Back to results
US 2026/0049998 A1Published application
Methods of detecting and treating cerebral aneurysms
AssigneesUniversity of Pittsburgh—Of the Commonwealth System of Higher Education, Upmc
InventorsRobert Max Friedlander, Michael Maurice McDowell, Aditya M. Mittal, Ali A. Alattar, Kamil W. Nowicki
Life of this patent
Estimating the term…
Abstract
The present disclosure provides a whole blood, protein-based diagnostic test for presence and evaluation of aneurysm status. Further, the present disclosure relates to methods of treating aneurysms.
Claims20
A method of classifying an aneurysm in a subject, the method comprising:
- receiving expression level data of one or more biomarkers in a biological sample obtained from the subject;
- inputting the expression level data into a machine learning model;
- analyzing the data using the machine learning model to generate one or more measures regarding (i) aneurysm presence and aneurysm rupture, (ii) aneurysm presence and aneurysm status, (iii) aneurysm presence and aneurysm location, or (iv) a combination thereof, and
- generating a diagnosis output based on the one or more measures that classify the aneurysm in the subject;
- wherein the machine learning model is trained based on:
- generating a first set of predictive models using a first training data set comprising expression data from a set of biomarkers across a patient cohort, wherein the first training data set is not age-matched;
- selecting a subset of the first set of predictive models, wherein each predictive model of the subset of the first set of predictive models is generated based on respective expression data associated with a respective subset of the set of biomarkers;
- deriving a second set of predictive models using a second training data set comprising expression data associated with the subsets of the set of biomarkers, wherein the second training data set is age-matched;
- and
- generating the machine learning model based on the second set of predictive models.
7 more elements
A method of training a machine learning model configured to detect an aneurysm, the method comprising:
- generating a first set of predictive models using a first training data set comprising expression data from a set of biomarkers across a patient cohort, wherein the first training data set is not age-matched;
- selecting a subset of the first set of predictive models, wherein each predictive model of the subset of the first set of predictive models is generated based on respective expression data associated with a respective subset of the set of biomarkers;
- deriving a second set of predictive models using a second training data set comprising expression data associated with the subsets of the set of biomarkers, wherein the second training data set is age-matched;
- and
- generating the machine learning model based on the second set of predictive models.
2 more elements
A system for predicting an aneurysm status in a subject, the method comprising:
- one or more processors;
- a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions for:
- (a) training a machine learning model on expression level data of one or more biomarkers in biological samples obtained from a cohort;
- (b) receiving expression level data of one or more biomarkers in a biological sample obtained from a subject;
- (c) inputting the expression level data into the machine learning model;
- and
- (d) predicting, using the machine learning model, the aneurysm status in the subject;
- wherein the machine learning model is trained based on:
- generating a first set of predictive models using a first training data set comprising expression data from a set of biomarkers across a patient cohort, wherein the first training data set is not age-matched;
- selecting a subset of the first set of predictive models, wherein each predictive model of the subset of the first set of predictive models is generated based on respective expression data associated with a respective subset of the set of biomarkers;
- deriving a second set of predictive models using a second training data set comprising expression data associated with the subsets of the set of biomarkers, wherein the second training data set is age-matched;
- and
- generating the machine learning model based on the second set of predictive models.
10 more elements
Dependencies are inferred from the claim text.