The intelligent monitoring system for oxygen concentrators

Abstract

The present invention delineates the intelligent oxygen concentrator monitoring system with at least a sensing device attached on an oxygen concentrator. The sensing device comprises six components: an oxygen concentration sensing unit, an oxygen flow sensing unit, an image capture unit, a temperature sensing unit, a sound sensing unit, and a vibration sensing unit. These units generate oxygen concentration data, oxygen flow data, image data, temperature data, sound data, and vibration data, respectively. It can also transmit the analytical results to at least one of the management and monitoring devices and designated customer contact devices. The server transmits the analytic results to at least one management and monitoring device and at least one designated customer contact device. The analytic results provide insights into the operational status of the oxygen concentrator, identify potential abnormalities, and predict failures to enable timely processing or replacement.

Description

[Technical Field]

[0001] This invention relates to a monitoring system, and more particularly to an intelligent oxygen concentrator monitoring system that can predict the usage status of an oxygen concentrator and handle malfunctions in real time, so as to achieve user safety and preventive maintenance of the machine. [Previous Technology]

[0002] Oxygen concentrators mainly concentrate the oxygen concentration in the atmosphere. Their function is to release gases with a higher oxygen concentration than those in the atmosphere. The oxygen concentration in the atmosphere is generally about 20%. If the oxygen concentration can be increased to about 30% to about 35%, it can help individuals relieve fatigue and release stress after strenuous exercise. In particular, patients with respiratory diseases, such as asthma, often need to use higher concentrations of oxygen for medical and health care purposes.

[0003] In general, medical institutions, such as hospitals and nursing homes, usually have an oxygen generator to produce an appropriate concentration of oxygen, thereby providing a more comfortable environment for general patients or maintaining the vital signs of critically ill patients. However, a large number of patients need to have an oxygen generator set in their home environment for their use.

[0004] Although the oxygen concentrators currently used at home are equipped with fault warning functions, if users or caregivers fail to detect the fault warnings in time, it can easily cause safety concerns for users. Moreover, existing oxygen concentrators usually only have the function of warning notification when a fault occurs. However, if a warning notification occurs, it means that the oxygen concentrator has already malfunctioned or broken. Repairing the oxygen concentrator at this time will also cause users to be unable to use it immediately, which will also cause safety concerns. In addition, oxygen concentrators can also be obtained through rental. Although most oxygen concentrators have basic fault detection, if oxygen concentrator rental companies can know the usage status or aging and malfunction potential of the oxygen concentrators in advance, they can carry out repairs or replacements in advance. This will not only make the oxygen concentrators easier to use, but also give users a better user experience.

[0005] Therefore, how to improve the above-mentioned deficiencies is the technical difficulty that the inventors of this case want to solve. [Summary of the Invention]

[0006] Therefore, in order to effectively solve the above problems, the main objective of the present invention is to provide an intelligent oxygen concentrator monitoring system that can predict the usage status of the oxygen concentrator and handle it in real time before a failure occurs, so as to achieve user safety and preventive maintenance of the machine.

[0007] Another objective of the present invention is to provide an intelligent oxygen concentrator monitoring system that can predict the decay time and schedule maintenance in advance, so as to provide users with multiple protections and reduce service interruptions caused by maintenance time.

[0008] To achieve the above objectives, the present invention provides an intelligent oxygen concentrator monitoring system, comprising: at least one sensing device, the sensing device being disposed on an oxygen concentrator, and the sensing device having a data compression unit, an oxygen concentration sensing unit, an oxygen flow sensing unit, an image capturing unit, a temperature sensing unit, a sound sensing unit, a vibration sensing unit, and an operation sensing unit; the oxygen concentration sensing unit senses the oxygen concentration delivered by the oxygen concentrator and generates at least one oxygen concentration sensing data; the oxygen flow sensing unit senses the oxygen delivery volume of the oxygen concentrator and generates at least one oxygen flow sensing data; the image capturing unit captures an image of an oxygen user and generates at least one image data; the temperature sensing unit senses the operating temperature of the oxygen concentrator and generates at least one temperature sensing data; and the sound sensing unit senses the operating sound of the oxygen concentrator and generates at least one sound sensing data. The system includes a vibration sensing unit that senses the vibration of the oxygen concentrator and generates at least one vibration sensing data point, an operation sensing unit that senses the operation of the oxygen concentrator and generates at least one operation sensing data point, and a service device having an information classification unit that receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data. The information classification unit analyzes the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data and generates at least one analysis information point. The service device transmits the analysis information to at least one management and monitoring device and at least one designated communication device. The management and monitoring device and the designated communication device learn about the usage status of the oxygen concentrator and any abnormalities before a malfunction from the analysis information and take appropriate action or replace the device.

[0009] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the sensing device further includes a data compression unit and a wireless communication unit. The data compression unit receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data. The data compression unit compresses the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data into a sensing compressed data and transmits the sensing compressed data to the wireless communication unit.

[0010] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes a data decompression unit, which receives the sensing compressed data of the wireless communication unit and decompresses it to restore the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data and transmits them to the information classification unit.

[0011] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes a data comparison unit, which receives oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data restored by the data decompression unit and generates comparison information after comparison.

[0012] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes a parameter data storage unit, and the data comparison unit retrieves at least one oxygen concentration data, at least one oxygen flow rate data, at least one skin color judgment data, at least one body temperature data, at least one body sound data, and at least one body vibration data from the parameter data storage unit, and compares them with oxygen concentration sensing data, oxygen flow rate sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data respectively to generate the comparison information, and the servo device transmits the comparison information to the management and monitoring device and the designated communication device.

[0013] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the data comparison unit retrieves oxygen concentration data, oxygen flow rate data, skin color judgment data, body temperature data, body sound data, and body vibration data from the parameter data storage unit, and compares them with oxygen concentration sensing data, oxygen flow rate sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data respectively to generate an abnormal information, and the server device transmits the comparison information and the abnormal information to the management and monitoring device and the designated communication device.

[0014] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes an information classification unit, which includes a normalization module, a feature selection module, and a cluster classification module. The normalization module receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data and performs normalization processing to obtain oxygen concentration normalized data, oxygen flow normalized data, image normalized data, temperature normalized data, sound normalized data, and vibration normalized data. The feature selection module receives the oxygen concentration normalized data, oxygen flow normalized data, image normalized data, temperature normalized data, sound normalized data, and vibration normalized data and extracts a plurality of independent features and at least one intersection feature. The cluster classification module receives the independent features and the intersection feature and classifies at least one classification result. The classification result and the independent features and the intersection feature are used for machine learning to generate the AI ​​information collection module.

[0015] According to an embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the information classification unit receives oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data. The oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data are normalized by the normalization module, and the feature selection module extracts the independent features and intersection features. The grouping and classification module classifies the data to obtain the classification result. The learned AI information collection module receives the classification result and the independent features and intersection features and makes a judgment to generate the analysis information.

[0016] According to one embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes a compensation module, which can generate at least one compensation data to the normalization module for normalization processing.

[0017] According to one embodiment of the intelligent oxygen concentrator monitoring system of the present invention, the servo device further includes an optimization module, which receives the classification result and converts the classification result into a data format that conforms to the parameter data storage unit and transmits it to the parameter data storage unit. The parameter data storage unit updates the oxygen concentration data, oxygen flow rate data, skin color judgment data, body temperature data, body sound data, and body vibration data according to the classification result. [Simplified Explanation of the Diagram]

[0046] Figure 1 is a block diagram of the intelligent oxygen concentrator monitoring system of the present invention.

[0047] Figure 2 is a block implementation diagram of the training model of the intelligent oxygen concentrator monitoring system of the present invention.

[0048] Figure 3 is a block implementation diagram of the intelligent oxygen concentrator monitoring system of the present invention monitoring the oxygen concentrator in an abnormal state.

[0049] Figure 4 is a schematic diagram of the first block of the intelligent oxygen concentrator monitoring system of the present invention monitoring the oxygen concentrator to a normal state.

[0050] Figure 5 is a schematic diagram of the second block of the intelligent oxygen concentrator monitoring system of the present invention, which monitors the oxygen concentrator to be in normal condition.

[0051] Figure 6 is a schematic diagram of the third-party implementation of the intelligent oxygen concentrator monitoring system of the present invention, which monitors the oxygen concentrator to be in normal condition.

Implementation Method

[0018] The above-mentioned objectives of the present invention and its structural and functional characteristics will be described with reference to the preferred embodiments shown in the accompanying drawings.

[0019] In the following, various applicable examples are listed and described in detail with reference to the accompanying drawings, etc., regarding the structure and technical content of the intelligent oxygen concentrator monitoring system of the present invention; however, the present invention is by no means limited to the listed embodiments, drawings or detailed descriptions.

[0020] Furthermore, those skilled in the art should understand that the listed embodiments and accompanying drawings are for reference and illustration only and are not intended to limit the present invention; any inventions that can be easily implemented based on the description are also considered to be within the scope of the spirit and intent of the present invention, and of course, such inventions are also included in the scope of the patent application of the present invention.

[0021] Furthermore, the directional terms mentioned in the following embodiments, such as "up," "down," "left," "right," "front," and "back," are only for reference to the directions shown in the accompanying illustrations. Therefore, the directional terms used are for illustrative purposes and not for limiting the present invention; moreover, in the following embodiments, the same or similar elements will be labeled with the same or similar element numbers.

[0022] First, please refer to Figure 1, which is a block diagram of the intelligent oxygen concentrator monitoring system of the present invention. As can be clearly seen from the figure, the intelligent oxygen concentrator monitoring system 1 includes at least one sensing device 2 and at least one servo device 3.

[0023] The sensing device 2 is installed on an oxygen concentrator 4, and the monitoring device 2 has an oxygen concentration sensing unit 20, an oxygen flow sensing unit 21, an image capturing unit 22, a temperature sensing unit 23, a sound sensing unit 24, a vibration sensing unit 25, and an operation sensing unit 26. The sensing device 2 is further provided with a data compression unit 27, a wireless communication unit 28, and a power supply unit 29. The data compression unit 27 is signal-connected to the oxygen concentration sensing unit 20, the oxygen flow sensing unit 21, the image capturing unit 22, the temperature sensing unit 23, the sound sensing unit 24, the vibration sensing unit 25, and the operation sensing unit 26. The wireless communication unit 28 is signal-connected to the data compression unit 27. The power supply unit 29 provides operating power to the sensing device 2.

[0024] The servo device 3 is wirelessly connected to the wireless communication unit 28 of the sensing device 2. The servo device 3 has an information classification unit 31. The information classification unit 31 has a normalization module 311, a feature selection module 312, a cluster classification module 313, and an AI information collection module 314 connected in sequence. The normalization module 311 is connected to a complement module 315, and the cluster classification module 313 is connected to an optimization module 316. The feature selection module 312 can use JIVE (Joint and Individual Variation Explained) or multi-view structural learning and other related technologies or software. The cluster classification module 313 can be unsupervised learning or a combination of unsupervised and semi-supervised learning.

[0025] Furthermore, the servo device 3 further includes a data comparison unit 32, a parameter data storage unit 33, and a data decompression unit 34. The data decompression unit 34 is signal-connected to the wireless communication unit 28 and the data comparison unit 32. The data comparison unit 32 is also signal-connected to the parameter data storage unit 33. The parameter data storage unit 33 stores at least one oxygen concentration data 331, at least one oxygen flow rate data 332, at least one skin color judgment data 333, at least one body temperature data 334, at least one body sound data 335, and at least one body vibration data 336.

[0026] The oxygen concentration data 331 is set such that an oxygen concentration of 90% to 96% is a normal concentration, an oxygen concentration of 80% to 90% is a warning concentration, and an oxygen concentration of 70% to 80% or greater than 96% is a danger.

[0027] The oxygen flow rate data 332 is set as follows: the oxygen flow rate is 2L / min to 3L / min as the standard flow rate, the oxygen flow rate is 1L / min to 2L / min as the low flow rate, the oxygen flow rate is greater than 3L / min as the high flow rate, and the oxygen flow rate is less than 1L / min as the abnormal flow rate.

[0028] The skin color judgment data 333 is set as follows: normal skin color is the standard skin color, while blue skin color, purple skin color and black skin color are warning skin colors.

[0029] The body temperature data 334 is set as follows: body temperature of 40℃±10% is the standard temperature, body temperature of 10%~20% above 40℃ is the high temperature, and body temperature of 20% above 40℃ is the abnormal temperature. The body sound data 335 is set as follows: body sound of 60dB~70dB is the standard sound, and body sound of 70dB is the abnormal sound. The body vibration data 336 is set as body acceleration change data.

[0030] The server device 3 is wirelessly connected to at least one management and monitoring device 5 and at least one designated communication device 6. The AI ​​information collection module 314 is an artificial intelligence model. The management and monitoring device 5 and the designated communication device 6 can also be wirelessly connected to the wireless communication unit 28 of the sensing device 2 at the same time. However, in this embodiment, the wireless communication unit 28 is mainly implemented by wirelessly connecting to the server device 3.

[0031] Please refer to Figure 2, which is a block diagram illustrating the implementation of the intelligent oxygen concentrator monitoring system of the present invention. In this embodiment, it represents the training process of the AI ​​information collection module 314 during the training phase. When the oxygen concentrator 4 provides concentrated oxygen to the user, it needs to provide normal oxygen concentration and flow rate. Furthermore, the oxygen concentrator 4 needs to provide concentrated oxygen at normal operating temperature, operating sound, and operating vibration. The oxygen concentration sensing unit 20 of the sensing device 2 senses the oxygen concentration delivered by the oxygen concentrator 4 and generates at least one oxygen concentration sensing data D1. The oxygen flow sensing unit 21 senses the oxygen flow rate delivered by the oxygen concentrator 4 and generates at least one oxygen flow sensing data D. 2. Simultaneously, the image capturing unit 22 continuously captures images of the oxygen concentrator user and generates at least one image data D3. The temperature sensing unit 23 senses the operating temperature of the oxygen concentrator 4 and generates at least one temperature sensing data D4. The sound sensing unit 24 senses the operating sound of the oxygen concentrator 4 and generates at least one sound sensing data D5. The vibration sensing unit 25 senses the vibration of the oxygen concentrator 4 and generates at least one vibration sensing data D6. The operation sensing unit 26 senses the operation of the oxygen concentrator 4 and generates at least one operation sensing data D7. The operation sensing data D7 may include the serial number, service life, number of uses, and usage time of the oxygen concentrator 4, as well as other machine information and machine operation information.

[0032] The oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, after being generated, are first transmitted to the data compression unit 27. The data compression unit 27 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, and compresses the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7 into a sensing compressed data CD1, and then compresses the sensing compressed data CD1 into a sensing compressed data CD1. D1 is transmitted to the wireless communication unit 28, which then transmits its sensed compressed data CD1 to the server device 3. The server device 3 receives the sensed compressed data CD1 and transmits it to the data decompression unit 34. The data decompression unit 34 receives the sensed compressed data CD1 and decompresses it to restore the oxygen concentration sensed data D1, oxygen flow sensed data D2, image data D3, temperature sensed data D4, sound sensed data D5, vibration sensed data D6, and operation sensed data D7. The data decompression unit 34 then transmits the oxygen concentration sensed data D1, oxygen flow sensed data D2, image data D3, temperature sensed data D4, sound sensed data D5, vibration sensed data D6, and operation sensed data D7 to the information classification unit 31.

[0033] After receiving the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, the information classification unit 31 first receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, and vibration sensing data D6 through the normalization module 311. The normalization module 311 aligns the aforementioned 6 types of data with the timestamp and performs normalization processing to obtain a normalized oxygen concentration data D11, a normalized oxygen flow data D21, a normalized image data D31, a normalized temperature data D41, a normalized sound data D51, and a normalized vibration data D61.

[0034] The feature selection module 312 receives the oxygen concentration data D11, oxygen flow rate data D21, image data D31, temperature data D41, sound data D51, and vibration data D61. The feature selection module 312 extracts independent features and common intersection features from the aforementioned 6 sets of data to obtain multiple independent features F1 and at least one intersection feature F2. The grouping and classification module 313 receives the independent features F1 and intersection features F2 and classifies them to obtain at least one classification result R1. The classification result R1 and the independent features F1 and intersection features F2 are used for machine learning to generate the AI ​​information collection module 314.

[0035] Furthermore, the selection of the oxygen concentrator 4 can be determined according to the use of the grouping and classification module 313. The grouping and classification module 313 first performs data grouping using unsupervised learning, and then classifies the grouped data to generate the classification result R1.

[0036] Please refer to Figure 3, which is a block diagram illustrating the implementation of the intelligent oxygen concentrator monitoring system of the present invention monitoring an oxygen concentrator in an abnormal state. When the oxygen concentrator 4 provides concentrated oxygen to the user, it needs to provide normal oxygen concentration and flow rate, and it also needs to provide concentrated oxygen at normal operating temperature and operating sound. The oxygen concentration sensing unit 20 of the monitoring device 2 senses the oxygen concentration delivered by the oxygen concentrator 4 and generates the oxygen concentration sensing data D1. The oxygen flow sensing unit 21 senses the oxygen delivery amount of the oxygen concentrator 4 and generates the oxygen flow sensing data D2. Simultaneously, the image capturing unit... The system continuously captures images of the user receiving oxygen and generates image data D3. The temperature sensing unit 23 senses the operating temperature of the oxygen concentrator 4 and generates temperature sensing data D4. The sound sensing unit 24 senses the operating sound of the oxygen concentrator 4 and generates sound sensing data D5. The vibration sensing unit 25 senses the vibration of the oxygen concentrator 4 and generates vibration sensing data D6. The operation sensing unit 26 senses the operation of the oxygen concentrator 4 and generates operation sensing data D7. The operation sensing data D7 may include the serial number, service life, number of uses, and usage time of the oxygen concentrator 4, as well as other machine information and operation information.

[0037] After the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7 are generated, they are first transmitted to the data compression unit 27. The data compression unit 27 receives the concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, and then compresses the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7. The measured data D7 is compressed into sensing compressed data CD1. The servo device 3 receives the sensing compressed data CD1 and transmits it to the data decompression unit 34. Then, the data decompression unit 34 receives the sensing compressed data CD1 and decompresses it to restore the concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7. The data decompression unit 34 then transmits the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7 to the data comparison unit 32.

[0038] The data comparison unit 32 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, and simultaneously retrieves the oxygen concentration data 331, oxygen flow sensing data 332, skin color judgment data 333, body temperature data 334, body sound data 335, and body vibration data 336 from the parameter data storage unit 33, and performs comparisons of oxygen concentration, oxygen flow, image, temperature, sound, and vibration, generating a comparison information CI. If the data comparison unit 32 finds that the oxygen concentration sensing data D1 and oxygen concentration data 331 are abnormal oxygen concentrations, or if the oxygen flow sensing data D2 and oxygen flow data 332 are abnormal oxygen concentrations, then the comparison unit 32 will perform a comparison of the oxygen concentration sensing data D1 and oxygen flow sensing data D2 and oxygen flow sensing data D7. When an abnormal situation occurs, such as abnormal oxygen flow, abnormal skin color when comparing image data D3 with skin color judgment data 333, abnormal temperature when comparing temperature sensing data D4 with body temperature data 334 and abnormal sound when comparing sound sensing data D5 with body sound data 335, or abnormal vibration when comparing vibration sensing data D6 with body vibration data 336, the data comparison unit 32 generates an abnormal information EI. At the same time, the data comparison unit 32 transmits the comparison information CI and the abnormal information EI along with the operation sensing data D7 to the management and monitoring device 5, and transmits the abnormal information EI to the designated communication device 6, so that the management and monitoring device 5 and the designated communication device 6 can know from the abnormal information EI that the oxygen concentrator 4 has malfunctioned and take immediate action.

[0039] The comparison information CI includes oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, and vibration sensing data D6, so that the management and monitoring device 5 can record the operation and information of the oxygen concentrator 4 through the comparison information CI and the operation sensing data D7.

[0040] Please refer to Figures 4 and 5, which are schematic diagrams of the first and second blocks of the intelligent oxygen concentrator monitoring system of the present invention monitoring the oxygen concentrator to a normal state. The data comparison unit 32 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7. Simultaneously, it retrieves the oxygen concentration data 331, oxygen flow data 332, skin color judgment data 333, body temperature data 334, body sound data 335, and body vibration data 336 from the parameter data storage unit 33 and performs comparisons of oxygen concentration, oxygen flow, image, temperature, sound, and vibration. If the data comparison unit 32 compares the oxygen concentration sensing data D1 with the oxygen flow data D2, the system will perform a comparison. When the concentration data 331 is normal oxygen concentration, and when the oxygen flow rate sensing data D2 and oxygen flow rate data 332 are normal oxygen flow rates, and when the image data D3 and skin color judgment data 333 are normal skin colors, and when the temperature sensing data D4 and body temperature data 334 are normal temperatures, and when the sound sensing data D5 and body sound data 335 are normal sounds, and when the vibration sensing data D6 and body vibration data 336 are normal vibrations, the data comparison unit 32 generates the comparison information CI. At the same time, the data comparison unit 32 transmits the comparison information CI and the operation sensing data D7 to the management and monitoring device 5, so that the management and monitoring device 5 can record the operation and information of the oxygen concentrator 4 through the comparison information CI and the operation sensing data D7.

[0041] Simultaneously, as shown in Figure 5, after the information classification unit 31 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, it first receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, and vibration sensing data D6 through the normalization module 311, and the normalization module 31... The aforementioned six types of data are aligned with the timestamp and normalized to obtain the normalized data of oxygen concentration D11, oxygen flow rate D21, image D31, temperature D41, sound D51, and vibration D61. The feature selection module 312 receives the normalized data of oxygen concentration D11, oxygen flow rate D21, image D31, temperature D41, sound D51, and vibration D61. 312. By extending the data features through feature engineering from the aforementioned 6 sets of formal data, independent features and common intersection features are extracted to obtain the independent feature F1 and intersection feature F2. The group classification module 313 receives the independent feature F1 and intersection feature F2 and classifies at least the classification result R1. The learned AI information collection module 314 receives the classification result R1 and the independent feature F1 and intersection feature F2 and judges to generate at least one analysis information PI. The server device 3 transmits the analysis information PI to the management and monitoring device 5 and the designated communication device 6, so that the management and monitoring device 5 and the designated communication device 6 can know the usage status of the oxygen concentrator 4 and the abnormalities before failure from the analysis information PI and take measures or replace it. This achieves the effect of predicting the usage status of the oxygen concentrator 4 and dealing with it in time before failure, so as to achieve the effects of user safety and preventive maintenance of the machine. It can also achieve the effect of estimating its degradation time and scheduling maintenance in advance, so as to provide users with multiple protections and reduce the service interruption caused by maintenance time.

[0042] Furthermore, when the information classification unit 31 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, if some of the aforementioned six types of sensing data are lost during transmission or data processing, the compensation module 315 can generate at least one compensation data D8. The generation of the compensation data is based on the oxygen concentration sensing data D1, the oxygen flow sensing data D2, the image data D3, or... The average or median value of temperature sensing data D4, sound sensing data D5, or vibration sensing data D6, or statistical values ​​representing their distribution, are transmitted to the normalization module 311 via the complementary value data D8. This allows the normalization module 311 to generate the oxygen concentration normalization data D11, oxygen flow rate normalization data D21, image normalization data D31, temperature normalization data D41, sound normalization data D51, and vibration normalization data D61, thus preventing the normalization module 311 from failing to generate normalized data.

[0043] Furthermore, the server device 3 can wirelessly connect to multiple sets of sensing devices 2, and each sensing device 2 is installed on a different oxygen concentrator 4. Therefore, the server device 3 can receive the comparison information CI and operational sensing data D7 generated by all sensing devices 2, and generate different analysis information PI for different comparison information CI and operational sensing data D7. The analysis information PI is then transmitted to the management and monitoring device 5 and different designated communication devices 6, thereby achieving the monitoring of different oxygen concentrators 4 through the server device 3. The control device 5 can also learn about the usage status and abnormalities before failure of each oxygen concentrator 4 from the different analysis information PI, and handle or replace them. This achieves the goal of predicting the usage status of the oxygen concentrator 4 and handling it in time before failure, so as to achieve the effects of user safety and preventive maintenance of the machine. The server device will also transmit the comparison information CI and the operation sensing data D7 to the management and monitoring device 5. The management and monitoring device 5 can record the operation and information of different oxygen concentrators 4 through the comparison information CI and the operation sensing data D7.

[0044] Please refer to Figure 6, which is a schematic diagram of the third-party block implementation of the intelligent oxygen concentrator monitoring system of the present invention for monitoring the oxygen concentrator to be in normal condition. After receiving the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7, the information classification unit 31 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, and vibration sensing data D6 through the normalization module 311. Furthermore, the normalization module 311 aligns the aforementioned six types of data with the timestamp and performs normalization processing, thereby obtaining the oxygen concentration normalized data D11, oxygen flow rate normalized data D21, image normalized data D31, temperature normalized data D41, sound normalized data D51, and vibration normalized data D61. The feature selection module 312 receives the oxygen concentration normalized data D11, oxygen flow rate normalized data D21, image normalized data D31, temperature normalized data D41, sound normalized data D51, and vibration normalized data D61, and the feature selection module 312 extracts independent features and common features from the aforementioned six sets of normalized data. The independent feature F1 and the intersection feature F2 are obtained from the characteristics of the intersection. The grouping and classification module 313 receives the independent feature F1 and the intersection feature F2 and classifies at least the classification result R1. The optimization module 316 receives the classification result R1 and converts the classification result R1 into a data format that conforms to the parameter data storage unit 33 and transmits it to the parameter data storage unit 33. The conversion method can be to calculate a new reference value using methods such as the distance of individual data or data sets. The parameter data storage unit 33 updates the oxygen concentration data 331 and the oxygen flow rate data 33 according to the classification result R1. 2. Skin color judgment data 333, body temperature data 334, body sound data 335, and body vibration data 336, so that when the data comparison unit 32 receives the oxygen concentration sensing data D1, oxygen flow sensing data D2, image data D3, temperature sensing data D4, sound sensing data D5, vibration sensing data D6, and operation sensing data D7 again, its data comparison unit 32 retrieves the updated oxygen concentration data 331, oxygen flow data 332, skin color judgment data 333, body temperature data 334, body sound data 335, and body vibration data 336 from the parameter data storage unit 33.

[0045] The present invention has been described in detail above. However, the above description is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the present invention. All equivalent changes and modifications made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A smart oxygen concentrator monitoring system, comprising: At least one sensing device is installed on an oxygen concentrator, and the sensing device includes a data compression unit, an oxygen concentration sensing unit, an oxygen flow sensing unit, an image capturing unit, a temperature sensing unit, a sound sensing unit, a vibration sensing unit, and an operation sensing unit. The oxygen concentration sensing unit senses the oxygen concentration delivered by the oxygen concentrator and generates at least one oxygen concentration sensing data point. The oxygen flow sensing unit senses the oxygen delivery rate of the oxygen concentrator and generates at least one oxygen flow sensing data point. The image capturing unit captures an image of a user receiving oxygen and generates at least one image data point. The temperature sensing unit senses the operating temperature of the oxygen concentrator and generates at least one temperature sensing data point. The sound sensing unit senses the operating sound of the oxygen concentrator and generates at least one sound sensing data point. The vibration sensing unit senses the vibration of the oxygen concentrator. The system includes an operating sensing unit that senses the operating status of the oxygen concentrator and generates at least one vibration sensing data point; and a servo device that has an information classification unit that receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operating sensing data, analyzes the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operating sensing data, and generates at least one analysis information point. The servo device transmits the analysis information to at least one management and monitoring device and at least one designated communication device. The management and monitoring device and the designated communication device learn about the usage status of the oxygen concentrator and any abnormalities before a malfunction from the analysis information and take appropriate action or replace the device.

2. The intelligent oxygen concentrator monitoring system as described in claim 1, wherein the sensing device further includes a data compression unit and a wireless communication unit, the data compression unit receiving the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data, and the data compression unit compressing the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data into a sensing compressed data, and transmitting the sensing compressed data to the wireless communication unit.

3. The intelligent oxygen concentrator monitoring system as described in claim 2, wherein the servo device further includes a data decompression unit, which receives and decompresses the sensing and compressed data from the wireless communication unit, and restores it to the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data, and transmits them to the information classification unit.

4. The intelligent oxygen concentrator monitoring system as described in claim 3, wherein the servo device further includes a data comparison unit, which receives oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data restored by the data decompression unit, and generates comparison information after comparison.

5. The intelligent oxygen concentrator monitoring system as described in claim 4, wherein the servo device further includes a parameter data storage unit, and the data comparison unit retrieves at least one oxygen concentration data, at least one oxygen flow rate data, at least one skin color judgment data, at least one body temperature data, at least one body sound data, and at least one body vibration data from the parameter data storage unit, and compares them with the oxygen concentration sensing data, oxygen flow rate sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data respectively, and generates the comparison information, and the servo device transmits the comparison information to the management and monitoring device and the designated communication device.

6. The intelligent oxygen concentrator monitoring system as described in claim 5, wherein the data comparison unit retrieves oxygen concentration data, oxygen flow rate data, skin color judgment data, body temperature data, body sound data, and body vibration data from the parameter data storage unit, and compares them with oxygen concentration sensing data, oxygen flow rate sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data respectively to generate an abnormality information, and the server device transmits the comparison information and abnormality information to the management and monitoring device and the designated communication device.

7. The intelligent oxygen concentrator monitoring system as described in claim 6, wherein the servo device further includes an information classification unit, which includes a normalization module, a feature selection module, a clustering and classification module, and an AI information collection module. The normalization module receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data, and performs normalization processing to obtain normalized oxygen concentration data, normalized oxygen flow data, normalized image data, normalized temperature data, normalized sound data, and normalized vibration data. The feature selection module receives the normalized oxygen concentration data, normalized oxygen flow data, normalized image data, normalized temperature data, normalized sound data, and normalized vibration data, and extracts multiple independent features and at least one intersection feature. The clustering and classification module receives the independent features and the intersection feature and classifies at least one classification result. The classification result and the independent features and the intersection feature are used for machine learning to generate the AI ​​information collection module.

8. The intelligent oxygen concentrator monitoring system as described in claim 7, wherein the information classification unit receives the oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, vibration sensing data, and operation sensing data. The oxygen concentration sensing data, oxygen flow sensing data, image data, temperature sensing data, sound sensing data, and vibration sensing data are normalized by the normalization module, and the feature selection module extracts the independent features and intersection features. The grouping and classification module then classifies the data to produce the classification result. The learned AI information collection module receives the classification result and the independent features and intersection features, and makes a judgment to generate the analysis information.

9. The intelligent oxygen concentrator monitoring system as described in claim 8, wherein the servo device further includes a compensation module, which is capable of generating at least one compensation data to the normalization module for normalization processing.

10. The intelligent oxygen concentrator monitoring system as described in claim 8, wherein the servo device further includes an optimization module, which receives the classification results and converts the classification results into a data format that conforms to the parameter data storage unit and transmits it to the parameter data storage unit, wherein the parameter data storage unit updates the oxygen concentration data, oxygen flow rate data, skin color judgment data, body temperature data, body sound data, and body vibration data according to the classification results.

Images