Medical fluoroscopy equipment quality control system

Abstract

The medical fluoroscopy equipment quality control system according to the present invention comprises: an inspection unit for performing an inspection according to a quality control item by instructing the medical fluoroscopy equipment to operate; an analysis unit in which artificial intelligence analyzes the inspection result from the inspection unit to generate a report, and compares the inspection result with a reference value determined for each quality control item to determine whether the inspection result is acceptable; a notification unit for generating a problem occurrence notification, an inspection cycle notification, and a report generation notification; a learning unit in which artificial intelligence learns a correction value if the user inputs the correction value after discovering an error in the analysis result from the analysis unit; and a storage unit for storing activity details of the inspection unit, the analysis unit, and the notification unit. The present invention is advantageous in that fluoroscopy equipment quality control is performed using artificial intelligence, thereby enabling more accurate quality control.

Description

Quality control system for medical fluoroscopy equipment

[0001] The present invention relates to a system for performing quality control of medical fluoroscopy equipment using artificial intelligence.

[0002]

[0003] As the use of specialized medical equipment, such as CT and MRI, using radiation is increasing in the medical field, the radiation exposure of medical professionals, who are frequently exposed to radiation, is also increasing. Consequently, legal regulations for quality control of specialized medical equipment are being introduced to manage the risks associated with radiation exposure for patients and medical professionals.

[0004] However, for fluoroscopy equipment such as C-arms, despite being radiation-using devices, they are not classified as special medical equipment, and therefore have relatively weak legal obligations for quality control. While quality control items are established for C-arms based on the examination cycle, these are often performed by individuals unfamiliar with the equipment. Furthermore, for abstract items such as whether the image is clearly identifiable, it is difficult for examiners to make accurate judgments.

[0005] (Patent Document 0001) Korean Patent No. 10-2559943 (October 24, 2017, System and Method for Image Quality Analysis and Recommended Shooting Conditions Based on Artificial Intelligence)

[0006]

[0007] The present invention aims to develop a system that performs quality control of the fluoroscopy equipment by introducing artificial intelligence for the safe use of the fluoroscopy equipment as described above.

[0008]

[0009] A quality control system for medical fluoroscopy equipment according to an embodiment of the present invention may include: an inspection unit that directs the operation of the medical fluoroscopy equipment to perform an inspection according to quality control items; an analysis unit that analyzes the inspection results of the inspection unit using artificial intelligence to generate a report and compares the inspection results with a standard value set for each quality control item to determine whether the result has passed; a notification unit that generates a problem occurrence notification, an inspection cycle notification, and a report generation notification; a learning unit that inputs a correction value when a user finds an error in the analysis results of the analysis unit and the artificial intelligence learns the correction value; and a storage unit that stores the activity history of the inspection unit, the analysis unit, and the notification unit.

[0010] In addition, the analysis unit may include a report writing unit that writes the inspection results and analysis results inspected according to a set inspection cycle in the form of a report.

[0011] In addition, the above inspection cycle may be one or more of one day, one week, three months, six months, one year, and three years, and the quality control items may be different according to the above inspection cycle.

[0012] In addition, the inspection unit may include a sample position adjustment unit that adjusts the position of a sample for performing an inspection according to the quality control items.

[0013] In addition, the above analysis unit can analyze one or more of the position, formation angle, area, and shade of the sample by forming a plurality of square separation areas of the same size by arranging a plurality of horizontal and vertical auxiliary lines at equal intervals on an image obtained by irradiating radiation.

[0014] In addition, the analysis unit can assign grades to the test results for each of the quality control items, calculate a sum by differentiating scores according to the grades, and then compare the calculated sum with a standard value to determine pass or fail.

[0015] In addition, the inspection unit includes phantom image evaluation as the quality control item, and the analysis unit, in the case of the phantom image evaluation, uses artificial intelligence to analyze the shape using the difference in brightness and the difference in contrast of a square separation area formed on the image to confirm the number of holes and the number of mesh patterns, and the learning unit, in the case of the phantom image evaluation, when the number of holes or the number of mesh patterns determined by the artificial intelligence of the analysis unit is different from the result determined by the user, the user can input a correction value and the artificial intelligence can learn the correction value.

[0016] In addition, in the case of the phantom image evaluation, if the number of holes or the number of mesh patterns determined by the artificial intelligence of the analysis unit according to the image observation distance is different from the result determined by the user, the artificial intelligence can learn the observation according to the distance based on the correction value when the user inputs a correction value.

[0017]

[0018] The quality control system for medical fluoroscopy equipment according to an embodiment of the present invention enables more accurate quality control by performing quality control of the fluoroscopy equipment using artificial intelligence.

[0019]

[0020] Figure 1 is a block diagram of a precision management system for medical fluoroscopy equipment according to an embodiment of the present invention.

[0021] Figure 2 is an example of an analysis diagram of the analysis section.

[0022] Figure 3 is an example of a visual evaluation method for a monitor for reading and shooting.

[0023] Figure 4 is an example of phantom image evaluation.

[0024]

[0025] A quality control system for medical fluoroscopy equipment according to an embodiment of the present invention may include an inspection unit that directs the operation of medical fluoroscopy equipment to perform an inspection according to quality control items; an analysis unit that analyzes the inspection results of the inspection unit using artificial intelligence to generate a report and compares the inspection results with a standard value set for each quality control item to determine whether the result has passed or failed; a notification unit that generates a problem occurrence notification, an inspection cycle notification, and a report generation notification; a learning unit that inputs a correction value when a user finds an error in the analysis results of the analysis unit and the artificial intelligence learns the correction value; and a storage unit that stores the activity history of the inspection unit, the analysis unit, and the notification unit.

[0026]

[0027] The following description of the present invention with reference to the drawings is not limited to specific embodiments, and various modifications and embodiments may be made. Furthermore, the following description should be understood to encompass all modifications, equivalents, and alternatives within the spirit and technical scope of the present invention.

[0028] In the following description, terms such as first, second, etc. are used to describe various components, and are not limited in meaning in themselves, but are used only for the purpose of distinguishing one component from another.

[0029] The same reference numbers used throughout this specification represent the same components.

[0030] As used herein, singular expressions include plural expressions unless the context clearly dictates otherwise. In addition, terms such as "comprise," "include," or "have" used herein should be interpreted to indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, and should be understood to not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0031] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by those of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0032] In addition, when describing with reference to the attached drawings, identical components will be assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted. When describing the present invention, if a detailed description of a related known technology is judged to unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

[0033] Hereinafter, the precision management system of a medical fluoroscopy device according to an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 4.

[0034]

[0035] FIG. 1 is a block diagram of a quality control system for medical fluoroscopy equipment according to an embodiment of the present invention, FIG. 2 is an example diagram of an analysis unit, FIG. 3 is an example diagram of a visual evaluation method for a monitor for reading and shooting, and FIG. 4 is an example diagram of a phantom image evaluation.

[0036] Referring to FIG. 1, the quality control system (1) of a medical fluoroscopy device according to an embodiment of the present invention may include an inspection unit (10), an analysis unit (20), a notification unit (30), a learning unit (40), and a storage unit (50).

[0037] At this time, the medical fluoroscopy equipment may be one of film, CR (Computed Radiography), and DR (Digital Radiography), but in this specification, the quality control of DR, particularly the quality control of C-arm, is described as an example.

[0038] The inspection unit (10) may direct the operation of medical fluoroscopy equipment to perform inspection according to quality control items.

[0039] In detail, it can instruct the operation of the equipment, such as moving the equipment forward and backward, turning the light on / off, irradiating with radiation, and taking images.

[0040] In addition, the inspection unit (10) can adjust the position of the sample to perform inspection according to the quality control items, including the sample position adjustment unit.

[0041] The analysis unit (20) may use artificial intelligence to analyze the inspection results of the inspection unit (10) to generate a report, and compare the inspection results with the standard values ​​set for each quality control item to determine whether or not it has passed.

[0042] Additionally, if it is judged as a failure after determining whether or not it has passed, the cause can be analyzed and a solution can be derived.

[0043] In addition, the analysis unit (20) may include a report writing unit that writes the inspection results and analysis results inspected according to the set inspection cycle in the form of a report.

[0044] Here, the inspection cycle may be one or more of one day, one week, three months, six months, one year, and three years, and the quality control items may be different according to the inspection cycle.

[0045] The notification unit (30) may generate a problem occurrence notification, an inspection cycle notification, and a report generation notification.

[0046] The learning unit (40) may be configured to have the artificial intelligence learn the correction value when the user inputs a correction value if an error is found in the analysis results of the analysis unit (20).

[0047] The storage unit (50) may store the activity history of the inspection unit (10), analysis unit (20), and notification unit (30).

[0048]

[0049] Hereinafter, the functions of the inspection unit (10), analysis unit (20), notification unit (30), learning unit (40), and storage unit (50) according to the quality control items of the fluoroscopy equipment will be described in detail.

[0050] For example, the quality control items according to the set inspection cycle of DR fluoroscopy equipment are as shown in Table 1 below.

[0051] Daily - Fluoroscopic equipment operation and inspection - Examination patient dose record Weekly - Reading room environment inspection 3 months - Clinical image evaluation - Management of monitors for reading and shooting (CRT) 6 months - Phantom image evaluation - Management of monitors for reading and shooting (LCD) 1 year - Irradiation field inspection - Automatic exposure device inspection and reproducibility test - Tube voltage, tube current test - Half-value test - Inspection of examiner and patient protective equipment - Quality management training for employees 3 years (regular inspection) - Grounding equipment confirmation test - External leakage current test - Reproducibility test of irradiation dose - Tube voltage, tube current test - Illuminance test - Difference test between X-ray irradiation field and light irradiation field - Irradiation field coincidence test - Incident irradiation dose rate test - Timer test (only for imaging devices) - Half-value test

[0052] Quality control items with a daily inspection cycle may include operation and inspection of fluoroscopy equipment and recording of patient doses. Operation and inspection of fluoroscopy equipment may include checking the status of the control panel, equipment movement, table detector, artifact occurrence, and other malfunctions.

[0053] In detail, the inspection unit (10) can issue commands to each part of the equipment to attempt movement or video recording to check whether the equipment is moving normally.

[0054] In addition, if a part does not operate according to the operation command of the inspection unit (10) or an artifact occurs in the acquired image, the cause can be analyzed using artificial intelligence and a solution can be derived.

[0055] At this time, if the problem can be solved programmatically, such as correction for system error or equipment error, the problem can be solved internally within the analysis unit (20), and if it is a problem that must be solved by a person, such as replacement of parts, a notification can be generated in the notification unit (30).

[0056] In addition, the operation of the equipment as described above and the occurrence of artifacts, as well as the radiation dose records of the examined patient, can be generated in real time.

[0057] The radiation dose record of the patient being examined is important to maintain a safe level by recording the radiation dose received by the patient. By registering the medical staff in charge with the device before using the device and then using it, the hand of the medical staff in charge confirmed in the image can be recognized by the analysis unit (20) and the cumulative radiation dose received by the medical staff can also be recorded.

[0058] At this time, the analysis unit (20) can analyze one or more of the sample's location, formation angle, area, and shade using artificial intelligence by forming multiple square separation areas of the same size by arranging multiple horizontal and vertical auxiliary lines at equal intervals on the image obtained by irradiating the radiation.

[0059] Therefore, when using a C-arm during surgery, as shown in Fig. 2, the analysis unit (20) can record the radiation dose received by the medical staff and the radiation dose received by the patient by analyzing the area of ​​the square separation area occupied by the medical staff's hand, the patient's affected area, external space, or medical equipment in the image obtained by irradiating the radiation.

[0060] Quality control items with a weekly inspection cycle may include inspection of the reading room environment.

[0061] The reading room environment inspection is an item that must be checked weekly, but can be checked in real time using the analysis department (20).

[0062] Specifically, the reading room environment should be inspected, including the reading room lighting, reading desk illuminance, ventilation, and temperature. Using artificial intelligence, illuminance, air density, and temperature can be monitored based on brightness. If these values ​​fall below the standard, an alarm (30) generates an alert, thereby maintaining the reading room environment at a certain level.

[0063] Quality control items with a 3-month inspection cycle may include clinical image evaluation and interpretation, and management of monitors for photography (CRT).

[0064] Clinical imaging evaluation may include upper gastrointestinal series and colonography.

[0065] A gastrointestinal series can check barium concentration, patient dose information, imaging markers, X-ray transparency, artifacts, esophageal images, compression images, prone images, double contrast images, and flow images.

[0066] Colonography can check barium concentration, patient dose information, imaging markers, X-ray transparency, examination pre-treatment, artifacts, examination technique, overhead photographs, instantaneous photographs, and mucosal coating.

[0067] In addition, clinical imaging evaluation can check the contrast agent used, fluoroscopy time, filming markers, X-ray transmittance, and artifacts.

[0068] The above items can be analyzed according to the evaluation content for each item using the artificial intelligence of the analysis unit (20) after obtaining data through the inspection unit (10).

[0069] At this time, the analysis unit (20) can assign grades to the test results for each item, calculate the sum by differentiating scores according to grades, and then compare the calculated sum with a standard value to determine pass or fail.

[0070] Taking the upper gastrointestinal series as an example, if there is an appropriate prone image, a grade A score of 6 points is given, if there is an image but it is not appropriate, a grade B score of 3 points is given, and if there is no image, a grade C score of 0 points is given.

[0071] Additionally, if a double-contrast image includes a well-covered gastric fundus image, it is given a grade A score of 8 points; if the image is present but inadequate, it is given a grade B score of 4 points; and if the image is absent, it is given a grade C score of 0 points.

[0072] In addition, in the case of elements that require a person to judge, such as whether a shooting cover was written on paper and pasted, if a person judges and then fills in the item, the analysis unit (20) can use it to calculate the score by judging only the grade of the item.

[0073] The analysis unit (20) can determine whether the score for each item is passed if it exceeds the standard score, or whether it is failed if it does not exceed the standard score, after adding up all the scores for each item as described above.

[0074] Additionally, if an F is included in the judgment grade, it may be judged as a failure even if the standard score is exceeded.

[0075] At this time, if the analysis unit (20) determines that it is grade A, but a person later checks the result and determines that it is grade B, the learning unit (40) learns this and then raises the judgment standard to determine the test result.

[0076] Management of monitors for reading or shooting can be categorized into visual and quantitative evaluations.

[0077] Visual evaluation can be conducted using the TG18-QC pattern. The TG18-QC pattern is a standard image pattern used to test image quality, and can be used to evaluate image resolution, contrast, noise levels, etc.

[0078] Referring to FIG. 3, the analysis unit (20) can conduct an evaluation by forming a square separation area on the pattern and comparing the same, and can also compare a rectangular area by combining the square separation areas as needed.

[0079] Additionally, the analysis unit (20) can vary the number of horizontal and vertical auxiliary lines depending on the image being evaluated, thereby varying the area of ​​the square separation area. Accordingly, in the case of image analysis for recording radiation dose as described above, the number of auxiliary lines can be increased to compare fine areas, and in the case of visual evaluation as shown in FIG. 3, the number of auxiliary lines can be reduced to compare wide areas.

[0080] Referring to Fig. 3(a), the visual evaluation can evaluate whether the pattern is well expressed in straightness without geometric distortion, whether the alignment intervals of the grid line bilattice structure in the center and four corners are equal and well observed in a square shape, and whether the black and white gradation changes of the gradation bars formed vertically on both sides are naturally and continuously observed.

[0081] In addition, it is possible to evaluate whether the luminance patches formed in a square shape in the central area as in Fig. 3(b) are clearly identifiable, whether the luminance difference can be clearly identified for each patch when observing the 5% luminance patch on the 0% luminance background at the bottom and the 95% luminance patch on the 100% luminance background, and whether the phrases in the three areas formed at the bottom can be clearly identified.

[0082] In the present invention, the TG18-QC pattern is used as a visual evaluation method for managing a monitor for reading or shooting, but any pattern that can be used for a visual evaluation method, such as a SMPTE pattern, can be used.

[0083] Quantitative evaluations include the luminance around the reading monitor, contrast response, luminance variation between multiple monitors, luminance uniformity, and maximum luminance variation, which can also be evaluated by comparing square separated areas of the image.

[0084] Quality control items with a 6-month inspection cycle may include phantom image evaluation and interpretation, and management of monitors (LCD) for shooting.

[0085] The phantom used to perform phantom image evaluation can be placed directly on the reading table by a person, and the sample position adjustment unit can adjust the position of the sample appropriately. Alternatively, the phantom can be installed on one side of the fluoroscopy equipment so that the sample position adjustment unit can automatically position the phantom on the reading table.

[0086] In the case of phantom image evaluation, it can be divided into a low-contrast evaluation that checks the number of holes as shown in Fig. 4(a) and a high-contrast evaluation that checks the number of mesh patterns observed in the measuring device as shown in Fig. 4(b).

[0087] Specifically, the analysis unit (20) can identify the number of holes and the number of mesh patterns by analyzing the shape using the contrast and luminance differences of the square separation area formed on the image. Through this, the analysis unit (20) can precisely analyze the phantom image evaluation, which is difficult to clearly determine with the naked eye, using artificial intelligence.

[0088] Additionally, if the number of holes or mesh patterns determined by the artificial intelligence of the analysis unit (20) is different from the result determined by the user later, the learning unit (40) can learn this.

[0089] For example, as shown in Figure 4(a), the low-contrast evaluation of phantom image evaluation considers a hole visible if it is similar in size to a clearly visible hole when observed on a monitor from a distance of 4 feet. However, the observation of an unclear hole can vary depending on the distance between the monitor and the observer, and thus, human and AI judgments may differ.

[0090] That is, the number of holes recognized by the AI ​​may be 4, but the number of holes recognized at a distance of 4 feet may be 3, so if the user inputs a correction value for this part, the AI ​​can learn to observe according to the distance based on the correction value.

[0091] Quality control items with a one-year inspection cycle may include field inspection, automatic exposure device inspection and reproducibility testing, tube voltage, tube current testing, half-value layer testing, inspection of inspector and patient protective equipment, and quality control training for personnel.

[0092] Field checks can be a method of measuring errors by aligning the test tool with the center of the light field.

[0093] The test tool used at this time can also be placed directly on the reading table by a person, and the sample position adjustment unit can adjust the position of the sample appropriately. Alternatively, the test tool can be installed on one side of the fluoroscopy equipment so that the sample position adjustment unit can automatically position the test tool on the reading table.

[0094] Once the position adjustment of the test tool is completed, the analysis unit (20) can measure the error angle or error length of the light irradiation field using horizontal and vertical auxiliary lines, and if the error exceeds the standard value, a non-conformity judgment can be made.

[0095] Additionally, the report writing department can create a report by showing in an image the extent to which the beam examined by the test tool deviates from the center line.

[0096] The automatic exposure device can check its status through the inspection unit (10), and a reproducibility test to evaluate accuracy and consistency when taking repeated shots under the same conditions can be performed by calculating the coefficient of variation, which is the standard deviation of the average of the exposure dose, through the analysis unit (20).

[0097] The tube voltage and tube current can be measured three or more times and averaged in the analysis unit (20), and the suitability can be determined by comparing the error with the reference value.

[0098] The half-value layer test is a test to check the thickness at which the initially measured dose value is reduced by half while changing the thickness of the material that allows radiation to pass through, and the analysis unit (20) can check whether the half-value layer exceeds the minimum half-value layer thickness.

[0099] Inspection of protective equipment involves first visually evaluating the appearance and then evaluating the size of the damaged part using a fluoroscopy device. When the protective equipment is placed on the reading table, the sample position adjustment unit adjusts the sample position appropriately, and then the analysis unit (20) uses artificial intelligence to find the damaged part of the protective equipment and determine whether it needs to be replaced.

[0100] Quality control items with a 3-year inspection cycle (regular inspection) may include grounding equipment verification test, external leakage current test, reproducibility test of irradiation dose, tube voltage, tube current test, illuminance test, difference test between X-ray irradiation field and light irradiation field, irradiation field coincidence test, incident irradiation dose rate test, timer test (limited to photographing equipment), and half-value test.

[0101] In the case of illuminance tests, incident irradiance rate tests, and timer tests, the analysis unit (20) can calculate the average value to check whether it is within the error range, and the analysis unit (20) can analyze whether the external leakage current is within the standard value and make a suitability judgment.

[0102] In addition, if human confirmation is required, such as whether grounding has been performed through a third-class grounding method or whether quality control training has been provided to workers, the items can be entered after a human judgment.

[0103] The report writing department can prepare reports based on the inspection and analysis results, as described above, according to the items required for each inspection cycle. Additionally, if any items are found to be unsuitable, the reasons for the unsuitability or solutions can be included in the report.

[0104] After checking the report, if the user confirms an incorrect judgment of the analysis unit (20), the user can correct it, and the learning unit (40) can learn the corrected content and use the learned content for the next analysis.

[0105] The notification unit (40) can generate a notification when a problem occurs during use of the equipment, so that the user can check it and take action, in the case of items that can be checked in real time, such as checking whether the equipment is operating or measuring real-time illuminance, even when using the viewing equipment.

[0106] In addition, unlike items that can be inspected in real time as described above, the notification unit (40) can generate a notification when the inspection cycle arrives in cases where the use of the equipment must be stopped when the inspection unit (10) conducts an inspection, such as a phantom image evaluation or clinical image evaluation, so that the user can be aware of the inspection progress time and proceed with the inspection.

[0107] Additionally, reports are automatically generated when inspection and analysis are completed on a periodic basis, so you can generate a notification when report generation is complete to allow users to check the report.

[0108] The storage unit (50) can store all data generated according to program operation, such as inspection records, analysis records, and notification records. Accordingly, the user can view the data stored in the storage unit (50) as needed.

[0109] As described above, the quality control system (1) of the medical fluoroscopy equipment of the present invention can automatically perform inspections according to the inspection cycle, and by analyzing the inspection results using artificial intelligence, the maintenance of the fluoroscopy equipment can be performed more accurately, thereby enabling the equipment to be used safely.

[0110]

[0111] Although the embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention can be implemented in other specific forms without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above are illustrative in all respects and are not limiting.

[0112]

[0113] [Explanation of symbols]

[0114] 1: Quality control system for medical fluoroscopy equipment

[0115] 10: Inspection Department

[0116] 20: Analysis Department

[0117] 30: Notification Department

[0118] 40: Learning Department

[0119] 50: Storage

Claims

1. Inspection department that directs the operation of medical fluoroscopy equipment and performs inspection according to quality control items; An analysis unit that analyzes the inspection results of the above inspection department using artificial intelligence to generate a report and compares the inspection results with the standard values ​​set for each quality control item to determine whether it passes or fails; Notification section that generates problem occurrence notifications, inspection cycle notifications, and report generation notifications; A learning unit in which the artificial intelligence learns the correction value when the user inputs a correction value if an error is found in the analysis results of the above analysis unit; and A quality control system for medical fluoroscopy equipment, including a storage unit that stores activity records of the inspection unit, analysis unit, and notification unit.

2. In paragraph 1, The above analysis unit, A quality control system for medical fluoroscopy equipment, including a report writing unit that writes the test results and analysis results examined according to a set test cycle in the form of a report.

3. In paragraph 2, The above inspection cycle can be one or more of one day, one week, three months, six months, one year, or three years. A quality control system for medical fluoroscopy equipment characterized by different quality control items according to the above inspection cycle.

4. In paragraph 1, The above inspection department, A quality control system for medical fluoroscopy equipment, characterized by including a sample position adjustment unit for adjusting the position of a sample for performing an inspection according to the above quality control items.

5. In paragraph 1, The above analysis unit, A quality control system for medical fluoroscopy equipment characterized in that it analyzes at least one of the position, formation angle, area, and shade of a sample by forming a plurality of square separation areas of the same size by arranging a plurality of horizontal and vertical auxiliary lines at equal intervals on an image obtained by irradiating radiation.

6. In paragraph 1, The above analysis unit, A quality control system for medical fluoroscopy equipment characterized by assigning grades to the test results for each of the above quality control items, calculating a sum by differentiating scores according to the grades, and then comparing the calculated sum with a standard value to determine pass or fail.

7. In paragraph 1, The above inspection department, The above quality control items include phantom image evaluation, The above analysis unit, In the case of the above phantom image evaluation, artificial intelligence analyzes the shape using the difference in brightness and contrast of the square separation area formed on the image, thereby confirming the number of holes and the number of mesh patterns. The above learning department, A medical fluoroscopy equipment quality control system characterized in that, in the case of the above phantom image evaluation, if the number of holes or the number of mesh patterns judged by the artificial intelligence of the analysis unit is different from the result judged by the user, the user inputs a correction value, and the artificial intelligence learns the correction value.

8. In paragraph 7, The above learning department, In the case of the above phantom image evaluation, if the number of holes or the number of mesh patterns judged by the artificial intelligence of the analysis unit according to the image observation distance is different from the result judged by the user, the user inputs a correction value, and the artificial intelligence learns observation according to the distance based on the correction value. A quality control system for medical fluoroscopy equipment.

Images