Medical equipment inspection system and medical equipment inspection method

Abstract

Providing technology that allows any healthcare professional to perform maintenance and inspection easily and efficiently. [Solution] The medical equipment inspection system comprises a storage unit that stores image data of the medical equipment in a normal state and a large number of image data of the medical equipment in an abnormal state; an imaging unit that can image the appearance of the medical equipment; and an inspection unit that inspects the medical equipment at the time of inspection by matching the image data of the appearance of the medical equipment at the time of inspection, which is captured by the imaging unit, with the image data stored in the storage unit.

Description

Technical Field

[0001] The present invention relates to an inspection system for medical devices including medical devices such as infusion pumps and a method for inspecting medical devices.

Background Art

[0002] The global medical device market is predicted to expand from approximately 78 trillion yen in 2023 to approximately 98 trillion yen in 2027 (from the Ministry of Economy, Trade and Industry). Medical devices require maintenance inspections to maintain their accuracy and performance, and proper implementation is regulated by the Medical Law and the Pharmaceutical and Medical Device Law in Japan. On the other hand, the inspections for increasing medical devices are diversifying and becoming more complex. Medical devices equipped with liquid crystal screens display various characters and numerical values, and it is difficult to understand their meanings without specialized knowledge. In addition, medical devices with complex structures or small-sized medical devices have low visibility, so corresponding experience is required to identify external damage.

[0003] In hospital maintenance inspections, there are regular inspections carried out by manufacturers and sellers and daily inspections carried out by users. Regular inspections are carried out at a frequency of 1 to 4 times a year, and daily inspections are carried out every day. Regular inspections are either outsourced to manufacturers and sellers or carried out by specialized staff in the in-hospital maintenance inspection department, while daily inspections are carried out by users, that is, an unspecified number of medical staff belonging to each department and auxiliary staff without medical qualifications (see Patent Document 1).

[0004] As an example of medical devices, infusion pumps are numerous in number and have a high operation rate from use to the end of use. There are also hospitals that operate them 3 to 4 times a day per unit. According to a survey by Yano Research Institute, the number of shipped infusion pumps in 2012 was 33,800 units, which increased to 40,890 units in 2019. Therefore, the number of daily inspections after use inevitably increases. In addition, the time required for daily inspections by manual work such as visual inspection becomes a work load. Therefore, there is a possibility that the quality of daily inspections may decline or that they may be used without being carried out.

[0005] The applicant's hospital possesses 8,384 medical devices, of which 625 are infusion pumps. They are operated at an operating rate of 90-105%. The Medical Device Management Office, a specialized department for maintenance and inspection, performs daily inspections on 150 units per day. Meanwhile, the chemotherapy center, which employs a large number of nurses and support staff, is expected to perform daily inspections on 90 infusion pumps per day. 70% of the daily inspections involve visual inspection (such as checking for blood / medication contamination, damage, and cracks), and require careful attention. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2008-287627 [Overview of the project] [Problems that the invention aims to solve]

[0007] Maintenance and inspection are crucial for maintaining the performance and accuracy of medical devices. However, many problems exist, as follows: Many medical devices have complex and uneven structures, making them difficult to see. • The latest medical devices display a lot of information on LCD screens, which can take time to understand. • Medical equipment with high utilization rates in hospitals will require more frequent routine inspections. • Frequent inspections increase workload. • The inspection takes time because there are many inspection items. • Inspections are performed manually, including visual checks for cracks and damage. • Manpower shortage due to population decline • Time must be allocated for training in inspection procedures. • Decreased visual acuity reduces the accuracy of visual inspections. • Variation in inspection skills among an unspecified number of staff members In view of the above circumstances, the object of the present invention is to provide a technology that allows any medical professional to perform maintenance and inspection easily and efficiently. [Means for solving the problem]

[0008] To achieve the above objective, the medical equipment inspection system according to the present invention comprises a storage unit that stores image data of the normal state of the appearance of the medical equipment and a large number of image data of the abnormal state thereof; an imaging unit capable of imaging the appearance of the medical equipment; and an inspection unit that inspects the medical equipment at the time of inspection by matching the image data of the appearance of the medical equipment at the time of inspection, which is captured by the imaging unit, with the image data stored in the storage unit.

[0009] This allows any healthcare professional to perform maintenance and inspections easily and efficiently.

[0010] The medical equipment inspection system according to the present invention may be configured such that the storage unit stores barcode data that identifies an individual medical equipment, the inspection unit reads a barcode that identifies an individual medical equipment, which is placed at a predetermined position on the appearance of the medical equipment at the time of inspection as captured by the imaging unit, identifies the individual medical equipment captured by the imaging unit by comparing the read barcode data with the barcode data stored in the storage unit, and links the identified medical equipment with the inspection result.

[0011] The medical equipment inspection system according to the present invention is a medical device having a display unit that displays predetermined information, and the inspection unit may be configured to inspect the medical equipment at the time of inspection by matching the image data of the information and appearance of the medical equipment displayed on the display unit and the image data of the medical equipment at the time of inspection, which are captured by the imaging unit, with the image data stored in the storage unit.

[0012] The medical equipment inspection system according to the present invention may be configured such that the imaging unit has an imaging screen that displays a framing line indicating the imaging range in which the external appearance of the medical equipment is captured.

[0013] The medical equipment inspection system according to the present invention may be configured such that the inspection unit stores image data of the medical equipment at the time of inspection in the storage unit, along with an inspection item table that associates inspection items with the inspection results of the medical equipment.

[0014] The medical equipment inspection system according to the present invention may be configured such that, when there is an inspection item with a poor inspection result, the inspection unit determines which department will repair the item according to the risk associated with the poor inspection result and notifies the department of the determination.

[0015] The medical equipment inspection system according to the present invention may have an input unit for inputting comments, and may be configured to store the comments input from the input unit for the inspection item list together with the inspection item list in the storage unit.

[0016] The medical equipment inspection system according to the present invention may further include a thermograph capable of imaging the medical equipment, the storage unit may store data of the medical equipment in a normal state, and a large amount of surface temperature data of the medical equipment in an abnormal state, and the inspection unit may be configured to inspect the medical equipment at the time of inspection by matching the surface temperature data of the medical equipment at the time of inspection that was imaged with the surface temperature data stored in the storage unit.

[0017] The medical equipment inspection system according to the present invention may further be configured to include a magnifying glass with a light source lens that can be attached to the imaging unit.

[0018] The medical equipment inspection system according to the present invention comprises a server device having the storage unit and the inspection unit, and a smart device having the imaging unit. The medical equipment inspection system according to the present invention further includes a mechanism on which the medical equipment to be inspected is placed and which rotates or moves the placed medical equipment, and the imaging unit may be configured to image the medical equipment being rotated or moved.

[0019] The inspection system for medical equipment according to the present invention may be configured to accumulate the repair history of the medical equipment and predict the repair of the medical equipment based on the accumulated big data.

[0020] The inspection system for medical equipment according to the present invention may be configured such that the medical equipment is a medical device capable of storing device information of the medical equipment, and based on the device information, predict the deterioration of the medical device.

[0021] The inspection system for medical equipment according to the present invention may be configured to calculate the usage frequency of the medical equipment from the intervals of the dates on which the medical equipment is imaged, and based on the calculated usage frequency, notify a change in the placement position or the inspection timing of the medical device.

[0022] The inspection system for medical equipment according to the present invention may be configured to diagnose deterioration or contamination of the medical equipment based on the intervals of the dates on which the medical equipment is imaged and the comparison of the image data imaged on each date.

[0023] The inspection system for medical equipment according to the present invention may be configured such that the imaging unit images a display screen of the medical equipment used at home as the appearance of the medical equipment. is something.

[0024] The inspection method for medical equipment according to the present invention stores image data of a normal state of the appearance of the medical equipment, further stores a large number of image data of an abnormal state of the appearance of the medical equipment, images the appearance of the medical equipment at the time of inspection, and inspects the medical equipment at the time of inspection by matching the image data of the appearance of the medical equipment at the time of inspection and the stored image data.

Effect of the Invention

[0025] [[ID=二十九]]According to the present invention, any medical staff can perform maintenance inspections easily and efficiently.

Brief Description of the Drawings

[0026] [Figure 1] This diagram shows the configuration of a medical equipment inspection system according to one embodiment of the present invention. [Figure 2] This diagram illustrates the diagnosis of normality or abnormality based on AI matching. [Figure 3] This diagram explains the mechanism for diagnosing whether an infusion pump is functioning correctly or not during inspection. [Figure 4] This figure shows an image of an infusion pump being scanned during inspection using a smart device, including a QR code (registered trademark). [Figure 5] This is a diagram to explain the inspection checklist. [Figure 6] This is a diagram to explain framing lines. [Figure 7] This image shows the infusion pump being imaged by a smart device during inspection. [Figure 8] This is a diagram illustrating an example of free text input. [Figure 9] This diagram shows a smart device with a thermographic camera attached. [Figure 10] This diagram shows a smart device with a magnifying glass equipped with a light source lens attached. [Figure 11] This diagram shows the configuration of an external inspection device having a rotating mechanism. [Figure 12] This diagram shows the configuration of a visual inspection device that has a belt conveyor. [Figure 13] This diagram illustrates an example of using AI to diagnose medical devices used at home. [Figure 14] This diagram illustrates an example of using AI analysis to predict repairs based on big data from inspection and repair history records kept in the inspection and repair department. [Figure 15] This diagram illustrates an example of using AI analysis to predict deterioration based on data stored in the internal memory of a medical device. [Figure 16] This diagram illustrates an example of calculating the frequency of medical device use from the interval between imaging days using a smart device, and then using that frequency to alert users about potential changes in the placement of medical devices or when inspections should be performed. [Figure 17] This diagram illustrates an example of using AI to diagnose the deterioration and spread of contamination in medical devices based on the interval between imaging days using smart devices and the changes in the acquired images over time. [Modes for carrying out the invention]

[0027] Embodiments of the present invention will be described below with reference to the drawings.

[0028] Figure 1 shows the configuration of a medical equipment inspection system according to one embodiment of the present invention.

[0029] As shown in Figure 1, the medical equipment inspection system 1 is configured such that a server device 2 and a smart device 3 carried by a medical professional inspecting the medical equipment can be connected via a network 4.

[0030] Medical equipment refers to medical devices and caregiving and welfare equipment defined by the Pharmaceuticals and Medical Devices Act that are used in medical settings such as hospitals and clinics, or in caregiving and welfare facilities, train stations, sports clubs, and home healthcare settings (for example, infusion pumps, endoscopes, medical devices with screen displays of pressure waveforms, ventilation volume waveforms and numerical values ​​for ventilators, small steel instruments, oxygen flow meters with pressure regulators, small steel instruments, AEDs, etc., as well as the following: caregiving and welfare equipment such as wheelchairs, walkers, crutches, and IV poles, surgical sets for different surgical procedures, inventory management and expiration dates for emergency carts, various treatment sets, intubation sets, puncture sets, IV sets which are medical supplies such as IV drips and medications, ventilator circuits, etc). The present invention is applicable to these medical devices. In the following embodiments, an infusion pump 5 will be used as an example of medical equipment.

[0031] Server device 2 is a computer system that provides services to smart devices 3 over network 4. The hardware of server device 2, although not shown in the diagram, consists of a processor (CPU), memory (RAM), storage, network interface (NIC), power supply unit, etc. The software of server device 2, although not shown in the diagram, consists of an operating system (OS), server applications, etc.

[0032] Smart device 3 includes tablet terminals and smartphones. In addition to smart devices, the present invention can also use information processing devices such as PCs. One example of smart device 3 is a portable computer with advanced functions that provides many functions such as communication, entertainment, and information processing. The hardware of smart device 3, although not shown in the figures, consists of a processor (CPU), GPU, memory (RAM), internal storage (ROM / Flash Memory), touchscreen display (OLED, LCD), camera (imaging unit), battery, various sensors, communication device, ports and interfaces, etc. The software of smart device 3, although not shown in the figures, has an operating system (OS) and various applications (native applications, web applications) installed. For example, email, browser, messaging applications, and inspection applications according to the present invention are installed.

[0033] Server device 2 includes a storage unit 21 that stores image data of the infusion pump 5 in a normal state as shown in Figure 2 and later, and also stores a large number of image data of various abnormal states of the infusion pump 5's appearance, and an inspection unit 22 that inspects the infusion pump 5 at the time of inspection by matching image data of the infusion pump 5's appearance during inspection, for example, after use, with image data stored in the storage unit 21 using AI. Specifically, the inspection unit 22 compares the image captured at the time of inspection with past images captured over time and the image captured at the time of inspection, and uses AI to match characters, numbers, lights, marks, and colors displayed on the LCD screen of the infusion pump 5, as well as the shape of the infusion pump 5, scratches, damage, defects, and drug / blood stains, to diagnose whether it is normal or abnormal. Note that part or all of the storage unit 21 may be located on the smart device 3 side.

[0034] For example, by using AI matching, as shown in Figure 2, it is possible to diagnose whether the infusion pump 5 is normal or abnormal by comparing its fit with an image of the infusion pump 5 within its frame (within the framing line described later) captured by the smart device 3 during inspection (right side of the figure). The mechanism involves the following comparison (see Figure 3).

[0035] (1) Matching the shape to fit damage and cracks

[0036] (2) Matching the type of stain to suit chemical solutions, blood stains, etc.

[0037] (3) Compatibility comparison by matching displays, lights, markings, numbers, etc. To use AI-based matching technology, the following steps are typically taken.

[0038] (1) Data collection Collection of normal and abnormal data: Collect images of more normal and abnormal conditions. This includes a variety of cases, such as characters, numbers, lights, marks, and colors displayed on the LCD screen of the infusion pump 5, as well as the shape of the infusion pump 5, scratches, damage, defects, and drug / blood stains.

[0039] (2) Data labeling Labeling: Accurately label the collected images to indicate which parts are normal and which are abnormal. For abnormal data, also label the type of abnormality (e.g., "dirt," "scratch," "damage," etc.).

[0040] (3) Model selection and design Selection of deep learning models: Convolutional neural networks (CNNs) are highly effective for image recognition. Furthermore, fine-tuning pre-trained models, such as those on ImageNet, can be expected to reduce training time and improve accuracy.

[0041] (4) Model training Training: A deep learning model is trained using labeled image data. Data augmentation is also effective in increasing data diversity. Retraining the model each time new data becomes available helps maintain and improve accuracy.

[0042] Figure 4 shows an image of the infusion pump 5 being captured by the smart device 3 during inspection.

[0043] As shown in Figure 4, the exterior of the infusion pump 5 is affixed with a label bearing a QR code (registered trademark) 51 for individual identification of the infusion pump 5. By capturing an image including this QR code (registered trademark) 51 with the smart device 3, information such as the management number of the infusion pump 5 can be recognized, thereby identifying the unique characteristics and information of the infusion pump 5 during inspection. For the identification of the inspector, the inspector can be identified by reading and recognizing the barcode on the identification card worn by the staff member using the smart device 3. This allows for the identification of the inspector's department, name, years of experience, number of times images have been taken, etc., and by linking this with the diagnostic results, the reliability of the diagnostic results can be measured. Note that in addition to QR codes (registered trademark), numbers, letters, symbols, etc., may also be used.

[0044] For this purpose, the storage unit 21 of the server device 2 stores QR code (registered trademark) data as barcode data to identify individual infusion pumps 5. The same applies to barcodes on identification cards. The inspection unit 22 reads the QR code (registered trademark) 51 that identifies the individual infusion pump 5, which is placed in a predetermined position on the exterior of the infusion pump 5 during inspection as captured by the smart device 3. By comparing the read QR code (registered trademark) data with the QR code (registered trademark) data stored in the storage unit 21, it identifies the individual infusion pump 5 captured by the smart device 3. It then makes it possible to read the individual number, manufacturing date, purchase date, model, manufacturer, past image history, etc., which are registered in the storage unit 21, and links the identified infusion pump 5 with the diagnostic result of whether it is normal or abnormal by matching it with AI. Note that in addition to QR codes (registered trademark), numbers, letters, symbols, etc., may also be used.

[0045] Furthermore, the inspection unit 22 stores the image data of the infusion pump 5 during inspection in the storage unit 21, along with an inspection item table that associates inspection items (e.g., dirt, damage / cracks / missing parts, power-on possible, AC power operation, battery level, etc.) with the inspection results of the infusion pump 5 (results of diagnosis of normal or abnormality through AI matching). Specifically, as shown in Figures 5(a) and (b), after the inspection is completed, the inspection unit 22 displays a "green ○" if the result on the inspection item table is normal and a "red ×" if it is abnormal. By recognizing the time elapsed since the inspection items and imaging intervals, the unit assigns the department responsible for inspection and repair. For example, as shown in Figure 5(a), if an × is marked for the dirt item, and the change in the extent of the dirt over time is small, the risk is considered low, and the department responsible can adequately handle it at the usage site. As shown in Figure 5(b), if the change in the extent of the dirt over time is large, the risk is considered high, and the smart device 3 displays that the medical equipment management room should handle it. In this embodiment, the AI ​​evaluates the risk of such a judgment, i.e., an "X," and the AI ​​recommends the appropriate department accordingly. In the inspection item list shown in Figure 5, if the AI ​​diagnosis indicates a successful match, a "O" is displayed on the smart device 3, the "model" and "management number" are displayed as a QR code (registered trademark), and the "date and time" is displayed as the imaging date and time on the smart device 3. In addition, in the inspection item list shown in Figure 5, the department recommended by the AI ​​is displayed on the smart device 3 with an "X." Note that in addition to a QR code (registered trademark), numbers, letters, symbols, etc. may also be used.

[0046] As shown in Figures 6(a) and 6(b), the imaging screen of the smart device 3 displays four framing lines 31 that indicate the imaging range that includes the exterior of the infusion pump 5. In Figure 6(a), the infusion pump 5 is outside the four framing lines 31, so part of the framing lines 31 are displayed in white, indicating that the infusion pump 5 is not recognized. In Figure 6(b), the infusion pump 5 is within the four framing lines 31, so all of the framing lines 31 are displayed as black lines, indicating that the infusion pump 5 is recognizable. This is performed by the inspection application installed on the smart device 3. By displaying it in this way, it is possible to reliably image the entire exterior of the infusion pump 5. When the infusion pump 5 is within the four framing lines 31, the framing lines 31 may be highlighted by lighting up or other means. This makes it possible to more reliably image the entire exterior of the infusion pump 5. Continuous shooting may be used to recognize the blinking state.

[0047] Figure 7 shows the infusion pump 5 being imaged by smart device 3 during inspection. The procedure for imaging with smart device 3 is as follows:

[0048] (1) Start the inspection application.

[0049] (2) When you start the camera, a frame (framing line) will appear on the camera screen.

[0050] (3) Adjust the angle and direction of the camera so that the infusion pump 5 is within the frame.

[0051] (4) The infusion pump management number and other information can be recognized by placing the QR code (registered trademark) within the frame.

[0052] (5) The frame lights up when the object is properly placed within the frame.

[0053] (6) Press the shutter to capture an image.

[0054] (7) The captured image data is saved to the smart device 3.

[0055] (8) The AI ​​makes a judgment by comparing the data with the stored image of the normal state through communication with server device 2.

[0056] (9) If normal: A circle (〇) will be displayed in the inspection item, and the image and inspection item list will be automatically saved.

[0057] (10) In case of abnormality: An "X" will be displayed next to the abnormal item, and the image and inspection list will be automatically saved.

[0058] (11) The inspection can continue unless the "End Inspection" button is pressed.

[0059] (12) Past captured images and inspection item lists can be viewed by performing the prescribed operation.

[0060] Smart device 3 can be used to perform such checks in various locations, such as medical equipment management rooms, chemotherapy centers, operating rooms, intensive care units, wards, nursing homes, home nursing stations, home healthcare settings, and ambulances.

[0061] Therefore, with the medical equipment inspection system 1 according to this embodiment, any medical professional can easily and efficiently perform maintenance and inspection in various locations.

[0062] The present invention is not limited to the embodiments described above, and can be modified and applied within the scope of its technical concept, and the scope of such implementation also falls within the technical scope of the present invention.

[0063] For example, since images and inspection results from past imaging are stored as history in the storage unit 21 of the server device 2, they can be viewed by the smart device 3. Furthermore, it may be possible to enable the input of comments of up to 20 characters in free text from the smart device 3. As shown in Figure 8, when the inspection is completed (a), the smart device 3 can select a management number (b). In addition, the image and inspection results of the management number selected in (b) can be displayed as in (c). At this point, it would be good to enable the input of comments. This makes it possible to record the on-site opinions of medical professionals.

[0064] For example, as shown in Figure 9, the surface temperature of a medical device can be imaged by attaching a thermographic camera 32 to the smart device 3. The storage unit 21 of the server device 2 stores surface temperature data of the medical device in a normal state, and also stores a large amount of surface temperature data of the medical device in an abnormal state. The inspection unit 22 can then inspect the medical device at the time of inspection by matching the surface temperature data of the medical device at the time of inspection, which is imaged by the thermographic camera 32 on the smart device 3, with the surface temperature data stored in the storage unit 21 using AI. This allows the condition of the medical device to be determined by displaying the surface temperature on the smart device 3 and comparing the effect of driving heat over time.

[0065] For example, as shown in Figure 10, by attaching a magnifying glass 33 with a light source lens to the smart device 3, it is possible to image the deterioration of a part of the medical device, such as the depth of a wound. The storage unit 21 of the server device 2 similarly stores a large amount of normal data and abnormal data, and the inspection unit 22 can then inspect the medical device at the time of inspection by matching it with AI. This allows for more accurate inspections by comparing the depth of wounds, etc., over time using the magnifying glass 33.

[0066] For example, as shown in Figure 11, a box-shaped external inspection device 91, similar to a microwave oven, may be used, which has a rotating mechanism 92 inside for rotating a medical device (e.g., an infusion pump 5) and a camera, like a smart device 3. When a medical device (e.g., an infusion pump 5) is placed inside the external inspection device 91, it may be configured to automatically rotate 360 ​​degrees while capturing images of the external appearance and enabling AI diagnosis. This makes it possible to compensate for the drawback of the smart device 3 only being able to capture images in one direction.

[0067] For example, as shown in Figure 12, a visual inspection device 101 may be adopted that automates the process from the entrance to the exit of the inspection department, using a conveyor belt 102, similar to that used in airport baggage inspection areas, to move medical devices (e.g., infusion pumps 5).

[0068] For example, the present invention makes it possible to diagnose the normal operation of medical devices used at home using AI. As shown in Figure 13, the smart device 3 reads QR codes (registered trademarks) 112, 113, and 114 in three of the four corners of the LCD screen 111, including the patient, to recognize the normal screen orientation of the device. This is because the QR codes (registered trademarks) cannot be recognized if the screen orientation is not normal. The three QR codes (registered trademarks) contain (1) patient information, (2) setting condition information, and (3) device information. The AI ​​diagnosis by the inspection unit 22 of the server 2 recognizes not only numbers and marks but also waveforms (pressure waveform, flow velocity waveform, ventilation volume, etc.) displayed on the LCD screen 111. Therefore, the memory unit 22 stores normal waveforms and numerous abnormal waveforms. It is also possible to read (1) patient information such as the patient's address and name, which is useful for transporting patients to medical institutions in the event of a disaster, (2) information on the setting conditions of medical devices such as ventilation volume, respiratory rate, and oxygen concentration in the case of a ventilator, and (3) medical device information such as the purchase date, manufacturer name, and serial number. Therefore, the memory unit 22 is linked to this information and the QR code (registered trademark).

[0069] Therefore, while it is common practice to make voice calls to the hospital to clarify any doubts when patients are at home, showing the following image data to medical professionals at this time allows for a more accurate assessment than communicating via voice or text. This assessment involves determining whether observation at home is possible, whether immediate hospitalization is necessary, or whether nighttime or weekend consultations are required. Example 1: Condition of postoperative wounds, etc. Example 2: Patient contact surface of tubes, etc. Example 3: Images of patients or their families who are concerned about medical equipment (worried about medical equipment and have questions or concerns). Furthermore, in the case of medical devices such as ventilators, the waveforms displayed on the LCD screen 111 contain important information (such as the presence or absence of sputum, the presence or absence of air leaks, the presence or absence of appropriate pressure, and the presence or absence of appropriate tidal volume). Therefore, AI diagnosis through waveform comparison over time makes it possible to evaluate the risk to the patient. For this reason, patient waveform data is stored in the memory unit 22.

[0070] For example, as shown in Figure 14, big data of inspection and repair history recorded in the inspection and repair department can be used to predict repairs using AI analysis. For instance, by using AI analysis by the inspection unit 22 to analyze the interval and frequency of occurrence of the infusion pump 5 from the previous repair to the current repair, it is possible to predict the probability and timing of the next repair.

[0071] Furthermore, as shown in Figure 15, the system may be configured to perform AI analysis to predict deterioration based on data stored in the internal memory unit (not shown) of the medical device (e.g., infusion pump 5). Various predictions are made based on the stored data within the infusion pump 5. For example, if the operating time exceeds a certain limit, the AI ​​can prompt a warning prompting the return of the device to a specialized department for performance and accuracy checks. If the motor rotation speed increases, the AI ​​can prompt a notification for motor or battery replacement. Since the acceleration sensor is a sensor that detects impacts to the medical device, the AI ​​can prompt a prediction of loosening or falling screws around the circuit board or damage to the exterior based on an increase in shock history. The AI ​​can also prompt a prediction of deterioration of bubble detection sensors, etc., based on changes in A / D values. This can also be done by sending data stored in the internal memory unit of the medical device (e.g., infusion pump 5) to the server device 2, and having the inspection unit 22 perform a diagnosis using AI.

[0072] For example, as shown in Figure 16, the inspection unit 22 of the server device 2 may be configured to calculate the frequency of use of medical devices from the interval between imaging dates stored in the storage unit 21 of the smart device 3, and to alert users to changes in the placement of medical devices or when inspections should be performed based on the frequency of use. The smart device 3 displays the frequency of use of each pump based on the interval between imaging dates and times. For example, the infusion pump 5 installed in a room close to the staff station has a high usage rate, while the one in the far back room has a low usage rate, so it is possible to prompt adjustments to the placement of the infusion pump 5. Since inspections to confirm the performance and accuracy of medical devices with high frequency of use are recommended, this function is also useful as an alert to encourage inspections.

[0073] For example, as shown in Figure 17, the inspection unit 22 of the server device 2 can perform AI diagnosis of whether or not medical equipment is deteriorating or contamination is spreading based on the interval between imaging days and the changes in the captured images over time, as stored in the storage unit 21 of the smart device 3. In other words, by analyzing the changes in the images over time, it is possible to predict future damage expansion (left in the figure). Furthermore, by analyzing the changes in the images over time, it may be possible to predict future circuit board damage risk by predicting that the spread of drug contamination will penetrate into gaps and corrode and degrade the internal circuit board of the infusion pump, and to prompt cleaning (right in the figure). [Explanation of Symbols]

[0074] 1. Medical equipment inspection system 2 Server devices 3. Smart device (including imaging unit) 4 Network 5. Infusion pump 21 Memory section 22 Inspection Department 31 Framing lines 51 QR Code (Registered Trademark)

Claims

1. A medical equipment inspection system comprising: a storage unit that stores image data of the medical equipment in a normal state and a large number of image data of the medical equipment in an abnormal state; an imaging unit capable of imaging the medical equipment's appearance; and an inspection unit that inspects the medical equipment at the time of inspection by matching the image data of the medical equipment's appearance at the time of inspection, captured by the imaging unit, with the image data stored in the storage unit.

2. A medical equipment inspection system according to claim 1, wherein the storage unit stores barcode data that identifies an individual medical equipment, and the inspection unit reads a barcode that identifies an individual medical equipment, which is provided at a predetermined position on the appearance of the medical equipment at the time of inspection as captured by the imaging unit, identifies the individual medical equipment captured by the imaging unit by comparing the read barcode data with the barcode data stored in the storage unit, and links the identified medical equipment with the inspection result.

3. A medical device inspection system according to claim 1 or 2, wherein the medical device is a medical device having a display unit for displaying predetermined information, and the inspection unit is a medical device inspection system that inspects the medical device at the time of inspection by matching the image data of the information and appearance of the medical device at the time of inspection, which is captured by the imaging unit, with the image data stored in the storage unit.

4. A medical device inspection system according to any one of claims 1 to 3, wherein the imaging unit has an imaging screen that displays a framing line indicating an imaging range that includes the external appearance of the medical device.

5. A medical equipment inspection system according to any one of claims 1 to 4, wherein the inspection unit stores image data of the medical equipment at the time of inspection in the storage unit together with an inspection item table that associates inspection items with the inspection results of the medical equipment.

6. A medical equipment inspection system according to claim 5, wherein the inspection unit determines which department will repair an inspection item if the inspection result is poor, according to the risk associated with the poor inspection item, and notifies the department of the determination.

7. A medical equipment inspection system according to claim 5 or 6, comprising an input unit for inputting comments, and storing the comments input from the input unit for the inspection item list together with the inspection item list in the storage unit.

8. A medical equipment inspection system according to any one of claims 1 to 7, further comprising a thermograph capable of imaging the surface temperature of the medical equipment, wherein the storage unit stores surface temperature data of the medical equipment in a normal state, and further stores a large number of surface temperature data of the medical equipment in an abnormal state, and the inspection unit inspects the medical equipment at the time of inspection by matching the surface temperature data of the medical equipment at the time of inspection that was imaged with the surface temperature data stored in the storage unit.

9. A medical equipment inspection system according to any one of claims 1 to 8, A medical equipment inspection system further comprising a magnifying glass with a light source lens that can be attached to the imaging unit.

10. A medical equipment inspection system according to any one of claims 1 to 9, comprising a server device having the storage unit and the inspection unit, and a smart device having the imaging unit.

11. A medical equipment inspection system according to any one of claims 1 to 10, further comprising a mechanism on which medical equipment to be inspected is placed and which rotates or moves the placed medical equipment, wherein the imaging unit images the rotating or moving medical equipment.

12. A medical equipment inspection system according to any one of claims 1 to 11, wherein the system accumulates the repair history of the medical equipment and predicts the repair of the medical equipment based on the accumulated big data.

13. A medical device inspection system according to any one of claims 1 to 12, wherein the medical device is a medical device capable of storing device information of the medical device, and the medical device inspection system predicts the deterioration of the medical device based on the device information.

14. A medical equipment inspection system according to any one of claims 1 to 13, wherein the system calculates the frequency of use of the medical equipment from the interval between days on which the medical equipment is imaged, and notifies the medical equipment of a change in its placement or the timing of its inspection based on the calculated frequency of use.

15. A medical equipment inspection system according to any one of claims 1 to 14, the medical equipment inspection system that diagnoses deterioration or spread of contamination of the medical equipment based on the interval between days on which the medical equipment is imaged and a comparison of the image data captured on each day.

16. A medical device inspection system according to any one of claims 1 to 15, wherein the imaging unit captures an image of the display screen of the medical device being used at home as the external appearance of the medical device.

17. A method for inspecting medical equipment, comprising: storing image data of the medical equipment in a normal state; storing a large number of image data of the medical equipment in an abnormal state; imaging the appearance of the medical equipment during inspection; and inspecting the medical equipment at the time of inspection by matching the image data of the medical equipment's appearance captured during inspection with the stored image data.

Images