System and method for determining triage categories
Embodiments disclosed herein provide a system, method, and computer program product for providing a triage classification system. The triage classification system uses a computer model that is developed using historical patient data. The developed computer model is applied to collected patient attribute data from a patient in a pre-hospital setting to generate a triage category. Based on the generated triage category, health care professionals can take desired actions, such as transporting the patient to a facility matching the generated triage category.
1 . A computer-implemented method for triaging patients in pre-hospital settings, the method comprising:
receiving, by a computer, patient data from devices in a pre-hospital setting, the patient data describing a patient in the pre-hospital setting, the computer having a machine learning (ML) model for assigning a triage category to the patient based on a plurality of input variables associated with the patient, wherein the ML model is trained with the plurality of input variables based on historical patient data that describe patients who had been assigned triage categories in at least one pre-hospital setting prior to the patients being transported or delivered into care of one or more trauma centers, each of the one or more trauma centers having a level designation corresponding to a respective one of the triage categories, wherein the ML model is trained using a subset of the historical patient data until the ML model is able to produce a correct triage classification corresponding to a respective subject of the assigned triage categories;
tuning the ML model according to a desired over triage rate or a desired under triage rate so that triage categories generated by the ML model are more likely to yield the desired over triage rate or under triage rate;
applying, by the computer, the ML model thus tuned to the patient data received from the devices in the pre-hospital setting, wherein the applying comprises processing the plurality of input variables associated with the patient and generating a triage category for the patient in the pre-hospital setting based on the processing; and
presenting, by the computer via a user interface, the triage category generated by the ML model for the patient in the pre-hospital setting for transporting or delivering the patient into care of a trauma center having a level designation corresponding to the triage category generated for the patient.
2 . The method according to claim 1 , wherein the ML model comprises a statistical model that includes a classifier configured for taking as input one or more attributes collected from the patient and determining a triage category for the patient based on the one or more attributes collected from the patient.
3 . The method according to claim 2 , wherein the classifier is configured for processing a number of input variables fewer than the plurality of input variables and for generating a triage category for the patient in the pre-hospital setting based on the number of input variables associated with the patient.
4 . The method according to claim 1 , wherein the at least one pre-hospital setting is in a geographic region, wherein the historical patient data were collected from at least one trauma center in the geographic region, and wherein the ML model is tailored to triage levels present in the geographic region based on the historical patient data collected from the at least one trauma center in the geographic region.
5 . The method according to claim 4 , wherein the historical patient data comprises the triage categories assigned to the patients in the at least one pre-hospital setting and a disposition reflecting accuracy of an initial triage level assigned in the at least one pre-hospital setting.
6 . The method according to claim 4 , wherein the historical patient data were collected from the at least one trauma center in the geographic region by elements within the at least one trauma center.
7 . The method according to claim 4 , wherein the historical patient data were collected from the at least one trauma center in the geographic region by a third party.
8 . A triage system for triaging patients in pre-hospital settings, the triage system comprising:
a processor;
a non-transitory computer-readable medium; and
instructions stored on the non-transitory computer-readable medium and translatable by the processor for:
receiving patient data from devices in a pre-hospital setting, the patient data describing a patient in the pre-hospital setting, the triage system having a machine learning (ML) model for assigning a triage category to the patient based on a plurality of input variables associated with the patient, wherein the ML model is trained with the plurality of input variables based on historical patient data that describe patients who had been assigned triage categories in at least one pre-hospital setting prior to the patients being transported or delivered into care of one or more trauma centers, each of the one or more trauma centers having a level designation corresponding to a respective one of the triage categories, wherein the ML model is trained using a subset of the historical patient data until the ML model is able to produce a correct triage classification corresponding to a respective subject of the assigned triage categories;
tuning the ML model according to a desired over triage rate or a desired under triage rate so that triage categories generated by the ML model are more likely to yield the desired over triage rate or under triage rate;
applying the ML model thus tuned to the patient data received from the devices in the pre-hospital setting, wherein the applying comprises processing the plurality of input variables associated with the patient and generating a triage category for the patient in the pre-hospital setting based on the processing; and
presenting, via a user interface, the triage category generated by the ML model for the patient in the pre-hospital setting for transporting or delivering the patient into care of a trauma center having a level designation corresponding to the triage category generated for the patient.
9 . The system of claim 8 , wherein the ML model comprises a statistical model that includes a classifier configured for taking as input one or more attributes collected from the patient and determining a triage category for the patient based on the one or more attributes collected from the patient.
10 . The system of claim 9 , wherein the classifier is configured for processing a number of input variables fewer than the plurality of input variables and for generating a triage category for the patient in the pre-hospital setting based on the number of input variables associated with the patient.
11 . The system of claim 8 , wherein the at least one pre-hospital setting is in a geographic region, wherein the historical patient data were collected from at least one trauma center in the geographic region, and wherein the ML model is tailored to triage levels present in the geographic region based on the historical patient data collected from the at least one trauma center in the geographic region.
12 . The system of claim 11 , wherein the historical patient data comprises the triage categories assigned to the patients in the at least one pre-hospital setting and a disposition reflecting accuracy of an initial triage level assigned in the at least one pre-hospital setting.
13 . The system of claim 11 , wherein the historical patient data were collected from the at least one trauma center in the geographic region by elements within the at least one trauma center.
14 . The system of claim 11 , wherein the historical patient data were collected from the at least one trauma center in the geographic region by a third party.
15 . A non-transitory computer program product for triaging patients in pre-hospital settings, the non-transitory computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor of a triage system for:
receiving patient data from devices in a pre-hospital setting, the patient data describing a patient in the pre-hospital setting, the triage system having a machine learning (ML) model for assigning a triage category to the patient based on a plurality of input variables associated with the patient, wherein the ML model is trained with the plurality of input variables based on historical patient data that describe patients who had been assigned triage categories in at least one pre-hospital setting prior to the patients being transported or delivered into care of one or more trauma centers, each of the one or more trauma centers having a level designation corresponding to a respective one of the triage categories, wherein the ML model is trained using a subset of the historical patient data until the ML model is able to produce a correct triage classification corresponding to a respective subject of the assigned triage categories;
tuning the ML model according to a desired over triage rate or a desired under triage rate so that triage categories generated by the ML model are more likely to yield the desired over triage rate or under triage rate;
applying the ML model thus tuned to the patient data received from the devices in the pre-hospital setting, wherein the applying comprises processing the plurality of input variables associated with the patient and generating a triage category for the patient in the pre-hospital setting based on the processing; and
presenting, via a user interface, the triage category generated by the ML model for the patient in the pre-hospital setting for transporting or delivering the patient into care of a trauma center having a level designation corresponding to the triage category generated for the patient.
16 . The computer program product of claim 15 , wherein the ML model comprises a statistical model that includes a classifier configured for taking as input one or more attributes collected from the patient and determining a triage category for the patient based on the one or more attributes collected from the patient.
17 . The computer program product of claim 16 , wherein the classifier is configured for processing a number of input variables fewer than the plurality of input variables and for generating a triage category for the patient in the pre-hospital setting based on the number of input variables associated with the patient.
18 . The computer program product of claim 15 , wherein the at least one pre-hospital setting is in a geographic region, wherein the historical patient data were collected from at least one trauma center in the geographic region, and wherein the ML model is tailored to triage levels present in the geographic region based on the historical patient data collected from the at least one trauma center in the geographic region.
19 . The computer program product of claim 18 , wherein the historical patient data comprises the triage categories assigned to the patients in the at least one pre-hospital setting and a disposition reflecting accuracy of an initial triage level assigned in the at least one pre-hospital setting.
20 . The computer program product of claim 18 , wherein the historical patient data were collected from the at least one trauma center in the geographic region by elements within the at least one trauma center.