Method and system for automating clinical data standards
View Patent ↗A method and a clinical data standards (CDS) automated system are provided for automating clinical data standards and generating study data tabulation model (SDTM) artifacts required for a regulatory submission process using a machine learning model and a natural language processing (NLP) engine with minimal user intervention. The CDS automated system extracts metadata from multiple raw datasets automatically using NLP and feeds the extracted metadata into the machine learning model; predicts automatic case report form (CRF) annotations on the extracted metadata and records new learnings onto the CDS automated system; maps one or more raw datasets against a target SDTM variable; generates an SDTM statistical analysis system (SAS) code, an SDTM specification, and one or more SDTM datasets; generates a define package; validates the generated define package and the SDTM artifacts generated throughout the entire cycle; and generates validation reports in real time.
1 . A method for automating clinical data standards and generating study data tabulation model (SDTM) artifacts required for a regulatory submission process using a machine learning model and a natural language processing engine, the method employing a clinical data standards automated system comprising at least one processor configured to execute computer program instructions for performing the method comprising:
extracting metadata from raw datasets automatically using natural language processing and feeding the extracted metadata into the machine learning model, by the clinical data standards automated system;
pre-filling the extracted metadata into a blank case report form (CRF), by the clinical data standards automated system;
predicting automatic case report form (CRF) annotations on the extracted metadata using the machine learning model, based on annotated metadata information available in the clinical data standards automated system and information acquired from previous learnings, and recording new learnings into the clinical data standards automated system with minimal user intervention, by the clinical data standards automated system, wherein the previous learnings are learnings recorded by the clinical data standards automated system based on the annotated metadata information manually acquired from a user into the blank case report form (CRF) at beginning of a clinical trial process, wherein the predicted CRF annotations and corresponding text and attributes are sent to an application in the clinical data standards automated system, and wherein the application inserts CRF annotations into the blank CRF pre-filled with the extracted metadata, based on prediction from the machine learning model;
mapping the raw datasets against a target SDTM variable based on the extracted metadata available in the clinical data standards automated system and information acquired from the previous learnings, by the clinical data standards automated system;
generating an SDTM statistical analysis system (SAS) code based on the mapping, the prediction, and derivations, an SDTM specification, and SDTM datasets based on the mapping, by the clinical data standards automated system;
generating a define package based on annotated metadata information and the mapping available in the clinical data standards automated system, by the clinical data standards automated system; and
validating the generated define package and SDTM artifacts generated during a development and review cycle, and generating validation reports in real time, by the clinical data standards automated system.
2 . The method of claim 1 , wherein the clinical data standards automated system predicts derivation logic from a derivation repository based on mapping combinations, study design, and the extracted metadata, and wherein the generated SDTM SAS code is executable outside the clinical data standards automated system in a SAS application.
3 . The method of claim 1 , wherein the natural language processing engine preprocesses text information before feeding the text information into the machine learning model.
4 . The method of claim 1 , wherein the define package is a Clinical Data Interchange Standards Consortium (CDISC) standard “define.xml” file and a reviewer's guide, and wherein the “define.xml” file is an extensible markup language (XML) document that describes structure and content of data collected during the clinical trial process.
5 . The method of claim 1 , wherein the machine learning model is used to identify the derivations and generate the study data tabulation model (SDTM) statistical analysis system (SAS) code based on the derivations.
6 . The method of claim 1 , wherein the SDTM artifacts comprise the generated define.xml files, stylesheets, the SDTM datasets, an annotated case report form (CRF), the reviewers guide, and supporting documents, and wherein the validations are based on the Clinical Data Interchange Standards Consortium (CDISC) or United States Food and Drug Administration (FDA)-published validation rules.
7 . A clinical data standards automated system for automating clinical data standards and generating study data tabulation model (SDTM) artifacts required for a regulatory submission process using a machine learning model and a natural language processing engine, the system comprising:
a non-transitory, computer readable storage medium configured to store computer program instructions defined by modules of the clinical data standards automated system; and
at least one processor communicatively coupled to the non-transitory computer readable storage media, the at least one processor configured to execute the computer program instructions defined by the modules of the clinical data standards automated system, the modules comprising:
a metadata extractor for extracting metadata from raw datasets automatically using natural language processing and feeding the extracted metadata into the machine learning model;
a case report form extractor for pre-filling the extracted metadata into a blank case report form (CRF);
an annotator for automatically predicting and annotating contents in the blank CRF and the extracted metadata, using the machine learning model, based on annotated metadata information available in the clinical data standards automated system and information acquired from previous learnings, and recording new learnings into the clinical data standards automated system with minimal user intervention, wherein the previous learnings are learnings recorded by the clinical data standards automated system based on the annotated metadata information manually acquired from a user into the blank case report form (CRF) at beginning of a clinical trial process, wherein the predicted CRF annotations and corresponding text and attributes are sent to an application in the clinical data standards automated system, and wherein the application inserts CRF annotations into the blank CRF pre-filled with the extracted metadata, based on prediction from the machine learning model;
a smart mapper for mapping the raw datasets against a target SDTM variable based on the extracted metadata available in the clinical data standards automated system and information acquired from the previous learnings;
a plurality of SDTM generators for generating an SDTM SAS code based on the mapping, the prediction, and derivations, an SDTM specification, and SDTM datasets based on the mapping;
a define package generator for generating a define package based on annotated metadata information and the mapping available in the clinical data standards automated system, wherein the define package is a Clinical Data Interchange Standards Consortium (CDISC) standard “define.xml” file and a reviewer's guide, and wherein the “define.xml” file is an extensible markup language (XML) document that describes structure and content of data collected during the clinical trial process; and
a validator for validating the generated define package and the SDTM artifacts generated during a development and review cycle, and generating validation reports in real time, wherein the SDTM artifacts comprise the generated define.xml files, stylesheets, the SDTM datasets, an annotated case report form (CRF), the reviewers guide, and supporting documents, and wherein the validations are based on the Clinical Data Interchange Standards Consortium (CDISC) or United States Food and Drug Administration (FDA)-published validation rules.
8 . The system of claim 7 , wherein the clinical data standards automated system predicts derivation logic from a derivation repository based on mapping combinations, study design, and the extracted metadata, and wherein the generated SDTM SAS code is executable outside the clinical data standards automated system in a SAS application.
9 . The system of claim 7 , wherein the natural language processing engine preprocesses text information before feeding the text information into the machine learning model.
10 . The system of claim 7 , wherein the define package is a Clinical Data Interchange Standards Consortium (CDISC) standard “define.xml” file and a reviewer's guide, and wherein the “define.xml” file is an extensible markup language (XML) document that describes structure and content of data collected during the clinical trial process.
11 . The system of claim 7 , wherein the machine learning model is used to identify the derivations and generate the study data tabulation model (SDTM) statistical analysis system (SAS) code based on the derivations.
12 . The system of claim 7 , wherein the SDTM artifacts comprise the generated define.xml files, stylesheets, the SDTM datasets, an annotated case report form (CRF), the reviewers guide, and supporting documents, and wherein the validations are based on the Clinical Data Interchange Standards Consortium (CDISC) or United States Food and Drug Administration (FDA)-published validation rules.
13 . A clinical data standards automated system for automating clinical data standards and generating study data tabulation model (SDTM) artifacts required for a regulatory submission process using a machine learning model and a natural language processing engine, the system comprising:
a non-transitory, computer readable storage medium configured to store computer program instructions defined by modules of the clinical data standards automated system; and
at least one processor communicatively coupled to the non-transitory computer readable storage media, the at least one processor configured to execute the computer program instructions defined by the modules of the clinical data standards automated system, the modules comprising:
a metadata extractor for extracting metadata from raw datasets automatically using natural language processing and feeding the extracted metadata into the machine learning model;
a case report form extractor for pre-filling the extracted metadata into a blank case report form (CRF);
an annotator for automatically predicting and annotating contents in the blank CRF and the extracted metadata, using the machine learning model, based on annotated metadata information available in the clinical data standards automated system and information acquired from previous learnings, and recording new learnings into the clinical data standards automated system with minimal user intervention;
a smart mapper for mapping the raw datasets against a target SDTM variable based on the extracted metadata available in the clinical data standards automated system and information acquired from the previous learnings;
a plurality of SDTM generators for generating an SDTM SAS code based on the mapping, the prediction, and derivations, an SDTM specification, and SDTM datasets based on the mapping, wherein the clinical data standards automated system predicts derivation logic from a derivation repository based on mapping combinations, study design, and the extracted metadata, and wherein the generated SDTM SAS code is executable outside the clinical data standards automated system in a SAS application;
a define package generator for generating a define package based on annotated metadata information and the mapping available in the clinical data standards automated system, wherein the define package is a Clinical Data Interchange Standards Consortium (CDISC) standard “define.xml” file and a reviewer's guide, and wherein the “define.xml” file is an extensible markup language (XML) document that describes structure and content of data collected during the clinical trial process; and
a validator for validating the generated define package and the SDTM artifacts generated during a development and review cycle, and generating validation reports in real time, wherein the SDTM artifacts comprise the generated define.xml files, stylesheets, the SDTM datasets, an annotated case report form (CRF), the reviewers guide, and supporting documents, and wherein the validations are based on the Clinical Data Interchange Standards Consortium (CDISC) or United States Food and Drug Administration (FDA)-published validation rules.
14 . The system of claim 13 , wherein the previous learnings are learnings recorded by the clinical data standards automated system based on the annotated metadata information manually acquired from a user into the blank case report form (CRF) at beginning of a clinical trial process, wherein the predicted CRF annotations and corresponding text and attributes are sent to an application in the clinical data standards automated system, and wherein the application inserts CRF annotations into the blank CRF pre-filled with the extracted metadata, based on prediction from the machine learning model.
15 . The system of claim 13 , wherein the natural language processing engine preprocesses text information before feeding the text information into the machine learning model.
16 . The system of claim 13 , wherein the define package is a Clinical Data Interchange Standards Consortium (CDISC) standard “define.xml” file and a reviewer's guide, and wherein the “define.xml” file is an extensible markup language (XML) document that describes structure and content of data collected during the clinical trial process.
17 . The system of claim 13 , wherein the machine learning model is used to identify the derivations and generate the study data tabulation model (SDTM) statistical analysis system (SAS) code based on the derivations.
18 . The system of claim 13 , wherein the SDTM artifacts comprise the generated define.xml files, stylesheets, the SDTM datasets, an annotated case report form (CRF), the reviewers guide, and supporting documents, and wherein the validations are based on the Clinical Data Interchange Standards Consortium (CDISC) or United States Food and Drug Administration (FDA)-published validation rules.