Personalized health risk assessment for critical care

• US 8,990,135 B2
• Filed: 10/05/2011
• Issued: 03/24/2015
• Est. Priority Date: 06/15/2010
• Status: Active Grant

• Current Assignee: Henry Ford Health System, Board of Regents of the University of Michigan (University of Michigan)
• Original Assignee: Henry Ford Health System, Board of Regents of the University of Michigan (University of Michigan)
• Inventors: Syed, Zeeshan H., Rubinfeld, Ilan S.
• Primary Examiner(s): Wong, Lut
• Assistant Examiner(s): BUSS, BENJAMIN J
• Application Number:
  US13/253,928
• Publication Number:
  US 20120059779A1
• Time in Patent Office: 1,266 Days
• Field of Search: 706/45, 706/924
• US Class Current: 706/45
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass
• CPC Class Codes: Y10S 706/924 Medical

Abstract

A method for assessing whether a patient is at risk of developing a clinical condition includes receiving training data representing a set of patient-related variables for each of a plurality of patients; generating model data based on the received training data; receiving target data representing the set of patient-related variables for a target patient; determining a risk level for the target patient of developing the clinical condition; and indicating the risk level of the target patient, where the set of patient-related variables consists of a first set of variables when the clinical condition is a mortality condition and a second set of variables when the clinical condition is a morbidity condition.

Claims

1. A computer-implemented method of assessing whether a patient is at risk of developing a clinical condition, the method comprising:

receiving from a database training data representing a set of patient-related variables for each of a plurality of patients;

generating model data via a computer based on the received training data;

receiving target data representing the set of patient-related variables for a target patient;

determining via the computer a risk level for the target patient of developing the clinical condition, wherein determining the risk level includes performing one or more calculations involving the received target data and the generated model data; and

indicating the risk level of developing the clinical condition,

wherein

i) when the clinical condition is a mortality condition, the set of patient-related variables consists of between three and ten variables selected from the group consisting of;

functional health status prior to surgery, American Society of Anesthesiologists (ASA) classification, preoperative serum albumin, age, presence of disseminated cancer, preoperative blood urea nitrogen (BUN), do-not-resuscitate (DNR) status, emergent versus non-emergent case, work relative value unit, and presence of ascites, and

ii) when the clinical condition is a morbidity condition, the set of patient-related variables consists of between three and ten variables selected from the group consisting of;

ASA classification, work relative value unit, preoperative serum albumin, emergent versus non-emergent case, functional health status prior to surgery, inpatient versus outpatient case, preoperative systemic sepsis, age, steroid use for chronic condition, and weight.

2. The method of claim 1, wherein receiving the training data from the database includes receiving the training data from a National Surgical Quality Improvement Program (NSQIP) database.

3. The method of claim 1, wherein the clinical condition is a mortality condition.

4. The method of claim 1, wherein the clinical condition is a morbidity condition.

5. A computer-implemented method of assessing whether a patient is at risk of developing a clinical condition, the method comprising:

receiving from a database training data representing a set of patient-related variables for each of a plurality of patients;

generating model data via a computer based on the received training data;

receiving target data representing the set of patient-related variables for a target patient;

determining via the computer a risk level for the target patient of developing the clinical condition, wherein determining the risk level includes performing one or more calculations involving the received target data and the generated model data; and

indicating the risk level of developing the clinical condition, wherein

i) when the clinical condition is a mortality condition, the set of patient-related variables consists of between three and five variables selected from the group consisting of;

functional health status prior to surgery, American Society of Anesthesiologists (ASA) classification, preoperative serum albumin, age, presence of disseminated cancer, preoperative blood urea nitrogen (BUN), do-not-resuscitate (DNR) status, emergent versus non-emergent case, work relative value unit, and presence of ascites, and

ii) when the clinical condition is a morbidity condition, the set of patient-related variables consists of between three and five variables selected from the group consisting of;

ASA classification, work relative value unit, preoperative serum albumin, emergent versus non-emergent case, functional health status prior to surgery, inpatient versus outpatient case, preoperative systemic sepsis, age, steroid use for chronic condition, and weight.

6. The method of claim 5, wherein receiving the training data from the database includes receiving the training data from a National Surgical Quality Improvement Program (NSQIP) database.

7. The method of claim 5, wherein the clinical condition is a mortality condition.

8. The method of claim 5, wherein the clinical condition is a morbidity condition.

9. A computer-implemented method of assessing whether a patient is at risk of developing a clinical condition, the method comprising:

receiving from a database training data representing a set of patient-related variables for each of a plurality of patients;

generating model data via a computer based on the received training data;

receiving target data representing the set of patient-related variables for a target patient;

determining via the computer a risk level for the target patient of developing the clinical condition, wherein determining the risk level includes performing one or more calculations involving the received target data and the generated model data; and

indicating the risk level of developing the clinical condition, wherein

i) when the clinical condition is a mortality condition, the set of patient-related variables consists of between three and five variables selected from the group consisting of;

functional health status prior to surgery, American Society of Anesthesiologists (ASA) classification, preoperative serum albumin, age, and presence of disseminated cancer, and

ii) when the clinical condition is a morbidity condition, the set of patient-related variables consists of between three and five variables selected from the group consisting of;

ASA classification, work relative value unit, preoperative serum albumin, emergent versus non-emergent case, and functional health status prior to surgery.

10. The method of claim 9, wherein receiving the training data from the database includes receiving the training data from a National Surgical Quality Improvement Program (NSQIP) database.

11. The method of claim 9, wherein the clinical condition is a mortality condition.

12. The method of claim 9, wherein the clinical condition is a morbidity condition.

13. A non-transitory computer-readable storage medium comprising computer-readable instructions to be executed on a processor of a system for assessing whether a patient is at risk of developing a clinical condition, the instructions when executed cause the processor to:

receive training data from a database representing a set of patient-related variables for each of a plurality of patients;

generate model data based on the received training data;

receive target data representing the set of patient-related variables for a target patient;

automatically determine a risk level for the target patient of developing the clinical condition, wherein determining the risk level includes causing the processor to perform one or more calculations involving the received target data and the generated model data; and

automatically indicate the risk level of developing the clinical condition,

wherein

i) when the clinical condition is a mortality condition, the set of patient-related variables consists of between three and ten variables selected from the group consisting of;

functional health status prior to surgery, American Society of Anesthesiologists (ASA) classification, preoperative serum albumin, age, presence of disseminated cancer, preoperative blood urea nitrogen (BUN), do-not-resuscitate (DNR) status, emergent versus non-emergent case, work relative value unit, and presence of ascites, and

ii) when the clinical condition is a morbidity condition, the set of patient-related variables consists of between three and ten variables selected from the group consisting of;

ASA classification, work relative value unit, preoperative serum albumin, emergent versus non-emergent case, functional health status prior to surgery, inpatient versus outpatient case, preoperative systemic sepsis, age, steroid use for chronic condition, and weight.