System and method for assessing the reliability of a patient's early warning score

Abstract

A system is provided for assessing the reliability of an early warning score (EWS). The system receives patient test data and determines the patient's EWS. A real-time feature extractor extracts features from the patient test data. A reliability score evaluator generates a reliability score for the EWS by processing the extracted features through a reliability score regression model. An inference engine generates an inference based on the reliability score and the extracted features. The inference may be displayed in a user interface. The reliability score regression model may be determined via deep learning training. A training portion of the system receives training datasets. A data annotator assigns reliability annotations to each training dataset. A training feature extractor generates extracted training features from the training datasets. A deep learning trainer generates a reliability score regression model using the extracted training features and the reliability annotations.

Description

[Technical field]

[0001] The present disclosure relates generally to systems and methods for assessing the reliability of an early warning score (EWS) determined for a patient. [Background technology]

[0002]

[0002] An Early Warning Score (EWS) assesses a patient's risk of developing a condition before that condition manifests. The EWS is determined based on patient data, which may be determined through review of the patient's measurements or patient records. The reliability of the determined EWS depends on multiple factors, such as the availability of measurements, the age of the measurements, and the signal quality index (SQI) of the measurements. Due to this variable reliability, physicians who receive an EWS assessment of their patients require a concomitant assessment of the reliability of the EWS assessment itself to aid in the decision-making process regarding potential interventions and treatments. Summary of the Invention [Problem to be solved by the invention]

[0003] Thus, there is a need in the art for systems and methods for assessing the reliability of an EWS and providing said assessment to medical professionals. [Means for solving the problem]

[0004]

[0003] The present disclosure is generally directed to a system and method for assessing the reliability of an Early Warning Score (EWS) determined for a patient through training and implementation of a reliability score regression model that processes features extracted from patient test data to determine a reliability score for the corresponding EWS.

[0005]

[0004] The system receives various patient test data corresponding to a patient and uses the patient test data to determine the patient's EWS. A real-time feature extractor generates one or more extracted features from the patient test data. A reliability score evaluator then generates a reliability score for the determined EWS by processing the extracted features through a reliability score regression model. The reliability score may then be displayed on a user interface.

[0006]

[0005] An inference engine can be used to generate one or more inferences based on the confidence scores and the extracted features. The inferences can be displayed in a user interface and provided as notifications to a medical professional.

[0007]

[0006] The confidence score regression model can be determined via deep learning training. A training portion of the system receives a plurality of training datasets. The training datasets are provided to a data annotator to evaluate the confidence of training EWSs of the training datasets. The data annotator assigns a confidence annotation to each training dataset based on the evaluation. The evaluation can be performed manually or automatically. A training feature extractor generates one or more extracted training features from each of the plurality of training datasets. The deep learning trainer generates the confidence score regression model using the extracted training features and the confidence annotations.

[0008] In general, in one aspect, a system for assessing an early warning score (EWS) of a patient is provided that includes a test data receiver configured to receive test data of the patient.

[0008]

[0009] The system further includes an EWS evaluator configured to determine the EWS based on patient test data, according to one example, at least a portion of which is collected by a patient monitor.

[0010]

[0009] The system further includes a real-time feature extractor configured to generate one or more extracted features from the patient examination data.

[0011]

[0010] The system further includes a confidence score evaluator configured to generate a confidence score corresponding to the EWS based on the one or more extracted features and the confidence score regression model.

[0012] According to one example, the system further includes an inference engine. The inference engine is configured to generate one or more inferences based on at least one of the confidence scores and the one or more extracted features. The inference engine may be further configured to display at least one of the one or more inferences via a user interface. The inference engine may be further configured to generate a notification corresponding to at least one of the one or more inferences.

[0013] According to another example, the system further includes a training data receiver configured to receive a plurality of training data sets, each of the plurality of training data sets including a training EWS and one or more training features.

[0014]

[0013] The system further includes a data annotator. The data annotator is configured to assign a confidence annotation to each of the plurality of training data sets based on the training EWS. According to one example, the data annotator assigns the at least one confidence annotation based on a user input. According to another example, the data annotator assigns the at least one confidence annotation based on a closeness to an EWS threshold. According to a further example, the data annotator assigns the at least one confidence annotation based on a count exceeding an EWS threshold. The count exceeding an EWS threshold may be determined for a predetermined period of time. According to yet another example, the data annotator assigns the at least one confidence annotation based on an EWS moving window.

[0015]

[0014] The system further includes a training feature extractor configured to generate one or more extracted training features from each of the plurality of training data sets. According to one example, the one or more extracted features include at least one of: measurement availability, measurement expiration (expiration), measurement interruption, feature age, feature value, short-term delta feature, long-term delta feature, and signal quality index (SQI).

[0016] According to one example, each of the one or more extracted features corresponds to one or more patient features based on the patient exam data, which may include at least one of heart rate, oxygen saturation, respiratory rate, temperature, diastolic blood pressure, systolic blood pressure, patient age, pulse pressure, approximate mean arterial pressure, and shock index.

[0017]

[0016] The system further includes a deep learning trainer configured to generate a confidence score regression model based on the confidence annotations and the one or more extracted training features.

[0018]

[0017] In general, in another aspect, a method is provided for assessing the reliability of an EWS of a patient. The method includes receiving a plurality of training data sets, where each of the plurality of training data sets includes a training EWS and one or more training features. The method further includes assigning a reliability annotation to each of the plurality of training data sets via a data annotator. The method further includes generating one or more extracted training features from each of the plurality of training data sets via a training feature extractor. The method further includes generating a reliability score regression model based on the reliability annotations and the one or more extracted training features via a deep learning trainer. The method further includes receiving patient test data. The method further includes determining an EWS based on the patient data via an EWS evaluator. The method further includes generating one or more extracted features from the patient test data via a real-time feature extractor. The method further includes generating a reliability score corresponding to the EWS based on the one or more extracted features and the reliability score regression model via a reliability score evaluator.

[0019]

[0018] In various implementations, a processor or controller may be associated with one or more storage media (collectively referred to herein as "memory", e.g., volatile and non-volatile computer memory such as RAM, PROM, EPROM, EEPROM, floppy disk, compact disk, optical disk, magnetic tape, SSD, etc.). In some implementations, the storage media may be encoded with one or more programs that, when executed on one or more processors and / or controllers, perform at least some of the functions described herein. The various storage media may be fixed within a processor or controller or may be portable, such that one or more stored programs can be loaded into a processor or controller to implement various aspects described herein. The term "program" or "computer program" is used herein in a generic sense to refer to any type of computer code (e.g., software or microcode) that can be used to program one or more processors or controllers.

[0020]

[0019] It should be understood that all combinations of the above concepts and additional concepts described in more detail below (where such concepts are not mutually inconsistent) are contemplated as part of the inventive subject matter disclosed herein. In particular, all combinations of subject matter recited in the claims appearing at the end of this disclosure are considered to be part of the inventive subject matter disclosed herein. It should also be understood that terms explicitly used in this specification and that may also appear in any disclosures incorporated by reference should be given the meaning that is most consistent with the particular concepts disclosed herein.

[0021]

[0020] These and other aspects of the various embodiments will be apparent from and elucidated with reference to the embodiments described hereinafter.

[0022]

[0021] In the drawings, like reference characters generally refer to the same parts throughout the different views. Also, the drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of various embodiments. [Brief description of the drawings]

[0023] [Figure 1] FIG. 1 is a system diagram of a system for assessing a patient's early warning score, according to an example. [Diagram 2] FIG. 2 is a system diagram of a training subsystem for a confidence score regression model, according to an example. [Diagram 3]

[0024] FIG. 3 is a user interface displaying early warning scores, confidence scores, inferences, and patient characteristics, according to one example. [Figure 4]

[0025] FIG. 4 is a flow chart of a method for determining a reliability score for an early warning score, according to an example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024]

[0026] The present disclosure is generally directed to a system and method for assessing the reliability of an Early Warning Score (EWS) determined for a patient through training and implementation of a reliability score regression model that processes features extracted from patient test data to determine a reliability score for the corresponding EWS.

[0025]

[0027] The system receives various patient test data corresponding to a patient and uses the patient test data to determine an EWS for the patient. The patient test data can be obtained from various patient measurements and from patient records. The patient test data can be the basis for determining various patient characteristics such as heart rate, oxygen saturation, respiratory rate, temperature, diastolic blood pressure, systolic blood pressure, patient age, pulse pressure, approximate mean arterial pressure, and shock index.

[0026]

[0028] A real-time feature extractor generates one or more extracted features from the patient test data. The extracted features describe the patient test data and the patient measurements used to determine the patient characteristics. The extracted features are therefore an important indicator of the reliability of the determined EWS. The extracted features may be a variety of characteristics such as measurement availability, measurement expiration, measurement interruption, feature age, feature value, short-term delta feature, long-term delta feature, and signal quality index (SQI).

[0027]

[0029] A confidence score evaluator then generates a confidence score for the determined EWS by processing the extracted features through a confidence score regression model. The confidence score can be normalized to be a number between 0 and 1, where 0 corresponds to the least reliable EWS and 1 corresponds to the most reliable EWS. The confidence score can then be displayed on a user interface or provided to a medical professional via some other practical means.

[0028]

[0030] An inference engine can be used to generate one or more inferences based on the confidence score and the extracted features. For example, if the confidence score is low and the extracted features indicate a low SQI for the heart rate measurement, the inference can suggest that the medical professional check the ECG leads and / or electrodes and then re-perform the measurement. The inference can be displayed on the user interface and provided to the medical professional by a notification. The notification can be visual and / or audible.

[0029]

[0031] The confidence score regression model can be determined via deep learning training. A training portion of the system receives a plurality of training data sets, each of which includes a training EWS and one or more training features.

[0030]

[0032] The training data sets are provided to a data annotator to assess the reliability of the training EWS of the training data sets. The data annotator assigns a reliability annotation to each training data set based on the assessment. The assessment may be performed manually, such as by a medical professional reviewing the training EWS and inputting an assessment of the reliability of the training EWS. In other examples, the assessment is performed automatically, such as by assessing the closeness of the training EWS to an EWS threshold, counting the number of times a set of training EWS exceeds an EWS threshold, or assessing the variability of a set of training EWS based on an EWS variation window.

[0031]

[0033] A training feature extractor generates one or more extracted training features from each of the multiple training datasets. A deep learning trainer generates a confidence score regression model using the extracted training features and the confidence annotations.

[0032]

[0034] FIG. 1 shows a system diagram of a system 100 for evaluating a patient's EWS 10. The system 100 includes a test data receiver 102. The test data receiver 102 receives patient test data 12 from one or more sources. The patient test data 12 may be derived from various patient measurements and / or patient records. For example, the patient test data 12 may include patient measurements such as heart rate, blood pressure, electrocardiogram signals, blood oxygen levels, etc. These measurements may be obtained by a variety of devices communicatively coupled to a patient monitor 300. The monitor 300 may then pass these measurements to the test data receiver 102 in the form of patient test data 12. The test data receiver 102 may also receive various patient records, which may include demographic information (age, ethnicity, etc.) and historical information regarding the health history of the patient and the patient's family. The test data receiver 102 may be configured as a wired and / or wireless communication component or subcomponent.

[0033]

[0035] The test data receiver 102 then communicates the patient test data 12 to the EWS evaluator 104. The EWS evaluator 104 determines an EWS 10 based on the patient test data 12. The EWS 10 represents the risk of a pre-occurring condition progressing. For example, as shown in FIG. 3, the EWS 10 may be a Hemodynamic Stability Index, which represents the risk of a patient developing hemodynamic instability. The EWS 10 may be normalized to be on a scale of 0 to 1, where 0 represents the lowest risk and 1 represents the highest risk. Although the EWS 10 is an important predictive and therapeutic tool, the reliability of the EWS 10 may vary based on a variety of factors. Thus, by providing a reliability score 16 associated with the EWS 10, medical professionals can quickly assess the predictive risk a patient faces and the reliability of the assessment before taking action regarding treatment and / or other interventional procedures. As shown in FIG. 3, the EWS 10 may be displayed on a user interface 18. In FIG. 3, EWS10 is a hemodynamic stability index score of 0.59.

[0034]

[0036] The EWS 10 may also provide notifications 20, such as visual and / or audio notifications, to the medical professional. The EWS 10 indications and / or notifications 20 may correspond to the criticality of the EWS 10.

[0035]

[0037] The test data receiver 102 also communicates the patient test data 12 to a real-time feature extractor 106. The real-time feature extractor 106 "extracts" features of the patient test data 12 that are related to the reliability of the EWS 10. The extracted features may be a variety of characteristics, such as availability of a measurement, expiration date of a measurement, interruption of a measurement, feature age, feature value, short-term delta feature, long-term delta feature, and SQI. Some or all of these extracted features may correspond to one or more patient characteristics 70, such as heart rate, oxygen saturation, respiratory rate, temperature, diastolic blood pressure, systolic blood pressure, patient age, pulse pressure, approximate mean arterial pressure, and shock index.

[0036]

[0038] Measurement availability may be a binary indication of whether a measurement (such as heart rate) is currently available. Similarly, measurement expiration may be a binary indication of whether a measurement is currently available. Additionally, measurement pause may be a binary indication of whether a measurement has stopped collecting data. Feature age indicates the period since the last measurement corresponding to that feature (such as 60 seconds since the last oxygen saturation measurement). Feature value indicates the value of the measurement corresponding to that feature (such as heart rate at 80 beats per minute). Short-term and long-term delta indicate the variability of a measurement (such as temperature) over a short or long period of time. SQI indicates the signal quality associated with the measurement corresponding to that feature on a scale of 0 to 1. By analyzing these extracted features 14 as a whole using a reliability score evaluator 108, the system 100 can quickly determine a reliability score 16 for the EWS 10.

[0037]

[0039] The real-time feature extractor 106 communicates the extracted features 14 to the reliability score evaluator 108. The reliability score evaluator 108 calculates a reliability score 16 by processing the extracted features 14 through a reliability score regression model 110. As described in more detail below, the reliability score regression model 110 is trained by a training subsystem 200 using a deep learning trainer 206 and a training dataset 50. Thus, the reliability score regression model 110 learns how particular extracted features 16 (and their associated values) correlate to the reliability of the EWS 10. For example, a high degree of short-term variability (delta) in a diastolic blood pressure measurement may correspond to a low reliability of the EWS 10 with respect to hemodynamic stability. As such, this measurement will result in a lower reliability score 16. Conversely, a low degree of long-term variability in a diastolic blood pressure measurement may correspond to a higher reliability of the EWS 10 related to hemodynamic stability and may result in a higher reliability score. On the other hand, the value of the patient's temperature would not be particularly indicative of the reliability of the EWS 10 with respect to hemodynamic stability and would therefore not result in an increase or decrease in the reliability score 16.

[0038]

[0040] The reliability score 16 may be normalized so that the reliability score 16 is between 0 and 1, where a reliability score 16 of 0 represents the lowest reliability while a reliability score 16 of 1 represents the highest reliability. For example, Figure 3 shows a user interface 114 displaying a hemodynamic score index with a reliability score 16 of 0.4 out of 1. As such, the hemodynamic score index has slightly less than medium reliability.

[0039]

[0041] Once the confidence score 16 has been calculated, the inference engine 112 generates one or more inferences 18 based on the confidence score 16 and the extracted features 14. The inferences 18 can provide the medical professional with further details regarding the confidence score 16 as well as provide the medical professional with suggestions to improve the confidence score 16. Figure 3 shows a user interface 114 displaying two inferences 18, "Update Lab Measurements" and "Check ECG Leads (Heart Rate SQI Low)". These inferences 18 provide the medical professional with suggestions to improve the hemodynamic score index from 0.4 to a number indicating a higher confidence by checking the ECG leads and updating the measurements.

[0040]

[0042] 3, the inferences 18 can be displayed on a user interface 114. The user interface 114 can include a variety of other information, such as the confidence score 16 and one or more patient characteristics 70. In one example, the user interface 114 is a touch screen integrated into the patient monitor 300. As another example, the user interface 114 can be an element of a smartphone, a personal computer, a tablet computer, or some other suitable device.

[0041]

[0043] Additionally, the inferences 18 may be provided to the medical professional as one or more notifications 20, such as visual and / or audio notifications. The visual / audio notification 20 of the inference 18 may correspond to the severity of the inference 18. For example, the inference 18 "Check ECG leads (heart rate SQI low)" may correspond to an audible tone (or series of audible tones) notification 20 emitted by the patient monitor 300. In this example, the notification 20 may include a flashing LED or a set of pixels within the user interface 114. The notification 20 may also be communicated to a device operated by the medical professional, such as a smart phone, personal computer, or tablet computer.

[0042]

[0044] 2 shows a system diagram of a training subsystem 200 for the deep learning training confidence score regression model 110. As shown in FIG 2, a training data receiver 208 receives multiple training data sets 50. Each training data set 50 includes a training EWS 52 and one or more training features 54. Similar to the extracted features 14, the training features 54 can include measurement availability, measurement expiration, measurement interruption, feature age, feature value, short term delta feature, long term delta feature, and SQI.

[0043]

[0045] As shown in FIG. 2, the training data set 50 is provided to a data annotator 202. The data annotator 202 is configured to assess the reliability of the training EWSs 52 in the training data set 50. The data annotator 202 assigns a reliability annotation 56 to each training data set 50 based on the assessment. The reliability annotation 56 may be binary (reliable or not) or may be in the form of a numerical scale. The assessment may be performed manually, such as by a trained medical professional reviewing the training EWS 52 and inputting an assessment of the reliability of the training EWS 52. For example, the trained medical professional may determine that the training EWS 52 is inaccurate based on variability of the training EWS 52 over a period of time. Additionally, the assessment of inaccuracy may be based on apparent discrepancies between the training EWS 52 and the training features 54 (e.g., feature values).

[0044]

[0046] Manual evaluation can be time consuming and expensive. Thus, in other examples, the data annotator 202 performs the evaluation automatically according to one or more programmable evaluation criteria. In one example, the data annotator 202 evaluates the reliability of the training EWS 52 based on the closeness of the training EWS 52 to an EWS threshold 60. In this example, the reliability annotation 56 may correspond to a decrease in reliability if the training EWS 52 is close to or exceeds the EWS threshold 60. The EWS threshold 60 may be an "optimal cutoff" for the training EWS 52.

[0045]

[0047] In another example, data annotator 202 evaluates the reliability of a set of training EWSs 52 by counting instances where the training EWSs 52 exceed the EWS threshold 60 over a predetermined period 66, such as 30 minutes. In a further example, data annotator 202 evaluates the reliability of a set of training EWSs 52 by evaluating the variation of the EWS training scores 52 over an EWS variation window 68, such as 30 minutes.

[0046]

[0048] Any aspect of the automatic data annotator 202 described above may be adjusted according to hospital and / or clinician preferences or definitions of reliability.

[0047]

[0049] While the data annotator 202 is assigning confidence annotations 56, the training feature extractor 204 generates one or more extracted training features 58 from each of the multiple training data sets 50. The training feature extractor 204 operates similarly to the real-time feature extractor 204 shown in FIG.

[0048]

[0050] The reliability annotations 56 and extracted training features 58 are provided to a deep learning trainer 206. The deep learning trainer 206 evaluates the correlation between each set of reliability annotations 56 and the extracted training features 58. For example, unavailability of an important measurement or a low SQI for an important measurement may be highly correlated with a low reliability annotation 56. Conversely, the age of the patient may not be correlated with the value of the corresponding reliability annotation 56. The deep learning trainer 206 uses these correlations to generate a reliability score regression model 110. As previously mentioned, the reliability score regression model 110 may be used in this case to generate a reliability score 16 for the EWS 10 based on one or more extracted features 14. For example, following the previous example, if one of the extracted features 14 indicates that an important measurement (such as heart rate) is unavailable or that the measurement has a low SQI value, the reliability score regression model 110 may generate a lower reliability score. In one example, the confidence score regression model 110 is an XGBoost regressor that allows estimation of a confidence score 16 based on one or more top contributing extracted features 14.

[0049]

[0051] In general, in another aspect, referring to FIG. 4, a method 500 for assessing the reliability of a patient's EWS is provided. The method 500 includes receiving 502 a plurality of training data sets, where each of the plurality of training data sets includes a training EWS and one or more training features. The method 500 further includes assigning 504 a reliability annotation to each of the plurality of training data sets via a data annotator. The method 500 further includes generating 506 one or more extracted training features from each of the plurality of training data sets via a training feature extractor. The method 500 further includes generating 508 a reliability score regression model based on the reliability annotations and the one or more extracted training features via a deep learning trainer. The method 500 further includes receiving 510 patient test data. The method 500 further includes determining 512 an EWS based on the patient data via an EWS evaluator. The method 500 further includes generating 514 one or more extracted features from the patient test data via a real-time feature extractor. The method 500 further includes generating 516 a confidence score corresponding to the EWS based on the one or more extracted features and the confidence score regression model via a confidence score evaluator.

[0050]

[0052] All definitions defined and used herein should be understood to control for any dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meaning of the defined terms.

[0051]

[0053] As used in the specification and claims, the singular forms "a," "an," and "the" should be understood to mean "at least one," unless expressly stated otherwise.

[0052]

[0054] The term "and / or" as used in the specification and claims should be understood to mean "either or both" of the elements so conjoined, i.e., elements that are conjunctive in some cases and disjunctive in other cases. Multiple elements listed with "and / or" should be construed in the same manner, i.e., "one or more" of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the "and / or" clause, whether related or unrelated to those specifically identified elements.

[0053]

[0055] As used herein and in the claims, "or" should be understood to have the same meaning as "and / or" as defined above. For example, when separating items in a list, "or" or "and / or" should be interpreted as inclusive, i.e., including not only the inclusion of at least one of a plurality of elements or a list of elements, but also including two or more, and optionally including additional unlisted items. Only terms clearly indicated otherwise, such as "only one of" or "exactly one of," or, when used in the claims, "consisting of," will refer to the inclusion of exactly one element of a plurality of elements or a list of elements. In general, the term "or" as used herein should be interpreted as indicating exclusive alternatives (i.e., one or the other, but not both) only when preceded by the terms "either," "one of," "only one of," or "exactly one of."

[0054]

[0056] As used herein in the specification and claims, the phrase "at least one" referring to a list of one or more elements should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed in the list of elements, and not excluding any combination of elements in the list of elements. This definition also allows for the optional presence of elements other than those specifically identified in the list of elements to which the phrase "at least one" refers, whether related or unrelated to those specifically identified elements.

[0055]

[0057] Also, unless expressly stated to the contrary, in any method recited in the claims herein that includes two or more steps or actions, the order of the method steps or actions should not be understood to be necessarily limited to the order in which the method steps or actions are recited.

[0056]

[0058] In the claims and in the specification above, all transitional phrases such as "having," "including," "carrying," "having," "containing," "involving," "holding," "comprising," and the like, shall be understood to be open-ended, i.e., meaning including but not limited to. Only the transitional phrases "consisting of" and "consisting essentially of" shall be closed or semi-closed transitional phrases, respectively.

[0057]

[0059] The above-described examples of the described subject matter can be implemented in any of numerous ways. For example, some aspects can be implemented using hardware, software, or a combination thereof. If any aspects are implemented at least partially in software, the software code can be executed on any suitable processor or collection of processors, whether located in a single device or computer, or distributed among multiple devices / computers.

[0058]

[0060] The present disclosure may be embodied as a system, method, and / or computer program product at any possible level of technical detailed integration. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions for causing a processor to perform aspects of the present disclosure.

[0059]

[0061] A computer-readable storage medium may be a tangible device capable of holding and storing instructions for use by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. A non-exhaustive list of more specific examples of computer-readable storage media includes portable computer diskettes, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), static random access memories (SRAMs), portable compact disk read-only memories (CD-ROMs), digital versatile disks (DVDs), memory sticks, floppy disks, mechanically encoded devices such as punch cards or raised structures in grooves with instructions recorded thereon, and any suitable combination of the above. As used herein, a computer-readable storage medium should not be construed as being a transitory signal itself, such as an electric signal transmitted over a wire, such as an electric wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a wave guide or other transmission medium (e.g., a light pulse passing through an optical fiber), or an electric signal transmitted over a wire.

[0060]

[0062] The computer readable program instructions described in the specification can be downloaded from a computer readable storage medium to each computer / processing device or to an external computer or storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may include copper transmission cables, optical transmission fiber, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer readable program instructions from the network and transfers the computer readable program instructions to a computer readable storage medium in each computing / processing device for storage.

[0061]

[0063] The computer readable program instructions for carrying out the processes of the present disclosure may be assembler instructions, instruction set architecture (ISA) instructions, machine language instructions, machine dependent instructions, microcode, firmware instructions, state setting data, configuration data for an integrated circuit, or may be either source code or object code written in any combination of one or more programming languages, including object oriented programming languages ​​such as Smalltalk, C++, and procedural programming languages ​​such as the "C" programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (e.g., via the Internet using an Internet Service Provider). In some examples, electronic circuitry including, for example, a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), can execute computer readable program instructions by utilizing state information of the computer readable program instructions to customize the electronic circuitry to implement aspects of the present disclosure.

[0062]

[0064] Aspects of the present disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to examples of the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer readable program instructions.

[0063]

[0065] The computer readable program instructions may be provided to a processor of a special purpose computer or other data processing device to produce a machine, such that the instructions executing via the processor of the computer or other programmable data processing device create means for performing the function / acts specified in the block or blocks of the flowcharts and / or block diagrams. These computer readable program instructions may also be stored on a computer readable storage medium that causes a computer, programmable data processing device, and / or other device to function in a particular manner, such that the computer readable storage medium having instructions stored thereon comprises an article of manufacture including instructions that perform the function / processing aspects specified in the flowcharts and / or block diagrams.

[0064]

[0066] The computer readable program instructions may be loaded onto a computer, other programmable data processing apparatus or other device and a sequence of operational steps may be executed on the computer, other programmable apparatus or other device to generate a computer-implemented process, the instructions executing on the computer, other programmable apparatus or other device performing the function / acts specified in the block or blocks of the flowcharts and / or block diagrams.

[0065]

[0067] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various examples of the present disclosure. In this regard, each block in the flowcharts or block diagrams may represent a module, segment, or part of instructions, including one or more executable instructions for implementing the specified logical functions. In some alternative implementations, the functions shown in the blocks may be performed out of the order shown in the figures. For example, two blocks shown in succession may in fact be executed substantially simultaneously, or the blocks may sometimes be executed in the reverse order, depending on the functionality involved. It should also be noted that each block of the block diagrams and / or flowchart diagrams, and combinations of blocks in the block diagrams and / or flowchart diagrams, may be implemented by a dedicated hardware-based system that executes the specified functions or operations, or executes a combination of dedicated hardware and computer instructions.

[0066]

[0068] Other configurations are within the scope of the following claims and other claims to which the applicant has rights.

[0067]

[0069] Although various examples have been described and illustrated herein, one of ordinary skill in the art will readily envision various other means and / or configurations for performing the functions described herein and / or obtaining one or more of the results and advantages described herein. Each such variation and / or modification is deemed to fall within the scope of the examples described herein. More generally, one of ordinary skill in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are exemplary, and that the actual parameters, dimensions, materials, and / or configurations will depend on the particular application or applications in which the teachings are used. One of ordinary skill in the art will recognize, or be able to ascertain using no more than routine experimentation, equivalents to the specific examples described herein. Thus, the examples described above are presented by way of example only, and it should be understood that, within the scope of the appended claims and equivalents thereof, examples other than those specifically described and claimed may be practiced. Examples of the present disclosure are directed to each individual feature, system, article, material, kit, and / or method described herein. Furthermore, any combination of two or more of such features, systems, articles, materials, kits, and / or methods is within the scope of the present disclosure, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent.

Claims

1. A system for evaluating a patient's Early Warning Score (EWS), comprising: a test data receiver for receiving patient test data; an EWS evaluator for determining the EWS based on the patient test data; a real-time feature extractor for generating one or more extracted features from the patient test data; a reliability score evaluator for generating a reliability score corresponding to the EWS based on the one or more extracted features and a reliability score regression model A system having the above components.

2. The system according to claim 1, further comprising an inference engine for generating one or more inferences based on at least one of the reliability score and the one or more extracted features.

3. The system according to claim 2, wherein the inference engine further displays at least one of the one or more inferences via a user interface.

4. The system according to claim 2, wherein the inference engine further generates a notification corresponding to at least one of the one or more inferences.

5. A training data receiver for receiving a plurality of training data sets, each of the plurality of training data sets having a training EWS and one or more training features; a data annotator for assigning a reliability annotation to each of the plurality of training data sets based on the training EWS; a training feature extractor for generating one or more extracted training features from each of the plurality of training data sets; a deep learning trainer for generating the reliability score regression model based on the reliability annotation and the one or more extracted training features The system according to claim 1, further comprising the above components.

6. The system according to claim 5, wherein the data annotator assigns at least one reliability annotation based on user input.

7. The system according to claim 5, wherein the data annotator assigns at least one reliability annotation based on proximity to an EWS threshold.

8. The system according to claim 5, wherein the data annotator assigns at least one reliability annotation based on an EWS threshold exceed count.

9. The system according to claim 8, wherein the EWS threshold exceed count is determined over a predetermined period.

10. The system according to claim 5, wherein the data annotator assigns at least one reliability annotation based on an EWS fluctuation window.

11. The system according to claim 1, wherein each of the one or more extracted features corresponds to one or more patient features based on the patient examination data.

12. The system according to claim 11, wherein the one or more patient features include at least one of heart rate, oxygen saturation, respiratory rate, body temperature, diastolic blood pressure, systolic blood pressure, patient age, pulse pressure, approximate mean arterial pressure, and shock index.

13. The system according to claim 1, wherein the one or more extracted features include at least one of availability of measurement values, expiration date of measurement values, interruption of measurement, elapsed time of features, feature values, short-term delta features, long-term delta features, and signal quality index (SQI).

14. The system according to claim 1, wherein at least a part of the patient examination data is collected by a patient monitor.

15. A method for evaluating the reliability of a patient's Early Warning Score (EWS), comprising: receiving a plurality of training data sets each including a training EWS and one or more training features; assigning a reliability annotation to each of the plurality of training data sets via a data annotator; generating one or more extracted training features from each of the plurality of training data sets via a training feature extractor; generating a reliability score regression model based on the reliability annotation and the one or more extracted training features via a deep learning trainer; receiving patient examination data; determining an EWS based on the patient examination data via an EWS evaluator; generating one or more extracted features from the patient examination data via a real-time feature extractor; and generating a reliability score corresponding to the EWS based on the one or more extracted features and the reliability score regression model via a reliability score evaluator and having the method.

Images