Data processing system and method for computer-assisted coding of natural language medical text
A system and method utilizing deep clinical knowledge represented as a knowledge-graph to complement and enhance Natural Language Processing (NLP) for efficient and high-quality computer assisted coding of medical text. One embodiment utilizes the International Classification of Diseases version-10 Procedural Coding System (ICD-10-PCS). The system uses multiple knowledge bases combined with direct mapping provided by the ICD-10-PCS standard to enhance the coverage of assigned code. The system identifies ICD-10-PCS code considering hierarchical mapping and identifies the code by individual ICD-10-PCS character.
1. A computer-controlled method for analyzing natural language clinical text describing a medical procedure, and for generating an accurate procedure code based on a procedural coding system that associates all known medical concepts to alphanumeric characters in a multi-axial coding structure that prohibits rule-based Natural Language Processing (NLP) and machine learning from generating an accurate procedure code, wherein the procedure code comprises a set of alphanumeric characters corresponding to the described medical procedure, the method comprising:
creating a background knowledge graph that models all known medical concepts as nodes in the graph and illustrates hierarchical relationships between the medical concepts;
mapping the background knowledge graph to the procedural coding system to associate each of the medical concepts with at least one alphanumeric character;
analyzing the medical text to determine key words and phrases identifying medical concepts related to the described medical procedure; and mapping each identified medical concept to the background knowledge graph to determine each character of the set of alphanumeric characters in the procedure code, wherein the procedure code is generated without utilizing inefficient, iterative trial-and-error techniques;
wherein creating the background knowledge graph includes mapping medical domain knowledge to ICD-10-PCS concepts to create the knowledge graph, which is utilized as background knowledge for precise ICD-10 Computer Assisted Coding (CAC);
wherein creating the background knowledge graph includes defining super classes of procedures, and classes and subclasses of procedures within each super class; and
wherein ICD-10-PCS defines a hierarchy of alphanumeric characters or sets of characters associated with each of the super classes, classes, and subclasses of procedures.
2. The computer-controlled method according to claim 1 , wherein the procedural coding system is International Classification of Diseases version 10 Procedural Coding System (ICD-10-PCS).
3. The computer-controlled method according to claim 1 , wherein the background knowledge graph represents in graphical format, all known medical concepts related to the described medical procedure, associated medical equipment, and relationships between the procedure and the medical equipment.
4. A data processing system for analyzing natural language clinical text describing a medical procedure, and for generating an accurate procedure code based on a procedural coding system that associates all known medical concepts to alphanumeric characters in a multi-axial coding structure that prohibits rule-based Natural Language Processing (NLP) and machine learning from generating an accurate procedure code, wherein the procedure code comprises a set of alphanumeric characters corresponding to the described medical procedure, the system comprising:
at least one processor coupled to a non-transitory memory that stores computer program instructions, wherein when the at least one processor executes the instructions, the system is caused to:
create a background knowledge graph that models all known medical concepts as nodes in the graph and illustrates hierarchical relationships between the medical concepts;
map the background knowledge graph to the procedural coding system to associate each of the medical concepts with at least one alphanumeric character;
analyze the medical text to determine key words and phrases identifying medical concepts related to the described medical procedure; and map each identified medical concept to the background knowledge graph to determine each character of the set of alphanumeric characters in the procedure code; wherein the procedure code is generated without utilizing inefficient, iterative trial-and-error techniques;
wherein the at least one processor is configured to create the background knowledge graph by mapping medical domain knowledge to ICD-10-PCS concepts to create the knowledge graph, which is utilized as background knowledge for precise ICD-10 Computer Assisted Coding (CAC);
wherein the at least one processor is configured to create the background knowledge graph by defining super classes of procedures, and classes and subclasses of procedures within each super class; and
wherein ICD-10-PCS defines a hierarchy of alphanumeric characters or sets of characters associated with each of the super classes, classes, and subclasses of procedures.
5. The data processing system according to claim 4 , wherein the procedural coding system is International Classification of Diseases version 10 Procedural Coding System (ICD-10-PCS).
6. The data processing system according to claim 4 , wherein the background knowledge graph represents in graphical format, all known medical concepts related to the described medical procedure, associated medical equipment, and relationships between the procedure and the medical equipment.
7. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations in a data processing system for analyzing natural language clinical text describing a medical procedure, and for generating an accurate procedure code based on a procedural coding system that associates all known medical concepts to alphanumeric characters in a multi-axial coding structure that prohibits rule-based Natural Language Processing (NLP) and machine learning from generating an accurate procedure code, wherein the procedure code comprises a set of alphanumeric characters corresponding to the described medical procedure, the operations comprising:
creating a background knowledge graph that models all known medical concepts as nodes in the graph and illustrates hierarchical relationships between the medical concepts;
mapping the background knowledge graph to the procedural coding system to associate each of the medical concepts with at least one alphanumeric character;
analyzing the medical text to determine key words and phrases identifying medical concepts related to the described medical procedure; and mapping each identified medical concept to the background knowledge graph to determine each character of the set of alphanumeric characters in the procedure code; wherein the procedure code is generated without utilizing inefficient, iterative trial-and-error techniques;
wherein the operation of creating the background knowledge graph includes mapping medical domain knowledge to ICD-10-PCS concepts to create the knowledge graph, which is utilized as background knowledge for precise ICD-10 Computer Assisted Coding (CAC);
wherein the operation of creating the background knowledge graph includes defining super classes of procedures, and classes and subclasses of procedures within each super class; and
wherein ICD-10-PCS defines a hierarchy of alphanumeric characters or sets of characters associated with each of the super classes, classes, and subclasses of procedures.
8. The non-transitory machine-readable medium according to claim 7 , wherein the procedural coding system is International Classification of Diseases version 10 Procedural Coding System (ICD-10-PCS).
9. The non-transitory machine-readable medium according to claim 7 , wherein the background knowledge graph represents in graphical format, all known medical concepts related to the described medical procedure, associated medical equipment, and relationships between the procedure and the medical equipment.