CN104915360B

CN104915360B - Medical monitoring system and alarm sound reduction method applied to same

Info

Publication number: CN104915360B
Application number: CN201410096114.0A
Authority: CN
Inventors: S·根克; 茹雨; R·J·科丰; R·Q·塔姆; J·A·米尔斯; B·A·弗里德曼
Original assignee: General Electric Co
Current assignee: General Electric Co
Priority date: 2014-03-14
Filing date: 2014-03-14
Publication date: 2020-04-07
Anticipated expiration: 2034-03-14
Also published as: CN104915360A

Abstract

The invention relates to a medical monitoring system and an alarm sound reduction method applied to the medical monitoring system. The medical monitoring system comprises an alarm sequence receiving unit, a data comparison unit and an alarm sound shielding execution unit. The alarm sequence receiving unit is used for receiving an alarm sequence in real time. The data comparison unit is used for identifying whether a preset key alarm mode is contained in the currently received alarm sequence, and the key alarm mode is composed of at least two alarm events. The alarm sound shielding execution unit is used for shielding the alarm sound triggered by the corresponding false alarm event according to the identification result. The invention also relates to an alarm sound reduction method applied to the medical monitoring system.

Description

Medical monitoring system and alarm sound reduction method applied to same

Technical Field

The present invention relates to medical monitoring systems, and more particularly, to a medical monitoring system that generates an alarm during a monitoring process and an alarm sound reduction method applied thereto to reduce the number of occurrences of alarm sounds triggered by false alarm events generated.

Background

In many medical systems (e.g., anesthesia systems applied in operating rooms, emergency care systems applied in intensive care units, etc.), there are medical monitoring systems, so as to monitor patients and medical instruments in real time while performing corresponding specific operations (e.g., performing anesthesia operations and emergency care operations), so as to help doctors or related personnel to know whether the states of the patients and medical instruments are good or not in real time.

For example, during the operation of the anesthesia system, the monitoring system of the anesthesia machine itself and other monitoring devices (such as a heart rate meter) configured around the monitoring system will be used as a complete medical monitoring system of the anesthesia system to perform real-time monitoring on the entire anesthesia operation process, such as monitoring whether the operation of the anesthesia machine itself is normal and monitoring whether physiological data of the patient under anesthesia is normal. The doctor or related personnel can know the state of each aspect in the anesthesia operation process in real time through the real-time monitoring display of the medical monitoring system, so that timely and accurate reaction can be made when problems occur.

In conventional anesthesia systems, there are numerous alarm mechanisms that are designed according to predetermined alarm criteria, such as those set by the U.S. food and drug administration. Generally speaking, when a monitored parameter falls within a predetermined alarm standard range during the monitoring process, the monitoring system generates a corresponding alarm event, which in turn triggers a corresponding alarm signal (including an alarm image and an alarm sound) to notify a doctor or a related person.

In order to reduce the misjudgment rate, the range of the alarm standard is usually set to be large so as to reduce the medical risk as much as possible. However, a large number of alarm signals may ensue, including both real and false alarm signals. Moreover, the false alarm signals are in a considerable proportion, which often causes a trouble to the doctor, i.e. the doctor is often disturbed by the false alarm signals, in particular the sound of the alarm, which affects the subsequent operation. Although the physician may manually input commands to mask off some of the alarms, the masking also results in subsequent occurrences of the same alarm event without sounding an alarm, which may lead to a medical accident.

Of course, many similar medical systems, other than the anesthesia systems and emergency care systems mentioned above, suffer from the same problems, which are not illustrated here. Therefore, a new technology is needed to solve the above contradiction, so as to ensure that the false alarm signal, especially the sound type false alarm signal, does not appear too much while the misjudgment rate is low.

Therefore, it is desirable to provide a new medical monitoring system and an alarm sound reduction method applied thereto to effectively reduce the number of occurrences of alarm sounds triggered by false alarm events generated.

Disclosure of Invention

One or more aspects of the present invention are now summarized to facilitate a basic understanding of the invention, wherein the summary is not an extensive overview of the invention and is intended neither to identify certain elements of the invention, nor to delineate the scope thereof. Rather, the primary purpose of the summary is to present some concepts of the invention in a simplified form prior to the more detailed description that is presented hereinafter.

One aspect of the present invention provides a medical monitoring system, comprising:

an alert sequence receiving unit for receiving an alert sequence in real time;

the data comparison unit is used for identifying whether a preset key alarm mode is contained in the currently received alarm sequence, and the key alarm mode is composed of at least two alarm events; and

and the alarm sound shielding execution unit is used for shielding the alarm sound triggered by the corresponding false alarm event according to the identification result.

Another aspect of the present invention is to provide an alarm sound reduction method applied to a medical monitoring system, including:

determining at least one critical alarm pattern based on historical data, the critical alarm pattern consisting of at least two alarm events;

receiving an alarm sequence generated by the medical monitoring system in real time;

comparing the alert sequence to the key alert pattern;

identifying a portion of the alert sequence that is identical to the critical alert pattern;

determining whether the identified alarm events of the same portion are false alarm events; and

the alarm sounds triggered by the corresponding false alarm events are masked.

Compared with the prior art, the medical monitoring system and the alarm sound reduction method applied to the medical monitoring system obtain the current alarm information by receiving the alarm sequence in real time, and then judge whether the current alarm sequence contains the preset key alarm mode. The key alarm pattern is a predetermined alarm pattern comprising at least two alarm events, typically a type of alarm pattern that occurs most frequently in the historical data or an empirically determined type of alarm pattern. After the part of the alarm sequence which is the same as the key alarm mode is identified, whether the identified key alarm mode is a false alarm event is judged, if so, the alarm sound triggered by the corresponding false alarm event is shielded according to a preset alarm sound shielding rule, for example, one shielding rule can shield the subsequent repeated alarm sound triggered by the current false alarm event, and the other shielding rule can shield the alarm sound triggered by the same key alarm mode which appears in the subsequent operation process. Therefore, the number of false alarm sounds can be greatly reduced, the interference of the false alarm sounds to the follow-up operation of a doctor is further reduced, and the medical diagnosis is ensured to be carried out quickly.

Drawings

The invention may be better understood by describing embodiments thereof in conjunction with the following drawings, in which:

fig. 1 is a schematic diagram of a medical system, exemplified by an anesthesia system.

Fig. 2 is a block diagram of one embodiment of the medical monitoring system of the present invention applied to the anesthesia system of fig. 1.

Fig. 3 is a flow chart of an alarm sound reduction method applied in the medical monitoring system of fig. 2.

Fig. 4 is a block diagram of an alarm event sample determination unit in the medical monitoring system of fig. 2.

Fig. 5 is a flow chart of a method performed by the alarm event sample determination unit of fig. 4.

Fig. 6 is a block diagram of a key alarm pattern recognition unit in the medical monitoring system of fig. 2.

FIG. 7 is a flow chart of a method performed by the key alarm pattern recognition unit of FIG. 6.

FIG. 8 is a process diagram for encoding an exemplary alert sequence.

Fig. 9 is a block diagram of a data comparison unit in the medical monitoring system of fig. 2.

FIG. 10 is a flow chart of a method performed by the data compare unit of FIG. 9.

FIG. 11 is a diagram illustrating a state machine operation performed by the data compare unit of FIG. 9.

Fig. 12 is a block diagram of an alarm sound-shielding execution unit in the medical monitoring system of fig. 2.

Detailed Description

While specific embodiments of the invention will be described below, it should be noted that in the course of the detailed description of these embodiments, in order to provide a concise and concise description, all features of an actual implementation may not be described in detail. It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions are made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.

Unless otherwise defined, technical or scientific terms used in the claims and the specification should have the ordinary meaning as understood by those of ordinary skill in the art to which the invention belongs. The use of "first," "second," and similar terms in the description and claims of the present application do not denote any order, quantity, or importance, but rather the terms are used to distinguish one element from another. The terms "a" or "an," and the like, do not denote a limitation of quantity, but rather denote the presence of at least one. The word "comprise" or "comprises", and the like, means that the element or item listed before "comprises" or "comprising" covers the element or item listed after "comprising" or "comprises" and its equivalent, and does not exclude other elements or items. "connected" or "coupled" and similar terms are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

Referring to fig. 1 and 2, fig. 1 is a schematic diagram of a medical system, which is an example of an anesthesia system 10. Fig. 2 is a block diagram of one embodiment of a portion of a medical monitoring system of the anesthesia system 10 of fig. 1 in which the present invention is implemented. It should be noted that, the anesthesia system 10 is merely used as an example to illustrate a specific application of the medical monitoring system of the present invention, and besides being applied to an anesthesia system, the medical monitoring system may also be applied to other medical systems, such as emergency care systems, and the like, and the specific application is similar, and it is not described here one by one, and the portion that needs local adjustment may be adjusted according to the situation by combining with the prior art.

In the example medical system of fig. 1, the anesthesia system 10 generally includes a computer 100, an anesthesia machine 200, and a patient monitoring device 300. Other ancillary equipment, such as power supplies and the like, not shown for ease of illustration of the medical monitoring system portion of the anesthesia system 10, may also be included. In the present embodiment, the three apparatuses are independent apparatuses, but they may be appropriately integrated as needed, for example, the computer 100 is directly integrated into the anesthesia machine 200, the patient monitoring apparatus 300 is integrated into the anesthesia machine 200, or the three apparatuses are integrated into one apparatus. In other words, the hardware configuration of the anesthesia system 10 can be adjusted as desired, regardless of the configuration.

Wherein, the computer 100 is used as a data learning module in the medical monitoring system of the anesthesia system 10. While the computer 100 is shown as a desktop computer in the embodiment of fig. 1, other embodiments may employ other types of data processing devices, such as microcontrollers, or may process data directly through a remote data processing center, and a desktop computer is shown for illustrative purposes only. Generally speaking, the computer 100 is used to collect historical data generated from past cases of other anesthesia machines and perform corresponding calculation processing on the historical data to obtain a part of data of interest, which is taken as a basis for subsequent operations, and the subsequent paragraphs will further describe the data processing process in detail. For example, when applied to an anesthesia system, the historical data is historical data generated in the past when anesthesia operations are performed on a large number of patients, and the historical data may be obtained from different types of anesthesia machines, or the historical data is from different countries, regions and the like, and the selection of the historical data can be adjusted according to actual needs.

The anesthesia machine 200 is used as an anesthesia module in a medical monitoring system of the anesthesia system 10. In the embodiment of fig. 1, the anesthesia machine 200 may include, as an example, a gas supply unit 201, an anesthetic vaporizer 202, a breathing circuit unit 203, a ventilator 204, a monitoring unit 205, and other auxiliary units (not numbered or shown). In other embodiments, the model and specific configuration of the anesthesia machine 200 can be adjusted according to actual needs, which is only an example. It is to be understood that the anesthesia machine 200 mainly functions to perform an anesthesia operation on a patient, and since the specific anesthesia operation is not a technical problem to be solved by the present invention, the specific procedure of the anesthesia operation will not be described in detail herein, and only a portion related to the technical features of the present invention, that is, the portion shown in fig. 2, will be described in detail, and other portions are not shown in fig. 2. In addition, as mentioned above, the medical monitoring system of the anesthesia system 10 in fig. 2 can be applied not only to anesthesia systems, but also to other medical systems, and only the medical monitoring system in fig. 2 needs to be configured in corresponding devices of other medical systems, and the working principle is similar, so the present invention will not be described in detail with respect to other types of medical systems.

The patient monitoring device 300 is used as a physiological monitoring module in a medical monitoring system of the anesthesia system 10. As an example, the patient monitoring device 300 may include a heart rate monitor 301, an electrocardiogram monitor 302, a blood pressure monitor 303, a respiration monitor 304, and the like. In other embodiments, other types of monitors can be added as needed to provide a comprehensive real-time understanding of the patient physiological data. As described above, the patient monitoring device 300 may also be integrated directly into the anesthesia machine 200.

Referring again to fig. 2, a diagram of one embodiment of the medical monitoring system of the anesthesia system 10 is shown. That is, the parts of the control units illustrated by the computer 100, the anesthesia machine 200 and the patient monitoring device 300 of the anesthesia system 10 of fig. 2 together constitute the medical monitoring system of the anesthesia system 10 for the purpose of effectively reducing the number of alarm sounds triggered by false alarm events that are generated, which is addressed by the present invention, while the other parts of the anesthesia system 10 are not shown in fig. 2.

In fig. 2, the computer 100 as a data learning module includes a historical data storage unit 110, an alarm event sample determination unit 120, a visualization unit 130, and a critical alarm pattern recognition unit 140. In the present embodiment, the above units are independent processing units, and in other embodiments, any two or more units of them may be integrated into one processing unit, or any one of them may be decomposed into two or more processing units.

The history data storage unit 110 is used to store history data generated from past cases. For example, in the case of an anesthesia system, these historical data are historical data generated when a certain number of anesthesia machines have performed an anesthesia operation in the past. Typically, the collected data often requires orders of magnitude, such as 200,000 cases of historical data, and the more historical data that is collected, the more accurate the results of the statistical analysis often become. For some anesthesia machines, the machine log file generated during the anesthesia operation can be pre-stored as the historical data required for collection by the historical data storage unit 110.

Each machine log file may contain a variety of information including, for example, time information, monitoring information, status information, alarm information, and the like. For example, a machine log generated by an anesthesia machine may include, but is not limited to, the following information: case number information, three anesthesia operation stage information, gas change event information, gas discharge event information, alarm event information, user operation information, ventilation mode information, anesthesia time length information, machine model and number information, anesthesia site information, and the like. The machine log file can be said to contain information about the machine itself and information generated during the entire anesthesia procedure, which will be used as the initial data sample to be used as the basis for further processing of subsequent data.

The alarm event sample determination unit 120 serves as a data processing unit for further calculating and determining one or more alarm event samples based on the initial historical data samples. Herein, "alarm event" represents such an event that contains alarm information in the historical data. For example, the alarm events include, but are not limited to, "APNEA" alarm event (APNEA alarm event), "ETCO 2 low" alarm event (low exhaled carbon dioxide alarm event), "ETCO 2 high" alarm event (high exhaled carbon dioxide alarm event), "exhalation low" alarm event (low exhaled carbon dioxide alarm event), "exhalation high" alarm event (high exhaled carbon dioxide alarm event), "O2 low" alarm event (low exhaled carbon dioxide alarm event), and so on. It should be noted that the symbol corresponding to the alarm event may be different according to different machine settings, and the above is only an example. The alarm events are generated in a corresponding state in the anesthesia operation process according to preset alarm standards so as to help doctors to make specific judgment.

Generally, the types of alarm events are very many, for example, more than 100 types, and the number of alarm events included is very large because the base of the historical data is very large. If statistical analysis calculations are performed for all types of alarm events, the amount of calculations will be large, the time consumed will be large, and the data obtained will not necessarily be ideal. Therefore, the alarm event sample determination unit 120 will determine one or more alarm event samples among all types of alarm events as the basis for subsequent data processing. In some embodiments, the "alarm event sample" is the one or more alarm events that occur most frequently among all types of alarm events. In other embodiments, the "alarm event sample" may also be one or more alarm events that are otherwise determined among all types of alarm events, such as by the experience of a physician, or the like. The following paragraphs will provide specific processing methods for determining the alarm event samples in conjunction with fig. 4 and 5.

The key alarm pattern recognition unit 140 serves as a data processing unit for further calculating and determining one or more key alarm patterns based on the above alarm event sample determinations and based on historical data samples. In this context, "alarm mode" means two or more consecutive alarm events, that is to say, in a chronologically generated sequence of alarm events, we define two or more consecutive alarm events as alarm mode, i.e. alarm mode is a combined mode consisting of at least two alarm events. For example, an alarm mode includes two consecutive alarm events (e.g., an "APNEA" alarm event followed by an "ETCO 2 low" alarm event).

The number of different types of alarm patterns in the historical data is also very large, and if statistical analysis calculations are performed on all types of alarm patterns, the calculation amount is large, the consumed time is also large, and the finally obtained data may not be ideal. Therefore, the key alarm pattern recognition unit 140 will determine one or more key alarm patterns among all types of alarm patterns as a basis for subsequent data processing. In some embodiments, the "critical alert mode" is the one or more alert modes that occur most frequently among all types of alert modes. In other embodiments, the "critical alarm mode" may also be one or more alarm modes that are otherwise determined among all types of alarm modes, such as through the experience of a physician or the like. The following paragraphs will give a specific processing method how to determine the key alarm pattern in conjunction with fig. 6 to 8.

The visualization unit 130 is used for displaying the processing results of the two

units

120 and 140. For example, the displayed results may be displayed by the monitoring unit 205 of the anesthesia machine 200, or may be displayed by a display device on the computer 100 or the patient monitoring device 300. The visualization unit 130 is an optional unit, which may not be provided in some embodiments.

Through the above calculations of the

units

110, 120, 140 in the computer 100, the critical alarm patterns required in the subsequent anesthetic operation will be obtained. In other embodiments, the computer 100 may be used as a data processing unit independent of the anesthesia system 10, and is dedicated to calculating and identifying the required critical alarm patterns, and then providing the identified critical alarm patterns as input data to the anesthesia machine 200, and the anesthesia machine 200 performs the subsequent anesthesia operation according to the critical alarm patterns.

With continued reference to FIG. 2, the anesthesia machine 200 is schematically illustrated as an alarm sound masking module that essentially performs the operations of the alarm sound reduction method of the present invention. The alarm sound shielding module includes an alarm sequence receiving unit 210, a data comparing unit 220, a notifying unit 230, a user command input unit 240, and an alarm sound shielding execution unit 250. In the present embodiment, the above units are independent processing units, and in other embodiments, any two or more units of them may be integrated into one processing unit, or any one of them may be decomposed into two or more processing units.

The alarm sequence receiving unit 210 is used for receiving an alarm sequence generated by an alarm generating unit (not shown) of the anesthesia machine 200 in real time. During the operation of the anesthesia machine 200, the monitoring system will determine to generate corresponding alarm events and trigger alarm signals according to the alarm requirements, wherein the generated alarm events will be received by the alarm sequence receiving unit 210 in sequence, i.e. forming an "alarm sequence". For example, an alarm sequence may include an "APNEA" alarm event, an "ETCO 2 high" alarm event, an "exposure low" alarm event, an "O2 low" alarm event, and so forth. Typically, different anesthesia procedures will produce different alarm sequences.

The data comparing unit 220 is configured to obtain a key alarm mode calculated by the computer 100 and obtain an alarm sequence received by the alarm sequence receiving unit 210 in real time, compare the key alarm mode with the alarm sequence received in real time, and determine whether a corresponding alarm event in the alarm sequence can be shielded during an anesthesia operation according to a comparison result, that is, an alarm sound corresponding to the key alarm mode is no longer triggered. The following paragraphs will present specific processing methods for how to determine masked alarm events in conjunction with fig. 9-11.

The notification unit 230 is used to notify the user of the comparison result of the data comparison unit 220, and may be a video notification method, an audio notification method, or both. In some embodiments, the notification unit 230 notifies directly through the monitoring unit 205 on the anesthesia machine 200. In other embodiments, the notification may also be made through displays and speakers on the computer 100 and the patient monitoring device 300.

The user command input unit 240 is used for receiving a command signal input by a user, for example, the input of the command signal is realized through a user interface on the anesthesia machine 200. Wherein the command signal at least comprises an alarm sound shielding command signal for shielding a corresponding alarm event, i.e. the corresponding alarm sound is not triggered during the subsequent anesthetic operation, but the video alarm can also be displayed. For example, when an alarm sound masking command signal is input, the corresponding control parameter of the alarm sound triggered by the corresponding alarm event is set to an invalid value, i.e., the current critical alarm mode does not trigger the subsequent repeated alarm sound, or the same critical alarm mode occurring in the subsequent operation process does not trigger the alarm sound again.

The alarm sound masking execution unit 250 is configured to execute an alarm sound masking command signal, which includes the above-mentioned command manually input by the user and also includes an automatically generated command. The following paragraphs will provide a specific processing method for specifically executing the alarm sound masking command signal in conjunction with fig. 3 and 12.

With continued reference to fig. 2, the patient monitoring device 300 is schematically illustrated as a physiological monitoring module for monitoring physiological data of a patient in real time. In this embodiment, the patient monitoring apparatus 300 includes a physiological data monitoring unit 310 and a physiological data storage unit 320. The physiological data monitoring unit 310 is used for monitoring physiological data of a patient in real time, for example, by monitoring apparatuses such as a heart rate monitor 301, an electrocardiogram monitor 302, a blood pressure monitor 303, a respiration monitor 304, and the like. The physiological data storage unit 320 is used for storing the real-time monitored data for subsequent data processing, such as automatic masking execution for the alarm sound masking execution unit 250, which will be described in detail with reference to fig. 3. In other embodiments, the physiological data monitoring unit 310 and the physiological data storage unit 320 can be a single integrated unit.

Referring to fig. 3, a flow chart of an alarm sound reduction method 400 applied to the medical monitoring system of the anesthesia system 10 of fig. 2 is shown. The method 400 includes the following steps.

In step 411, the history data is collected by the history data storage unit 110, so as to be used as an initial data basis for subsequently obtaining the key alarm mode. As mentioned above, the historical data contains a very large amount of data, for example, including millions of machine log files, which may be obtained from one type of anesthesia machine or from a plurality of types of anesthesia machines, and the place where the sampled anesthesia machine works may be in one place or in several different places, and the collection of data may be selected according to actual needs, and is not limited to the collection of data of a certain type.

In step 412, one or more alarm event samples are determined by the alarm event sample determination unit 120 based on the historical data. In some embodiments, the one or more alarm event samples are the most frequently occurring alarm events of all alarm events in the historical data. In other embodiments, the alarm event samples may be determined based on other criteria or experience. This step 412 specifically includes several sub-steps, which will be described in further detail in subsequent paragraphs.

In step 413, the data processing results of the alarm event sample determination unit 120 are displayed in a video format, for example, sometimes a doctor needs to know the information. In other embodiments, step 413 may be omitted.

In step 414, one or more critical alarm patterns are determined by the critical alarm pattern recognition unit 140 based on the historical data and the determined alarm event samples. Also, the data processing result of this step can be selectively displayed by the visualization process of step 413. This step 414 specifically includes several sub-steps, which will be described in further detail in subsequent paragraphs. In some embodiments, the above steps 411-414 are implemented by corresponding units in the computer 100, and the subsequent steps are implemented by corresponding units in the anesthesia machine 200. In other embodiments, the steps 411-414 may be implemented in the anesthesia machine 200 or via a remote data center.

In step 421, the alarm events generated by the anesthesia machine 200 in real time are received by the alarm sequence receiving unit 210 in the form of an alarm sequence in time sequence. Meanwhile, the data comparison unit 220 continuously compares the last number of alarm events in the current alarm sequence equal to the length of the critical alarm pattern with the critical alarm pattern after the number of alarm events in the alarm sequence is equal to the length of the critical alarm pattern (i.e., the number of alarm events therein). This step 421 specifically includes several sub-steps, which will be described in detail in subsequent paragraphs.

In step 422, the data comparing unit 220 determines in real time whether the last number of alarm events in the current alarm sequence equal to the length of the key alarm pattern is the same as the key alarm pattern, i.e. the last alarm event in the current alarm sequence is the alarm event in the key alarm pattern. If yes, go to

steps

423 and 424; if the result is negative, the step 421 is returned to continue the above determination.

In step 423, if the determination result is yes, the user is notified of the comparison result by the notification unit 230, for example, by video or audio. After the user knows the comparison result, the user can perform corresponding operations as required. For example, if it is empirically determined that the alarm events in the alarm sequence that are identical to the critical alarm pattern are all false alarm events, an alarm sound masking command signal may be input to mask off the alarm sound triggered by the critical alarm pattern during the anesthetic operation (although the video alarm may not be masked), as described above, the masked alarm sound may include the alarm sound triggered by the current critical alarm pattern and the alarm sound triggered by the same critical alarm pattern during the subsequent operation. According to the above description, in some embodiments, the key alarm mode is the alarm mode with the highest frequency among all the alarm modes, so that as long as the corresponding alarm sequence is determined to be a false alarm event and the corresponding alarm sound is masked, the frequency of the occurrence of the alarm sound triggered by the false alarm event can be effectively reduced, the interference of the false alarm sound on the subsequent operation of the doctor is greatly reduced, and the rapid medical diagnosis is ensured. In other embodiments, the key alarm pattern may also be determined based on other criteria, such as from the experience of the physician. In addition, sometimes, although the doctor knows the comparison result, the doctor cannot judge whether the alarm event is true or false according to own experience, so that the doctor may not perform any operation.

In step 424, the alarm sound masking performing unit 250 determines whether an alarm sound masking command signal input by a user is received within a preset time period, for example, 1 minute. If so, step 425 is entered, and if not, step 426 is entered.

In step 425, the alarm sound masking performing unit 250 masks off the alarm sound triggered by the corresponding alarm event based on the alarm sound masking command signal input by the user. In some embodiments, some non-critical alarm events may sound only once, while some critical alarm events may sound several times intermittently until the user takes the appropriate action, i.e., the repetitive alarm sound mechanism described above. In some embodiments, the alarm sound masking performing unit 250 masks all subsequent repeated alarm sounds except the first alarm sound triggered by the current key alarm mode with the repeated alarm sound mechanism based on the alarm sound masking command signal, and in other embodiments, the alarm sound masking performing unit 250 masks all alarm sounds triggered by the same key alarm mode occurring during subsequent operations based on the alarm sound masking command signal. After this step is finished, the procedure will return to step 421 again to repeat the above steps until the whole anesthesia operation is finished.

In step 426, the alarm sound masking implementing unit 250 will read the physiological data acquired by the patient monitoring device 300 in real time. That is, if the user does not perform the manual alarm sound masking operation in step 424, the alarm sound masking performing unit 250 performs an automatic alarm sound masking processing operation.

In step 427, the alarm sound masking performing unit 250 analyzes the real-time acquired physiological data and determines whether the same alarm event in the current alarm sequence as the alarm event in the critical alarm mode is a false alarm event by an appropriate medical criterion. Since the medical standard is not the technical gist of the present invention, the medical standard is not specifically described here. If yes, step 425 will be executed, and if no, step 421 will be returned to. In some embodiments,

steps

423 and 424 may be omitted, i.e., only the automatic determination mode is retained, so that step 426 is directly entered after step 422 is finished. In other embodiments,

steps

426 and 427 may be omitted, i.e., only the manual execution mode is retained, so that step 424 is ended and step 425 is directly entered or step 421 is returned.

As described above, the method 400 may identify in real time during an anesthesia procedure whether the same portion of the generated alert sequence exists as the determined critical alert mode and optionally notify the user after identification. In one aspect, the user may empirically determine whether the alarm event is false for the same portion of the alarm sequence as the critical alarm pattern and input a corresponding alarm sound masking command signal accordingly. On the other hand, if the user has not done anything, the method 400 further determines whether an alarm event for the same portion of the alarm sequence as the critical alarm pattern is false by analyzing the real-time monitored patient physiological data and generates a corresponding alarm sound masking command signal accordingly. Once the corresponding alarm sound shielding command signal is executed, the corresponding alarm sound is shielded in the subsequent operation process, so that the number of false alarm sounds occurring in the subsequent operation process can be greatly reduced.

Referring to fig. 4 and 5 together, a block diagram of the alarm event sample determination unit 120 and a detailed flowchart of the step 412 executed correspondingly are shown. In the present embodiment, the alarm event sample determination unit 120 includes a data structure extraction subunit 121, an alarm information extraction subunit 122, a data filtering subunit 123, and a sample data determination subunit 124. Correspondingly, the step 412 specifically includes the following sub-steps.

In sub-step 4121, the data structure extraction sub-unit 121 constructs a data extraction structure. As described above, since the original historical data contains a very large amount of information, many data are irrelevant data, and therefore a data extraction structure related to the interested data needs to be constructed to extract the interested part of the historical data. For example, in a non-limiting embodiment, one data extraction structure includes, but is not limited to, the following information: case number, case time length, case phase information (e.g., initial phase of anesthesia, stable phase of anesthesia, end phase of anesthesia), alarm event, etc. This is merely an example, and the specific data extraction structure may be determined according to the information to be extracted.

In sub-step 4122, the alarm information extraction sub-unit 122 extracts all information that the data in the history data matches with the data extraction structure based on the constructed data extraction structure. After the step, the data volume for subsequent processing is greatly reduced.

In sub-step 4123, the data filtering sub-unit 123 further filters the extracted data according to some filtering rules. For example, one screening rule is that when the time length of a case is less than a preset time period, the information corresponding to the case is screened out; another screening rule is that if the case phase information is the anesthesia stable phase, the information corresponding to the case is screened out. After this step, the amount of data for subsequent processing is further reduced. The above screening rules are merely exemplary and other embodiments may be adapted. In addition, this step may also be omitted in some embodiments.

In sub-step 4124, the sample data determination subunit 124 determines an alarm event sample based on the filtered historical data. In a non-limiting embodiment, all alarm events are statistically calculated in the filtered historical data, and one or more of the alarm events with the highest frequency of occurrence are selected as the alarm event sample. For example, there are 100 alarm events statistically in the historical data, wherein the two alarm events with the highest frequency of occurrence are determined as the alarm event samples, such as the "ETCO 2" alarm event and the "APNEA" alarm event. While the "ETCO 2" alarm event may also specifically include sub-alarm events such as an "ETCO 2HIGH ON" alarm event, an "ETCO 2LOW ON" alarm event, etc., the "APNEA" alarm event may also specifically include sub-alarm events such as an "APNEA Vol ON" alarm event, an "APNEA-SYSTEM ON" alarm event, an "APNEA CO2 On" alarm event, an "APNEA FOR MORE THAN2MINUTES ON" alarm event, etc. It should be noted that the symbols such as "ETCO 2" are machine symbols, which represent different types of alarm events, and the machine symbols written by different anesthesia machines may be different, which is only an example.

Referring to fig. 6 and 7 together, a block diagram of the key alarm pattern recognition unit 140 and a detailed flowchart of the step 414 executed correspondingly are shown. In this embodiment, the key alarm pattern recognition unit 140 includes a data encoding subunit 141, a data pattern recognition subunit 142, a data pattern statistics subunit 143, and a key pattern determination subunit 144. Correspondingly, this step 414 specifically includes the following sub-steps.

In sub-step 4141, the data encoding sub-unit 141 encodes the alarm event corresponding to each case based on the filtered historical data and the data encoding rule preset according to the alarm event sample. Since the machine language is relatively complex and inconvenient for calculation, the step is used for coding the machine language. In the present embodiment, the machine language is coded as a single letter and a single number. In other embodiments, the encoding may be performed as other types of codes or without encoding.

FIG. 8 illustrates an example of coding, shown in the leftmost box, of local alarm events occurring in a case, whose machine languages are "LOW EXPIR MV-Vent on" and "LOW EXPIR MV-System on" …, respectively, after first alphabetical coding, are coded as "x", "x", "e", "V", "S", "C", "x", "x". In this embodiment, the two determined alarm event samples are an "ETCO 2" alarm event and an "APNEA" alarm event, so the corresponding machine languages "ETCO 2HIGH On", "ETCO 2LOW On", "APNEA Vol On", "APNEA-SYSTEM", "APNEA CO2 On", "APNEA FOR THAN2MINUTES On" are encoded as: "E", "E", "V", "S", "C", "A", and other non-alarm event sample alarm events are all encoded as "x". After the second numeric encoding, the letter codes "E", "E" are encoded as "2", and "V", "S", "C", "a" are encoded as "1". Thus, the final encoded result is "x", "x", "2", "1", "1", "1", "x", "x", "x". In this embodiment, encoding is performed twice, and in other embodiments, encoding may be performed once or more.

In sub-step 4142, the data pattern identification subunit 142 generates all alert patterns based on a preset pattern length range, e.g., 2 to 6. The "pattern length range" herein refers to a range of the number of alarm events included in the alarm pattern, which must be greater than or equal to 2. For example, if the pattern range is 2 to 6, then an alarm pattern is provided for as many consecutive alarm events as are in the range of 2 to 6. In the example of fig. 8, digitally encoded alert patterns are included: "2111","211","111","21","11".

In sub-step 4143, the data pattern statistics sub-unit 143 will perform statistics on all identified alarm patterns and analyze the frequency of occurrence of each alarm pattern in the historical data. For example, one alarm mode "211" (i.e., one "ETCO 2" alarm event followed by two "APNEA" alarm events) occurs most frequently, such as about 80%.

In sub-step 4144, the critical mode determination subunit 144 determines one or more critical alarm modes based on the statistical results described above. For example, the digitally encoded alert pattern "211" is the most frequently occurring alert pattern. Thus, in digital encoding, the alert mode "211" may be identified as a critical alert mode. Further, the digitally encoded alarm pattern "211" includes 32 kinds of alarm patterns in the letter code pattern, for example, "eVV", "eVS", "eVC", "evaa", "eSS", "eSV", "eSC", "evaa", "eCA", "eCC 64", "eCS", "eCA", "eAV", "eAS", "eAC", "eAA", "EVV", "EVS", "EVC", "eVA", "eSS", "ESV", "eSC", "ESA", "eCC", "ECV", "eCS", "eCA", "EAV", "EAS", "EAC", "EAA", respectively. The alarm patterns in 32 may be further ranked in terms of frequency of occurrence, and then one or more letter-coded key alarm patterns may be determined therefrom, for example, two of the alarm patterns "eVS" and "eCS" having the highest frequency of occurrence may be determined as key alarm patterns. In other embodiments, the determination of the critical alarm pattern may also be determined based on other requirements or based on experience, not just the manner in which the highest frequency occurs as described above.

Referring to fig. 9 and fig. 10 together, a block diagram of the data comparison unit 220 and a detailed flowchart of the step 421 executed correspondingly are shown. In this embodiment, the data comparing unit 220 includes a data encoding subunit 221 and a data comparing subunit 222. Correspondingly, this step 421 specifically includes the following sub-steps.

In sub-step 4211, during the anesthesia operation performed by the anesthesia machine 200, the data encoding sub-unit 221 performs real-time receiving and encoding on the alarm sequence received by the alarm sequence receiving unit 210, and the encoding process is the same as the above-mentioned letter encoding process, which is not described herein again. The data encoding subunit 221 will generate a corresponding letter code, e.g., "ESe V S C …"

In sub-step 4212, the data comparison sub-unit 222 first determines whether the number of alarm events in the current alarm sequence is greater than or equal to the length of the critical alarm pattern, and if not, returns to step 4211, and if so, proceeds to step 4213. For example, assuming that the key alarm patterns are "eVS" and "eCS", the length of the key alarm pattern is 3. Therefore, when the number of alarm events in the alarm sequence is greater than or equal to 3, step 4213 is entered, otherwise step 4211 is returned.

In sub-step 4213, the data comparison sub-unit 222 compares the last consecutive alarm event of the current encoded alarm sequence with the encoded key alarm pattern. That is, when the number of alarm events in an alarm sequence is equal to or greater than the length of a critical alarm pattern, the last alarm event in the alarm sequence that occurred and the number of alarm events equal to the length of the critical alarm pattern will be compared to the critical alarm pattern.

As described above, in step 422 of fig. 3, the data comparison unit 220 will determine whether the last alarm events in the alarm sequence that occurred and are equal in number to the length of the critical alarm pattern are the same as the critical alarm pattern. In some embodiments, the data comparison unit 220 employs a state machine to implement the comparison determination described above. For example, referring to fig. 11, a state machine corresponding to a critical alarm pattern "eVS" can identify whether a certain alarm sequence in the alarm sequence is the same as the critical alarm pattern "eVS". Since the state machine is a commonly used algorithm, it is not specifically described here. In other embodiments, other algorithms may be used.

Referring to fig. 12, a block diagram of the alarm sound masking execution subunit 250 is shown. The alarm sound masking execution subunit 250 includes a masking command determination subunit 251, a physiological data reading subunit 252, a false alarm analysis subunit 253, and an alarm sound masking subunit 254. The sub-units perform the

steps

424, 426, 427, and 425, respectively. In other embodiments, two or more of the above subunits may be integrated into one unit.

In the embodiment given above, the anesthesia system 10 is taken as a specific example to describe the technical solution of the medical monitoring system (fig. 2) of the present invention applied in the medical system. It should be emphasized that the anesthesia system 10 is merely an example of a specific application for convenience of describing the technical solution of the present invention, and the medical monitoring system and the alarm sound reduction method thereof of the present invention can also be applied to other medical systems, such as the emergency care system in intensive care unit, etc. mentioned above, and are not limited to the anesthesia system.

While the invention has been described in conjunction with specific embodiments thereof, it will be understood by those skilled in the art that many modifications and variations may be made to the invention. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit and scope of the invention.

Claims

1. A medical monitoring system, characterized in that the medical monitoring system comprises:

an alert sequence receiving unit for receiving an alert sequence in real time;

an alarm sound shielding execution unit, configured to shield an alarm sound triggered by a false alarm event corresponding to an alarm event in a preset key alarm mode according to the recognition result of whether the preset key alarm mode is included in the currently received alarm sequence,

wherein the medical monitoring system further comprises a data learning module for obtaining the key alarm pattern based on the historical data, the key alarm pattern being one or more alarm patterns with highest frequency of occurrence among all alarm patterns in the historical data.

2. The medical monitoring system of claim 1, wherein the data learning module comprises:

a history data storage unit for storing history data;

an alarm event sample determination unit for determining at least two alarm event samples, which are alarm events with the highest frequency of occurrence among all alarm events in the historical data; and

a key alarm pattern recognition unit for determining the key alarm pattern, which is the alarm pattern with the highest frequency of occurrence among all alarm patterns containing only the alarm event samples.

3. The medical monitoring system of claim 1, wherein the medical monitoring system further comprises:

a notification unit for issuing a notification signal to a user when the data comparison unit identifies that the key alarm pattern matches an alarm event in a currently received alarm sequence; and

a user command input unit for receiving an alarm sound shielding command signal input by a user;

the alarm sound shielding execution unit shields the alarm sound triggered by the corresponding false alarm event based on the alarm sound shielding command signal input by the user.

4. The medical monitoring system of claim 1 or 3, wherein the medical monitoring system further comprises:

the physiological monitoring module is used for monitoring physiological data of a patient to be monitored in real time;

the alarm sound shielding execution unit analyzes whether an alarm event matched with the key alarm mode in the currently received alarm sequence is a false alarm event or not based on the real-time monitored physiological data, and shields the alarm sound triggered by the corresponding false alarm event when the judgment result is yes.

5. A method for reducing alarm sounds for use in a medical monitoring system, the method comprising:

comparing the alert sequence to the key alert pattern;

determining whether an alarm event of a portion of the alarm sequence that is the same as the critical alarm pattern is a false alarm event; and

masking the alarm sound triggered by the corresponding false alarm event,

wherein the step of determining the critical alarm mode comprises:

collecting historical data;

determining at least two alarm event samples, wherein the alarm event samples are the alarm events with the highest occurrence frequency in all the alarm events in the historical data; and

the critical alarm pattern is determined, which is the most frequently occurring alarm pattern of all alarm patterns containing only the alarm event sample.

6. The alarm sound reduction method of claim 5, wherein the step of determining whether the identified alarm events of the same portion are false alarm events comprises:

issuing a notification signal to a user and receiving an alarm sound masking command signal input by the user upon identifying that the key alarm pattern matches an alarm event in a currently received alarm sequence;

the step of masking the alarm sound triggered by the corresponding false alarm event is accomplished based on an alarm sound masking command signal input by a user.

7. The alarm sound reduction method of claim 5 or 6, wherein the step of determining whether the identified alarm events of the same portion are false alarm events comprises:

monitoring physiological data of a patient to be monitored in real time; and

analyzing whether an alarm event in the currently received alarm sequence that matches the critical alarm pattern is a false alarm event based on the real-time monitored physiological data.

8. The alarm sound reduction method of claim 5, wherein the medical monitoring system applied thereto comprises a medical monitoring system applied to an anesthesia system or an emergency care system.