Back to results
US 12,597,520 B2Granted
System and method for automated diagnosis
AssigneesThe Regents of the University of Michigan, Empowr-Me LLC
InventorsPrashant Mahajan, Dilip Dubey
Life of this patent
Estimating the term…
Abstract
An automated diagnosis method executed by one or more computing devices including: a) receiving one or more data inputs inputted by a user into a user computing device; b) converting data from the one or more data inputs into stored data by one or more processors and storing within one or more storage mediums; c) determining one or more severity indexes based on the stored data with the one or more processors; d) predicting a presence of one or more medical conditions based on the one or more severity indexes by the one or more processors; and e) identifying a presence, an absence, or both of the one or more medical conditions by the one or more processors and transmitting for output to the user computing device.
Claims20
An automated diagnosis method executed by one or more computing devices comprising:
- a) receiving a plurality of data inputs including (i) one or more data inputs inputted by a user into a user computing device, (ii) one or more data inputs from one or more sensing devices, and (iii) one or more data inputs from one or more other applications;
- wherein the one or more sensing devices detect and measure one or more vital signals one or more physiological signals, or both;
- wherein the one or more sensing devices are separate from the user computing device;
- wherein the user computing device is a mobile computing device including a screen, a microphone, and a camera;
- wherein the one or more data inputs inputted by the user include the user vocally describing one or mot oms being experienced by the user using the microphone of the user computing device;
- wherein the one or more data inputs inputted by the user include the user uploading one or more images, one or more videos, or both showing at least one of the one or more symptoms experienced by the user using the camera of the user computing device;
- and
- wherein the one or more other applications include one or more fitness and health applications, one or more photograph applications, one or more navigation applications, or a combination thereof;
- b) converting a plurality of data from the plurality of data inputs from the user the one of more sensing devices, and the one or more other applications into a stored data by one or more processors and storing within one or more storage mediums;
- wherein the converting is automatically executed after receiving the plurality of inputs;
- wherein at least some of the plurality of data is unstructured data and the converting includes converting the unstructured data to a structured data;
- and
- wherein the converting of the unstructured data to the structured data includes natural language processing;
- speech to text;
- and data cleaning to decode at least some of the one or more data inputs received as the one or more images and/or the one or more videos, and a voice;
- c) determining one or more severity indexes based on the stored data with the one or more processors, wherein the determining of the one or more severity indexes includes converting the stored data in one or more dynamic digital profiles into the one or more severity indexes, and wherein the stored data in a dynamic digital profile of an event is used to determine an emergency severity index value based at least in part on an event time and an event duration;
- wherein the determining of the one or more severity indexes is automatically executed upon the converting of the plurality of data being completed;
- wherein the stored data and the one or more dyna mic digital profiles are stored within one or more databases;
- wherein the one or more dynamic digital profiles include a dynamic digital profile of a patient and the dynamic digital profile of the event;
- wherein the dynamic digital profile of the patient utilizes information from health history and trends, including medical history, demographics, regular symptoms, regular vitals, regular behavior, regular feeling, location GPS, regular visuals, and regular voice;
- and
- wherein the dynamic digital profile of the event utilizes event information including one or more chief complaints an arrival mode, the event time the event duration, event vitals, environment, and weather;
- d) predicting a presence of one or more medical conditions based on the one or more severity indexes by the one or more processors;
- wherein the predicting the presence of the one or more medical conditions is automatically executed upon the determining the one or more severity indexes;
- wherein the predicting be presence of the one or more medical conditions includes performing one or more condition tests, which include comparing the stored data, the one or more severity indexes, or both to one or more known conditions and filtering for those with common data patterns;
- wherein the one or more condition tests execute one or more machine learning models;
- and
- wherein the one or more machine learning models include one or more regression models;
- e) identifying the presence of the one or more medical conditions to the user by transmitting an output containing the presence of the one or more medical conditions to the user computing device via the one or more processors onto a user interface of the user computing device;
- and
- wherein the output indicates a severity of the one or more medical conditions and provides one or more recommendations with respect to medical care and attention, including whether immediate medical attention is necessary.
28 more elements
An automated diagnosis method executed by one or more computing devices comprising:
- a) receiving a plurality of data inputs (i) inputted by a user into a user interface of a user computing device, (ii) received from one or more sensing devices, and (iii) received from one or more other applications;
- wherein the plurality of data inputs include one or more vitals, a presence or an absence of critical symptoms, and a presence or an absence of current symptoms;
- wherein the one or more sensing devices detect and measure the one or more vital signals, one or more physiological signals, or both;
- wherein the one or more sensing devices are separate from the user computing device;
- wherein the user computing device is a mobile computing device with the user interface which includes a screen, a microphone and a camera;
- wherein one or more data inputs inputted by the user include the user vocally describing one or more symptoms being experienced by the user using the microphone of the user computing device;
- wherein the one or more data inputs inputted by the user include the user uploading one or more images, one or more videos, or both showing at least one of the one or more symptoms using the camera of the user computing device;
- and
- wherein the one or more other applications include one or more fitness and health applications, one or more photograph applications, one or more navigation applications, or a combination thereof;
- b) converting a plurality of data from the plurality of data inputs from the user, the one or more sensing devices, and the one or more other applications into a stored data by one or more processors and storing within one or more storage mediums, wherein the stored data is stored as part of a dynamic digital profile of a patient and a dynamic digital profile of an event;
- wherein the converting is automatically executed after receiving the plurality of inputs;
- wherein at least some of the plurality of data is unstructured data and the converting includes converting the unstructured data to a structured data;
- wherein the converting the unstructured data to the structured data includes natural language processing:
- speech to text;
- and data cleaning to decode at least some of the one or more data inputs received as the one or more images and/or the one or more videos, and a voice;
- wherein the dynamic digital profile of the patient and the dynamic digital profile of the event are stored within one or more databases;
- c) determining one or more severity indexes based on the stored data with the one or more processors by converting the stored data in the dynamic digital profile of the patient and the dynamic digital profile of the event into the one or more severity indexes, and wherein the stored data in the dynamic digital profile of the event is used to determine an emergency severity index value based at least in part on an event time and an event duration;
- wherein the determining of the one or more severity indexes is automatically executed upon the converting being completed;
- wherein the dynamic digital profile of the patient utilizes information from health history and trends including medical history, demographics regular symptoms, regular vitals, regular behavior, regular feeling, location GPS, regular visuals, and regular voice;
- wherein the dynamic digital profile of the patient includes the stored data related to each occurrence when the user submits the one or more data inputs into the user interface;
- wherein the dynamic digital profile of the event utilizes event information including one or more chief complaints, arrival mode, the event time, the event duration, event vitals, environment, and weather;
- wherein the dynamic digital profile of the event includes the stored data related to one or more occurrences when the user submits the one or more data inputs into the user interface which result in a diagnosis of the one or more medical conditions;
- d) predicting a presence of one or more medical conditions based on the one or more severity indexes by the one or more processors;
- wherein the predicting the presence of the one or more medical conditions is automatically executed upon the determining of the one or more severity indexes;
- wherein the predicting the presence of the one or more medical conditions includes performing one or more condition tests, which include comparing the stored data, the one or more severity indexes, or both to one or more known conditions and filtering for those with common data patterns;
- wherein the one or more condition tests execute one or more machine learning models;
- and
- wherein the one or more machine learning models include one or more regression models;
- e) identifying the presence of the one or more medical conditions to the user, based on the predicting the presence of the one or more medical conditions, by transmitting an output containing the presence of the one or more medical conditions to the user interface of the user computing device via the one or more processors;
- and
- wherein the output indicates a severity of the one or more medical conditions and provides one or more recommendations with respect to medical care and attention, including whether immediate medical attention is necessary.
28 more elements
An automated diagnosis method executed by one or more computing devices comprising:
- a) receiving a plurality of data inputs (i) inputted by a user into a user interface of a user computing device, (ii) received from one or more sensing devices, and (iii) received from one or more other applications;
- wherein the plurality of data inputs include one or more vitals, a presence or an absence of critical symptoms, and a presence or an absence of current symptoms;
- wherein the one or more sensing devices detect and measure one or more vital signals, one or more physiological signals, or a combination thereof;
- wherein the one or more sensing devices are separate from the user computing device;
- wherein the user computing device is a mobile computing device with the user interface which includes a screen, a microphone, and a camera;
- wherein one or more data inputs inputted by the user include a user vocally describing one or more symptoms experienced by the user using the microphone of the user computing device;
- wherein one or more data inputs inputted by the user include the user uploading one or more images, one or more videos, or both showing at least one of the one or more symptoms using the camera of the user computing device;
- and
- wherein the one or more other applications include one or more fitness and health applications, photograph applications, navigation applications, or a combination thereof;
- b) converting a plurality of data from the plurality of data inputs into a stored data by one or more processors and storing within one or more storage mediums, wherein the stored data is stored as part of a dynamic digital profile of a patient and a dynamic digital profile of an event;
- wherein the converting is automatically executed after receiving the plurality of inputs;
- wherein at least some of the plurality of data inputs are unstructured data and the converting includes converting the unstructured data to a structured data;
- wherein the converting the unstructured data to the structured data includes natural language processing:
- speech to text;
- and data cleaning to decode one or more inputs received as the one or more images and/or the one or more videos, and a voice;
- wherein the dynamic digital profile of the patient and the dynamic digital profile of the event are stored within one or more databases;
- c) determining one or more severity indexes based on the stored data with the one or more processors by converting the stored data in the dynamic digital profile of the patient and the dynamic digital profile of the event into the one or more severity indexes, and wherein the stored data in the dynamic digital profile of the event is used to determine an emergency severity index value and the stored data in the dynamic digital profile of the patient is used to determine a patient severity index based at least in part on an event time and an event duration;
- wherein the determining of the one or more severity indexes is automatically executed upon the converting of the plurality of data being completed;
- wherein the dynamic digital profile of the patient utilizes information from health history and trends including medical history, demographics, regular symptoms, regular vitals, regular behavior, regular feeling, location GPS, regular visuals, and regular voice;
- wherein the dynamic digital profile of the patient includes the stored data related to each occurrence when the user submits the one or more data inputs into the user interface;
- wherein the dynamic digital profile of the event utilizes event information including one or more chief complaints, arrival mode, the event time, the event duration, event vitals, environment, and weather;
- wherein the dynamic digital profile of the event includes the stored data related to one or more occurrences when the user submits the one or more data inputs into the user interface which result in a diagnosis of the one or more medical conditions;
- d) predicting a presence of one or more medical conditions based on the one or more severity indexes by the one or more processors, wherein the predicting the presence of the one or more medical conditions includes the steps of:
- i) automatically inputting the patient severity index and the emergency severity index into a default model to determine a probability level;
- ii) automatically comparing the probability level to a threshold level obtained from a threshold index table;
- ii-a) if the probability level is greater than or equal to the threshold level, an output of a diagnosis risk and/or medical recommendation is provided;
- ii-b) if the probability level is less than the threshold level, model selection is automatically performed from a model toolbox;
- ii-b-i) model execution is automatically performed on a model selected during the model selection, and an updated probability is determined;
- iii-b-ii) the updated probability is automatically compared to the threshold level;
- ii-b-iii-a) if the updated probability is greater than or equal to the threshold level, the output of a diagnosis risk and/or medical recommendation is provided;
- and
- ii-b-ii-b) if the updated probability level is less than the threshold level, no diagnosis risk or medical recommendation is provided;
- wherein the predicting the presence of the one or more medical conditions is automatically executed upon the determining of the one or more severity indexes;
- wherein the predicting the presence of the one or more medical conditions includes performing one or more condition tests which include comparing the stored data, the one or more severity indexes, or both to one or more known conditions and filtering for those with common data patterns;
- wherein the one or more condition tests execute one or more machine learning models;
- and
- wherein the one or more machine learning models include one or more regression models;
- e) identifying the presence of the one or more medical conditions to the user, based on the predicting the presence of the one or more medical conditions, by transmitting the output containing the presence of the one or more medical conditions to the user interface of the user computing device via the one or more processors;
- and
- wherein the output indicates a severity of the one or more medical conditions and provides one or more recommendations with respect to medical care and attention, including whether immediate medical attention is necessary.
37 more elements
Dependencies are inferred from the claim text.