Logic form identification of medical clinical trials Clint Tustison.

Slides:



Advertisements
Similar presentations
Semantic Business Management November 5, 2009 Paul Haley Automata, Inc. (412)
Advertisements

Artificial Intelligence: Natural Language and Prolog
Slide 1 Insert your own content. Slide 2 Insert your own content.
An Overview of the Integration of the UCSF Dept. of Radiology Teaching File with MIRC Wyatt M. Tellis University of California San Francisco Departments.
Relational Database and Data Modeling
Requirements. UC&R: Phase Compliance model –RIF must define a compliance model that will identify required/optional features Default.
Language Specification using Metamodelling Joachim Fischer Humboldt University Berlin LAB Workshop Geneva
0 - 0.
Native Americans of North Carolina Introduction Task Process Resources Evaluation Conclusion Teacher.
Introduction to Compiler Construction
Chapter 2-2 A Simple One-Pass Compiler
LABELING TURKISH NEWS STORIES WITH CRF Prof. Dr. Eşref Adalı ISTANBUL TECHNICAL UNIVERSITY COMPUTER ENGINEERING 1.
Session # 2 SWE 211 – Introduction to Software Engineering Lect. Amanullah Quadri 2. Fact Finding & Techniques.
March 1, 2009 Dr. Muhammed Al-Mulhem 1 ICS 482 Natural Language Processing Semantics (Chapter 17) Muhammed Al-Mulhem March 1, 2009.
LabVIEW Crash Course Presented by:.
Access Lesson 13 Programming in Access Microsoft Office 2010 Advanced Cable / Morrison 1.
Artificial Intelligence in the 21 st Century S. Lucci / D. Kopec Chapter 5: Logic in Artificial Intelligence 1.
Logic Programming Two possible work modes: 1.At the lab: Use SICstus Prolog. To load a prolog file (*.pl or *.pro extension) to the interpreter, use: ?-
Machine Translation II How MT works Modes of use.
September 12 1 An Algorithm for: Explaining Algorithms Tomasz Müldner.
The 20th International Conference on Software Engineering and Knowledge Engineering (SEKE2008) Department of Electrical and Computer Engineering
1 Logic Programming School of Informatics, University of Edinburgh Logic Programming in 50 Minutes The purpose of this lecture is to explain why logic.
CH-4 Ontologies, Querying and Data Integration. Introduction to RDF(S) RDF stands for Resource Description Framework. RDF is a standard for describing.
Artificial Intelligence By: David Hunt Lee Evans Jonathan Moreton Rachel Moss.
Improving System Safety through Agent-Supported User/System Interfaces: Effects of Operator Behavior Model Charles SANTONI & Jean-Marc MERCANTINI (LSIS)
1 Programming Languages (CS 550) Mini Language Interpreter Jeremy R. Johnson.
New Opportunities for Load Balancing in Network-Wide Intrusion Detection Systems Victor Heorhiadi, Michael K. Reiter, Vyas Sekar UNC Chapel Hill UNC Chapel.
Bio-Medical Interaction Extractor Syed Toufeeq Ahmed ASU.
ANTLR in SSP Xingzhong Xu Hong Man Aug Outline ANTLR Abstract Syntax Tree Code Equivalence (Code Re-hosting) Future Work.
RDF Tutorial.
CSE 425: Semantic Analysis Semantic Analysis Allows rigorous specification of a program’s meaning –Lets (parts of) programming languages be proven correct.
1 Pass Compiler 1. 1.Introduction 1.1 Types of compilers 2.Stages of 1 Pass Compiler 2.1 Lexical analysis 2.2. syntactical analyzer 2.3. Code generation.
COMP 6703 eScience Project Semantic Web for Museums Student : Lei Junran Client/Technical Supervisor : Tom Worthington Academic Supervisor : Peter Strazdins.
CS 330 Programming Languages 09 / 18 / 2007 Instructor: Michael Eckmann.
XML on Semantic Web. Outline The Semantic Web Ontology XML Probabilistic DTD References.
CS 330 Programming Languages 09 / 16 / 2008 Instructor: Michael Eckmann.
Introduction & Overview CS4533 from Cooper & Torczon.
Mapping Physical Formats to Logical Models to Extract Data and Metadata Tara Talbott IPAW ‘06.
1.3 Executing Programs. How is Computer Code Transformed into an Executable? Interpreters Compilers Hybrid systems.
CS-EE 481 Spring Founders Day, 2005 University of Portland School of Engineering Project Pocket Gopher Conversational Learning Agent Team Josh Jones.
© Janice Regan, CMPT 128, Jan CMPT 128 Introduction to Computing Science for Engineering Students Creating a program.
Some Thoughts on HPC in Natural Language Engineering Steven Bird University of Melbourne & University of Pennsylvania.
XP 1 CREATING AN XML DOCUMENT. XP 2 INTRODUCING XML XML stands for Extensible Markup Language. A markup language specifies the structure and content of.
Chapter 1 Introduction Dr. Frank Lee. 1.1 Why Study Compiler? To write more efficient code in a high-level language To provide solid foundation in parsing.
GLOSSARY COMPILATION Alex Kotov (akotov2) Hanna Zhong (hzhong) Hoa Nguyen (hnguyen4) Zhenyu Yang (zyang2)
LANGUAGE TRANSLATORS: WEEK 3 LECTURE: Grammar Theory Introduction to Parsing Parser - Generators TUTORIAL: Questions on grammar theory WEEKLY WORK: Read.
Resource Description Framework (RDF) Course: Electronic Document Team member: Ding Feng Ding Wei Wang Ling Date:
Semantically Processing The Semantic Web Presented by: Kunal Patel Dr. Gopal Gupta UNIVERSITY OF TEXAS AT DALLAS.
CPS 506 Comparative Programming Languages Syntax Specification.
1. 2 Preface In the time since the 1986 edition of this book, the world of compiler design has changed significantly 3.
3.2 Semantics. 2 Semantics Attribute Grammars The Meanings of Programs: Semantics Sebesta Chapter 3.
Semantics In Text: Chapter 3.
Programming Language Descriptions. What drives PL Development? Computers are “in charge” of extremely important issues Execute a program literally. Exercise.
A Portrait of the Semantic Web in Action Jeff Heflin and James Hendler IEEE Intelligent Systems December 6, 2010 Hyewon Lim.
Overview of Compilation Prepared by Manuel E. Bermúdez, Ph.D. Associate Professor University of Florida Programming Language Principles Lecture 2.
Chapter 04 Semantic Web Application Architecture 23 November 2015 A Team 오혜성, 조형헌, 권윤, 신동준, 이인용.
Introduction to Compiler Construction
Lexical and Syntax Analysis
Introduction to Parsing (adapted from CS 164 at Berkeley)
Overview of Compilation The Compiler Front End
Overview of Compilation The Compiler Front End
课程名 编译原理 Compiling Techniques
Lecture 2: General Structure of a Compiler
Compilers B V Sai Aravind (11CS10008).
Semantics In Text: Chapter 3.
Lecture 8 Information Retrieval Introduction
Semantic Markup for Semantic Web Tools:
ONTOMERGE Ontology translations by merging ontologies Paper: Ontology Translation on the Semantic Web by Dejing Dou, Drew McDermott and Peishen Qi 2003.
Owen Rambow 6 Minutes.
Presentation transcript:

Logic form identification of medical clinical trials Clint Tustison

2 Introduction  The what … Identify and extract logic forms from medical clinical trials (in)eligibility criteria  The why … Understand the data Match up the information with other data, i.e., patients ’ medical records  The how … Syntactic parser Cognitive modeling architecture

3 Process Syntactic Parser Cognitive modeling engine Clinical Trials (input) Predicate Calculus Post- Processing (output) Text- processing

4 Input  ClinicalTrials.gov  Sponsored by NIH and other federal agencies, private industry  8,800 current trials online  3,000,000 page views per month  Purpose, eligibility, location, more info.

5 Text processing  Convert trials to.xml format Eligibility Criteria Inclusion criteria: Adenocarcinoma of the pancreas.

6 Process: Input Syntactic Parser Clinical Trials (input) A criterion equals adenocarcinoma of the pancreas. Cognitive modeling engine Predicate Calculus Post- Processing (output)

7 Syntactic parser  Link-Grammar Parser Characteristics  Syntactic dependency parse  Constraints for determining grammaticality  Links give clues on how to process constituents Benefits  written in C  very fast  Robust - ability to process spelling errors  Free -  Can be easily integrated with other applications

8 Process: Syntactic Parser Syntactic parser A criterion equals adenocarcinoma of the pancreas Xp Wd Js----+ | | +--Ds Ss Os Mp Ds--+ | | | | | | | | | | LEFT-WALL a criterion.n equals.v adenocarcinoma[?].n of the pancreas.n.

9 Intelligent Processing  Soar Architecture Model and theory of cognition used in AI programming Translates syntactic parse to logic output by reading links Benefits  Goal-directed problem solving  Agent-based architecture  Ability to learn  Proven in multiple applications  Natural Language-Soar  Tactical Air-Soar  Nasa Test Director-Soar

10 Process: Intelligent processing (M1 ^idea N5 ^idea N4 ^idea N3 ^idea N2) (N5 ^annotation feat-dumped ^annotation seq-dumped ^annotation seq-prep ^aug N4 ^nuc pancreas ^wcount 7) (N4 ^annotation seq-dumped ^annotation seq-prep ^aug N3 ^nuc adenocarcinoma ^of N5 ^wcount 4) (N3 ^ext N2 ^int N4 ^nuc equals ^wcount 3) (N2 ^annotation feat-dumped ^annotation seq-dumped ^annotation seq-prep ^aug N3 ^nuc criterion ^wcount 2)

11 Tools: Representation  Predicate Logic Formal properties, allow for wide range of applications, usable crosslinguistically Vocabulary, syntax, semantics  First-order: quantification over individuals (FOPC)  Higher-order: quantification over relations, etc.

12 Process: Logic Output Predicate Calculus criterion(N2) & adenocarcinoma(N4) & pancreas(N5) & equals(N2,N4) & of(N4,N5). Syntactic Parser Cognitive modeling engine Clinical Trials (input) A criterion equals adenocarcinoma of the pancreas. Post- Processing (output)

13 Post-processing  Prolog axioms Remove elements not included in language of the criterion). Format elements needed in output (ampersands).  Reduce(Z, Y) :- member(Criterm, Y), functor(Criterm, criterion, 1), arg(1, Criterm, Critvar), member(Predterm, Y), functor(Predterm, Xterm, 1), arg(1, Predterm, Predvar), member(Equalsterm, Y), functor(Equalsterm, equals, 2), arg(1, Equalsterm, Critvar), arg(2, Equals, Critvar, Predvar), delete(Y, Criterm, Z2), delete(Z2, Equalsterm, Z).  Turns previous statement: criterion(N2) & adenocarcinoma(N4) & pancreas(N5) & equals(N2,N4) & of(N4,N5).  Into: adenocarcinoma(N4) & pancreas(N5) & of(N4,N5).

14 Output Eligibility Criteria Inclusion Criteria: Adenocarcinoma of the pancreas pancreas(N5) & adenocarcinoma(N4)& of(N4,N5)..

15 Results/Conclusion  Data can be matched up with patients’ medical records to determine if they meet criteria posted in the clinical trial.  Disadvantages Grammar is difficult to write Only one parsed output per utterance  Advantages Fast Robust Implementation in other languages Can be easily integrated with other applications/corpora