Methods and systems for improved prediction of hemodynamic instability by improving training data

Abstract

A method for generating a training dataset, comprising: receiving historical electronic medical records for a plurality of subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the condition types; extracting one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting; labeling, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types; removing any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; and gathering the remaining subjects in a training dataset.

Description

METHODS AND SYSTEMS FOR IMPROVED PREDICTION OF HEMODYNAMIC INSTABILITY BY IMPROVING TRAINING DATAField of the Disclosure

[0001] The present disclosure relates generally to methods and systems for generating an improved training dataset utilized to train a prediction model.Background

[0002] For critically ill patients, timely intervention to treat or prevent hemodynamic instability is crucial to patient outcomes. Unfortunately, the early warning signs of hemodynamic instability can be easily missed. Clinical expertise to recognize these signs may be scarce, and early signs of instability may not be obvious from simple monitoring of individual vitals.

[0003] Existing methods for detecting hemodynamic instability are primarily focused on simple rules that can be easily used by busy clinicians without automated assistance. For example, the PALS (Pediatric Advanced Life Support) guidelines provide age-adjusted normal ranges for common vitals, including heart rate, respiratory rate, and blood pressure. Vitals outside of normal ranges are considered as a sign of potential hemodynamic instability, requiring additional clinical attention. However, rule-based methods and systems are typically unable to detect potential hemodynamic instability from the simple monitoring of individual vitals, and do a poor job distinguishing among age-dependent signs of instability.

[0004] Accordingly, a hemodynamic stability index was developed to provide early warnings on the upcoming onset of hemodynamic instability. Although most hemodynamically unstable patients suffer from severe hypotension and inadequate organ perfusion to support normal organ functions, there exist multiple underlying shock types causing the instability, including cardiogenic shock (critical reduction of the heart’s pumping capacity), hypovolemic shock (severe blood loss or loss of fluids) and septic shock (dysregulated response to an infection resulting in life- threatening organ dysfunctions). Other less common shock types include anaphylactic shock and neurogenic shock, although they have a low prevalence in the ICU (less than 5%).

[0005] Patients belonging to different shock types often receive different types of hemodynamic interventions. For example, cardiogenic shock patients receive inotropes, including dobutamine and milrinone, or mechanical cardiac support, including IABP (intra-aortic balloonpump), VA-ECMO (venoarterial extracorporeal membrane oxygenation), and LVAD (left ventricular assistant device). Hypovolemic shock patients receive packed red blood cells transfusion (in cases of hemorrhage) or fluid bolus (in cases of hypovolemia). Septic shock patients receive vasopressors (norepinephrine, phenylephrine, dopamine, vasopressin, epinephrine) and overlapping antibiotics. Given the discriminative power of these interventions, they can be used to label different shock types as well as the onset time of the shock.

[0006] However, prior labeling of shock types only makes use of the first intervention received by each patient. This is problematic because the patient often receives their first hemodynamic interventions before ICU admission or within the first six hours of ICU admission. For patients that receive their first interventions before ICU admission, a clinician or prediction model may not have access to any input variables (vitals, laboratory measurements or ventilator settings) to make predictions. For patients that receive their first interventions within the first six hours of ICU admission, the laboratory measurements were most likely to be absent. Besides the first intervention, the same patient can also receive subsequent hemodynamic interventions. By only making use of the first intervention, the system will not fully utilize the other interventions received by the patient for model training.Summary of the Disclosure

[0007] Accordingly, there is a continued need for systems and methods that more generate an improved training dataset which can be utilized to train an improved prediction model.

[0008] Various embodiments and implementations herein are directed to a method and system configured to generate an improved training dataset. The system receives historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising two or more condition types, where treatment for the condition comprises a different type of treatment for each one of the two or more condition types. The system extracts one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting. An extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment. The system labels, with a condition type label, the subject as having a specific one of the plurality ofcondition types, based on the subj ect receiving the one of the different types of treatment associated with that specific one of the plurality of condition types. The subject is labeled with the condition type label when the received treatment type is associated with the earliest extracted intervention segment beginning first after the subject’s admission to the health care setting. The system removes any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label, and finally gathers the remaining subjects, each associated with a condition type label, in a training dataset.

[0009] Generally, in one aspect, a method for generating a training dataset is provided. The method comprises: (i) receiving historical electronic medical records (EMR) for a plurality of case subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the plurality of condition types and control subjects who don’t have the condition; (ii) extracting, from the received EMR, one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting, wherein an extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and wherein the intervention segment is associated with the one of the different types of treatment are administered to the subject, and further wherein the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment; (iii) labeling, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting; (iv) removing any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; and (v) gathering the remaining subjects, each associated with a condition type label, in a training dataset.

[0010] According to an embodiment, labeling the subject as having a specific one of the plurality of condition types further comprises labeling the subject as having a second, different specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with second, different specific one of the plurality ofcondition types at the same time as receiving the one of the different types of treatment associated with the first specific one of the plurality of condition types.

[0011] According to an embodiment, the condition is hemodynamic instability. According to an embodiment, the plurality of condition types are cardiogenic shock, hypovolemic shock, and septic shock. According to an embodiment, the type of treatment for cardiogenic shock comprises one or more of an inotrope and mechanical cardiac support, the type of treatment for hypovolemic shock comprises one or more of a transfusion or fluid bolus, and the type of treatment for septic shock comprises one or more of a vasopressor and an overlapping antibiotic.

[0012] According to an embodiment, the method further includes training, using the training dataset, a hemodynamic status model configured to identify a hemodynamic status of the subject. According to an embodiment, the hemodynamic status is a hemodynamic stability index, and hemodynamic status model is trained with the training dataset to predict whether a subject is likely to experience one or more of cardiogenic shock, hypovolemic shock, and septic shock.

[0013] According to an embodiment, the condition is respiratory distress. According to an embodiment, the plurality of condition types are hypoxemic respiratory failure and hypercapnic respiratory failure. According to an embodiment, the type of treatment for hypoxemic respiratory failure comprises mechanical ventilation, and the type of treatment for hypercapnic respiratory failure comprises nasal ventilation.

[0014] Also provided is a system for generating a training dataset. The system includes: (i) an EMR database comprising historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the plurality of condition types; and (ii) a processor configured to: extract, from the EMR, one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting, wherein an extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and wherein the intervention segment is associated with the one of the different types of treatment are administered to the subject, and further wherein the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment; label, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one ofthe different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting; remove any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; and gather the remaining subjects, each associated with a condition type label, in a training dataset.

[0015] It should be appreciated that all combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are contemplated as being part of the inventive subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are contemplated as being part of the inventive subject matter disclosed herein. It should also be appreciated that terminology explicitly employed herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.

[0016] These and other aspects of the various embodiments will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.Brief Description of the Drawings

[0017] In the drawings, like reference characters generally refer to the same parts throughout the different views. The figures showing features and ways of implementing various embodiments and are not to be construed as being limiting to other possible embodiments falling within the scope of the attached claims. Also, the drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the various embodiments.

[0018] FIG. 1 is a flowchart of a method for generating a training dataset, in accordance with an embodiment.

[0019] FIG. 2 is a schematic representation of a training system, in accordance with an embodiment.

[0020] FIG. 3 is a schematic representation of a treatment timeline, in accordance with an embodiment.

[0021] FIG. 4 is a schematic representation of a treatment timeline, in accordance with an embodiment.

[0022] FIG. 5 is a schematic representation of a treatment timeline, in accordance with an embodiment.

[0023] FIG. 6 is a flowchart of a method for training a model using a training dataset, in accordance with an embodiment.Detailed Description of Embodiments

[0024] The present disclosure describes various embodiments of a system and method configured to generate an improved training dataset used to train a hemodynamic stability index prediction model. More generally, Applicant has recognized and appreciated that it would be beneficial to generate improved training datasets, thereby training more accurate prediction models. Accordingly, a training system receives historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising two or more condition types, where treatment for the condition comprises a different type of treatment for each one of the two or more condition types. The system extracts one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting. An extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment. The system labels, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types. The subject is labeled with the condition type label when the received treatment type is associated with the earliest extracted intervention segment beginning first after the subject’s admission to the health care setting. The system removes any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label, and finally gathers the remaining subjects, each associated with a condition type label, in a training dataset.

[0025] According to an embodiment, the systems and methods described or otherwise envisioned herein can, in some non-limiting embodiments, be implemented as an element for a commercial product for patient analysis or monitoring, such as the Philips® IntelliVue® system (available from Koninklijke Philips NV, the Netherlands), or any suitable system. For example, the system and method can be implemented within existing or future clinical decision support systems (CDS), applications, and devices. However, the disclosure is not limited to these devices or systems, and thus disclosure and embodiments disclosed herein can encompass any device or system capable of generating a training dataset for a prediction model.

[0026] Referring to FIG. 1 , in one embodiment, is a flowchart of a method 100 for generating a training dataset using a training system 200. The methods described in connection with the figures are provided as examples only, and shall be understood not limit the scope of the disclosure. The training system can be any of the systems described or otherwise envisioned herein. The training system can be a single system or multiple different systems.

[0027] At step 110 of the method, a training system 200 is provided. Referring to an embodiment of a training system 200 as depicted in FIG. 2, for example, the system comprises one or more of a processor 220, memory 230, user interface 240, communications interface 250, and storage 260, interconnected via one or more system buses 212. It will be understood that FIG. 2 constitutes, in some respects, an abstraction and that the actual organization of the components of the system 200 may be different and more complex than illustrated. Additionally, training system 200 can be any of the systems described or otherwise envisioned herein. Other elements and components of training system 200 are disclosed and / or envisioned elsewhere herein.

[0028] At step 120 of the method, the training system receives or obtains patient information from an electronic medical record (EMR) database or system 270, for a plurality of patients or subjects. The patient information can be any information about the patient that the training system can or may utilize to generate a training dataset as described or otherwise envisioned herein. According to an embodiment, the patient information comprises one or more of demographic information about the patient, a diagnosis for the patient, medical history of the patient such as treatment information, and / or any other information. For example, demographic information may comprise information about the patient such as name, age, body mass index (BMI), and any other demographic information. The diagnosis for the patient may be any information about a medicaldiagnosis for the patient, historical and / or current. The medical history of the patient may be any historical admittance or discharge information, historical treatment information, historical diagnosis information, historical exam or imaging information, and / or any other information (although in some embodiments, a patient’s medical history may not be available). Other patient information that can be received by the training system includes lab test results. For example, the lab tests may be an analysis of blood gases, electrolytes, biomarkers, and / or any other types of lab tests. Yet another example of patient information received by the training system includes vital sign information for the patient. The vital sign information can be any vital sign of the patient such as heart rate, respiration rate, blood pressure, temperature, and / or any other information.

[0029] According to an embodiment, the received patient information may be associated with timestamps indicating time of admittance, discharge, treatment, measurement, lab test, vital sign, and / or any other aspect of the patient’s history or treatment. The patient information can include patient data and medical records that cover a period of time. For example, the patient information can include patient data and medical records for an entirety of a patient’s hospitalization to date, for a predetermined previous time, or any other period of time.

[0030] The patient information may be received or obtained from one or a plurality of different sources. According to an embodiment, the patient information is received from, retrieved from, or otherwise obtained from the electronic medical record database or system 270. The EMR database or system may be local or remote. The EMR database or system may be a component of the training system, or may be in local and / or remote communication with the training system. The received patient information may be utilized immediately, or may be stored in local or remote storage for use in further steps of the method.

[0031] At step 130 of the method, the training system utilizes the patient information received from the EMR database or system 270 to create one or more invention segments for subjects in the plurality of subjects. An intervention segment comprises a period of time during which a subject is treated with administration of a treatment. According to an embodiment, an intervention segment begins and ends based on predetermined criteria. For example, an intervention segment may begin when a particular type of treatment is administered, or may begin within a particular timeframe around administration of that particular type of treatment. That intervention segment may end, for example, when the particular type of treatment is no longer administered, or followinga gap of a predetermined time period between an initial treatment administration and a subsequent treatment administration. For example, the gap of a predetermined time period may be measured in hours, such as one hour, six hours, 12 hours, 24 hours, or any other time period. According to an embodiment, an intervention segment comprises a particular treatment or type of treatment for a particular condition or type of condition.

[0032] According to an embodiment, the training system extracts one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting. An extracted intervention segment may begin when any one of two or more different types of treatment, for a condition, are administered to the subject. The extracted intervention segment is associated with the one of the two or more different types of treatment administered to the subject, such that the type of treatment administered to the subject indicates the condition type experienced by the subject. According to an embodiment, the extracted intervention segment ends when there is a predetermined gap between: (1) administration of the first type of treatment and (2) a subsequent administration to the subject of a same or a different one of the different types of treatment. The predetermined gap may be measured in hours, such as one hour, six hours, 12 hours, 24 hours, or any other time period.

[0033] According to one non-limiting embodiment, the condition is hemodynamic instability or shock. There exist multiple shock types causing the instability, including cardiogenic shock (critical reduction of the heart’s pumping capacity), hypovolemic shock (severe blood loss or loss of fluids) and septic shock (dysregulated response to an infection resulting in life-threatening organ dysfunctions). According to an embodiment, while there may be other shock types causing the instability - such as anaphylactic shock and neurogenic shock - they may optionally be omitted since their prevalence is so low. However, according to an alternative embodiment, these and other shock types may be included.

[0034] Patients belonging to different shock types receive different types of hemodynamic interventions. Specifically: (1) cardiogenic shock patients receive inotropes, including dobutamine and milrinone, or mechanical cardiac support, including IABP (intra-aortic balloon pump), VA- ECMO (venoarterial extracorporeal membrane oxygenation), and / or LVAD (left ventricular assistant device), among other treatments; (2) hypovolemic shock patients receive packed red blood cells (PRBC) transfusion (in cases of hemorrhage) and fluid bolus (in cases of hypovolemia),among other treatments; 3) septic shock patients receive vasopressors (norepinephrine, phenylephrine, dopamine, vasopressin, epinephrine) and / or overlapping antibiotics, among other treatments. Given the discriminative power of these interventions, these different treatments can be used to label different shock types as well as the onset time of the shock.

[0035] According to an embodiment, for each shock type, the intervention segment starts when any of the interventions listed in the criteria starts (e.g. inotropes or MCS for cardiogenic shock, vasopressors with overlapping antibiotics for septic shock, etc.). The intervention segment continues until there is a gap of more than 12 hours between consecutive interventions. Therefore, each intervention segment is indexed by a label such as shock type, patient id, start time, and / or end time, among other possible labels. Note that multiple intervention segments can be generated from each patient.

[0036] According to an embodiment, treatment for cardiogenic shock includes inotropes, including dobutamine and milrinone, or mechanical cardiac support, including IABP (intra-aortic balloon pump), VA-ECMO (venoarterial extracorporeal membrane oxygenation), and / or LVAD (left ventricular assistant device), among other treatments.

[0037] For example, both dobutamine and milrinone are given as infusions. The onset time of each dobutamine / milrinone can be charted from the EMR. Although the offset time is not explicitly charted, it can be inferred by taking the minimum between 1) the onset time of the next administration of the same drug; and 2) the onset time of the current administration plus 2 hours, although other embodiments are possible. Given multiple records of inotropes administration over time for the same patient, the system can group those records with in-between gap time less than 12 hours in the same intervention segment. For mechanical cardiac support, the intervention often lasts for a few days or even longer. Therefore, for each patient, the system can restrict the labeling window up to the onset time of the first MCS. In this way, multiple intervention segments can be extracted for a patient.

[0038] According to an embodiment, treatment for hypovolemic shock includes packed red blood cells (PRBC) transfusion (in cases of hemorrhage) and fluid bolus (in cases of hypovolemia), among other treatments. According to an embodiment, a patient is determined as experiencing hypovolemic shock if the patient receives more than 800 cc packed red blood cells within 24 hours although other measurements and time periods are possible. The onset time of each PRBC administration record can be charted from the EMR. The offset time can be inferred as theminimum of 1) the onset time of the next PRBC administration; and 2) the onset time of the current PRBC administration plus four hours. Those records with in-between gap time less than 12 hours can be grouped as the same intervention segment, although other time periods are possible.

[0039] According to an embodiment, treatment for septic shock includes vasopressors (norepinephrine, phenylephrine, dopamine, vasopressin, epinephrine) and overlapping antibiotics, among other treatments. According to an embodiment, similarly to how the system utilizes inotropes administration records to create intervention segments for cardiogenic shock, the system can also create intervention segments for septic shock using vasopressor administration. However, for septic shock, the system may also require the administration of overlapping antibiotics, which is satisfied if either 1) the onset time of vasopressors is between the start and end time of antibiotics administration; or 2) the onset time of vasopressors is earlier than the start time of antibiotics administration but the time gap is less than 2 days, although other time periods are possible.

[0040] According to an embodiment, the system creates intervention segments by combining the intervention criteria of all shock types. An intervention segment may begin when any of the intervention criteria used to label shock types - as described or otherwise envisioned herein - is satisfied. The intervention segment may continue until there is a gap of more than 12 hours inbetween consecutive interventions.

[0041] Referring to FIG. 3, according to an embodiment, is an illustration 300 of extraction of multiple intervention segments based on application or administration of a treatment such as inotropes / MCS / PRBC / vasopressors, or other treatment. According to an embodiment, a subject experiences treatment as shown by a treatment timeline 310. At a first time point, the subject is treated with a treatment which results in the creation or extraction of a first intervention segment 320. The intervention segment 320 begins at the time of treatment. The intervention segment 320 ends when treatment ends, or after the treatment begins, provided that the next intervention segment 340 (defined by administration of a treatment such as inotropes / MCS / PRBC / vasopressors, or other treatment) begins after a predetermined time period following the first intervention segment. Here, the predetermined time period is 12 hours although other time periods are possible. Notably, the second treatment can be the same as or different from the first treatment.

[0042] According to another non-limiting embodiment, the condition is respiratory distress or failure. There exist multiple respiratory distress or failure, including but not limited to hypoxemicrespiratory failure and hypercapnic respiratory failure. Thus, there are at least a hypoxemic respiratory failure condition type and a hypercapnic respiratory failure condition type. According to an embodiment, the two or more different condition types are treated sufficiently differently such that the mode of treatment can be utilized to determine the condition type experienced by the subject. For example, the treatment for hypoxemic respiratory failure can comprise mechanical ventilation such as invasive ventilation or a face mask or other noninvasive ventilation, and the treatment for hypercapnic respiratory failure can comprise nasal ventilation. Intervention segments can be generated as described or otherwise envisioned herein.

[0043] According to other non-limiting embodiments, in addition to or as an alternative to hemodynamic shock and respiratory failure, the condition can be another medical condition provided that the condition comprises two or more condition types with an associated treatment that enables the system to distinguish between the two or more condition types.

[0044] At step 140 of the method, the training system labels the subject with a condition type label. A subject is labeled with the condition type label as having a specific one of the plurality of condition types. For example, the condition type label may be cardiogenic shock, hypovolemic shock, or septic shock when the condition is circulatory shock. As another example, the condition type label may be hypoxemic respiratory failure or hypercapnic respiratory failure when the condition is respiratory distress or failure.

[0045] According to an embodiment, the label is assigned based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting.

[0046] According to an embodiment, for each shock type, the system may only consider those intervention segments where the onset time is also identified in the second step. Consider, for example, a scenario where, starting from hour 0, the patient first receives inotropes for 6 hours. While the patient is still receiving inotropes, the patient is further given vasopressors (with overlapping antibiotics) at hour 6. Following step one, hour 0 will contribute one sample for cardiogenic shock and hour 6 will contribute one sample for septic shock. However, only hour 0 will be identified as the onset set for purposes of step 140. The system need only make use of samples at hour 0 because the patient has not received any hemodynamic intervention. In contrast,the patient has received inotropes for 6 hours at hour 6 and it will be inconsistent to use samples extracted after the intervention onset to build models to predict intervention onset.

[0047] Referring to FIG. 4, in one embodiment, is an illustration 400 of application or administration of multiple treatments to the same subject. According to an embodiment, a subject experiences treatment as shown by a treatment timeline 410. At a first time point, the subject is treated with inotropes at 420 for a time period of some hours. While the patient is still receiving inotropes, the patient begins to receive vasopressors with overlapping antibiotics at 430. The inotropes administration stops while the patient has received vasopressors for a few hours. While the patient is still receiving vasopressors, the patient starts to receive PRBC at 440. The question is, for example, how should the system create training samples from this patient to train the shock type prediction model?

[0048] According to an embodiment, there are at least three options for creating training samples from this patient to train the shock type prediction model.

[0049] For the first option, Option #1, the variables collected at tq (e.g., the intervention segment created) will be labeled as cardiogenic shock class, the variables collected at t2(e.g., the intervention segment created) will be labeled as septic shock, and t3(e.g., the intervention segment created) will be labeled as hypovolemic shock.

[0050] For the second option, Option #2, since the three intervention segments overlap, the system can only use variables at tj and label it as all three classes (cardiogenic shock, septic shock, and hypovolemic shock).

[0051] For the third option, Option #3, the system only uses variables collected at tj (e.g., the intervention segment created) and it is labelled as the cardiogenic shock class, since that is the treatment administered atAccording to this option, the system utilizes only the first type of interventions received by the patient within the intervention segment. If the patient received both inotropes and vasopressors with overlapping antibiotics at the start of the intervention segment, then the patient will be assigned to both cardiogenic and septic class.

[0052] According to an embodiment, Option #3 is the best way to label intervention segments and subjects, and thus the best way to create training data resulting in a superior trained prediction model. Thus, the methods and systems described or otherwise envisioned herein utilize Option #3 for creating training samples from a patient to train a prediction model.

[0053] At step 150 of the method, the system removes subjects in the plurality of subjects, according to one or more exclusion criteria. This results in a plurality of remaining subjects, surviving the exclusion, each associated with a condition type label. The exclusion criteria may be predetermined, or may be determined experimentally.

[0054] According to an embodiment, intervention onset times occurring earlier than six hours after ICU admission can be excluded. Other exclusion criteria may include: (1) removing patients less than 18 years old; (2) removing patients with a do-not-resuscitate (DNR) record; (3) removing patients who are not staying in the ICU at the time of intervention, and other exclusion criteria. The parameters of these exclusion criteria may be modified as well. For example, according to one embodiment, the system may remove patients above 18 years old.

[0055] According to an embodiment, the system may require that the patient stay in the ICU for at least six hours at the intervention onset. Therefore, samples with intervention onset time earlier than six hours after ICU admission can be excluded. Referring to FIG. 5, for example, is an illustration 500 of samples or intervention segments prior to a predetermined time period being removed or excluded. For example, treatment or intervention segment or subject 510 is prior to ICU admission 520, and thus is excluded. Similarly, treatment or intervention segment or subject 530 is less than six hours following ICU admission 520, and thus is excluded. However, treatment or intervention segment or subject 550 is at least or more than six hours 540 following ICU admission 520, and is thus included.

[0056] At step 160 of the method, the subjects remaining after applying the exclusion criteria, each of which are associated with a condition type label, and optionally associated with one or more intervention segments, are gathered into a training dataset. This can be any process for generating a dataset, including forming a specific data structure such as a database, table, or any other data structure comprising the data described or otherwise envisioned herein. For example, the subjects remaining after applying the exclusion criteria, including their association with a condition type label, can be saved in a database or table for utilization. Once generated, the database or other data structure can be utilized immediately or may be stored for downstream use.

[0057] At step 170 of the method, a predictive model is trained with the generated training dataset. According to an embodiment, a hemodynamic status model is trained with a generated training dataset to identify or predict a hemodynamic status of a subject. According to anotherembodiment, a respiratory status model is trained with a generated training dataset to identify or predict a respiratory status of a subject. Other predictive models are possible.

[0058] Referring to FIG. 6, in one embodiment, is a flowchart of a method 600 for training a predictive model, such as a hemodynamic status model. At step 610 of the method, a training system - which may be training system 200 or any other system - receives or obtains the training dataset generated according to the methods and systems described or otherwise envisioned herein, such as via the method described in conjunction with FIG. 1. Thus, the training data can comprise a plurality of subjects and their association with a condition type label. The training data can comprise any other patient information such as demographic information about the patients, diagnoses for the patients, medical history of the patients, treatment information, and / or any other information. The training data may be stored in and / or received from one or more databases. The database may be a local and / or remote database. For example, the training system may comprise a database of training data, such as the electronic medical record database or system 270.

[0059] According to an embodiment, the training system may comprise a data pre-processor or similar component or algorithm configured to process the received training data. For example, the data pre-processor analyzes the training data to remove noise, bias, errors, and other potential issues. The data pre-processor may also analyze the input data to remove low-quality data. Many other forms of data pre-processing or data point identification and / or extraction are possible.

[0060] At step 620 of the method, the system trains the model, which will be the algorithm utilized in analyzing the input information as described or otherwise envisioned. The model is trained using the training data set according to known methods for training a model. According to an embodiment, the model is trained, using the processed training dataset, to predict a condition status of a patient, such as a hemodynamic status, a respiratory status, and / or another status.

[0061] At step 630 of the method, the trained model of the system is stored for future use. According to an embodiment, the model may be stored in local or remote storage.

[0062] EXAMPLE

[0063] The following is an example implementation of the generation of a training dataset according to the methods described or otherwise envisioned herein. In this non-limiting example, the training dataset is generated for training a hemodynamic status prediction model. The example is provided only as one possible example or implementation or embodiment of the methodsdescribed or otherwise envisioned herein, and thus does not limit the scope of the methods or systems.

[0064] According to an embodiment, the methods described or otherwise envisioned herein were utilized to extract a shock type cohort from two different preexisting datasets of subjects and associated information. TABLE 1 identifies the sample size of each class.

[0065] TABLE 1. Sample size of extracted shock type cohorts.

[0066] Analysis of the extracted shock type cohorts showed that, when the distribution of key markers of each shock type across different shock types were compared from Dataset #1 and Dataset #2 for the different shock types, the aspartate aminotransferase (AST) results were highest for cardiogenic shock patients (as cardiogenic shock patients often have history of acute myocardial infarction, which leads to elevated level of AST). Further, hematocrit was lowest for hypovolemic shock patients (due to blood loss), and WBC was highest for septic patients (due to response to infection). These results are consistent with existing clinical literature, demonstrating that formation of the training dataset comprising the extracted shock type cohorts was successful.

[0067] Referring to FIG. 2 is a schematic representation of a training system 200. System 200 may be any of the systems described or otherwise envisioned herein, and may comprise any of the components described or otherwise envisioned herein. It will be understood that FIG. 2 constitutes, in some respects, an abstraction and that the actual organization of the components of the system 200 may be different and more complex than illustrated.

[0068] According to an embodiment, system 200 comprises a processor 220 capable of executing instructions stored in memory 230 or storage 260 or otherwise processing data to, for example, perform one or more steps of the method. Processor 220 may be formed of one or multiple modules. Processor 220 may take any suitable form, including but not limited to a microprocessor, microcontroller, multiple microcontrollers, circuitry, field programmable gate array (FPGA), application-specific integrated circuit (ASIC), a single processor, or plural processors.

[0069] Memory 230 can take any suitable form, including a non-volatile memory and / or RAM. The memory 230 may include various memories such as, for example LI, L2, or L3 cache or system memory. As such, the memory 230 may include static random access memory (SRAM), dynamic RAM (DRAM), flash memory, read only memory (ROM), or other similar memory devices. The memory can store, among other things, an operating system. The RAM is used by the processor for the temporary storage of data. According to an embodiment, an operating system may contain code which, when executed by the processor, controls operation of one or more components of system 200. It will be apparent that, in embodiments where the processor implements one or more of the functions described herein in hardware, the software described as corresponding to such functionality in other embodiments may be omitted.

[0070] User interface 240 may include one or more devices for enabling communication with a user. The user interface can be any device or system that allows information to be conveyed and / or received, and may include a display, a mouse, and / or a keyboard for receiving user commands. In some embodiments, user interface 240 may include a command line interface or graphical user interface that may be presented to a remote terminal via communication interface 250. The user interface may be located with one or more other components of the system, or may located remote from the system and in communication via a wired and / or wireless communications network.

[0071] Communication interface 250 may include one or more devices for enabling communication with other hardware devices. For example, communication interface 250 may include a network interface card (NIC) configured to communicate according to the Ethernet protocol. Additionally, communication interface 250 may implement a TCP / IP stack for communication according to the TCP / IP protocols. Various alternative or additional hardware or configurations for communication interface 250 will be apparent.

[0072] Storage 260 may include one or more machine-readable storage media such as readonly memory (ROM), random-access memory (RAM), magnetic disk storage media, optical storage media, flash-memory devices, or similar storage media. In various embodiments, storage 260 may store instructions for execution by processor 220 or data upon which processor 220 may operate. For example, storage 260 may store an operating system 261 for controlling various operations of system 200.

[0073] It will be apparent that various information described as stored in storage 260 may be additionally or alternatively stored in memory 230. In this respect, memory 230 may also be considered to constitute a storage device and storage 260 may be considered a memory. Various other arrangements will be apparent. Further, memory 230 and storage 260 may both be considered to be non-transitory machine-readable media. As used herein, the term non-transitory will be understood to exclude transitory signals but to include all forms of storage, including both volatile and non-volatile memories.

[0074] While system 200 is shown as including one of each described component, the various components may be duplicated in various embodiments. For example, processor 220 may include multiple microprocessors that are configured to independently execute the methods described herein or are configured to perform steps or subroutines of the methods described herein such that the multiple processors cooperate to achieve the functionality described herein. Further, where one or more components of system 200 is implemented in a cloud computing system, the various hardware components may belong to separate physical systems. For example, processor 220 may include a first processor in a first server and a second processor in a second server. Many other variations and configurations are possible.

[0075] According to an embodiment, the patient risk system comprises an electronic medical record (EMR) database or system 270. Alternatively, the EMR database or system 270 may be a local or remote database or system and thus the patient risk system may be in direct and / or indirect communication with the EMR database or system 270.

[0076] According to an embodiment, storage 260 of system 200 may store one or more algorithms, modules, and / or instructions to carry out one or more functions or steps of the methods described or otherwise envisioned herein. For example, the system may comprise, among other instructions or data, intervention segment identification instructions 262, labeling instructions 263, model training instructions 264, and a trained prediction model 265.

[0077] According to an embodiment, the intervention segment identification instructions 262 direct the system to create one or more invention segments for a plurality of subjects, as described or otherwise envisioned herein. An intervention segment comprises a period of time during which a subject is treated with administration of a treatment. According to an embodiment, an intervention segment begins and ends based on predetermined criteria. For example, anintervention segment may begin when a particular type of treatment is administered, or may begin within a particular timeframe around administration of that particular type of treatment. That intervention segment may end, for example, when the particular type of treatment is no longer administered, or following a gap of a predetermined time period between an initial treatment administration and a subsequent treatment administration. Thus, an extracted intervention segment may begin when any one of two or more different types of treatment, for a condition, are administered to the subject. The extracted intervention segment is associated with the one of the two or more different types of treatment administered to the subject, such that the type of treatment administered to the subject indicates the condition type experienced by the subject. According to an embodiment, the extracted intervention segment ends when there is a predetermined gap between: (1) administration of the first type of treatment and (2) a subsequent administration to the subject of a same or a different one of the different types of treatment.

[0078] According to an embodiment, the labeling instructions 263 direct the system to label subjects with a condition type label, as described or otherwise envisioned herein. A subject is labeled with the condition type label as having a specific one of the plurality of condition types. For example, the condition type label may be cardiogenic shock, hypovolemic shock, or septic shock when the condition is hemodynamic shock. As another example, the condition type label may be hypoxemic respiratory failure or hypercapnic respiratory failure when the condition is respiratory distress or failure. According to an embodiment, the label is assigned based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting.

[0079] According to an embodiment, the model training instructions 264 direct the system to train a predictive model, such as a hemodynamic status model, using the training dataset generated according to the methods and systems described or otherwise envisioned herein, such as via the method described in conjunction with FIG. 1. Thus, the training data can comprise a plurality of subjects and their association with a condition type label. The training data can comprise any other patient information such as demographic information about the patients, diagnoses for the patients, medical history of the patients, treatment information, and / or any other information. The trainingdata may be stored in and / or received from one or more databases. The database may be a local and / or remote database. The system trains the model, which will be the algorithm utilized in analyzing the input information as described or otherwise envisioned. The model is trained using the training data set according to known methods for training a model. According to an embodiment, the model is trained, using the processed training dataset, to predict a condition status of a patient, such as a hemodynamic status, a respiratory status, and / or another status. Following training, the trained model of the system is stored for future use. According to an embodiment, the model may be stored in local or remote storage. Thus, following training, system 200 comprises a trained prediction model 265.

[0080] Within the context of the disclosure herein, aspects of the embodiments may take the form of a computer program product embodied in one or more non-transitory computer-readable media having computer readable program code embodied thereon. Thus, according to one embodiment is a non-transitory computer-readable storage medium comprising computer program code instructions which, when executed by a processor, enables the processor to carry out a method including: (i) receiving historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the plurality of condition types; (ii) extracting, from the received EMR, one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting, wherein an extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and wherein the intervention segment is associated with the one of the different types of treatment are administered to the subject, and further wherein the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment; (iii) labeling, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting; (iv) removing any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; (v) gathering the remainingsubjects, each associated with a condition type label, in a training dataset; and / or training, using the training dataset, a model.

[0081] It should be appreciated that all combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are contemplated as being part of the inventive subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are contemplated as being part of the inventive subject matter disclosed herein. It should also be appreciated that terminology explicitly employed herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.

[0082] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0083] The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0084] The phrase “and / or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, 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 elements specifically identified.

[0085] As used herein in the specification and in the claims, the phrase “at least one,” in reference 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 within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified.

[0086] As used herein, although the terms first, second, third, etc. may be used herein to describe various elements or components, these elements or components should not be limited by these terms. These terms are only used to distinguish one element or component from another element or component. Thus, a first element or component discussed below could be termed a second element or component without departing from the teachings of the inventive concept.

[0087] Unless otherwise noted, when an element or component is said to be “connected to,” “coupled to,” or “adjacent to” another element or component, it will be understood that the element or component can be directly connected or coupled to the other element or component, or intervening elements or components may be present. That is, these and similar terms encompass cases where one or more intermediate elements or components may be employed to connect two elements or components. However, when an element or component is said to be “directly connected” to another element or component, this encompasses only cases where the two elements or components are connected to each other without any intermediate or intervening elements or components.

[0088] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of’ and “consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively.

[0089] It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.

[0090] 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. When any aspect is implemented at least in part in software, the software code can be executed on any suitable processor or collection of processors, whether provided in a single device or computer or distributed among multiple devices / computers.

[0091] The present disclosure can be implemented as a system, a method, and / or a computer program product at any possible technical detail level of integration. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.

[0092] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium can be, for example, but is 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 foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium comprises the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0093] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0094] Computer readable program instructions for carrying out operations of the present disclosure can be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, comprising an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable programinstructions can execute entirely on the user’s computer, partly on the user’s computer, as a standalone software package, partly on the user’s computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, comprising a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some examples, electronic circuitry comprising, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) can execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0095] 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 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.

[0100] The computer readable program instructions can be provided to a processor of a, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture comprising instructions which implement aspects of the function / act specified in the flowchart and / or block diagram or blocks.

[0101] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, otherprogrammable apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0102] The flowchart 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 flowchart or block diagrams can represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the blocks can occur out of the order noted in the Figures. For example, two blocks shown in succession can, in fact, be executed substantially concurrently, or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0103] Other implementations are within the scope of the following claims and other claims to which the applicant can be entitled.

[0104] While several inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and / or structures for performing the function and / or obtaining the results and / or one or more of the advantages described herein, and each of such variations and / or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and / or configurations will depend upon the specific application or applications for which the inventive teachings is / are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems,articles, materials, kits, and / or methods, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

Claims

ClaimsWhat is claimed is:

1. A method for generating a training dataset, comprising: receiving historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the plurality of condition types; extracting, from the received EMR, one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting, wherein an extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and wherein the intervention segment is associated with the one of the different types of treatment are administered to the subject, and further wherein the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment; labeling, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting; removing any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; and gathering the remaining subjects, each associated with a condition type label, in a training dataset.

2. The method of claim 1, wherein the labeling the subject as having a specific one of the plurality of condition types further comprises labeling the subject as having a second, different specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with second, different specific one of the plurality ofcondition types at the same time as receiving the one of the different types of treatment associated with the first specific one of the plurality of condition types.

3. The method of claim 1, wherein the condition is hemodynamic instability.

4. The method of claim 3, wherein the plurality of condition types are cardiogenic shock, hypovolemic shock, and septic shock.

5. The method of claim 4, wherein the type of treatment for cardiogenic shock comprises one or more of an inotrope and mechanical cardiac support, the type of treatment for hypovolemic shock comprises one or more of a transfusion or fluid bolus, and the type of treatment for septic shock comprises one or more of a vasopressor and an antibiotic.

6. The method of claim 1, further comprising the step of training, using the training dataset, a hemodynamic status model configured to identify a hemodynamic status of the subject.

7. The method of claim 6, wherein the hemodynamic status is a hemodynamic stability index, and hemodynamic status model is trained with the training dataset to predict whether a subject is likely to experience one or more of cardiogenic shock, hypovolemic shock, and septic shock.

8. The method of claim 1, wherein the condition is respiratory distress.

9. The method of claim 8, wherein the plurality of condition types are hypoxemic respiratory failure and hypercapnic respiratory failure.

10. The method of claim 9, wherein the type of treatment for hypoxemic respiratory failure comprises mechanical ventilation, and the type of treatment for hypercapnic respiratory failure comprises nasal ventilation.

11. A system for generating a training dataset, comprising:an EMR database comprising historical electronic medical records (EMR) for a plurality of subjects treated for a condition comprising a plurality of condition types, wherein treatment for the condition comprises a different type of treatment for each one of the plurality of condition types; and a processor configured to: extract, from the EMR, one or more intervention segments for each of the plurality of subjects beginning from admission to a health care setting, wherein an extracted intervention segment begins when any one of the different types of treatment are administered to the subject, and wherein the intervention segment is associated with the one of the different types of treatment are administered to the subject, and further wherein the extracted intervention segment ends when there is a predetermined gap between administration of the identified different type of treatment and a subsequent administration to the subject of a same or a different one of the different types of treatment; label, with a condition type label, the subject as having a specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with that specific one of the plurality of condition types, where the received one of the different type of treatment is associated with an earliest extracted intervention segment beginning first after the subject’s admission to the health care setting; remove any subjects in the plurality of subjects according to one or more exclusion criteria to result in a plurality of remaining subjects each associated with a condition type label; and gather the remaining subjects, each associated with a condition type label, in a training dataset.

12. The system of claim 1, wherein the labeling the subject as having a specific one of the plurality of condition types further comprises labeling the subject as having a second, different specific one of the plurality of condition types, based on the subject receiving the one of the different types of treatment associated with second, different specific one of the plurality ofcondition types at the same time as receiving the one of the different types of treatment associated with the first specific one of the plurality of condition types.

13. The system of claim 11, wherein the condition is hemodynamic instability.

14. The system of claim 13, wherein the plurality of condition types are cardiogenic shock, hypovolemic shock, and septic shock.

15. The system of claim 14, wherein the type of treatment for cardiogenic shock comprises one or more of an inotrope and mechanical cardiac support, the type of treatment for hypovolemic shock comprises one or more of a transfusion or fluid bolus, and the type of treatment for septic shock comprises one or more of a vasopressor and an antibiotic.

Images