DASH

News and Events

The Fourth International Symposium on Languages in Biology and Medicine (LBM 2011)

Health Informatics, Presentation, Publication, SYSLAB

Sumithra Velupillai attended The Fourth International Symposium on Languages in Biology and Medicine (LBM 2011) at the Nanyang Technological University in Singapore, December 14-15.

At LBM 2011, Sumithra presented the paper Automatic Classification of Factuality Levels – A Case Study on Swedish Diagnoses and the Impact of Local Context. In this paper, The Stockholm EPR Diagnosis-Factuality Corpus is described and used for building automatic classifiers. The classifier obtains promising results (best overall results are 0.699 average F-measure using all classes, 0.762 F-measure using merged classes), using simple local context features. Preceding context is more useful than posterior, although best results are obtained using a window size of 4. Lower levels of certainty are more problematic than higher levels, which was also the case for the human annotators in creating the corpus. A manual error analysis shows that conjunctions and other higher-level features are common sources of errors.

Happy LBM 2011 participants after a wonderful banquet at Chef Chan's Restaurant

Research visit within the Interlock project at DBMI, UCSD

Health Informatics, Presentation, SYSLAB

Sumithra is visiting Dr. Wendy Chapman and Danielle Mowery at the Division for Biomedical Informatics at UC San Diego during November within the Interlock-project.

Sumithra and Danielle have been working on a qualitative, comparative study on models and characterizations of uncertainties and negations in annotated clinical documentation in both English and Swedish electronic patient records.

On November 23, Sumithra gave a talk at the department, describing the IT for Health group at DSV as well as her research on certainty classification of Swedish clinical diagnostic statements.

The Interlock project will continue with further collaboration between the researchers in the respective groups, and during Sumithra’s visit several ideas for future experiments have emerged!

Some snacks before Sumithra's presentation

Dr. Wendy Chapman introducing Sumithra

Danielle, Sumithra and Wendy

Research visit at NICTA, Australia

Health Informatics, Information Systems, SYSLAB

I have been invited to NICTA, National Institute for Information and Communications Technology, Melbourne and Canberra, Australia to work on a collaborative project together with Dr. Hanna Suominen and Dr. David Martinez. We are working with clinical text mining of invasive fungal infections, together with a physician from the Alfred Health Hospital in Melbourne. This project is related to the NICTA research discipline of machine learning and funded by the NICTA business area of eHealth.

We started our collaboration by working in the NICTA Melbourne office for a week (Sept. 7th – 14th) – now we are based in the NICTA Canberra office the remaining time. I am very happy to have been given this wonderful opportunity and am very thankful to my colleagues here who are amazingly generous, welcoming and fantastic to work with!

Sumithra, Hanna and David having a coffee break
… trying to look serious …
Hanna and David working hard at the office

 

SYSLAB seminar series: Analyzing the Integration between Requirements and Models in Model Driven Development, held by Iyad Zikra

Enterprise Modelling, Information Systems, SYSLAB

On Wednesday, May 4th 2011, the SYSLAB seminar series was held by Iyad
Zikra, entitled Analyzing the Integration between Requirements
and Models in Model Driven Development. Iyad presented a survey on the state of the art of the relationship between requirements and model driven development (MDD). Different approaches were identified, such as using Natural Language Processing (NLP) techniques or guidelines, for dealing with this gap. NLP in this context means to parse the requirements that are written in natural language and transform them to some model format. The overall goal with this work is understand and minimize the gap between requirements and MDD, where NLP techniques will be studied further. The survey will be published in EMMSAD 2011.

Paper accepted to the XXIII International Conference of the European Federation for Medical Informatics (MIE 2011)

Health Informatics, Information Systems, Publication, Research Paper, SYSLAB

The paper Factuality Levels of Diagnoses in Swedish Clinical Text which I’ve written together with Hercules Dalianis and Maria Kvist has been accepted to the XXIII International Conference of the European Federation for Medical Informatics (MIE 2011), which will be held in Oslo, Norway 28th–31st August 2011. The paper describes work on an annotation project where clinical diagnoses were annotated for factuality levels, for the purpose of building automated systems for enhanced future information access. The paper will be included in the MEDLINE indexed proceedings, published by IOS Press in Amsterdam.

CIKM ’10 and 3rd ESAIR

Health Informatics, Information Systems, SYSLAB

October 27th-30th, I attended the 19th ACM Conference on Information and Knowledge Management (CIKM ’10), and the Third Workshop on Exploiting Semantic Annotations in Information Retrieval (ESAIR 2010), in Toronto, Canada. The CIKM ’10 conference had over 600 participants from all over the world, with tracks covering the topic areas Information Retrieval, Knowledge Management and Databases. The ESAIR workshop consisted of 17 position papers presented as posters, a one-minute boaster session and break-out-groups with discussions. I presented the position paper Semantic Annotations in Clinical Documentation — Exploring Potentials for Future Information Retrieval at ESAIR. The poster can be viewed either in the corridor in front of my office at DSV, or here: http://www.sics.se/jussi/esair2010/poster/APP-Velupillai.pdf. The position paper is available at the ACM Digital Library: http://portal.acm.org/citation.cfm?doid=1871962.1871968.

Joint article: HEXAnord at the 2nd Louhi Workshop on Text and Data Mining of Health Documents in L.A., June 5 2010

SYSLAB

Within the HEXAnord network, a joint article has been written where Finnish and Swedish Intensive Care Unit nursing notes have been analyzed both qualitatively and quantitatively. The article, Characteristics and Analysis of Finnish and Swedish Clinical Intensive Care Nursing Narratives, was submitted and accepted after a double blind peer-review process to the 2nd Louhi Workshop on Text and Data Mining of Health Documents in L.A., June 5 2010, which is organized by Hercules Dalianis, Martin Hassel, Gunnar Nilsson and Sumithra Velupillai.

The work has been performed by HEXAnord participants from Finland, Norway, Lithuania and Sweden: Helen Allvin, Elin Carlsson, Hercules Dalianis, Riitta Danielsson-Ojala, Vidas Daudaravi?ius, Martin Hassel, Dimitrios Kokkinakis, Heljä Lundgren-Laine, Gunnar Nilsson, Øystein Nytrø, Sanna Salanterä, Maria Skeppstedt, Hanna Suominen, Sumithra Velupillai.

Abstract:

We present a comparative study of Finnish and Swedish free-text nursing narratives from intensive care. Although the two languages are linguistically very dissimilar, our hypothesis is that there are similarities that are important and interesting from a language technology point of view. This may have implications when building tools to support producing and using health care documentation.
We perform a comparative qualitative analysis based on structure and content, as well as a comparative quantitative analysis on Finnish and Swedish Intensive Care Unit (ICU) nursing narratives. Our findings are that ICU nursing narratives in Finland and Sweden have many properties in common, but that many of these are challenging when it comes to developing language technology tools.

Paper accepted to the 7th international conference on Language Resources and Evaluation (LREC)

SYSLAB

The paper How Certain are Clinical Assessments? Annotating Swedish Clinical Text for (Un)certainties, Speculations and Negations which I’ve written together with Hercules Dalianis has been accepted to the seventh international conference on Language Resources and Evaluation (LREC), which will be held in Valetta, Malta 19th–21th May 2010 The experiments described in the paper have been part of the KEA project. The paper describes initial findings on an annotation project where clinical assessments were annotated for certain and uncertain language.