HealTAC PhD student/early postdoc forum

 

While there have been a number of successful applications of NLP technologies to clinical and patient-generated narrative, there are still a wide variety of challenges which are currently addressed through PhD and early postdoc research, including dealing with implicit free-text expressions, lack of shared resources, managing patient privacy and regulations, translation to clinical practice, several socio-technical (e.g. studying data as a human artefact) and medical humanities challenges (what is in the patient narrative), etc. The HealTAC PhD student/early postdoc forum aims to bring together graduate students and recent graduates with research topics focused on these challenges, with the main aim to

  • provide an opportunity to “show and tell” current PhD research in this domain, OR
  • provide an opportunity to pitch fellowship proposal ideas for recent graduates; AND
  • receive constructive feedback on graduate research efforts.

The event will be organised as a special session with an expert panel including several experienced researchers, international colleagues, industrial and user/clinical representatives. Each presenter will prepare a 10 min presentation followed by 10 min questions and constructive feedback from the expert panel (and audience) about significance and novelty of their research, proposed methodologies and approaches, as well as about presentation. The participants will be also encouraged to prepare posters so that they can discuss their work during the conference (no need to submit a separate poster submission).

Details from the last year’s PhD forum are available here.

 

Fellowship forum

  • Natalia Viani. Exploring alternative approaches to address temporal information extraction for clinical use-cases

 

PhD forum

  • Mark Ormerod. Assessing the Interpretability of Sentence-Level Clinical Notes Diagnosis Models
  • Glorianna Jagfeld. Talking about personal recovery in bipolar disorder
  • Denis Newman-Griffis. Kickstarting NLP for whole-person function information with representation learning and data analysis
  • Julia Walsh. Using spontaneously generated online patient experiences to improve healthcare for patients and providers
  • Daphné Chopard. Text Self-Normalization: An Adaptive and Self-sufficient Approach to Abbreviation Expansion

 

Expert panel

  • Prof Prof Hongfang Liu, Mayo Clinic
  • Dr Anoop Shah, University College London
  • Prof Stephane Meystre, Medical University of South Carolina
  • Dr Nigel Collier, University of Cambridge
  • Dr Beatrice Alex, University of Edinburgh

 

Submissions (closed)

Contributions should consists of up to 4 pages and should present ongoing PhD research (at any stage), clearly specifying the aim and objectives of research, as well as methodology and resources. These submissions will be treated separately from the other submissions.

Please use the HealTAC paper template and submit proposals via EasyChair.

 

Organiser

  • Sumithra Velupillai, King’s College London