Presentation Title

P-32 Using Evidence-based medicine summaries to help answer health economic questions

Presenter Status

Department of Engineering and Computer Science

Location

Buller Hallway

Start Date

31-10-2014 1:30 PM

End Date

31-10-2014 3:00 PM

Presentation Abstract

High quality meta-analyses, systematic reviews, and structured literature reviews are extremely useful for understanding the quality, and strength of published findings. However, high quality review studies, are time consuming and many published studies are suboptimal – lacking rigor, statistical power, or sufficiently specified models, a particular concern for cost-related studies. ACRES (Automatic Clinical Result Extraction and Summarization) is a machine learning-based software program designed to read abstracts from PubMed, extract the key trial elements, compute ratios (e.g., absolute risk reduction (ARR)) for proposed treatments, and generate summaries for the purpose of evidence-based medicine decision making. By generating detailed summaries and three ranking categories for PubMed search results, ACRES reduced time spent examining irrelevant papers, and was 4% more accurate in identifying relevant papers than was a systematic review on diabetes education and cost that was conducted by humans in 2008.

This document is currently not available here.

Share

COinS
 
Oct 31st, 1:30 PM Oct 31st, 3:00 PM

P-32 Using Evidence-based medicine summaries to help answer health economic questions

Buller Hallway

High quality meta-analyses, systematic reviews, and structured literature reviews are extremely useful for understanding the quality, and strength of published findings. However, high quality review studies, are time consuming and many published studies are suboptimal – lacking rigor, statistical power, or sufficiently specified models, a particular concern for cost-related studies. ACRES (Automatic Clinical Result Extraction and Summarization) is a machine learning-based software program designed to read abstracts from PubMed, extract the key trial elements, compute ratios (e.g., absolute risk reduction (ARR)) for proposed treatments, and generate summaries for the purpose of evidence-based medicine decision making. By generating detailed summaries and three ranking categories for PubMed search results, ACRES reduced time spent examining irrelevant papers, and was 4% more accurate in identifying relevant papers than was a systematic review on diabetes education and cost that was conducted by humans in 2008.