected at least one conversion formula to a second extracted image to generate a third extracted image for each contrast agent concentration from the second extracted image.

[0047] According to another feature of the present invention, a predetermined conversion formula for each blood vessel thickness may be calculated based on a CTA projection image of a calibration phantom, but is not limited thereto.

[0048] According to another feature of the present invention, the third extracted image may be an image in which the received CTA projection image is aligned with each of the third extracted images according to concentration, but is not limited thereto.

[0049] According to another feature of the present invention, the processor may be based on at least one of filtered back-projection (FBP), algebraic reconstruction technique (ART), maximum likelihood expectation maximization (ML-EM), and automap, but is not limited thereto.

[0050] According to another feature of the present invention, a processor may be configured to generate a rendering image of a blood vessel using the first extracted image as input to a second model trained to generate a rendering image using the first extracted image as input.

[0051] According to another feature of the present invention, the processor may be further configured to correct the quality of the rendered image.

[0052] According to another feature of the present invention, the correction may include at least one of restoration, enhancement, registration, and linear correction, but is not limited thereto.

[0053] The present invention will be explained in more detail below through examples. However, since these examples are merely illustrative of the present invention, the scope of the present invention should not be interpreted as being limited by these examples. Effects of the invention

[0054] The present invention can provide high-accuracy three-dimensional rendering images using CTA projection images taken based on a low concentration of contrast agent, and furthermore, the present invention can predict vascular disease in an individual based on this.

[0055] Accordingly, the present invention can minimize side effects caused by the contrast agent.

[0056] Furthermore, since the present invention can generate various machine learning training data based on contrast agent concentrations using calibration phantoms and computational simulations, data that was difficult to obtain due to ethical and personal information concerns can be constructed more easily and utilized for machine learning training, thereby improving the accuracy and reliability of machine learning.

[0057] In addition, since the model of the present invention is trained based on various data, it is possible to generate a three-dimensional image by correcting for high-density surgical data loss due to noise and beam hardening phenomena caused by the movement of artifacts, which are conventional limitations, even with a single projection image taken from a CTA device, thereby solving the problem of double coating (over-coating).

[0058] For example, in existing CTA methods, even if vascular data is extracted, a process of removing bone and high-density information with similar HU (Hounsfield unit) values ​​by medical staff or medical technicians had to be included. However, the present invention eliminates this process and automatically extracts accurate vascular and high-density procedure data by machine learning, and generates a 3D rendering image based on this, thereby maximizing the efficiency of image generation.

[0059] In addition, the present invention can also provide high-accuracy predictions for vascular diseases based on high-accuracy and high-precision three-dimensional rendering images generated through the aforementioned process, thereby assisting medical staff in making rapid diagnoses.

[0060] The effects according to the present invention are not limited to those exemplified above, and various other effects are included in this specification. Brief explanation of the drawing

[0061] FIG. 1a illustrates an exemplary system based on an information method for angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 1b is a block diagram showing the configuration of a device for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 1c is a block diagram of the configuration of a medical device according to one embodiment of the present invention. FIG. 1d is a schematic diagram of a user interface of a medical device and a medical imaging device according to one embodiment of the present invention. FIG. 2 is a flowchart of a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 3 is a schematic diagram of a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 4 is a schematic diagram of the learning process of a first model in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 5 is a schematic diagram of a calibration phantom and a conversion formula used in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. FIG. 6 is a schematic diagram of vascular disease prediction in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention. Specific details for implementing the invention

[0062] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims.

[0063] As used in this specification, the term "or" means "and / or" unless otherwise stated.

[0064] As used herein, the term “about” refers to a normal margin of error for each value that is readily known to those skilled in the art. In this specification, the designation of an “about” value or parameter includes an example relating to the value or parameter itself. Furthermore, unless otherwise stated or otherwise evident from the context, the term “about” indicates a range of values ​​corresponding to within 10% in either direction (greater than or less than) a mentioned reference value.

[0065] As used herein, the term “patient or individual” refers interchangeably to any single animal requiring treatment, more preferably a mammal (including such non-human animals, e.g., cats, dogs, horses, rabbits, zoo animals, cattle, pigs, sheep, and non-human primates). In various embodiments of this specification, the patient referred to may be a human.

[0067] One of the present invention Examples Contrast agent according to Reduction System based on a method of providing information on angiography for

[0069] FIG. 1a illustrates an exemplary system based on an information method for angiography for contrast agent reduction according to one embodiment of the present invention.

[0070] First, referring to FIG. 1a, the information providing system (1000) for angiography for contrast agent reduction may be a system configured to provide information related to computed tomography angiography (CTA) of an individual based on various data collected from the individual.

[0071] At this time, the angiography information provision system (1000) may be composed of a device (100) for providing information on angiography for contrast agent reduction configured to generate a rendering image of a projection image based on data collected from an individual, namely clinical data, a CTA projection image (310), and CTA contrast agent concentration data (320), and further, to predict a vascular disease of an individual based on the generated rendering image and the aforementioned clinical data, a medical staff device (200) for transmitting and receiving information on angiography, and a computed tomography device (medical imaging device, 300).

[0072] At this time, the device (100) for providing information on angiography for contrast agent reduction, the medical staff device (200), and the medical imaging device (300) can transmit and receive various information via wired and wireless communication.

[0073] More specifically, the device (100) for providing information on angiography for contrast agent reduction, the medical staff device (200), and the medical imaging device (300) can communicate via wired transmission and reception directly connected by a cable, but preferably, they can communicate wirelessly with the cable omitted.

[0074] Accordingly, the device (100) for providing information on angiography for contrast agent reduction, the medical staff device (200), and the medical imaging device (300) may be connected to a network for wireless communication, and the network may be a closed network such as a LAN (Local Area Network) and a WAN (Wide Area Network), or an open network such as the Internet, and may be a short-range wireless communication such as Bluetooth, NFC (Near Field Communication), RFID (Radio-Frequency Identification), Wi-Fi and / or Zigbee, but is not limited thereto.

[0075] A device (100) for providing information on angiography for contrast agent reduction according to one embodiment of the present invention may include a general-purpose computer, a laptop, and / or a data server, etc., which generates a rendering image (3D image) of a blood vessel based on medical image data of an individual collected from a medical device (200) and / or a medical imaging device (300), namely, a computed tomography angiography (CTA) projection image, CTA contrast agent concentration data, and clinical data, and further performs various calculations to predict vascular disease of the individual.

[0076] More specifically, with reference to FIG. 1b, a block diagram showing the configuration of a device for providing information for angiography for contrast agent reduction according to one embodiment of the present invention is shown. The device for providing information for angiography for contrast agent reduction (100) may include a communication interface (110), a memory (120), an I / O interface (130), and a processor (140), and each component may communicate with one or more communication buses or signal lines.

[0077] The communication interface (110) may refer to a communication unit (110) and can be connected to a medical staff device (200) and a medical imaging device (300) via a wired / wireless communication network to exchange data. For example, the communication interface (110) can receive various data (clinical data and medical imaging data) about an object in real time from the medical staff device (200) and / or the medical imaging device (300). As another example, the communication interface (110) can transmit various medical data and medical images related to an object from the medical staff device (200).

[0078] Meanwhile, a communication interface (110) that enables the transmission and reception of such data includes a wired communication port (111) and a wireless circuit (112), wherein the wired communication port (111) may include one or more wired interfaces, for example, Ethernet, Universal Serial Bus (USB), FireWire, etc. Additionally, the wireless circuit (112) may transmit and receive data with an external device via an RF signal or an optical signal. Furthermore, wireless communication may use at least one of a plurality of communication standards, protocols, and technologies, such as GSM, EDGE, CDMA, TDMA, Bluetooth, Wi-Fi, VoIP, Wi-MAX, or any other suitable communication protocol.

[0079] The memory (120) is used in the device (100) for providing information on angiography for contrast agent reduction and can store various data derived and generated.

[0080] More specifically, the memory (120) can store a prediction model and can also store various data extracted or generated (predicted) from the aforementioned prediction model. For example, the memory (120) may include various algorithms, parameters, functions, extracted (generated) data, etc. included in the model, but is not limited thereto.

[0081] Furthermore, the memory (120) may include all various data related to the output data. For example, the memory (120) may include various extracted data for the process of generating a rendering image of an object, but is not limited thereto.

[0082] Additionally, the memory (120) can store various data collected from a medical device (200) or a medical imaging device (300), such as clinical data and medical images.

[0083] More specifically, medical images may include images derived (generated) from medical imaging devices such as CT, MRI, and X-ray, but are not limited thereto, and may include all various medical images used to identify and evaluate a disease of an individual. However, most preferably, the medical images may be computed tomography angiography (CTA) projection images.

[0084] In various embodiments, the memory (120) may include a volatile or non-volatile recording medium capable of storing various data, commands, and information. For example, the memory (120) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM, SRAM, ROM, EEPROM, PROM, network storage, cloud, and blockchain database.

[0085] In various embodiments, the memory (120) may store at least one configuration of an operating system (121), a communication module (122), a user interface module (123), and one or more applications (124).

[0086] An operating system (121) (e.g., embedded operating systems such as LINUX, UNIX, MAC OS, WINDOWS, VxWorks, etc.) may include various software components and drivers for controlling and managing general system operations (e.g., memory management, storage device control, power management, etc.) and may support communication between various hardware, firmware, and software components.

[0087] The communication module (123) can support communication with another device through the communication interface (110). The communication module (120) may include various software components for processing data received by the wired communication port (111) or wireless circuit (112) of the communication interface (110).

[0088] The user interface module (123) can receive requests or inputs from a user, such as a keyboard, touch screen, keyboard, mouse, microphone, etc., through the I / O interface (130) and provide a user interface on the display.

[0089] The application (124) may include a program or module configured to be executed by one or more processors (140). Here, the application for providing information about angiography may be implemented on a server farm.

[0090] The I / O interface (130) can connect at least one of the input / output devices (not shown), such as a display, keyboard, touch screen, and microphone of the device (100) for providing information on angiography for contrast agent reduction, to the user interface module (123). The I / O interface (130) can receive user input (e.g., voice input, keyboard input, touch input, etc.) together with the user interface module (123) and process commands based on the received input.

[0091] The processor (140) is connected to the communication interface (110), memory (120), and I / O interface (130) to control the overall operation of the device (100) for providing information on angiography for contrast agent reduction, and can execute various commands to extract data related to angiography through an application or program stored in the memory (120).

[0092] The processor (140) may correspond to a computing device such as a CPU (Central Processing Unit) or an AP (Application Processor). Additionally, the processor (140) may be implemented in the form of an Integrated Chip (IC), such as a System on Chip (SoC) that integrates various computing devices. Alternatively, the processor (140) may include a module for computing artificial neural network (AI, machine learning) models, such as a Neural Processing Unit (NPU).

[0093] Referring again to FIG. 1a, a device (100) for providing information on angiography for contrast agent reduction receives clinical data, CTA medical imaging data, and CTA contrast agent concentration data for an object from a medical device (200) and / or a medical imaging device (300), generates a 3D rendering image from the received data, predicts vascular disease of the object, and can provide this to the medical device (200) and / or the medical imaging device (300).

[0094] In this way, data provided from the device (100) for providing information on angiography for contrast agent reduction may be provided as a web page through a web browser installed on a medical staff device (200) and / or a medical imaging device (300), or provided in the form of an application or program. In various embodiments, such data may be provided in a form included in a platform in a client-server environment.

[0095] First, the medical device (200) is an electronic device that requests information regarding angiography of an object and provides a user interface for displaying related information data (image data), and may include at least one of a smartphone, a tablet PC (Personal Computer), a laptop and / or a PC.

[0096] More specifically, with reference to FIG. 1c, a block diagram of the configuration of a medical device according to one embodiment of the present invention is illustrated. The medical device (200) may include a memory interface (210), one or more processors (220), and a peripheral interface (230). Various components within the medical device (200) may be connected by one or more communication buses or signal lines.

[0097] The memory interface (210) is connected to the memory (250) and can transmit various data to the processor (220). Here, the memory (250) may include at least one type of storage medium among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM, SRAM, ROM, EEPROM, PROM, network storage, cloud, and blockchain database.

[0098] In various embodiments, memory (250) may store at least one of an operating system (251), a communication module (252), a graphical user interface module (GUI) (253), a sensor processing module (254), a telephone module (255), and an application module (256). Specifically, the operating system (251) may include instructions for processing basic system services and instructions for performing hardware operations. The communication module (252) may communicate with at least one of one or more other devices, computers, and servers. The graphical user interface module (GUI) (253) may process a graphical user interface. The sensor processing module (254) may process sensor-related functions (e.g., processing voice input received using one or more microphones (292)). The telephone module (255) may process telephone-related functions. The application module (256) may perform various functions of a user application, such as electronic messaging, web browsing, media processing, exploration, imaging, and other processing functions.

[0099] In addition, the medical device (200) may store one or more software applications (256-1, 256-2) associated with any one type of service in the memory (250). At this time, the application (256-1) may provide information about angiography to the medical device (200).

[0100] In various embodiments, the memory (250) can store a digital assistant client module (257) (hereinafter, DA client module) and, accordingly, can store commands for performing client-side functions of the digital assistant and various user data (258) (e.g., user-customized vocabulary data, preference data, user's electronic address book, to-do list, other lists, etc.).

[0101] Meanwhile, the DA client module (257) can obtain voice input, text input, touch input and / or gesture input from the user through various user interfaces (e.g., I / O subsystem (240)) provided in the medical device (200).

[0102] Additionally, the DA client module (257) can output data in the form of audiovisual and tactile elements. For example, the DA client module (257) can output data consisting of a combination of at least two of voice, sound, notifications, text messages, menus, graphics, videos, animations, and vibrations. Furthermore, the DA client module (257) can communicate with a digital assistant server (not shown) using a communication subsystem (280).

[0103] In various embodiments, the DA client module (257) may collect additional information about the surrounding environment of the medical device (200) from various sensors, subsystems, and peripheral devices to construct the context associated with the user input. For example, the DA client module (257) may provide context information along with the user input to a digital assistant server to infer the user's intent. Here, the context information that may accompany the user input may include sensor information, e.g., lighting, ambient noise, ambient temperature, images, videos, etc. of the surrounding environment. As another example, the context information may include the physical state of the medical device (200) (e.g., device orientation, device location, device temperature, power level, speed, acceleration, motion patterns, cellular signal strength, etc.). As yet another example, the context information may include information related to the software state of the medical device (200) (e.g., processes running on the medical device (200), installed programs, past and present network activity, background services, error logs, resource usage, etc.).

[0104] In various embodiments, the memory (250) may include additional or deleted instructions, and furthermore, the medical device (200) may include additional configurations in addition to the configuration shown in FIG. 1c, or exclude some configurations.

[0105] The processor (220) can control the overall operation of the medical device (200) and can execute various commands to implement various data interfaces for angiography by running an application or program stored in memory (250).

[0106] The processor (220) may correspond to a computing device such as a CPU (Central Processing Unit) or an AP (Application Processor). Additionally, the processor (120) may be implemented in the form of an integrated chip (IC), such as a System on Chip (SoC) that integrates various computing devices such as a Neural Processing Unit (NPU).

[0107] The peripheral interface (230) is connected to various sensors, subsystems, and peripheral devices and can provide data to enable the medical device (200) to perform various functions. Here, it can be understood that the medical device (200) performing a function is performed by the processor (220).

[0108] The peripheral interface (230) may receive data from a motion sensor (260), a light sensor (light sensor) (261), and a proximity sensor (262), thereby enabling the medical device (200) to perform orientation, light, and proximity sensing functions. As another example, the peripheral interface (230) may receive data from other sensors (263) (positioning system—GPS receiver, temperature sensor, biometric sensor), thereby enabling the medical device (200) to perform functions related to the other sensors (263).

[0109] In various embodiments, the medical device (200) may include a camera subsystem (270) connected to a peripheral interface (230) and an optical sensor (271) connected thereto, thereby enabling the medical device (200) to perform various shooting functions such as taking photos and recording video clips.

[0110] In various embodiments, the medical device (200) may include a communication subsystem (280) connected to a peripheral interface (230). The communication subsystem (280) is composed of one or more wired / wireless networks and may include various communication ports, radio frequency transceivers, and optical transceivers.

[0111] In various embodiments, the medical device (200) includes an audio subsystem (290) connected to a peripheral interface (230), and the audio subsystem (290) includes one or more speakers (291) and one or more microphones (292), so that the medical device (200) can perform voice-operated functions, such as voice recognition, voice replication, digital recording, and telephone functions.

[0112] In various embodiments, the medical device (200) may include an I / O subsystem (240) connected to a peripheral interface (230). For example, the I / O subsystem (240) may control a touch screen (243) included in the medical device (200) through a touch screen controller (241). As an example, the touch screen controller (241) may detect user contact and movement or interruption of contact and movement using any one of a plurality of touch sensing technologies, such as capacitive, resistive, infrared, surface acoustic wave technology, proximity sensor array, etc. As another example, the I / O subsystem (240) may control other input / control devices (244) included in the medical device (200) through other input controller(s) (242). As an example, other input controller(s) (242) can control one or more pointer devices such as buttons, rocker switches, thumb wheels, infrared ports, USB ports, and styluses.

[0113] Referring again to FIG. 1a, the medical device (200) can receive images of blood vessels generated from the device (100) for providing information on angiography for contrast agent reduction, and prediction data for vascular diseases derived based thereon, as the medical device (200) includes the configuration described above.

[0114] The medical imaging device (300) may refer to medical imaging devices such as radiography, magnetic resonance imaging (MRI), ultrasound, endoscopy, thermography, and nuclear medicine imaging, but preferably may refer to a computerized or computed tomography (CT) device. More specifically, the medical imaging device (300) may refer to a CT device capable of irradiating a target area of ​​the human body from multiple directions using X-rays and a computer, collecting transmitted X-rays with a detector, and reconstructing the difference in X-ray absorption for that area into an image using a computer (computation) through mathematical techniques.

[0115] Furthermore, the medical imaging device (300) may be a device capable of providing captured medical images to a device (100) for providing information regarding angiography for contrast agent reduction, displaying them on the medical imaging device (300), and adjusting them. Accordingly, the medical imaging device (300) may include a smartphone, tablet PC (Personal Computer), laptop and / or PC, etc. as described above, but is not limited thereto, and may further include a general-purpose computer and / or data server, etc., capable of providing data about an object. Furthermore, the medical imaging device (300) may include a DB capable of storing captured medical images.

[0116] Meanwhile, in order to display images and related data, this medical imaging device (300) may include the same configuration as the medical staff device (200) described above in FIG. 1c. Accordingly, the medical imaging device (300) may include an interface that allows a user to directly input clinical data about an object, data related to medical images, and data related to the object's vascular disease from a keyboard, touch screen, keyboard, mouse, and microphone, and through this, receive data related to the object's blood vessels from the user and provide it to the device (100) for providing information on angiography for contrast agent reduction and / or the medical staff device (200).

[0117] In this regard, referring to FIG. 1d, a schematic diagram of a user interface for a medical device and a medical imaging device according to one embodiment of the present invention is shown.

[0118] Referring to FIG. 1d, a medical device (200) and a medical imaging device (300) used in an information provision system for angiography according to one embodiment of the present invention can receive various data generated and derived from a device (100) for providing information for angiography for contrast agent reduction. Accordingly, the medical device (200) and the medical imaging device (300) can display the received data on an interface.

[0119] More specifically, the medical staff device (200) and the medical imaging device (300) may be configured to display a rendering image (220) generated from the device (100) for providing information on angiography for contrast reduction and prediction data (230) for vascular disease predicted from the device (100) for providing information on angiography for contrast reduction on a user interface screen (UI, 210).

[0120] Furthermore, on the user interface screen (210), not only the aforementioned rendering image (220) and prediction data (230), but also various data used in the device (100) for providing information on angiography for contrast agent reduction can be displayed. For example, on the user interface screen (210), various clinical data related to the subject and the blood vessel extraction image generated from the device (100) for providing information on angiography for contrast agent reduction can be displayed, but is not limited thereto.

[0121] Accordingly, the medical staff device (200) and the medical imaging device (300) can more easily obtain and verify various data generated and predicted from the device (100) for providing information on angiography for contrast agent reduction.

[0122] Ultimately, the medical staff device (200) and the medical imaging device (300) receive various data generated and predicted from the device (100) for providing information on angiography for contrast agent reduction through the user interface screen (210), and the received data can be displayed through the user interface screen (210) and provided to the user.

[0123] Referring again to FIG. 1a, various medical image data captured (generated) by the medical imaging device (300) may be provided to the information providing device (100) for angiography for contrast agent reduction as well as to the medical staff device (200). For example, not only the projected image captured by the medical imaging device (300) but also the raw data stored in the medical imaging device (300) in relation thereto may be transmitted and used by the information providing device (100) for angiography for contrast agent reduction and the medical staff device (200), but is not limited thereto.

[0124] Ultimately, an information provision system for angiography based on a device for providing information for angiography for contrast agent reduction according to one embodiment of the present invention includes a device for providing information for angiography for contrast agent reduction (100), a medical staff device (200), and a medical imaging device (300). Accordingly, it receives CTA medical images and CTA contrast agent concentration data from the medical imaging device (300), and simultaneously receives clinical data from the medical staff device (200). It generates a rendering image related to the blood vessels of an individual based on the data received from the device for providing information for angiography for contrast agent reduction (100), and further predicts and determines various information (data) related to the vascular disease of the individual, and provides various information (data) generated (derived) from the device for providing information for angiography for contrast agent reduction (100) to the medical staff device (200) and / or the medical imaging device (300). However, not limited to this, the operation of predicting and determining vascular disease in an object based on received data may also be performed in a medical device (200).

[0126] One of the present invention Examples Contrast agent according to Reduction Method of providing information on angiography for

[0127] Hereinafter, with reference to FIGS. 2 to 6, a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention and a detailed process thereof will be described.

[0128] First, FIG. 2 is a flowchart of a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention.

[0129] Referring to FIG. 2, a method for providing information on angiography for contrast reduction according to one embodiment of the present invention is a method for providing information on angiography for contrast reduction performed by a processor, and may include the steps of: receiving a computed tomography angiography (CTA) projection image of an individual and CTA contrast agent concentration data (S210); selecting one of a first model for each contrast agent concentration that is trained to generate a first extraction image of a blood vessel using a projection image of CTA as input based on the contrast agent concentration data (S220); generating a first extraction image using the received projection image as input to the selected first model (S230); and generating a rendering image from the first extraction image.

[0130] First, the CTA projection image in the receiving step (S210) may refer to a two-dimensional projection image obtained by taking an X-ray-based CTA, but is not limited thereto and may include all various medical images capable of extracting blood vessels. Furthermore, the CTA projection image may refer to a projection image taken as a single shot after the contrast agent is injected. Generally, imaging techniques based on conventional CTA projection images were performed based on two CT images obtained through a second shooting after the injection of the contrast agent, and this conventional method results in double covering due to the second shooting. Furthermore, the conventional method has the disadvantage that artifacts may occur if the subject moves during the second shooting period. However, according to one embodiment of the present invention, a three-dimensional blood vessel image (3D rendering image) containing accurate blood vessel information can be generated through a single shot when the contrast agent exhibits maximum contrast enhancement in the region of interest. Furthermore, the contrast agent for the CTA projection image in the present invention may be at a low concentration. Generally, conventional reference contrast agent concentrations used in CTA systems can cause damage to the skin and gastrointestinal tract, and may lead to fatal reactions such as anaphylaxis. Accordingly, the present invention can overcome conventional side effects associated with contrast agent concentration by acquiring CTA projection images based on a contrast agent concentration of 90% or less of the reference concentration and generating three-dimensional rendering images based thereon. At this time, the reference concentration of the contrast agent may vary depending on various conditions such as individual characteristics, the composition of the contrast agent, and the manufacturer of the contrast agent. The contrast agent concentration of the present invention may be 90% or less of the reference concentration typically used by medical professionals, and preferably 50% or less.

[0131] Next, in the selection step (S220), the first model is a model trained to generate a first extracted image of a blood vessel using a projection image of CTA as input, and may exist according to contrast agent concentration. For example, the first model may include at least one of a 0% first model, a 10% first model, a 20% first model, a 30% first model, a 40% first model, a 50% first model, a 60% first model, a 70% first model, an 80% first model, a 90% first model, and a 100% first model with respect to a reference contrast agent concentration, but is not limited thereto. Accordingly, the present invention allows a specific first model corresponding to the received CTA contrast agent concentration data to be selected, and since the specific first model is trained to be specialized for a specific concentration, it may have higher reliability and accuracy than a model trained based on various contrast agent concentration data.

[0132] Furthermore, since the first model is a machine learning method, learning based on various prior data may be required. Accordingly, the present invention may include a learning step prior to the selection step (S220) for learning the first model.

[0133] More specifically, the first model may be trained based on a simulation image of blood vessel data and a data set containing the same. First, the present invention may further include the steps of: generating a second extracted image of a blood vessel from a CTA projection image received prior to the selection step (S220); obtaining blood vessel thickness from the second extracted image; converting the image intensity of the second extracted image based on blood vessel thickness to generate a third extracted image of a contrast agent concentration from the second extracted image; applying the third extracted image of a contrast agent concentration to a computer simulation of a CTA system to generate at least one simulation image of a contrast agent concentration; matching the received CTA projection image to each of the simulation images of a contrast agent concentration to generate a first data set of a contrast agent concentration; and training each of the first models of a contrast agent concentration based on the first data set of a contrast agent concentration. At this time, the extracted blood vessel data, i.e., the second extracted image, may refer to a three-dimensional image, but is not limited thereto, and may include all blood vessel-related data extracted from the CTA projection image.

[0134] The step of generating a second extracted image may include the step of removing bone data within the received CTA projection image and then applying a threshold value. More specifically, if the received CTA projection image is an image generated from dual-energy CT, information regarding bone that is similar in image intensity to the contrast agent may not be completely removed, and due to beam hardening, information regarding procedures on high-density materials such as metal may be lost. Accordingly, the present invention may include the step of removing bone and extracting information regarding procedures on high-density materials such as metal by applying a threshold value, as described above. Accordingly, by including the step of removing bone data within the received CTA projection image and then applying a threshold value as described above, the present invention allows for the use of various CTA projection images without restrictions on the CT scanning method.

[0135] The step of acquiring blood vessel thickness may include the step of measuring blood vessel thickness within the second extracted image from one or more lateral directions, and the step of calculating an average value for the one or more measured blood vessel thicknesses. In this case, the measurement of blood vessel thickness may involve measuring various blood vessel regions rather than a single measurement of a specific area. More specifically, the blood vessel data within the second extracted image may be three-dimensional data, and the thickness of the blood vessel data within the second extracted image may be measured. First, the blood vessel thickness may be measured based on the lateral side of the transverse plane, and the measurement of blood vessel thickness may be performed from one or more lateral directions; the one or more measured blood vessel thicknesses may be calculated as an average value, and the calculated average value may be used as the final blood vessel thickness. In this case, the blood vessel thickness may be calculated separately for each blood vessel region, and accordingly, there may be one or more blood vessel thicknesses.

[0136] The step of generating a third extracted image may include selecting at least one of a predetermined transformation formula for each blood vessel thickness based on blood vessel thickness, and applying the selected at least one transformation formula to the second extracted image to generate a third extracted image for each contrast agent concentration from the second extracted image.

[0137] Furthermore, the selection of the transformation formula may be based on the measured blood vessel thickness, and a transformation formula corresponding to that thickness may be selected. Since the transformation formula includes the correlation (ratio) of image intensity according to contrast agent concentration by thickness, when the transformation formula is applied to the second extracted image, images according to contrast agent concentration with transformed image intensity may be generated.

[0138] Meanwhile, the conversion formula is a predetermined and fixed formula for each thickness, which can be derived through a calibration phantom. That is, the predetermined conversion formula for each blood vessel thickness of the present invention is a conversion formula calculated based on CTA projection images of the calibration phantom. In this case, the calibration phantom of the present invention is a tool capable of representing blood vessels of various thicknesses within the body and is composed of acrylic and aluminum; acrylic is an equivalent material capable of representing soft tissue, and aluminum may be an equivalent material capable of representing hard tissue, i.e., bone, but is not limited thereto.

[0139] In the step of generating the third extracted image, the third extracted image may be an image in which the CTA projection image received is aligned with each of the third extracted images according to concentration.

[0140] The step of generating a simulation image may further include the step of generating the simulation image as two-dimensional data, the step of converting the two-dimensional data into a binary image, and the step of generating an extraction map from the binary image. In this case, computational simulation refers to a simulation that implements an environment identical to an actual medical imaging system (illuminator, detector, irradiation conditions, etc.), and a virtual projection image including two-dimensional or three-dimensional data can be generated based on the extracted vascular data. Furthermore, in order to distinguish targets (vasculars) more clearly within the virtual projection image, the virtual vascular projection image can be converted into a binary image, that is, an image composed of two values ​​such as black and white or 0 and 1. Accordingly, the present invention can distinguish targets more clearly, extract them, and generate an extraction map.

[0141] Ultimately, since the present invention includes steps for learning the first model described above prior to the selection step, a large number of training data can be easily obtained, and based on this, the first model can be trained to have high accuracy and reliability.

[0142] Next, in the step of generating a first extracted image (S230), a specific first model corresponding to the received CTA contrast agent concentration data is selected, and a first extracted image can be generated from the CTA projection image. At this time, the first extracted image may be a projection image in which the image intensity of the vascular region is enhanced (strengthened) from the CTA projection image for a low concentration of contrast agent. That is, since the first model of the present invention includes a plurality of models learned based on various contrast agent concentrations, the image intensity within the image can be enhanced without restriction on the contrast agent concentration of the received CTA projection image, thereby generating a projection image of the vascular region.

[0143] Next, in the step of generating a rendering image (S240), the rendering image may refer to a three-dimensional graphic image with a sense of depth generated by applying shadows, colors, density, etc., by considering data such as the shape, position, and lighting of the two-dimensional data (image).

[0144] The step (S240) of generating such a rendering image may be based on at least one of Filtered Back-Projection (FBP), Algebraic Reconstruction Technique (ART), Maximum Likelihood Expectation Maximization (ML-EM), and Automap, but is not limited thereto, and may include all various methods capable of converting a 2D image into a 3D image.

[0145] However, the step (S240) of generating a rendering image of the present invention may preferably be a step based on a machine learning method.

[0146] Accordingly, the step of generating a rendering image (S240) may include the step of generating a rendering image of a blood vessel by inputting the first extracted image to the second learning model. At this time, the second learning model may be a model trained to generate a rendering image by inputting the first extracted image, but is not limited thereto.

[0147] In addition, the step of generating a rendered image (S240) may further include a step of correcting such loss and errors in data that may occur during the rendering image conversion. That is, the step of generating a rendered image (S240) may further include a step of correcting the quality of the rendered image, wherein the correction may include at least one of restoration, enhancement, registration, and linear correction, but is not limited thereto.

[0148] Through the above process, the method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention can generate high-quality vascular rendering images based on CTA images of low-concentration contrast agents. That is, the present invention can minimize the side effects acting on an individual by conventional contrast agents.

[0149] In addition, since the present invention is based on a calibration phantom capable of representing internal blood vessels of various thicknesses, it can provide image intensity standards for blood vessels in various medical images, and accordingly, can provide uniform medical image results.

[0150] Furthermore, the present invention enables the easy acquisition of data in various cases through computational simulation of medical imaging and clinical data, which are difficult to obtain due to numerous regulations and restrictions such as personal information protection. Accordingly, the present invention allows for the training of a machine learning model using the numerous data acquired through the aforementioned computational simulation, and thus, the machine learning model of the present invention can derive results with high reliability and accuracy.

[0151] Meanwhile, a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention can provide not only a rendering image but also predict and provide the vascular disease of an individual based thereon.

[0152] Accordingly, a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention may further include the step of receiving clinical data of an individual, and may further include the step of predicting the probability of vascular disease of the individual based on the clinical data and a rendered image. In this case, the clinical data may refer to data including various test results, prescription data, treatment records, and personal information regarding the individual stored within a medical institution.

[0153] More specifically, the step of predicting the probability may include a step of extracting features within a rendered image and a step of predicting the probability of vascular disease in an individual using clinical data and features as input to a third model, wherein the third model may be a model trained to predict the probability of vascular disease in an individual using clinical data and features as input, but is not limited thereto.

[0154] Ultimately, the present invention generates and provides high-quality rendering images, and based thereon, can provide predictive information regarding vascular diseases in an object with high accuracy and reliability.

[0155] FIG. 3 is a schematic diagram of a method for providing information on angiography for contrast agent reduction according to an embodiment of the present invention. For convenience of explanation, the description will be made with reference to FIG. 4 to 6.

[0156] Referring to FIG. 3, the method for providing information on angiography for contrast agent reduction according to the present invention inputs a CTA projection image into a first model to generate a first extracted image, and can generate a rendering image based on the generated first extracted image.

[0157] More specifically, the CTA projection image can be input into one of the first models (10a, 10b, 10c, 10d) based on CTA contrast agent concentration data. In this case, the CTA projection image may refer to a medical image derived from a medical imaging device such as a CT, and various images capable of extracting blood vessels, not just CT, may be used. Such a CTA projection image may be received and utilized from a medical institution or a medical imaging device such as a CT. Furthermore, the CTA contrast agent concentration data may also be received in the same manner as the CTA projection image. Moreover, the CTA projection image received in the method for providing information on angiography for contrast agent reduction according to an embodiment of the present invention (hereinafter, the present invention) may be a single image taken after the contrast agent is injected, but is not limited thereto. In addition, the projection image received in the present invention may be an image taken after injecting a low concentration of contrast agent, preferably at a concentration of 90% or less of the standard contrast agent concentration, but is not limited thereto. Furthermore, the CTA projection image received in the present invention may utilize a conventional medical image recorded by taking two or more images, but preferably may be an image taken once after the contrast agent is injected.

[0158] The first model may include a first model (10a, 10b, 10c, 10d) for each contrast agent concentration, and is a model trained to generate a first extracted image by taking a projected image of CTA as input.

[0159] In this regard, referring to FIG. 4, a schematic diagram of the learning process of a first model in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention is shown.

[0160] The first model of the present invention can be trained based on training data generated from CTA projection images, i.e., the first data set (11a, 11b).

[0161] First, a second extracted image of a blood vessel can be generated from a CTA projection image. At this time, the CTA projection image used for generating training data is an image generated by CTA imaging techniques such as the direct subtraction method and the dual-energy method. It may be identical to the CTA projection image input to the first model in FIG. 3 described above, but is not limited thereto, and may be retrospective data stored in a medical institution or medical imaging device. Furthermore, the CTA projection image may include not only the image but also related data (data regarding blood vessel and contrast agent concentration). Furthermore, the CTA system can generate two-dimensional and three-dimensional images, and accordingly, the retrospective CTA projection image stored in a medical institution or medical imaging device may include both two-dimensional and three-dimensional images. Ultimately, the CTA projection images used for generating training data may include not only images of a specific object but also images of one or more diverse objects stored within a medical institution or medical imaging device, but are not limited thereto. Accordingly, the received CTA projection images used for generating training data may include one or more CTA projection images, and one or more virtual projection images may be generated based thereon.

[0162] Furthermore, a method for generating a second extracted image, that is, a method for extracting vascular data within a CTA projection image, may be based on manual methods (hand worked), subtraction, thresholding, and machine learning (deep learning) methods. For example, if the CTA projection image is a set of images with or without a contrast agent, vascular data can be extracted through subtraction based on the presence or absence. Furthermore, if the CTA projection image is based on a dual-energy method, bone data within the CTA projection image can be removed, and then a thresholding factor suitable for vascular data can be applied to extract the data. Furthermore, the extracted vascular and procedure data may be three-dimensional image data.

[0163] Next, the blood vessel thickness can be obtained from the generated second extracted image. More specifically, the blood vessel data within the second extracted image may be three-dimensional data, and the thickness of the blood vessel data within the second extracted image may be measured. First, the blood vessel thickness may be measured based on the side of the transverse plane, and the blood vessel thickness may be measured in one or more lateral directions. The one or more measured blood vessel thicknesses are calculated as an average value, and the calculated average value may be used as the final blood vessel thickness. At this time, the blood vessel thickness may be calculated separately for each blood vessel region, and thus, there may be one or more blood vessel thicknesses.

[0164] Next, based on the calculated final blood vessel thickness, a transformation formula corresponding to that thickness is selected, and the selected transformation formula can be applied to the second extracted image. Accordingly, the image intensity of the blood vessel data in the second extracted image is transformed, and a third extracted image can be generated from the second extracted image. At this time, since the transformation formula includes the correlation (ratio) of image intensity according to contrast agent concentration by thickness, when the transformation formula is applied to the second extracted image, images according to contrast agent concentration with transformed image intensity can be generated.

[0165] Meanwhile, this conversion formula is a predetermined and fixed conversion formula for each thickness, which can be derived through a calibration phantom.

[0166] More specifically, with reference to FIG. 5, a schematic diagram of a correction phantom and a conversion formula used in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention is shown.

[0167] The conversion formula used in the present invention can be derived through a calibration phantom. A calibration phantom is a tool for evaluating and calibrating the performance of medical imaging devices such as CT and MRI, and it incorporates equipment that represents the human body to establish a standard for measurement. The calibration phantom of the present invention is a tool capable of representing blood vessels of various thicknesses within the body and is composed of acrylic and aluminum; acrylic is an equivalent material capable of representing soft tissue, and aluminum may be an equivalent material capable of representing hard tissue, i.e., bone, but is not limited thereto.

[0168] The correction phantom of the present invention includes a cylindrical groove having a diameter of 1 to 100 mm within a frame made of acrylic, and a contrast agent can be injected into the groove. At this time, the grooves may be arranged in a checkerboard pattern, but are not limited thereto. For example, the grooves of the correction phantom of the present invention may be freely arranged within the acrylic frame in various forms such as 3 x 3, 4 x 4, MXN, etc., depending on the convenience of the user.

[0169] Furthermore, the calibration phantom of the present invention includes a structure made of aluminum spaced apart from the groove. As described above, the structure made of aluminum can represent bone, which is a hard tissue within the body, and can be formed at a certain distance from the groove so as not to interfere with the groove containing the contrast agent. Although the structure made of aluminum is depicted in the form of a small rectangular column to distinguish it from the groove, it is not limited thereto and can be configured in various forms, such as the same shape as the groove. That is, since the calibration phantom of the present invention includes a groove representing blood vessels of various sizes and a structure representing bone, the relative image intensity values ​​in the blood vessels and bone can be compared by injecting a contrast agent into the groove and imaging it with an actual CT. At this time, the injected contrast agent can be injected at various concentrations, and accordingly, image intensity data according to various blood vessel thicknesses and contrast agent concentrations can be obtained from the calibration phantom of the present invention.

[0170] CT scanning of the calibration phantom is performed under the same conditions and environment as CTA scanning of an actual patient, thereby enabling the acquisition of a CTA projection image of the calibration phantom. The acquired CTA projection image of the calibration phantom is subjected to the same method (algorithm and parameters) as the second extraction image generated from the object's CTA projection image in Fig. 4, thereby generating an extraction image of the groove and aluminum structure within the calibration phantom (hereinafter referred to as the calibration phantom extraction image) from the CTA projection image of the calibration phantom. Accordingly, image intensity data according to contrast agent concentration for each thickness can be obtained from the generated calibration phantom extraction image, and an equation is derived from the correlation function graph between contrast agent concentration and image intensity, which can be used as a transformation formula for each blood vessel thickness.

[0171] Ultimately, the present invention can convert the image intensity of a contrast agent from a projected image of a blood vessel having various thicknesses by including a conversion formula for the correction phantom of FIG. 5 described above.

[0172] Referring again to FIG. 4, the transformation formula described in FIG. 5 above is applied to the second extracted image, thereby generating a third extracted image according to contrast agent concentration from the second extracted image. At this time, the third extracted image may be a three-dimensional image of a vascular region like the second extracted image, but is not limited thereto, and may be a data set generated by aligning the transformed data with an existing CTA projection image. Accordingly, the third extracted image may include an image that is identical in structure and format to the CTA projection image, but differs only in the contrast agent concentration within the CTA projection image. That is, the third extracted image of the present invention may be a data set including a CTA projection image, a CTA projection image according to contrast agent concentration that differs only in the contrast agent concentration, and a three-dimensional vascular extracted image.

[0173] Next, the third extracted images for each contrast agent concentration can be applied to computer simulation of the CTA system to generate (acquire) various simulation images, namely digital reconstructed radiography (DRR). More specifically, the third extracted images for each contrast agent concentration can be applied to computer simulation modeling of the CTA system to generate at least one simulation image (virtual vascular projection image) for each contrast agent concentration. At this time, computer simulation is a method of predicting the behavioral state of an actual system using a model, and can be modeled by computer simulation software. For example, the computer simulation model of the present invention is software to which radiation physics is applied, and may be based on mathematical probabilistic algorithms such as the Monte Carlo Method, and may be a simulator such as GEANT4, EGS4, MCNP, and FLUKA, but is not limited thereto.

[0174] Next, a CTA projection image can be matched to each of the generated simulation images for each contrast agent concentration to create a first dataset (11a, 11b) for each contrast agent concentration, and based on this, a first model (10a, 10e) for each contrast agent concentration can be trained. For example, a first model (10a) based on 100% contrast agent concentration may be a model trained based on a simulation image having an image intensity of 100% contrast agent concentration and its first dataset (11a), and a first model (10e) based on N% contrast agent concentration may be a model trained based on a simulation image having an image intensity of N% contrast agent concentration and its first dataset (11b).

[0175] At this time, the present invention may further include a masking step to improve the accuracy of the target object, i.e., vascular data, during model training. More specifically, the simulation images generated by computational modeling for different contrast agent concentrations may be three-dimensional data. Accordingly, the present invention may further include a step of regenerating (converting) the three-dimensional simulation images into two-dimensional data, converting the generated two-dimensional data into a binary image, and generating an extraction map from the converted binary image. That is, the present invention may mask data excluding the target vascular data so as not to interfere with the model's training. Ultimately, by including the masking step, the present invention enables the model to learn the target vascular data more clearly. Meanwhile, as described above, the present invention may further include a masking step. Such a masking step may not be included depending on whether it is necessary. However, if included, the simulation images for each contrast agent concentration are generated as two-dimensional extraction maps, so the first model can be trained based on the received CTA projection images (original data) and the dataset based on each extraction map.

[0176] According to the above process, the first model of the present invention is trained based on simulation images of vascular data according to various contrast agent concentrations, and based on this, projection images of blood vessels according to contrast agent concentrations can be generated from CTA projection images. That is, the first model may be a model trained to generate a first extracted image of a blood vessel using a CTA projection image as input, and the first model may exist for each contrast agent concentration.

[0177] Furthermore, the present invention is not limited by personal information and other regulations (constraints), and can build a number of training databases from a minimum amount of data (CTA projection images) and train a first model based thereon.

[0178] Referring again to FIG. 3, the present invention allows one of the first models (10a, 10b, 10c, 10d) to be selected based on received CTA contrast agent concentration data, and the received CTA projection image to be input into the selected specific first model to generate a first extracted image. More specifically, as described above in FIG. 5, the first models may exist according to the contrast agent concentration, for example, the first models may include at least one of the following with respect to the reference contrast agent concentration: a 0% first model (10d), a 10% first model, a 20% first model, a 30% first model, a 40% first model, a 50% first model, a 60% first model, a 70% first model, an 80% first model (10c), a 90% first model (10b), and a 100% first model (10a). Accordingly, a first model corresponding to the received CTA contrast agent concentration is selected, and a first extracted image can be generated from the CTA projection image. At this time, the first extracted image is a projection image of a blood vessel, and may be an image in which the image intensity of the blood vessel region is enhanced (strengthened) from the CTA projection image based on a low concentration of contrast agent.

[0179] Next, a rendering image can be generated from the first extracted image. At this time, the rendering image may be generated by inputting the first extracted image to a second model trained to generate a rendering image using the first extracted image as input. At this time, at least one of restoration, enhancement, registration, and linear attenuation correction may be applied to the first extracted image as a post-processing step to improve image quality, and after the application thereof, the process of generating the rendering image may be performed. Accordingly, the present invention can generate a more distinct and clear three-dimensional image of blood vessels and procedure data. In addition, since the rendering image is based on the first extracted image, which is a projection image of the blood vessel, it may be a rendering image of the blood vessel (blood vessel region).

[0180] Furthermore, the present invention may generate a three-dimensional rendering image based on a first extracted image, and the generation, i.e., the reconstruction method, may be based on at least one of an analytic method-based filtered back-projection (FBP), an iterative method-based algebraic reconstruction technique (ART), a statistical method-based maximum likelihood expectation maximization (ML-EM) technique, and a machine learning method-based automap, but is not limited thereto, and most preferably, may be a method based on a machine learning method (second model) as described above.

[0181] Ultimately, through the aforementioned process, the present invention can generate a three-dimensional rendering image from a CTA projection image of a low concentration of contrast agent, and the generated three-dimensional rendering image is generated based on a machine learning method learned from a concentration-specific simulation image generated through computational simulation, thereby providing a more accurate and reliable three-dimensional image without restrictions on the concentration of the contrast agent.

[0182] Meanwhile, a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention can not only generate images but also predict and provide the probability of prevalence of vascular disease in an individual based thereon.

[0183] More specifically, with reference to FIG. 6, a schematic diagram of vascular disease prediction in a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention is shown.

[0184] The vascular disease prediction of the present invention may be based on a three-dimensional rendering image (final rendering image) and clinical data. In this case, the clinical data may be received from a medical institution and its server along with the CTA projection image, but is not limited thereto.

[0185] In the vascular disease prediction of the present invention, features of a rendered image are extracted and can be used to predict vascular disease in an individual. More specifically, the features within the rendered image may include various image processing conditions and may include image features such as enhancement, histogram, and texture, but are not limited thereto, and may include all various element features used in image processing. For example, the features within the rendering image used for predicting vascular disease according to the present invention may include, but are not limited to, an image model (pixel value representation method), sampling, quantization, storage amount of digital image signals, storage format of digital image signals (bmp, pcx, tiff, gif, jpg, png), pixel value conversion, spatial domain conversion (filtering), mapping, affine transformation, interpolation, nonlinear warping, frequency transformation, point processing, spatial processing, frequency transform processing, morphology, image segmentation, edge representation, region representation, and image compression.

[0186] These features within the image can be extracted based on the aforementioned parameters, but are not limited thereto, and various methods such as first-order and second-order based probabilistic methods and transform-based feature extraction methods may be applied.

[0187] Furthermore, in order to extract meaningful features among the features extracted by the aforementioned method, the dimension can be reduced, and LASSO, ridge, elastic net, machine learning, and manual removal methods may be applied, but are not limited thereto; however, most preferably, a machine learning method (third model) may be used.

[0188] In the case of the machine learning model (the third model), the input can be input in a multi-channel form. Additionally, the features input to the machine learning model may undergo image augmentation preprocessing such as linear transformation, distortion methods, and domain transformation, but are not limited thereto. Furthermore, normalization may be applied to the input signals, which can reduce the problem of local minima and improve the learning speed of the machine learning model.

[0189] Furthermore, by adding a layer such as CAM (class activation mapping) to the training model, results can be derived, making it possible to explain the derivation of results. That is, the third model of the present invention may be an Explainable Artificial Intelligence (XAI), but is not limited thereto.

[0190] Ultimately, a method for providing information on angiography for contrast agent reduction according to one embodiment of the present invention may further receive clinical data of an individual and predict and provide the probability of onset or prevalence of vascular disease in the individual based on the generated rendering image and the further received clinical data. Furthermore, features extracted from the generated rendering image may be utilized. For example, if a machine learning model is used to predict the probability of vascular disease, features within the rendering image may be extracted, and the probability of vascular disease in the individual may be predicted by inputting the clinical data and features to a third model. At this time, the third model may be a model trained to predict the probability of vascular disease in the individual using the clinical data and extracted features as input.

[0191] Accordingly, since the invention can provide not only rendering images of blood vessels but also information on related vascular diseases, it can assist medical staff and improve diagnostic accuracy.

[0192] Although embodiments of the present invention have been described in more detail with reference to the attached drawings, the present invention is not necessarily limited to these embodiments and may be modified in various ways within the scope of the technical spirit of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical spirit of the present invention, and the scope of the technical spirit of the present invention is not limited by these embodiments. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of protection of the present invention shall be interpreted by the claims below, and all technical spirits within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0194] 100: Device for providing information on angiography for contrast agent reduction 110: Communication interface (communication section) 111: Wired communication port 112: Wireless circuit 113: I / O Interface 114: Processor 120: Memory 121: Operating System 122: Communication Module 123: User Interface Module 124: Application 200: Medical staff device 210: Memory Interface and User Interface Screen 220: Processor and Rendering Images 230: Peripheral Interface and Predictive Data 240: I / O Subsystem 241: Touchscreen Controller 242: Other Input Controllers 243: Touch screen 244: Other Input Control Devices 250: Memory 251: Operating System 252: Communication Module 253: GUI Module 254: Sensor processing module 255: Phone module 256: Applications 256-1, 256-2: Application 257: Digital Assistant Client Module 258: User data 260: Motion sensor 261: Light sensor 262: Proximity sensor 263: Other sensors 270: Camera subsystem 271: Optical sensor 280: Communications subsystem 290: Audio Subsystem 291: Speaker 292: Microphone 300: Medical imaging device 310: CTA Projection Image 320: CTA Contrast Agent Concentration Data 350: Interface screen 1000: Information provision system for angiography

Claims

Claim 1 A method for providing information on angiography for contrast agent reduction performed by a processor, comprising: receiving a computed tomography angiography (CTA) projection image and CTA contrast agent concentration data of an individual; selecting one of a first model for each contrast agent concentration trained to generate a first extracted image of a blood vessel using the projection image of CTA as input based on the contrast agent concentration data; generating a first extracted image using the received projection image as input to the selected first model; and generating a rendering image from the first extracted image, wherein prior to the selecting step, a step of generating a second extracted image of a blood vessel from the received CTA projection image; a step of obtaining a blood vessel thickness from the second extracted image; a step of converting the image intensity of the second extracted image based on the blood vessel thickness to generate a third extracted image for each contrast agent concentration from the second extracted image; and the third extracted image for each contrast agent concentration to a CTA system A method for providing information on angiography for contrast agent reduction, further comprising: a step of generating at least one simulation image for each contrast agent concentration by applying it to a computer simulation; a step of generating a first data set for each contrast agent concentration by matching the received CTA projection image to each of the simulation images for each contrast agent concentration; and a step of learning each of the first models for each contrast agent concentration based on the first data set for each contrast agent concentration. Claim 2 A method for providing information on angiography for contrast agent reduction, comprising, in addition to the step of receiving clinical data of the subject in claim 1. Claim 3 A method for providing information on angiography for contrast agent reduction, further comprising the step of predicting the probability of vascular disease of the subject based on the clinical data and the rendering image in claim 2. Claim 4 A method for providing information on angiography for contrast agent reduction, wherein the step of predicting the probability comprises: a step of extracting features within the rendering image; and a step of predicting the probability of a vascular disease for the subject using the clinical data and the features as inputs to a third model trained to predict the probability of a vascular disease for the subject using the clinical data and the features as inputs. Claim 5 A method for providing information on angiography for contrast agent reduction, wherein the CTA projection image is a single projection image taken after the contrast agent is injected. Claim 6 A method for providing information on angiography for contrast agent reduction, wherein the contrast agent concentration is 90% or less of the standard contrast agent concentration. Claim 7 A method for providing information on angiography for contrast agent reduction, wherein the first model according to the contrast agent concentration is at least one of the 0% first model, 10% first model, 20% first model, 30% first model, 40% first model, 50% first model, 60% first model, 70% first model, 80% first model, 90% first model, and 100% first model relative to the reference contrast agent concentration. Claim 8 delete Claim 9 A method for providing information on angiography for contrast agent reduction, wherein the step of generating the second extracted image comprises removing bone data within the received CTA projection image and then applying a threshold value. Claim 10 A method for providing information on angiography for contrast agent reduction, wherein the step of acquiring the blood vessel thickness in the second extracted image comprises the step of measuring the blood vessel thickness in one or more lateral directions, and the step of calculating an average value for one or more of the measured blood vessel thicknesses. Claim 11 A method for providing information on angiography for contrast agent reduction according to claim 1, wherein the step of generating the third extracted image comprises: selecting at least one of a predetermined conversion formula for each blood vessel thickness based on the blood vessel thickness; and applying at least one of the selected conversion formulas to the second extracted image to generate a third extracted image for each contrast agent concentration from the second extracted image. Claim 12 In claim 11, the above-determined conversion formula for each blood vessel thickness is a method for providing information on angiography for contrast agent reduction, calculated based on a CTA projection image of a calibration phantom. Claim 13 A method for providing information on angiography for contrast agent reduction, wherein, in claim 1, the third extracted image is an image in which the CTA projection image received on each of the third extracted images according to concentration is aligned. Claim 14 A method for providing information on angiography for contrast agent reduction, wherein the step of generating the rendering image is based on at least one of Filtered Back-Projection (FBP), Algebraic Reconstruction Technique (ART), Maximum Likelihood Expectation Maximization (ML-EM), and Automap. Claim 15 A method for providing information on angiography for contrast agent reduction, wherein the step of generating the rendering image comprises the step of generating a rendering image of the blood vessel using the first extracted image as input to a second model trained to generate a rendering image using the first extracted image as input. Claim 16 A method for providing information on angiography for contrast agent reduction, wherein the step of generating the rendering image in claim 1 further includes the step of correcting the quality of the rendering image. Claim 17 In claim 16, the above correction comprises at least one of restoration, enhancement, registration, and linear correction, a method for providing information on angiography for contrast agent reduction. Claim 18 The apparatus comprises a communication unit configured to receive computed tomography angiography (CTA) projection images and CTA contrast agent concentration data of an object, and a processor connected to communicate with the communication unit, wherein the processor is configured to select one of a first model for each contrast agent concentration that is trained to generate a first extracted image of a blood vessel using a projection image of CTA as input based on the contrast agent concentration data, generate a first extracted image using the received projection image as input to the selected first model, and generate a rendering image from the first extracted image; prior to the selection of one of the first models for each contrast agent concentration, generate a second extracted image of a blood vessel from the received CTA projection image, obtain blood vessel thickness from the second extracted image, convert the image intensity of the second extracted image based on the blood vessel thickness to generate a third extracted image for each contrast agent concentration from the second extracted image, and the third extracted image for each contrast agent concentration A device for providing information on angiography for contrast agent reduction, further configured to apply to a computer simulation of a CTA system to generate at least one simulation image for each contrast agent concentration, match the received CTA projection image to each of the simulation images for each contrast agent concentration to generate a first data set for each contrast agent concentration, and learn each of the first models for each contrast agent concentration based on the first data set for each contrast agent concentration. Claim 19 In claim 18, the communication unit is further configured to receive clinical data of the object, a device for providing information on angiography for contrast agent reduction. Claim 20 In claim 19, the processor is further configured to predict the probability of vascular disease of the subject based on the clinical data and the rendering image, a device for providing information on angiography for contrast agent reduction. Claim 21 A device for providing information on angiography for contrast agent reduction, wherein the processor extracts features within the rendering image and is configured to predict the probability of vascular disease of the subject using the clinical data and the features as input to a third model trained to predict the probability of vascular disease of the subject using the clinical data and the features as input. Claim 22 In claim 18, the above CTA projection image is a single projection image taken after the contrast agent is injected, a device for providing information on angiography for contrast agent reduction. Claim 23 In claim 18, the device for providing information on angiography for contrast agent reduction, wherein the contrast agent concentration is 90% or less of the reference contrast agent concentration. Claim 24 In claim 18, the first model according to the contrast agent concentration is at least one of the first model with a concentration of 0%, the first model with a concentration of 10%, the first model with a concentration of 20%, the first model with a concentration of 30%, the first model with a concentration of 40%, the first model with a concentration of 50%, the first model with a concentration of 60%, the first model with a concentration of 70%, the first model with a concentration of 80%, the first model with a concentration of 90%, and the first model with a concentration of 100%, with respect to the reference contrast agent concentration, a device for providing information on angiography for contrast agent reduction. Claim 25 delete Claim 26 In claim 18, the processor is configured to remove data regarding bone within the received CTA projection image and then apply a threshold value, for a device providing information on angiography for contrast agent reduction. Claim 27 In claim 18, the device for providing information on angiography for contrast agent reduction is configured such that the processor measures the blood vessel thickness in the second extracted image in one or more lateral directions and calculates an average value for one or more of the measured blood vessel thicknesses. Claim 28 A device for providing information on angiography for contrast agent reduction, wherein, in claim 18, the processor is configured to select at least one of a predetermined conversion formula for each blood vessel thickness based on the blood vessel thickness, and to apply at least one of the selected conversion formulas to the second extracted image to generate a third extracted image for each contrast agent concentration from the second extracted image. Claim 29 In claim 28, the above-determined conversion formula for each blood vessel thickness is a device for providing information on angiography for contrast agent reduction, calculated based on a CTA projection image of a calibration phantom. Claim 30 In claim 18, the device for providing information on angiography for contrast agent reduction, wherein the third extracted image is an image in which the CTA projection image received on each of the third extracted images according to concentration is aligned. Claim 31 In claim 18, the processor is a device for providing information on angiography for contrast agent reduction, based on at least one of filtered back-projection (FBP), algebraic reconstruction technique (ART), maximum likelihood expectation maximization (ML-EM), and automap. Claim 32 In claim 18, the processor is configured to generate a rendering image of the blood vessel using the first extracted image as input to a second model trained to generate a rendering image using the first extracted image as input, and is a device for providing information on angiography for contrast agent reduction. Claim 33 In claim 18, the processor is further configured to correct the quality of the rendering image, a device for providing information on angiography for contrast agent reduction. Claim 34 A device for providing information for angiography for contrast agent reduction, wherein the correction comprises at least one of restoration, enhancement, registration, and linear correction.

Images