Natural Language Processing Correction Based on Treatment Plan

Patent metadata

Field	Value
Publication number	US-2016110501-A1
Application number	US-201414514563-A
Country	US
Kind code	A1
Filing date	Oct 15, 2014
Priority date	Oct 15, 2014
Publication date	Apr 21, 2016
Grant date	—

Abstract

An approach is provided in which an information handling system extracts treatment segments from documents corresponding to a patient and uses cognitive analysis to identify common treatment properties of a subset of the treatment segments. The information handling system combines the subset of treatment segments into a treatment aggregation that corresponds to a treatment history of the patient. In turn, the information handling system ingests the treatment aggregation into a domain for subsequent processing.

First claim

1 . A method implemented by an information handling system that includes a memory and a processor, the method comprising:
extracting, by the processor, a plurality of treatment segments from a plurality of documents corresponding to a patient; identifying one or more common treatment properties based upon cognitive analysis of the plurality of treatment segments, wherein the one or more common treatment properties correspond to a subset of the plurality of treatment segments; combining the subset of treatment segments into a treatment aggregation, wherein the treatment aggregation corresponds to a treatment history of the patient; and ingesting the treatment aggregation into the information handling system.

2 . The method of claim 1 wherein the cognitive analysis further comprises:
evaluating one or more cognitive relationship entries, wherein a first one of the one or more cognitive relationship entries associates a first drug to a second drug;
determining that a first one of the plurality of treatment segments comprises the first drug and determining that a second one of the plurality of treatment segments comprises the second drug; and
including the first treatment segment and the second treatment segment into the subset of treatment segments based upon the first cognitive relationship entry.

3 . The method of claim 1 wherein the treatment aggregation corresponds to a chronological treatment history of the patient, the method further comprising:
generating a chronological composition of the treatment aggregation, wherein the chronological composition comprises a treatment timeline indicator corresponding to the chronological treatment history, one or more first drug indicators corresponding to the first drug, and one or more second drug indicators corresponding to the second drug; and
adding the chronological composition to a chronological graph.

4 . The method of claim 1 further comprising:
determining that a selected one of the plurality of treatment segments is mutually exclusive to the treatment aggregation based upon one or more cognitive relationship entries, wherein the selected treatment segment includes a first start date;
determining an estimated treatment end date of the treatment aggregation based upon the first start date; and
adding the estimated treatment end date to the treatment aggregation.

5 . The method of claim 4 wherein the treatment aggregation comprises a second start date based upon one of the subset of treatment segments, the method further comprising:
determining a treatment time of the treatment aggregation based upon the second start date and the estimated treatment end date;
comparing the treatment time to one or more clinical guidelines that correspond to the treatment aggregation;
adjusting the estimated treatment end date based upon the comparison; and
generating a notification in response to the adjusting of the estimated treatment end date.

6 . The method of claim 1 wherein the plurality of documents are written in a natural language context, the method further comprising:
parsing a selected one of the plurality of documents into a plurality of sentence parts; and
deriving a corresponding one of the plurality of treatment segments based upon natural language processing analysis of the plurality of sentence parts.

7 . The method of claim 1 further comprising:
determining that a first one of the plurality of treatment segments comprises a first drug and determining that a second one of the plurality of treatment segments comprises the first drug; and
removing the second treatment segment from the plurality of treatment segments.

8 . The method of claim 1 wherein the treatment segment comprises data corresponding to a treatment administered to the patient, and wherein the treatment segment is selected from the group consisting of one or more sentences, a phrase, a term, and an n-gram.

9 . An information handling system comprising:
one or more processors; a memory coupled to at least one of the processors; and a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions of:
extracting a plurality of treatment segments from a plurality of documents corresponding to a patient;
identifying one or more common treatment properties based upon cognitive analysis of the plurality of treatment segments, wherein the one or more common treatment properties correspond to a subset of the plurality of treatment segments;
combining the subset of treatment segments into a treatment aggregation, wherein the treatment aggregation corresponds to a treatment history of the patient; and
ingesting the treatment aggregation into the information handling system.

10 . The information handling system of claim 9 wherein the one or more processors perform additional actions comprising:
evaluating one or more cognitive relationship entries, wherein a first one of the one or more cognitive relationship entries associates a first drug to a second drug;
determining that a first one of the plurality of treatment segments comprises the first drug and determining that a second one of the plurality of treatment segments comprises the second drug; and
including the first treatment segment and the second treatment segment into the subset of treatment segments based upon the first cognitive relationship entry.

11 . The information handling system of claim 9 wherein the treatment aggregation corresponds to a chronological treatment history of the patient, and wherein the one or more processors perform additional actions comprising:
generating a chronological composition of the treatment aggregation, wherein the chronological composition comprises a treatment timeline indicator corresponding to the chronological treatment history, one or more first drug indicators corresponding to the first drug, and one or more second drug indicators corresponding to the second drug; and
adding the chronological composition to a chronological graph.

12 . The information handling system of claim 9 wherein the one or more processors perform additional actions comprising:
determining that a selected one of the plurality of treatment segments is mutually exclusive to the treatment aggregation based upon one or more cognitive relationship entries, wherein the selected treatment segment includes a first start date;
determining an estimated treatment end date of the treatment aggregation based upon the first start date; and
adding the estimated treatment end date to the treatment aggregation.

13 . The information handling system of claim 12 wherein the treatment aggregation comprises a second start date based upon one of the subset of treatment segments, and wherein the one or more processors perform additional actions comprising:
determining a treatment time of the treatment aggregation based upon the second start date and the estimated treatment end date;
comparing the treatment time to one or more clinical guidelines that correspond to the treatment aggregation;
adjusting the estimated treatment end date based upon the comparison; and
generating a notification in response to the adjusting of the estimated treatment end date.

14 . The information handling system of claim 9 wherein the plurality of documents are written in a natural language context, and wherein the one or more processors perform additional actions comprising:
parsing a selected one of the plurality of documents into a plurality of sentence parts; and
deriving a corresponding one of the plurality of treatment segments based upon natural lang

Assignees

• IBM

Inventors

• Allen Corville O
• Dettman Elizabeth T
• Freed Andrew R
• Payne Michael T
• Schroeder Michael W

Classifications

• G06F19/322 Primary

  Physics · mapped topic

• G16Z99/00

  Subject matter not provided for in other main groups of this subclass · CPC title

• G16H10/60 Primary

  for patient-specific data, e.g. for electronic patient records · CPC title