rss
J Am Med Inform Assoc 16:590-595 doi:10.1197/jamia.M3095
  • Focus on i2b2 Obesity NLP Challenge

A System for Classifying Disease Comorbidity Status from Medical Discharge Summaries Using Automated Hotspot and Negated Concept Detection

Table 1

Macro-averaged F1 textual and intuitive scores for our best-performing submission in the 2008 i2b2 Obesity Challenge Task. In addition to presenting scores for each comorbidity within each task, our system scores are included, which was the basis for ranking team submissions by the i2b2 Challenge organizers

Macro-Averaged F1
Textual Intuitive
System 0.598 0.634
Asthma 0.483 0.970
CAD 0.600 0.630
CHF 0.710 0.612
Depression 0.962 0.935
Diabetes 0.641 0.915
Gallstones 0.632 0.961
GERD 0.486 0.579
Gout 0.959 0.981
Hypercholesterolemia 0.484 0.912
Hypertension 0.827 0.899
Hypertriglyceridemia 0.727 0.876
OA 0.945 0.631
Obesity 0.489 0.973
OSA 0.654 0.653
PVD 0.955 0.623
VI 0.675 0.725
  • CAD = coronary artery disease; CHF = congestive heart failure; GERD = gastroesophageal reflux disease; OA = osteo arthritis; OSA = obstructive sleep apnea; PVD = peripheral vascular disease; VI = venous insufficiency.

Access policy for JAMIA

All content published in JAMIA is deposited with PubMed Central by the publisher with a 12 month embargo. Authors/funders may pay an Unlocked fee of $2,000 to make the article free on the JAMIA website and PMC immediately on publication.

All content older than 12 months is freely available on this website.

AMIA members can log in with their JAMIA user name (email address) and password or via the AMIA website.