Flatiron Health, Inc.

Sharang Phadke

Nisha Singh

Guy Amster

Will Shapiro

Lauren Sutton

Angel Leung

Janet Donegan

Dominic Green

Alexander Padmos

Chun-Che Wang

Addison Shelley

Brett Wittmershaus

Edward Liu

Lucy He

Michael Waskom

Natalie Roher

John Ritten

James Gippetti

Melissa Estevez

Sankeerth Garapati

Aaron Cohen

Suganya Sridharma

Described herein are techniques of using machine learning to automatically extract clinical variable values for subjects from clinical record data. The techniques designate certain clinical variables as hybrid variables that can be assigned values by machine learning model prediction. The techniques process, using a machine learning model trained to predict a value of a hybrid variable, clinical record data associated with a subject to obtain a predicted hybrid variable value and an associated confidence score. The techniques set the value of the hybrid variable for the subject to the predicted hybrid variable value when the model prediction is of sufficiently high confidence.

Machine learning extraction of clinical variable values for subjects from clinical record data

A computer-implemented system for identifying a patient for a trial may include at least one processor. The at least one processor may be programmed to receive an indication of a selected trial, the selected trial being associated with a testing status criterion; access a plurality of patient records associated with a patient of a plurality of patients; determine, using a machine learning model and based on unstructured information from one at least one of the patient records, a likelihood of an occurrence of genomic testing for the patient; determine a genomic testing status of the patient based on the determined likelihood of the occurrence of genomic testing; determine that the genomic testing status satisfies the testing status criterion; and include the patient in a subset of the plurality of patients based on the genomic testing status satisfying the testing status criterion.

Systems and methods for determining a genomic testing status

A model-assisted system for extracting patient information. A processor may be programmed to access a database storing one or more medical records associated with a patient and determine, using a first machine learning model and based on unstructured information included in the one or more medical records, whether the patient is associated with a condition. The processor may further be programmed to identify a date associated with the patient and determine, using a second machine learning model and based on the unstructured information, whether the patient is associated with the condition relative to the date. The processor may generate an output indicating whether the patient is associated with the condition and whether the patient is associated with the condition relative to the date.

Systems and methods for extracting dates associated with a patient condition

A computer-implemented system for determining a genomic testing status of a patient may include at least one processor programmed receive, from a source, unstructured information from a plurality of patient records associated with a patient; determine, using a first machine learning model, a primary patient record from among the plurality of patient records, wherein at least a portion of information represented in the primary patient record correlates to genomic testing; determine, using a second machine learning model and based on unstructured information from one at least one of the patient records, a likelihood of an occurrence of genomic testing for the patient; determine a genomic testing status of the patient based on the determined likelihood of the occurrence of genomic testing; and display a user interface comprising an indicator of the genomic testing status of the patient and a link to the primary patient record.

Growing community of inventors

Plainsboro, NJ, United States of America

Sharang Phadke