Back to results
US 2008/0288227 A1Published application
Algorithms to predict clinical response, adherence, and shunting with thiopuriness
AssigneeUniversity of Michigan
InventorsPeter Higgins, Ji Zhu, Akbar Waljee, Sijian Wang, Joel Joyce
Life of this patent
Estimating the term…
Abstract
A method of using a variable set from complete blood counts and blood chemistry panels to generate a machine learned algorithm for determining the effectiveness of thiopurine treatment on inflammatory bowel disease (IBD) patients using CART, boosted trees, random forest classification, RuleFit and/or logistic regression analysis.
Claims23
A method of determining the effectiveness of thiopurine treatment in inflammatory bowel disease (IBD) comprising:
- receiving a set of inputs including at least three of white blood cell count (WBC), hemoglobin (HGB), hematocrit (HCT), platelet count (PLT), red blood cell count (RBC), mean corpuscular volume (MCV). mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), mean platelet volume (MPV), sodium level (SOD), potassium level (POT), chloride (CHLOR), carbon dioxide (CO2), urea nitrogen (UN), creatinene level (CREAT), glucose level (GLUC), calcium level (CAL), protein level (PROT), albumin level (ALB), aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALK), bilirubin total (TBIL), neutrophil count (NEUT), lymphocyte count (LYMPH), moncyte counte (MONO), eosinophil count (EOS), basophil count (BASO), or age of patient (Age) and
- determining at least one of a clinical response probability, an adherence probability, or a shunting probability based on the set of inputs.
A method of creating a model for determining the effectiveness of thiopurine treatment in inflammatory bowel disease (IBD) comprising:
- receiving a first set of inputs including at least one value for a set of variables including at least three of white blood cell count (WBC), hemoglobin (HGB), hematocrit (HCT), platelet count (PLT), red blood cell count (RBC), mean corpuscular volume (MCV). mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), mean platelet volume (MPV), sodium level (SOD), potassium level (POT), chloride (CHLOR), carbon dioxide (CO2), urea nitrogen (UN), creatinene level (CREAT), glucose level (GLUC), calcium level (CAL), protein level (PROT), albumin level (ALB), aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALK), bilirubin total (TBIL), neutrophil count (NEUT), lymphocyte count (LYMPH), moncyte counte (MONO), eosinophil count (EOS), basophil count (BASO), or age of patient (Age), and
- generating a set of conditions from a random forest tree model using the first set of inputs as independent variables and using one of clinical response, adherence or shunting as a dependent variable.
A method of determining the effectiveness of thiopurine treatment in inflammatory bowel disease (IBD) comprising:
- receiving a first set of inputs including at least one value for a set of variables including at least three of white blood cell count (WBC), hemoglobin (HGB), hematocrit (HCT), platelet count (PLT), red blood cell count (RBC), mean corpuscular volume (MCV). mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), mean platelet volume (MPV), sodium level (SOD), potassium level (POT), chloride (CHLOR), carbon dioxide (CO2), urea nitrogen (UN), creatinene level (CREAT), glucose level (GLUC), calcium level (CAL), protein level (PROT), albumin level (ALB), aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALK), bilirubin total (TBIL), neutrophil count (NEUT), lymphocyte count (LYMPH), moncyte counte (MONO), eosinophil count (EOS), basophil count (BASO), or age of patient (Age), and
- determining one of clinical response, adherence or shunting for a patient by analyzing the inputs using random forest model conditions.
A computer-storage medium having computer-executable instructions for performing operations comprising:
- storing an input set comprising at least three of white blood cell count (WBC), hemoglobin (HGB), hematocrit (HCT), platelet count (PLT), red blood cell count (RBC), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), mean platelet volume (MPV), sodium level (SOD), potassium level (POT), chloride (CHLOR), carbon dioxide (CO2), urea nitrogen (UN), creatinene level (CREAT), glucose level (GLUC), calcium level (CAL), protein level (PROT), albumin level (ALB), aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALK), bilirubin total (TBIL), neutrophil count (NEUT), lymphocyte count (LYMPH), moncyte counte (MONO), eosinophil count (EOS), basophil count (BASO), or age of patient (Age);
- storing a regression and classification tree model generated algorithm;
- calculating at least one of a clinical response probability, an adherence probability, or a shunting probability based on the inputs using the generated algorithm;
- displaying at least one of a clinical response probability, an adherence probability, or a shunting probability on a display device.
1 more element
Dependencies are inferred from the claim text.