Patient data mining for lung cancer screening
First Claim
1. A lung cancer screening system comprising:
- a database including patient population information in a structured format and for a patient population associated with lung cancer and a domain knowledge base including knowledge information about lung cancer;
an individual patient record of an individual patient; and
a processor configured, using a diagnosis module program tangibly embodied on a program storage device, to screen the individual patient record for a diagnosis of malignant lung cancer, the diagnosis screening comprising determining a current state of the individual patient from a combination of values from multiple variables of the individual patient and determining whether or not the current state of the individual patient indicates the diagnosis of the malignant lung cancer as a function of the individual patient record, the patient population information in the structured format, and the knowledge information about lung cancer from the domain knowledge base of the database, the processor configured to determine a probability less than 100 percent of the diagnosis that the patient is of the current state, the probability being a function of the individual patient record and the knowledge information about lung cancer;
wherein the processor is configured to, for a first one of the values of a first one of the variables, receive different possible values from different sources for the first value of the individual patient, identify a conflict between the different sources, reconcile the conflict between the different possible values for the first value based on a knowledge base, the probability being a function of the reconciled conflict between the different possible values;
wherein the processor further comprises a modeling module for extracting only patient records from the patient population information for patients with similar characteristics as the individual patient, and for analyzing the extracted patient population information of the database to determine a trend in the patients of the patient population with the similar characteristics of the individual patient as determined by the individual patient record, the modeling module predicting a progression of lung cancer in the patient based on the determined trend and predicting lung cancer risk.
3 Assignments
0 Petitions
Accused Products
Abstract
A system and method for lung cancer screening is provided. The system includes a database including structured patient information for a patient population and a domain knowledge base including information about lung cancer; an individual patient record; and a processor for analyzing the patient record with data from the database to determine if a patient has indications of lung cancer. The method includes the steps of inputting patient-specific data into a patient record; performing at least one lung cancer screening procedure on a patient, wherein at least one result from at least one procedure is inputted into the patient record in a structured format; and analyzing the patient record with a domain knowledge base to determine if the patient has indications of lung cancer.
129 Citations
14 Claims
-
1. A lung cancer screening system comprising:
-
a database including patient population information in a structured format and for a patient population associated with lung cancer and a domain knowledge base including knowledge information about lung cancer; an individual patient record of an individual patient; and a processor configured, using a diagnosis module program tangibly embodied on a program storage device, to screen the individual patient record for a diagnosis of malignant lung cancer, the diagnosis screening comprising determining a current state of the individual patient from a combination of values from multiple variables of the individual patient and determining whether or not the current state of the individual patient indicates the diagnosis of the malignant lung cancer as a function of the individual patient record, the patient population information in the structured format, and the knowledge information about lung cancer from the domain knowledge base of the database, the processor configured to determine a probability less than 100 percent of the diagnosis that the patient is of the current state, the probability being a function of the individual patient record and the knowledge information about lung cancer; wherein the processor is configured to, for a first one of the values of a first one of the variables, receive different possible values from different sources for the first value of the individual patient, identify a conflict between the different sources, reconcile the conflict between the different possible values for the first value based on a knowledge base, the probability being a function of the reconciled conflict between the different possible values; wherein the processor further comprises a modeling module for extracting only patient records from the patient population information for patients with similar characteristics as the individual patient, and for analyzing the extracted patient population information of the database to determine a trend in the patients of the patient population with the similar characteristics of the individual patient as determined by the individual patient record, the modeling module predicting a progression of lung cancer in the patient based on the determined trend and predicting lung cancer risk. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A lung cancer screening system comprising:
-
a database including patient population information in a structured format and for a patient population and a domain knowledge base including knowledge information about lung cancer; an individual patient record of an individual patient; a processor configured, using a first program comprising a diagnosis module program tangibly embodied on a first program storage device, to determine whether or not the individual patient has indications of lung cancer as a function of the patient record and data from the database, the determination being part of screening for whether a current state of the individual patient indicates a diagnosis of malignant lung cancer, the current state determined from a combination of values from multiple variables of the individual patient; and a data miner configured, using a second program tangibly embodied on a second program storage device, to populate the database in the structured format with the patient population information representing values of variables indicative or not of malignant lung cancer by data mining patient records in both structured and unstructured formats; wherein the processor is configured to, for a first one of the values of a first one of the variables, receive different possible values from different sources for the first value of the individual patient, identify a conflict between the different sources, reconcile the conflict between the different possible values for the first value based on a knowledge base; wherein the processor further comprises a modeling module for extracting only patient records from the patient population information for patients with similar characteristics as the individual patient, and for analyzing the extracted patient population information of the database to determine a trend in the patients of the patient population with the similar characteristics of the individual patient as determined by the individual patient record, the modeling module predicting a progression of lung cancer in the patient based on the determined trend and predicting lung cancer risk, the prediction of the lung cancer risk being a function of the reconciled conflict between the different possible values.
-
-
14. A lung cancer screening system comprising:
-
a database including patient population information in a structured format and for a patient population and a domain knowledge base including knowledge information about lung cancer; an individual patient record of an individual patient; a processor configured, using a diagnosis module program tangibly embodied on a program storage device, to determine that the individual patient has indications of a diagnosis of lung cancer as a function of a combination of multiple sources in the patient record and as a function of data from the database, the determination being part of screening for whether or not a current state of the individual patient includes malignant lung cancer; and a data miner for creating the individual patient record by mining values of variables indicative or not of malignant lung cancer from sources in both structured and unstructured formats; wherein the processor is configured to, for a first value of a first variable of the indications of the individual patient, receive different possible values from the multiple sources, identify a conflict between the multiple sources, and reconcile the conflict between the different possible values for the first value based on a knowledge base; wherein the processor further comprises a modeling module for extracting only patient records from the patient population information for patients with similar characteristics as the individual patient, and for analyzing the extracted patient population information of the database to determine a trend in the patients of the patient population with the similar characteristics of the individual patient as determined by the individual patient record, the modeling module predicting a progression of lung cancer in the patient based on the determined trend and predicting lung cancer risk, the prediction of the lung cancer risk being a function of the reconciled conflict between the different possible values.
-
Specification