Patient monitoring device
Through the patient monitoring model, query is generated and medical images related to health parameters is extracted, which solves the problem of combining ultrasound and patient monitoring workflows in the intensive care environment, realizes data collaboration between clinical decision support tools and patient monitors, and improves the Accuracy of nursing decisions.
Patent Information
- Application Number
- CN202380034224.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-07-28
- Filing Date
- 2023-04-10
- Publication Date
- 2024-11-26
AI Technical Summary
In intensive care settings, prior art is difficult to effectively combine ultrasound and patient monitoring workflows, resulting in information not being optimally stored or extracted, affecting the synergy between the output of clinical decision support tools and the data between the patient monitor.
By determining health parameters using patient monitoring models and using these parameters to generate search queries, the medical image database is queried to extract relevant matching medical images, thereby displaying ultrasound images and measurements related to the output of clinical decision support tools on the patient monitor.
The patient monitor is able to automatically select and display medical image sequences related to the prediction of clinical decision support tools, assisting clinicians in interpreting results, improving the quality of care and accuracy of decision-making.
Smart Images

Figure CN119032405A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a patient monitoring device, and a computer-implemented method performed by the patient monitoring device. Background Art
[0002] The disclosure herein relates to monitoring patients in a clinical setting, such as a hospital, clinic, nursing home, or any other medical facility. In such a facility, patients are typically monitored in an automated manner. For example, vital signs, such as heart rate, blood pressure, and SpO2 rate, etc., may be monitored using sensors attached to the patient.
[0003] In addition, clinical decision support tools are becoming more and more popular in hospital environments. Such tools (or models) exchange data with databases such as electronic medical records (EMR) or patient monitoring databases during a patient's hospital stay and generate predictions based on the monitored vital signs. For example, the prediction can involve a deterioration event and can be used to trigger an alarm on a patient monitor. Patient monitors such as bedside monitors, remote health monitors (for remote monitoring of patients), and central stations (where data from multiple patient monitoring systems are displayed) are typically the primary vital sign visualization dashboard for critical care teams. Summary of the invention
[0004] In a clinical setting, care is often centered around workflow. For example, there is an ultrasound workflow associated with acquiring, processing, and storing ultrasound images obtained during an ultrasound exam. Ultrasound workflows are well defined in cardiology and radiology and often revolve around modality worklists and Digital Imaging and Communications in Medicine DICOM studies, which include multiple images.
[0005] There are also separate workflows associated with monitoring patients in a critical care setting, for example involving obtaining, storing, and processing vital sign measurements.
[0006] Workflows are developed to standardize care and ensure that continuity of care quality is met across patients. Workflows also ensure strict record keeping.
[0007] A disadvantage of workflows is that they can be difficult to combine, which can result in information being stored or retrieved suboptimally. The present disclosure aims to improve upon the above-mentioned problem.
[0008] Thus, according to a first aspect, there is provided a computer-implemented method performed by a patient monitoring device. The method comprises: determining a health parameter of a subject using a patient monitoring model; and determining a search query using one or more outputs of the patient monitoring model, querying a database of medical images with the search query to extract matching medical images associated with the determined health parameter.
[0009] According to a second aspect, a device is provided, comprising: a memory including instruction data representing an instruction set; and a processor configured to communicate with the memory and execute the instruction set, wherein the instruction set, when executed by the processor, causes the processor to perform the following operations: determine health parameters of an object using a patient monitoring model; and determine a search query using one or more outputs of the patient monitoring model, and use the search query to query a database of medical images to extract matching medical images related to the determined health parameters.
[0010] According to a third aspect, a patient monitoring device is provided, comprising the apparatus of the second aspect, and further comprising: a display. A processor is further caused to perform the following operations: obtaining the matching medical image from the database of medical images; and sending an instruction to the display so that the display makes the matching medical image available for presentation on the display.
[0011] According to a fourth aspect, there is provided a computer program product comprising a computer-readable medium having computer-readable code embodied therein, the computer-readable code being configured such that, when executed by a suitable computer or processor, the computer or processor is caused to perform the method of the first aspect.
[0012] Therefore, methods and apparatus are provided for selecting a sequence of medical images for display on a patient monitoring display in an automated manner. As described above, the disclosure herein is made in response to the challenge of combining the workflow of ultrasound and patient monitoring in a critical care environment. The disclosure herein provides a novel approach to the challenge of enabling a patient monitor to "pull" relevant ultrasound data from a server. An advantage of the method is that it enables a patient monitor to obtain / "pull" relevant images in response to a prediction or calculation made by a clinical decision support tool, so as to provide images relevant to the prediction / calculation made by the clinical decision support tool and assist the clinician in interpreting the results.
[0013] These and other aspects will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Example embodiments will now be described, by way of example only, with reference to the following drawings, in which:
[0015] Figure 1 shows an apparatus according to some embodiments herein;
[0016] Figure 2 A patient monitoring device according to some embodiments herein is shown;
[0017] Figure 3 Methods according to some embodiments of the present invention are shown;
[0018] Figure 4 An example system according to some embodiments herein is shown; and
[0019] Figure 5 Shown in Figure 4 An example signal flow between components is shown in FIG. DETAILED DESCRIPTION
[0020] As described above, the disclosure of this article is inspired by the cross-value stream initiative in ultrasound and patient monitoring. The inventors of this article recognize that there are challenges in the workflow of combining ultrasound and patient monitoring in a critical care environment. Ultrasound workflows are well defined in, for example, cardiology and radiology, and include modality worklists and DICOM studies including multiple images. Similar to computed tomography (CT) / magnetic resonance (MR), the measurement results produce a separate entity known as a structured report. The data is stored in an institutional DICOM storage entity or a picture archiving and communication system (PACS) archive. In a critical care environment, as more care providers become proficient in acquisition, and as new technologies help improve the reproducibility and continuous nature of echo-derived vital signs, these workflows are slowly adopted. At present, dedicated storage of image data in critical care tends to occur occasionally and is not usually stored permanently, but when it is stored, it is usually in a separate DICOM storage entity online.
[0021] On the other hand, patient monitoring information is stored in various locations including bedside as well as central information centers. Patient monitoring information tends to be stored using industry standard formats such as the Health Level Seven (HL7) format, which is a set of international standards for transmitting clinical and administrative data between software applications used by healthcare providers.
[0022] Patient monitors such as bedside monitors, remote health monitors, and central stations tend to be the de facto vital sign visualization dashboards for care teams (such as critical care teams) in hospitals. As described in more detail below, the disclosure herein relates to improved patient monitors, remote health monitors, and / or central stations that are capable of displaying medical image(s) relevant to the current state of the monitored patient(s). Thus, the disclosure herein enables "monitoring drag": for example, the ability of a patient monitor to filter and display relevant image data for monitoring insights.
[0023] In short, the invention herein enables a patient monitoring system to filter and extract ultrasound images and measurements that are clinically relevant to the output of a clinical decision support tool. In short, as described above, during a patient's hospital stay, the clinical decision support tool exchanges data with a patient information center and generates predictions, such as predictions of exacerbation events. The output of the clinical decision support tool is used to determine a list of DICOM query parameters that are used to obtain images related to the predicted exacerbation event. In some embodiments, the parameter list can be generated as part of the output of the tool, or performed separately by a predefined lookup table or a user-configurable lookup table.
[0024] Thus, a mechanism is provided for "dragging" relevant images associated with the output of a clinical decision making tool to a patient monitor.
[0025] Now go to Figure 1 In some embodiments, there is an apparatus 100 according to some embodiments herein. Typically, the apparatus may form part of a computer device or system, such as a laptop, desktop computer or other computing device. In some embodiments, the apparatus 100 may form part of a distributed computing arrangement or cloud.
[0026] The apparatus includes: a memory 104 including instruction data representing an instruction set; and a processor 102 (e.g., processing circuitry or logic) configured to communicate with the memory and execute the instruction set. Generally, the instruction set, when executed by the processor, can cause the processor to perform any embodiment of the method 300 described below.
[0027] Embodiments of the apparatus 100 may be used, for example, to monitor a patient in a clinical environment such as a hospital. More specifically, the set of instructions, when executed by a processor, causes the processor to perform the following operations: determine a health parameter of an object using a patient monitoring model; and determine a search query using one or more outputs of the patient monitoring model, querying a database of medical images using the search query to extract matching medical images associated with the determined health parameter.
[0028] The processor 102 may include one or more processors, processing units, multi-core processors, or modules configured or programmed to control the device 100 in the manner described herein. In a particular embodiment, the processor 102 may include a plurality of software and / or hardware modules, each of which is configured to perform or be used to perform a single or multiple steps of the method described herein. The processor 102 may include one or more processors, processing units, multi-core processors, and / or modules configured or programmed to control the device 100 in the manner described herein. In some embodiments, for example, the processor 102 may include a plurality of (e.g., interoperable) processors, processing units, multi-core processors, and / or modules configured for distributed processing. Those skilled in the art will appreciate that such processors, processing units, multi-core processors, and / or modules may be located in different locations and may perform different steps of the method described herein and / or different parts of a single step.
[0029] The memory 104 is configured to store program code that can be executed by the processor 102 to perform the methods described herein. Alternatively or additionally, one or more memories 104 may be external to the device 100 (i.e., separate or remote from the device 100). For example, one or more memories 104 may be part of another device. The memory 104 may be used to store determined health parameters, search queries, matching medical images, and / or any other information or data received, calculated, or determined by the processor 102 of the device 100 or from any interface, memory, or device external to the device 100. The processor 102 may be configured to control the memory 104 to store determined health parameters, search queries, matching medical images, and / or any other information or data.
[0030] In some embodiments, the memory 104 may include multiple sub-memories, each of which may store one instruction data. For example, at least one sub-memory may store instruction data representing at least one instruction in an instruction set, and at least one other sub-memory may store instruction data representing at least one other instruction in the instruction set.
[0031] Will understand, Figure 1 Only components necessary to illustrate this aspect of the present disclosure are shown, and in actual implementations, the device 100 may include additional components than those shown. For example, the device 100 may include a battery or other power source for powering the device 100 or a unit for connecting the device 100 to a mains power source. The device 100 may also include a communication interface, for example for wireless communication with other computing devices.
[0032] Apparatus 100 may be included in a patient monitor, such as Figure 2The patient monitor 200 shown in FIG. The patient monitor 200 includes a display 202. In this example, the processor 102 is also caused to obtain a matching medical image from a database of medical images and to send instructions to the display so that the display makes the matching medical image available for presentation on the display.
[0033] For example, the display 202 may include a computer screen, such as a liquid crystal display (LCD), or any other type of display. The patient monitor may also include other components, such as a stand 204 and / or a user input device, such as a keyboard, a mouse, one or more buttons, or any other input device that enables a user to interact with the device, for example, to cause the display to display a matching image.
[0034] In addition, the patient monitoring device 200 may include one or more sensors, for example, for obtaining vital signs or other information about the patient being monitored. Examples of sensors include, but are not limited to, SpO2 monitors, heart rate monitors, and blood pressure monitors.
[0035] Go to Figure 3 , there is a computer-implemented method 300 performed by a patient monitoring device. For example, an embodiment of the method 300 may be performed by an apparatus such as the apparatus 100 described above.
[0036] In brief summary, in a first step 302, the method 300 includes determining a health parameter of a subject using a patient monitoring model. In a second step 304, the method 300 includes determining a search query using one or more outputs of the patient monitoring model, and querying a database of medical images using the search query to extract matching medical images associated with the determined health parameter.
[0037] As used herein, for example, a patient monitoring device (or patient monitor) can be a bedside monitor or a central station, such as a central computer unit for receiving and displaying information for monitoring multiple patients. In other embodiments, the patient monitoring device can be a remote monitoring device, such as a remote health monitor. An example of a remote health monitor is a remote intensive care unit (ICU) monitor.
[0038] As described above, in step 302, a health parameter of the subject is determined using a patient monitoring model.
[0039] As used herein, a patient monitoring model may be a clinical decision tool, a clinical decision support algorithm, or any other model or process that may be used to process monitoring information from a patient.
[0040] For example, a patient monitoring model may take one or more vital signs as input and output a health parameter. The health parameter may be determined, for example, the health parameter may be calculated or predicted by the patient monitoring model. The patient monitoring model may determine the health parameter in any manner, for example, using a lookup table, a sequence of thresholds or if / then statements, an equation, using machine learning, or in any other manner.
[0041] In some embodiments, the health parameter is a score. Thus, the score can be derived from patient data available from a patient monitoring device. For example, the score can be any metric used to determine whether the patient's condition is improving, worsening, or stable. The score can be associated with a specific structure, parameter, measurement time, or parameter value. Examples of scores include, but are not limited to: a hemodynamic stability index score; a hypotension prediction index score; an acute heart failure prediction index; a score associated with a patient monitoring alarm or a patient monitoring trend.
[0042] Examples of patient monitoring models include, but are not limited to, models for determining a hemodynamic stability index score, a hypotension prediction index score, or an acute heart failure prediction index.
[0043] A major need for ultrasound in critical care is to help manage the longitudinal management of the patient's cardiovascular physiology to improve hemodynamics. An example of this is the patient-specific Frank-Starling relationship of the heart, which describes the relationship between cardiac preload and output. Ultrasound is a trusted non-invasive tool to measure this in real time on a beat-by-beat basis.
[0044] However, the skilled artisan will appreciate that these are merely examples and that the method 300 is equally applicable to any patient monitoring model.
[0045] In step 304 , the method includes using one or more outputs of the patient monitoring model to determine a search query with which to query a database of medical images to extract matching medical images associated with the determined health parameter.
[0046] Although some examples herein describe ultrasound (US) images, the medical images in the database of medical images can be of any image modality. Example medical images include, but are not limited to, computed tomography (CT) images (e.g., from a CT scan), such as a C-arm CT image, a spectral CT image or a phase contrast CT image, an X-ray image (e.g., from an X-ray scan), a magnetic resonance (MR) image (e.g., from an MR scan), an ultrasound (US) image (e.g., from an ultrasound scan), a fluoroscopic image, a nuclear medicine image, or any other medical image.
[0047] Medical images may be stored in a format that includes meta information (e.g., metadata). For example, in the form of an image header. The skilled person will appreciate that this is merely an example, and that the metadata may be stored in other ways, for example, using application-related strings, images, event hyperlinks, coded objects, and / or macros.
[0048] In some embodiments, the medical image may be in Digital Imaging and Communications in Medicine (DICOM) format. The DICOM format is described in the following document: DanduRavi Varma (2012) entitled "Managing DICOM images: Tips and tricks for the radiologist"; Indian J Radiol Imaging, Jan-Mar 2012; 22(1):4-13. Various tools and coding languages may be used to read and / or search DICOM metadata. As an example, in MathWorks, the following command may be used to read metadata from a DICOM file "Filama.dcm": metadata=dicominfo('Filename.dcm'). It should be understood that this is merely an example, and that medical images may be stored in other formats, such as the Flexible Image Transfer System (FITS) format or any other format that includes searchable meta information.
[0049] In some embodiments, the medical images may be stored in an institutional DICOM storage entity or a Picture Archiving and Communication System (PACS) archive. Thus, the database of medical images may be a DICOM database, a PACS database, or any other database containing medical images. The database of medical images may be referred to as an electronic medical record (EMR) database or a patient monitor database. The database of medical images may be stored or hosted locally, for example forming part of a hospital network. Alternatively, a patient database of medical images may be made available via the cloud.
[0050] The medical image may be associated with an accompanying structured report that includes measurements and / or annotations (such as calculations) of features in the associated medical image. The structured report may be generated in a standardized manner, for example as part of an ultrasound workflow.
[0051] In step 302, one or more outputs of the patient monitoring model are used to determine a search query that is used to query a database of medical images to extract matching medical images associated with the determined health parameter. In this sense, a matching medical image is a medical image that satisfies the search query.
[0052] In some examples, at least one of the one or more outputs of the patient monitoring model is formatted in a manner for direct use as a Digital Imaging and Communications in Medicine (DICOM) search query.For example, the patient monitoring model may be configured to output a search query.
[0053] Using the mitral valve as an example, a query may include the term "mitral valve," or additionally (or alternatively) may specify a particular image view that indirectly ensures visualization of the mitral valve (e.g., "apical 4 chamber (A4C) view"). Additional options may be filtered by ultrasound (US) acquisition parameters such as depth, focal zone, mode (Bmode / color / PW), etc. For example, an A4C view acquired at a depth of 18 cm (versus, say, 8 cm depth) is more likely to fully include visualization of the mitral valve.
[0054] The DICOM Query Format can be found in the following online document: "DICOM PS3.3 2022b - InformationObject Definitions", which is a DICOM Standards Committee (2022) For example, section: DICOMPS3.17 2022b-Explanatory Information: “N.3.5 Left Ventricle Output” relates to the DICOM standard for left ventricle output.
[0055] The following code shows a python example of querying a DICOM study using a DICOM tag (PatientName) using the DICOM web standard. In this example, "params" can be output by the patient monitoring model for querying.
[0056]
[0057]
[0058]
[0059] In some embodiments, at least one of the one or more outputs of the patient monitoring model is formatted as a string or an image, and the string or image is used in the search query.
[0060] In other examples, one or more outputs of a patient monitoring model may be converted into a search query using a lookup table, a machine learning (ML) model, a mapping process, a mathematical function, or any other method of converting an output into a DICOM search query. It should be understood that the lookup table and mapping function may be pre-configured by a human engineer who may specify appropriate DICOM queries for different situations. Similarly, an ML model (such as a neural network) or the like may be trained using training data comprising training examples, wherein each training example comprises i) an example output of a patient monitoring model and ii) a ground truth DICOM query that should be performed for the example output.
[0061] In this way, the output of the patient monitoring model can be formatted as a query against a database to search for and extract medical images related to health parameters.
[0062] After step 304, method 300 may further include obtaining a matching medical image from a database of medical images and sending instructions to a display to make the matching medical image available for presentation on the display. The matching medical image may be presented automatically or based on a user request of a patient monitor.
[0063] In some embodiments, step 304 may be performed in response to certain criteria being met (eg, in response to a trigger).
[0064] For example, step 304 may be performed in response to a determined health parameter triggering an alarm on a patient monitoring device. In such an example, when an alarm is triggered due to a health parameter (or its value), the patient monitor may automatically query a database of medical images to obtain medical images associated with the alarm. These may then be displayed on the patient monitor to help the clinician determine whether action is required.
[0065] In another example, step 304 may be triggered at another threshold of the health parameter rather than the threshold that triggers an alarm on the patient monitor. For example, a first threshold may be used to trigger performance of step 304, and a second threshold may be used to trigger an alarm.
[0066] In some embodiments, step 304 may be performed in response to the determined health parameter indicating a deterioration of the patient. For example, the health parameter (or its value) may be compared to a standard, and if the standard (likely) indicates a deterioration of the patient, this may trigger the performance of step 304. Thus, if the patient begins to deteriorate, the patient monitor may pull in medical images relevant to the particular deteriorating condition, thereby providing all relevant information to the clinician on the patient monitor.
[0067] In this way, information generated by an image processing workflow (eg, a US workflow) can be integrated with a patient monitoring workflow to provide clinicians with the most relevant information about the patient and help them make the most informed decisions regarding patient care.
[0068] In another embodiment, more than one query may be formatted. For example, as part of step 202 (e.g., as part of a step of determining a health parameter of a subject using a patient monitoring model), the patient monitoring model may query a database of medical images. For example, method 300 may include generating a first output to generate a first search query, the results of which are used by the patient monitoring model to generate a health parameter that may be displayed to a user (which may be considered a second output, an updated output, or a final output in this embodiment). As a result of the second output, a second query for the search may also be generated according to step 304. In this manner, the initial output of the patient monitoring model may be used to query a database of medical images in order to obtain information for determining a final health parameter.
[0069] Now turn to Figure 4 , which shows how the patient monitor 200 can be interfaced with other elements in a hospital system in order to perform the method 300 described above according to some embodiments of the present invention. In this embodiment, the medical image is a US image stored in the DICOM image format. It should be understood that these are merely examples, and Figure 4 The examples in are equally applicable to other imaging modalities and other patient monitoring models.
[0070] Figure 4 The embodiment in FIG. 1 shows the following elements:
[0071] One or more ultrasound imaging systems 402
[0072] The ultrasound imaging system proceeds through its standard DICOM workflow, which includes the creation of studies based on unique information about the patient, the acquisition of series and images 404, and the transmission of raw images and the resulting structured reports.
[0073] A database of medical images in the form of a DICOM storage server 406
[0074] A DICOM Storage Server can be any institutionally configured DICOM repository that supports Service Class Providers (SCPs) and Service Class Users (SCUs).
[0075] Patient monitoring device 200 (e.g., bedside monitor)
[0076] Equipment for visualizing integrated data
[0077] Patient Monitoring Model 408 (also referred to herein as Patient Monitoring Clinical Decision Support Tool)
[0078] exist Figure 4 In the example of , the patient monitoring model is a hemodynamic stability index (HSI) calculation model. As described above, the patient monitoring model performs a method that combines patient data to determine health parameters 410 (e.g., provide a prediction and / or score about the patient's health) and generates a searchable string 412 that can be used to perform queries on the DICOM space.
[0079] DICOM SCP and Controller Unit
[0080] Unit for managing streams of searchable strings and performing SCP queries
[0081] In step 414 , matching images from the query may be sent and displayed on the patient monitor 200 .
[0082] Thus, in this manner, the patient monitoring system is able to filter and extract ultrasound images and measurements that are clinically relevant to the output of a patient monitoring model (eg, algorithm).
[0083] Now turn to Figure 5 , which illustrates an embodiment of the present disclosure. In this embodiment, the medical image is a US image in DICOM format, and a patient monitoring model is used to determine the HSI score. It should be understood that these are merely examples, and Figure 5 The examples in are equally applicable to other imaging modalities and other patient monitoring models.
[0084] Step 502: The US acquisition system 402 stores the images / measurements to the DICOM server 406. The ultrasound system is used according to the standard DICOM workflow to generate studies, series, and images. The image processing process and / or the expert reviewer generates measurements based on these images to generate one or more observations encapsulated in a structured report. For a typical cardiac surgery patient in the ICU, three different ultrasound systems operated by different users generate six separate ultrasound studies, with a total of 80 movie loops containing 1600 image frames. Since this is a heart valve surgery patient, the automatic image analysis algorithm measures parameters such as the velocity time integral of the aorta, and the expert reviewer evaluates parameters such as mitral regurgitation volume (SR code meaning (0008,0104) aortic regurgitation volume (PISA)) with SR finding sites such as "left ventricular outflow tract" (code value T-32650). These measurements are directly linked to the SOP instance UID of the secondary capture annotation generated by the algorithm. Many other measurements are made from the ultrasound data such as the right pulmonary artery diameter (code value 18021-6) on the right pulmonary artery (finding site code value 44200).
[0085] Step 504: The patient monitoring server 512 interacts with the HSI patient monitoring model 408 to generate predictions. The ultrasound system exports these studies and their associated reports for long-term storage in a dedicated DICOM storage server (eg, institutional PACS). All six of the above studies can be exported and stored.
[0086] Step 506: During the patient's hospital stay, the HSI patient monitoring model 408 exchanges data with the patient information center 512 and generates a prediction that the patient's blood pressure will drop sharply within the next 15 minutes (e.g., in accordance with step 302 above). The HSI patient monitoring model 408 recognizes that, combined with the presence of surgery and the increased heart rate, this event may involve the patient's mitral valve function. Therefore, the HSI patient monitoring model sends (in step 508) a list of DICOM query parameters along with its deterioration prediction. As described above, this parameter list can be generated as part of the HSI patient monitoring model output, or performed separately by a predefined lookup table or a user-configurable lookup table (e.g., in accordance with step 304 above).
[0087] Step 508: The patient monitoring controller uses the parameters to query the DICOM server, which returns a list of measurements and five image sequences (SOP instances). In the case of measurements, image phase labels are included so specific parts of the cardiac cycle can be highlighted.
[0088] Step 510: The patient monitoring controller formats the image pointer list to the bed monitor 200, and the bedside monitor 200 presents the image according to the data extracted from the DICOM server.
[0089] In the above embodiments, HSI is given as an example of a prediction score derived from patient data. As mentioned above, this is only an example, and Figure 5The system can work with another score, such as the hypotension prediction index (HPI) developed by Edwards. See the paper: Davies SJ, Vistisen ST, Jian Z, et al., titled: “Ability of anarterial waveform analysis derived hypotension prediction index to predictfuture hypotensive events in surgical patients.” Anesthesia and Analgesia, 2020;130:352-9. In addition, see the paper: Hatib F, Jian Z, Buddi S, et al., titled: “Machine-learning algorithm to predict hypotension based on high-fidelity arterial pressure waveform analysis,” Anesthesiology, 2018;129:663-74; and Wijnberge M, Geerts BF, Hol L, et al., “Effect of a machine learning derived early warning system for intraoperative hypotension vs standard care on depth and duration of intraoperative hypotension during elective noncardiac surgery,” JAMA 2020;323:1052-60. Another example of a score that can be used is an acute heart failure prediction index, such as described in the following paper: Claire Zhao et al., titled: "Predicting Disposition Decision, Mortality, and Readmission for Acute Heart Failure Patients in the Emergency Department Using Vital Sign, Laboratory, Echocardiographic, and Other Clinical Data," Blood Circulation, Vol. 138, November 6, 2018, Issue Suppl_1.
[0090] Turning now to other embodiments, there is provided a computer program product comprising a computer readable medium having computer readable code embodied therein, the computer readable code being configured to, when executed by a suitable computer or processor, cause the computer or processor to perform one or more of the methods described herein.
[0091] It will therefore be appreciated that the present disclosure also applies to computer programs, in particular to computer programs on or in a carrier, suitable for putting the embodiments into practice. The program may be in the form of source code, object code, a code intermediate source and object code, such as in partially compiled form, or in any other form suitable for use in the implementation of the methods according to the embodiments described herein.
[0092] It should also be understood that such a program can have many different architectural designs. For example, the program code that implements the function of the method or system can be subdivided into one or more subroutines. Many different ways of allocating functions between these subroutines will be apparent to those skilled in the art. The subroutines can be stored together in an executable file to form a self-contained program. Such an executable file can include computer executable instructions, such as processor instructions and / or interpreter instructions (e.g., Java interpreter instructions). Alternatively, one or more or all of the subroutines can be stored in at least one external library file and, for example, statically or dynamically linked to the main program at runtime. The main program contains at least one call to at least one subroutine. The subroutines can also include function calls to each other.
[0093] The carrier of a computer program may be any entity or device capable of carrying the program. For example, the carrier may include a data storage device (such as a ROM, such as a CD ROM or a semiconductor ROM) or a magnetic recording medium (such as a hard disk). In addition, the carrier may be a transmissible carrier, such as an electrical or optical signal, which may be transmitted via an electric or optical cable or by radio or other means. When the program is embodied in such a signal, the carrier may be constituted by such a cable or other device or unit. Alternatively, the carrier may be an integrated circuit in which the program is embedded, the integrated circuit being suitable for executing or for executing the relevant method.
[0094] According to the study of the drawings, the present disclosure and the appended claims, those skilled in the art can understand and implement variations to the disclosed embodiments when practicing the principles and techniques described herein. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit can implement the functions of several items recorded in the claims. The mere fact that certain measures are cited in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously. The computer program may be stored or distributed on a suitable medium, such as an optical storage medium or solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems. Any figure mark in the claims should not be interpreted as limiting the scope.
Claims
1. A computer-implemented method performed by a patient monitoring device, the method comprising: using a patient monitoring model to determine health parameters of a subject; as well as One or more outputs of the patient monitoring model are used to determine a search query that is used to query a database of medical images to extract matching medical images that are associated with the determined health parameter.
2. The method according to claim 1, wherein: The step of determining a search query using one or more outputs of the patient monitoring model is performed in response to: The determined health parameter triggers an alarm on the patient monitoring device; and / or The determined health parameter is indicative of a deterioration of the patient.
3. The method according to claim 1 or 2, wherein: The medical images in the database of medical images are associated with accompanying structured reports, the structured reports comprising measurements and / or annotations of features in the associated medical images; and wherein the search query is formatted to search for the measurements and / or annotations.
4. The method according to any one of claims 1 to 3, wherein: At least one of the one or more outputs of the patient monitoring model is formatted as a string or an image; And wherein the character string or image is used in the search query.
5. The method according to any one of claims 1 to 3, wherein: At least one of the one or more outputs of the patient monitoring model is formatted in a manner to be directly usable as a Digital Imaging and Communications in Medicine (DICOM) search query.
6. A method according to any one of the preceding claims, wherein: The step of using one or more outputs of the patient monitoring model to determine the search query comprises: The one or more outputs of the patient monitoring model are converted to the search query using: a lookup table, a machine learning model, or a mapping process.
7. A method according to any one of the preceding claims, wherein: The determined health parameter is a score derived from patient data available from the patient monitoring device.
8. A method according to any one of the preceding claims, wherein: The scores are: hemodynamic stability index score; Hypotension Predictive Index score; Acute Heart Failure Prediction Index; or Scores associated with patient monitoring alerts or patient monitoring trends.
9. A method according to any one of the preceding claims, wherein: The medical image contains meta-information.
10. The method according to any one of the preceding claims, wherein: The medical image is in DICOM image format.
11. The method according to any one of the preceding claims, further comprising: Obtaining the matching medical image from a database of medical images; as well as An instruction is sent to a display to make the matching medical image available for presentation on the display.
12. An apparatus comprising: a memory including instruction data representing an instruction set; as well as A processor configured to communicate with the memory and execute the instruction set, wherein the instruction set, when executed by the processor, causes the processor to perform the following operations: using the patient monitoring model to determine a health parameter of the subject; and One or more outputs of the patient monitoring model are used to determine a search query that is used to query a database of medical images to extract matching medical images that are associated with the determined health parameter.
13. A patient monitoring device comprising the apparatus according to claim 12, and further comprising: display; and wherein the processor is further caused to perform the following operations: Obtaining the matching medical image from a database of medical images; and Instructions are sent to the display to cause the display to enable the matching medical image for presentation on the display.
14. The patient monitoring device according to claim 13, wherein: The processor is also caused to perform the method according to any one of claims 2 to 10.
15. A computer program product comprising a computer readable medium having computer readable code embodied therein, the computer readable code being configured such that when executed by a suitable computer or processor, the computer or processor is caused to perform the method according to any one of claims 1 to 11.
