18375914. MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES simplified abstract (GOOGLE LLC)
Contents
- 1 MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES
Organization Name
Inventor(s)
Uri N. Lerner of Los Altos CA (US)
Natan Potikha of Tel Aviv (IL)
Ayelet Benjamini of Kfar Saba (IL)
MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES - A simplified explanation of the abstract
This abstract first appeared for US patent application 18375914 titled 'MATCHING UNSTRUCTURED TEXT TO CLINICAL ONTOLOGIES
Simplified Explanation
The computer-implemented method described in the abstract involves matching unstructured text to ontology entities in a clinical ontology. Here is a simplified explanation of the patent application:
- Receiving clinical notes associated with a patient
- Extracting text spans from the unstructured text in each clinical note using a neural network
- Matching each text span with a respective output ontology entity from an ontology using a text matcher
- Outputting data defining the text spans and the respective output ontology entity for each text span
Potential Applications
This technology could be applied in healthcare settings for automatically identifying and categorizing clinical conditions mentioned in patient notes.
Problems Solved
This technology solves the problem of manually sorting through large amounts of unstructured clinical text to identify relevant clinical conditions.
Benefits
The benefits of this technology include increased efficiency in analyzing clinical notes, improved accuracy in identifying clinical conditions, and potential for better patient care outcomes.
Potential Commercial Applications
"Automated Clinical Condition Identification in Unstructured Text" could be a valuable tool for healthcare providers, electronic health record companies, and medical researchers looking to streamline data analysis processes.
Possible Prior Art
One possible prior art could be the use of natural language processing techniques in healthcare to extract information from unstructured text data. Another could be the use of neural networks for text analysis in medical contexts.
Unanswered Questions
How does this method handle variations in language or terminology used in clinical notes?
The method described in the abstract uses a neural network and text matcher to extract and match text spans to ontology entities. However, it is unclear how the system accounts for variations in language or terminology that may be used in clinical notes.
What is the accuracy rate of matching text spans to ontology entities using this method?
While the abstract mentions matching text spans to ontology entities, it does not provide information on the accuracy rate of this matching process. It would be important to know how reliable and precise the system is in identifying and categorizing clinical conditions based on unstructured text.
Original Abstract Submitted
A computer-implemented method for matching unstructured text to ontology entities in a clinical ontology is described. The method includes receiving one or more clinical notes associated with a patient; for each of the one or more clinical notes: extracting, using a neural network, one or more text spans from unstructured text in each clinical note, each of the one or more text spans identifying a respective input phrase in the unstructured text; for each of the one or more text spans, matching, using a text matcher, the text span with a respective output ontology entity from an ontology, the respective output ontology entity relating to a clinical condition of the patient; and outputting data defining the one or more text spans and the respective output ontology entity for each of the one or more text spans.
