Back to results
US 11,694,815 B2Granted
Intelligent ranking of sections of clinical practical guidelines
AssigneeInternational Business Machines Corporation
InventorsJoao H. Bettencourt-Silva, Natalia Mulligan, Marco Luca Sbodio, Theodora Brisimi
Life of this patent
Estimating the term…
Abstract
Embodiments for implementing intelligent ranking of sections of clinical practice guidelines by a processor. One or more clinical practice guidelines (CPGs) may be identified and analyzed. A score may be assigned to one or more sections of the CPGs according evidential data identified in one or more data sources. The one or more sections of the CPGs may be ranked according to the scoring.
Claims20
A method for implementing intelligent ranking of sections of clinical practice guidelines by a processor, comprising:
- receiving a selection, via a user interface of a computing device, of one or more clinical practice guidelines (CPGs) as text data;
- executing machine learning logic to analyze the one or more CPGs to generate one or more CPG models, wherein generating the one or more CPG models includes splitting the one or more CPGs into one or more sections of the CPGs based on a correlation of the text data between the one or more sections of the CPGs, and wherein the splitting of the one or more CPGs into the one or more sections of the CPGs includes:
- ingesting the text data from a data source,
- performing a topic extraction operation on the text data to identify key topic terms,
- generating a word embeddings model of the text data,
- performing a query of a bag-of-terms nearest to the key topic terms, and
- generating the one or more sections of the CPGs according to an output of the query;
- subsequent to receiving the selection of the one or more CPGs, retrieving evidential data for the one or more CPGs from a plurality of data sources by parsing each of the plurality of data sources to identify and extract the evidential data relevant to the one or more CPGs among irrelevant data to the one or more CPGs in the plurality of data sources, wherein retrieving the evidential data includes automatically generating systematic reviews of a particular field of the one or more CPGs, using database search tools to perform the parsing, and instantiating instances of Artificial Intelligence (AI) logic by the machine learning logic to identify and analyze the evidential data;
- assigning a score to the one or more sections of CPGs according to the evidential data identified in the plurality of data sources;
- ranking the one or more sections of the CPGs according to the scoring;
- and
- outputting, via the user interface of the computing device, the ranking of the one or more sections of the CPGs according to the scoring.
9 more elements
A system for implementing intelligent ranking of sections of clinical practice guidelines, comprising:
- one or more computers with executable instructions that when executed cause the system to:
- receive a selection, via a user interface of a computing device, of one or more clinical practice guidelines (CPGs) as text data;
- execute machine learning logic to analyze the one or more CPGs to generate one or more CPG models, wherein generating the one or more CPG models includes splitting the one or more CPGs into one or more sections of the CPGs based on a correlation of the text data between the one or more sections of the CPGs, and wherein the splitting of the one or more CPGs into the one or more sections of the CPGs includes:
- ingesting the text data from a data source,
- performing a topic extraction operation on the text data to identify key topic terms,
- generating a word embeddings model of the text data,
- performing a query of a bag-of-terms nearest to the key topic terms, and generating the one or more sections of the CPGs according to an output of the query;
- subsequent to receiving the selection of the one or more CPGs, retrieve evidential data for the one or more CPGs from a plurality of data sources by parsing each of the plurality of data sources to identify and extract the evidential data relevant to the one or more CPGs among irrelevant data to the one or more CPGs in the plurality of data sources, wherein retrieving the evidential data includes automatically generating systematic reviews of a particular field of the one or more CPGs, using database search tools to perform the parsing, and instantiating instances of Artificial Intelligence (AI) logic by the machine learning logic to identify and analyze the evidential data;
- assign a score to the one or more sections of CPGs according to the evidential data identified in the plurality of data sources;
- rank the one or more sections of the CPGs according to the scoring;
- and
- output, via the user interface of the computing device, the ranking of the one or more sections of the CPGs according to the scoring.
9 more elements
A computer program product for implementing intelligent ranking of sections of clinical practice guidelines by a processor, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
- an executable portion that receives a selection, via a user interface of a computing device, of one or more clinical practice guidelines (CPGs) as text data;
- an executable portion that executes machine learning logic to analyze the one or more CPGs to generate one or more CPG models, wherein generating the one or more CPG models includes splitting the one or more CPGs into one or more sections of the CPGs based on a correlation of the text data between the one or more sections of the CPGs, and wherein the splitting of the one or more CPGs into the one or more sections of the CPGs includes:
- ingesting the text data from a data source,
- performing a topic extraction operation on the text data to identify key topic terms,
- generating a word embeddings model of the text data,
- performing a query of a bag-of-terms nearest to the key topic terms, and generating the one or more sections of the CPGs according to an output of the query;
- an executable portion that, subsequent to receiving the selection of the one or more CPGs, retrieves evidential data for the one or more CPGs from a plurality of data sources by parsing each of the plurality of data sources to identify and extract the evidential data relevant to the one or more CPGs among irrelevant data to the one or more CPGs in the plurality of data sources, wherein retrieving the evidential data includes automatically generating systematic reviews of a particular field of the one or more CPGs, using database search tools to perform the parsing, and instantiating instances of Artificial Intelligence (AI) logic by the machine learning logic to identify and analyze the evidential data;
- an executable portion that assigns a score to the one or more sections of CPGs according to the evidential data identified in the plurality of data sources;
- an executable portion that ranks the one or more sections of the CPGs according to the scoring;
- and
- an executable portion that outputs, via the user interface of the computing device, the ranking of the one or more sections of the CPGs according to the scoring.
8 more elements
Dependencies are inferred from the claim text.