Call for Papers

There are many healthcare datasets consisting of both structured and unstructured information, which provide a challenge for artificial intelligence and machine learning researchers seeking to extract knowledge from data. Existing healthcare datasets include electronic medical records, large collections of complex physiological information, medical imaging data, genomics, as well as other socio-economic and behavioral data. In order to perform data-driven analysis or build causal and inferential models using these datasets, challenges such as integrating multiple data types, dealing with missing data, and handling irregularly sampled data need to be addressed. While these challenges must be considered by researchers working with healthcare data, a larger problem involves how to best ensure that the hypotheses posed and types of knowledge discoveries sought are relevant to the healthcare community. Clinical perspectives from medical professionals are required to ensure that advancements in healthcare data analysis result in positive impact to point-of-care and outcome-based systems.

This workshop builds upon the success of previous Knowledge Discovery in Healthcare Data (KDH) workshops. It welcomes contributions providing insight on the extent to which AI techniques have successfully penetrated the healthcare field, interaction among AI techniques to achieve successful learning healthcare systems, and distinctions between AI and non-AI models needed in modern healthcare environments. The focus of the workshop is on issues in data extraction and assembly, knowledge discovery, decision support for healthcare providers, and personalised self-care aids for patients. A workshop highlight will be the Blood Glucose Level Prediction (BGLP) Challenge, in which researchers will compare the efficacy of different machine learning prediction approaches on a standard set of data from patients with type 1 diabetes.

Topics

Contributions are welcome in areas including, but not limited to, the following:

• Knowledge discovery and data analytics
  • Predictive and prescriptive analyses of healthcare data
  • Detecting and extracting hidden information from healthcare data
  • Extracting causal relationships from healthcare data
  • Artificial neural network models or deep learning approaches for healthcare data analytics
  • Active, transfer and reinforcement learning in healthcare
  • Applications of probabilistic analysis in medicine
  • Physiological data analysis
  • Mathematical model development in biology and medicine, modeling of disease interaction and progression
  • Development of novel diagnostic and prognostic tests utilizing quantitative data analysis
  • Handling uncertainty in large healthcare datasets: dealing with missing values and non-uniformly sampled data
  • Novel visualization techniques
• Data extraction, organization and assembly
  • Knowledge-driven and data-driven approaches for information retrieval and data mining
  • Multilevel data integration in healthcare, e.g. behavioral data, diagnoses, vitals, radiology imaging, doctor's notes, phenotype, and different omics data, including multi-agent approaches
  • Integration and use of medical ontologies
  • Knowledge abstraction, classification, and summarization from literature or electronic health records
  • Biomedical data generation and curation
• Personalisation and decision support
  • Patient empowerment through personalised patient-centered systems
  • Clinical decision support systems
  • Applications of IoT (wearables, sensors) in healthcare
• Blood glucose level prediction
  • System description papers detailing results of the BGLP Challenge
  • Scientific papers presenting new research in machine learning for blood glucose level prediction

Submission and Format

Submissions can be made as:

1. Long papers (7 pages + 1 page references): Long papers should present original research work and be no longer than eight pages in total: seven pages for the main text of the paper (including all figures but excluding references), and one additional page for references. Papers reporting on original research in blood glucose level prediction, but not BGLP Challenge system description papers, should be formatted as long papers and submitted by the deadline for all workshop papers.
2. Short papers (4 pages + 1 page references): Short papers may report on works in progress, descriptions of available datasets, as well as data collection efforts. Position papers regarding potential research challenges are also welcomed. Short paper submissions should be no longer than five pages in total: four pages for the main text of the paper (including all figures but excluding references), and one additional page for references. BGLP Challenge system description papers should be formatted as short papers; however, these papers have their own submission deadline.

Both long and short papers must be formatted according to ECAI guidelines and submitted electronically through EasyChair: https://easychair.org/conferences/?conf=kdh2020.

Publication

Proceedings: The papers accepted for KDH 2020 will be published in the CEUR-WS.org international proceedings volume. This proceedings volume will be published electronically and indexed by Google Scholar and DBLP.