Methods and apparatus for applying user corrections to medical fact extraction
First Claim
1. A method of analyzing a text documenting a patient encounter, the method comprising:
- extracting from a first portion of the text, using a fact extraction component implemented via at least one processor, a first set of one or more medical facts representing one or more abstract semantic concepts;
receiving information identifying a change made by a user to the first set of medical facts extracted from the first portion of the text, wherein the information identifying the change identifies at least one user-specified abstract semantic concept;
updating the fact extraction component based on the at least one user-specified abstract semantic concept and the first portion of the text, wherein the updating comprises;
identifying features included in the first portion of the text, the features including one or more tokens and/or one or more N-grams appearing in the first portion of the text;
identifying, from among the features identified in the first portion of the text, a first set of features to associate with the at least one user-specified abstract semantic concept, the first set of features including one or more features, the one or more features of the first set including one or more tokens and/or one or more N-grams; and
updating the fact extraction component to associate the first set of features with the at least one user-specified abstract semantic concept; and
extracting, using the fact extraction component following the updating and using the at least one processor, a second set of one or more medical facts from a second portion of the text, wherein extracting the second set comprises, in response to identifying in the second portion of the text a second set of features corresponding to the first set of features, extracting from the second portion of the text one or more facts corresponding to the at least one user-specified abstract semantic concept, the second set of features including one or more features.
1 Assignment
0 Petitions
Accused Products
Abstract
Techniques for applying user corrections to medical fact extraction may include extracting a first set of one or more medical facts from a first portion of text documenting a patient encounter. A correction to the first set of medical facts may be received from a user. The correction may identify a fact that should be associated with the first portion of the text. A second set of one or more medical facts may be extracted from a second portion of the text based at least in part on the user'"'"'s correction to the first set of medical facts. Extracting the second set of facts may include extracting one or more facts similar to the identified fact from the second portion of the text.
134 Citations
21 Claims
-
1. A method of analyzing a text documenting a patient encounter, the method comprising:
-
extracting from a first portion of the text, using a fact extraction component implemented via at least one processor, a first set of one or more medical facts representing one or more abstract semantic concepts; receiving information identifying a change made by a user to the first set of medical facts extracted from the first portion of the text, wherein the information identifying the change identifies at least one user-specified abstract semantic concept; updating the fact extraction component based on the at least one user-specified abstract semantic concept and the first portion of the text, wherein the updating comprises; identifying features included in the first portion of the text, the features including one or more tokens and/or one or more N-grams appearing in the first portion of the text; identifying, from among the features identified in the first portion of the text, a first set of features to associate with the at least one user-specified abstract semantic concept, the first set of features including one or more features, the one or more features of the first set including one or more tokens and/or one or more N-grams; and updating the fact extraction component to associate the first set of features with the at least one user-specified abstract semantic concept; and extracting, using the fact extraction component following the updating and using the at least one processor, a second set of one or more medical facts from a second portion of the text, wherein extracting the second set comprises, in response to identifying in the second portion of the text a second set of features corresponding to the first set of features, extracting from the second portion of the text one or more facts corresponding to the at least one user-specified abstract semantic concept, the second set of features including one or more features. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. Apparatus comprising:
-
at least one processor; and at least one memory storing processor-executable instructions that, when executed by the at least one processor, perform a method of analyzing a text documenting a patient encounter, the method comprising; extracting from a first portion of the text, using a fact extraction component, a first set of one or more medical facts representing one or more abstract semantic concepts; receiving information identifying a change made by a user to the first set of medical facts extracted from the first portion of the text, wherein the information identifying the change identifies at least one user-specified abstract semantic concept; updating the fact extraction component based on the at least one user-specified abstract semantic concept and the first portion of the text, wherein the updating comprises; identifying features included in the first portion of the text, the features including one or more tokens and/or one or more N-grams appearing in the first portion of the text; identifying, from among the features identified in the first portion of the text, a first set of features to associate with the at least one user-specified abstract semantic concept, the first set of features including one or more features, the one or more features of the first set including one or more tokens and/or one or more N-grams; and updating the fact extraction component to associate the first set of features with the at least one user-specified abstract semantic concept; and extracting, using the fact extraction component following the updating and using the at least one processor, a second set of one or more medical facts from a second portion of the text, wherein extracting the second set comprises, in response to identifying in the second portion of the text a second set of features corresponding to the first set of features, extracting from the second portion of the text one or more facts corresponding to the at least one user-specified abstract semantic concept, the second set of features including one or more features. - View Dependent Claims (19)
-
-
20. At least one non-transitory computer-readable storage medium encoded with computer-executable instructions that, when executed, perform a method of analyzing a text documenting a patient encounter, the method comprising:
-
extracting from a first portion of the text, using a set of one or more rules and/or one or more statistical models implemented via at least one processor, a first set of one or more medical facts representing one or more abstract semantic concepts; receiving information identifying a change made by a user to the first set of medical facts extracted from the first portion of the text, wherein the information identifying the change identifies at least one user-specified abstract semantic concept; updating the set of rules and/or the one or more statistical models based on the at least one user-specified abstract semantic concept and the first portion of the text, wherein the updating comprises; identifying features included in the first portion of the text, the features including one or more tokens and/or one or more N-grams appearing in the first portion of the text; identifying, from among the features identified in the first portion of the text, a first set of features to associate with the at least one user-specified abstract semantic concept, the first set of features including one or more features, the one or more features of the first set including one or more tokens and/or one or more N-grams; and updating the set of rules and/or the one or more statistical models to associate the first set of features with the at least one user-specified abstract semantic concept; and extracting, using the set of rules and/or the one or more statistical models following the updating and using the at least one processor, a second set of one or more medical facts from a second portion of the text, wherein extracting the second set comprises, in response to identifying in the second portion of the text a second set of features corresponding to the first set of features, extracting from the second portion of the text one or more facts corresponding to the at least one user-specified abstract semantic concept, the second set of features including one or more features. - View Dependent Claims (21)
-
Specification