Methods and apparatus for presenting alternative hypotheses for medical facts
View Patent ↗Techniques for presenting alternative hypotheses for medical facts may include identifying, using at least one statistical fact extraction model, a plurality of alternative hypotheses for a medical fact to be extracted from a portion of text documenting a patient encounter. At least two of the alternative hypotheses may be selected, and the selected hypotheses may be presented to a user documenting the patient encounter.
1. A method of analyzing a text documenting a patient encounter, the method comprising:
analyzing the text to identify a set of one or more features of at least a portion of the text;
correlating the set of one or more features to a set of alternative hypotheses for an abstract semantic concept representing an intended semantic meaning of the at least a portion of the text;
computing, using at least one statistical model implemented using at least one processor and for each of at least some of the alternative hypotheses of the set, one or more measures of estimated likelihood that the respective alternative hypothesis accurately represents the intended semantic meaning of the at least a portion of the text;
identifying hypotheses, including some or all of the at least some of the alternative hypotheses, to be presented to a user, wherein identifying the hypotheses comprises, for each of the at least some of the alternative hypotheses of the set:
in response to determining that the one or more measures of estimated likelihood for the alternative hypothesis satisfy one or more criteria, identifying that the alternative hypothesis is to be presented to the user; and
presenting the identified hypotheses, to the user documenting the patient encounter, as alternative hypotheses for a medical fact that could be extracted from the text.
2. The method of claim 1 , wherein at most one of the alternative hypotheses accurately represents the intended semantic meaning of the at least a portion of the text.
3. The method of claim 1 , wherein each of the alternative hypotheses represents a different semantic meaning.
4. The method of claim 1 , wherein the alternative hypotheses comprise a first hypothesis corresponding to a first concept and a second hypothesis corresponding to a second concept, wherein the first concept is a parent of the second concept.
5. The method of claim 1 , further comprising:
receiving input indicating a hypothesis selected by the user from among the presented hypotheses; and
designating the hypothesis indicated by the input as accurately representing the intended semantic meaning of the at least a portion of the text.
6. The method of claim 1 , wherein the presenting is performed in response to a selection by the user of the portion of the text.
7. The method of claim 1 , wherein:
the identified hypotheses comprise a first hypothesis and a second hypothesis of the at least some of the alternative hypotheses, and
the presenting comprises:
presenting the first hypothesis to the user; and
in response to a selection by the user of the presented first hypothesis, presenting the second hypothesis to the user.
8. The method of claim 1 , wherein
the identifying comprises including in the identified hypotheses a first hypothesis of the at least some of the alternative hypotheses having the highest computed measure of estimated likelihood, and at least one additional hypothesis of the at least some of the alternative hypotheses; and
wherein the presenting comprises:
presenting the first hypothesis in association with the text, and
in response to a selection by the user of the portion of the text and/or of the presented first hypothesis, presenting the at least one additional hypothesis.
9. An 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:
analyzing the text to identify a set of one or more features of at least a portion of the text;
correlating the set of one or more features to a set of alternative hypotheses for an abstract semantic concept representing an intended semantic meaning of the at least a portion of the text;
computing, using at least one statistical model implemented using at least one processor and for each of at least some of the alternative hypotheses of the set, one or more measures of estimated likelihood that the respective alternative hypothesis accurately represents the intended semantic meaning of the at least a portion of the text;
identifying hypotheses, including some or all of the at least some of the alternative hypotheses, to be presented to a user, wherein identifying the hypotheses comprises, for each of the at least some of the alternative hypotheses of the set:
in response to determining that the one or more measures of estimated likelihood for the alternative hypothesis satisfy one or more criteria, identifying that the alternative hypothesis is to be presented to the user; and
presenting the identified hypotheses, to the user documenting the patient encounter, as alternative hypotheses for a medical fact that could be extracted from the text.
10. The apparatus of claim 9 , wherein at most one of the alternative hypotheses accurately represents the intended semantic meaning of the at least a portion of the text.
11. The apparatus of claim 9 , wherein each of the alternative hypotheses represents a different semantic meaning.
12. The apparatus of claim 9 , wherein the alternative hypotheses comprise a first hypothesis corresponding to a first concept and a second hypothesis corresponding to a second concept, wherein the first concept is a parent of the second concept.
13. The apparatus of claim 9 , wherein the method further comprises:
receiving input indicating a hypothesis selected by the user from among the presented hypotheses; and
designating the hypothesis indicated by the input as accurately representing the intended semantic meaning of the at least a portion of the text.
14. The apparatus of claim 9 , wherein:
the identified hypotheses comprise a first hypothesis and a second hypothesis of the at least some of the alternative hypotheses, and
the presenting comprises:
presenting the first hypothesis to the user; and
in response to a selection by the user of the presented first hypothesis, presenting the second hypothesis to the user.
15. At least one non-transitory computer-readable storage medium encoded with computer-executable instructions that, when executed, perform a method comprising:
accessing a text documenting a patient encounter;
analyzing the text to identify a set of one or more features of at least a portion of the text;
correlating the set of one or more features to a set of alternative hypotheses for an abstract semantic concept representing an intended semantic meaning of the at least a portion of the text;
computing, using at least one statistical model implemented using at least one processor and for each of at least some of the alternative hypotheses of the set, one or more measures of estimated likelihood that the respective alternative hypothesis accurately represents the intended semantic meaning of the at least a portion of the text;
identifying hypotheses, including some or all of the at least some of the alternative hypotheses, to be presented to a user, wherein identifying the hypotheses comprises, for each of the at least some of the alternative hypotheses of the set:
in response to determining that the one or more measures of estimated likelihood for the alternative hypothesis satisfy one or more criteria, identifying that the alternative hypothesis is to be presented to the user; and
presenting the identified hypotheses, to the user documenting the patient encounter, as alternative hypotheses for a medical fact that could be extracted from the text.
16. The at least one non-transitory computer-readable storage medium of claim 15 , wherein at most one of the alternative hypotheses accurately represents the intended semantic meaning of the at least a portion of the text.
17. The at least one non-transitory computer-readable storage medium of claim 15 , wherein each of the alternative hypotheses represents a different semantic meaning.
18. The at least one non-transitory computer-readable storage medium of claim 15 , wherein the alternative hypotheses comprise a first hypothesis corresponding to a first concept and a second hypothesis corresponding to a second concept, wherein the first concept is a parent of the second concept.
19. The at least one non-transitory computer-readable storage medium of claim 15 , wherein the method further comprises:
receiving input indicating a hypothesis selected by the user from among the presented hypotheses; and
designating the hypothesis indicated by the input as accurately representing the intended semantic meaning of the at least a portion of the text.
20. The at least one non-transitory computer-readable storage medium of claim 15 , wherein:
the identified hypotheses comprise a first hypothesis and a second hypothesis of the at least some of the alternative hypotheses, and
the presenting comprises:
presenting the first hypothesis to the user; and
in response to a selection by the user of the presented first hypothesis, presenting the second hypothesis to the user.