.

Dr Dympna O'Sullivan

Dympna O'Sullivan

Phone number
0121 204 3464
Email
d.m.osullivan@aston.ac.uk
Room number
MB213E

Profile
I joined the Department of Computer Science at Aston University as a lecturer in December 2008. Prior to that I spent 18 months working as a post-doctoral fellow at the University of Ottawa, Canada. In 2006 I completed a PhD in health informatics at University College Dublin, Ireland.

Qualifications & Education

  • BSc with First Class Honours in Computer Science, University College Dublin, Ireland, 2002
  • PhD in Health Informatics, University College Dublin, Ireland, 2006
  • Post doctoral fellowship in Health Informatics, University of Ottawa, Canada, 2008

Employment

  • 2008 – Current: Lecturer in Computer Science, Department of Computer Science, School of Applied Science and Engineering, Aston University.
  • 2006 – 2008: Post doctoral fellow in Health Informatics, University of Ottawa, Canada

 

Teaching activity

  • Strategic Information Systems (Module CS3260)

Research interests

I am a member of the KEG (Knowledge Engineering Group) at the Department of Computer Science at Aston University.

My research is in the area of health informatics and clinical decision support systems. I am currently interested in the following areas:

Retrieving online biomedical literature

The online biomedical literature is vast with approximately 30,000 articles published annually. Furthermore many current repositories are charactersed by poor indexing and querying methodologies. As a result clinicians often find it very difficult to easily retrieve information relevant for a particular presentation and patient, especially at the point of care where such information is likely to be most useful. I am interested in resolving some such issues by developing ontology-based information retrieval methodologies that can retrieve and emphasize presentation and patient specific aspects of clinical evidence rather than only aspects relating to the specified disease. I am also interested in developing methodologies for accurately ranking and presenting relevant aspects of clinical evidence for improved discrimination at the point of care.

Automation of clinical practice guidelines

Clinical practice guidelines (CPG) primarily exist in non-digitized formats, either as long text-based documents summarizing relevant clinical evidence or as summarized ‘flowcharts’ outlining diagnoses, differential diagnoses and recommended interventions.  As a result it is often difficult to directly link CPG to clinical practice.
I am interested in integrating CPG with clinical workflow by developing methodologies for automating and implementing CPG. CPG attribute and values pairs, diagnoses and recommended interventions, are encoded as sets of rules and/or constraints which are executed in real-time on collected clinical data. This research allows for the practice of evidence-based medicine as well as provides improved methodologies for clinical data entry and clinical data accuracy.

Biomedical data mining

I am interested in developing applications for clinical decision support that utilize data mining techniques to support clinicians. Support is provided either by presenting exemplar patient cases for comparative purposes or by calculating a diagnostic predication given a current patient state.  I have performed work in this area using medical imagery (similar patient cases) and paediatric asthma (prediction system) as sample domains.

PhD Supervision
I am available to supervise suitable applicants who are interested in performing research in the health informatics domain, particularly those with an interest in my specified research areas.