OVERVIEW

The goal of the first workshop on Knowledge Discovery in Healthcare Data is to foster discussion and present progress on research efforts that leverage large amounts of observational data (clinical, biological, physiological) to expedite discovery in medicine. The workshop is intended to encourage a cross-disciplinary exchange of ideas between medical researchers and the artificial intelligence community.

Healthcare datasets consisting of both structured and unstructured information provide a challenge for artificial intelligence and machine learning researchers seeking to extract knowledge from data. Rich healthcare datasets exist, including 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 models using these datasets, challenges need to be addressed, such as integrating multiple data types, dealing with missing data and handling irregularly sampled data. While these challenges need to be taken into account by researchers working with healthcare data, a larger problem involves how to best ensure the hypotheses posed and types of knowledge discoveries sought are relevant to the healthcare community. Clinical perspectives from medical care professionals are required to assure that advancements in healthcare data analysis results in positive impact to eventual point-of-care and outcome-based systems.

The process of discovery in medicine starts with a small set of observations and many pre-clinical and clinical trials on different patient population cohorts. Heterogeneous environments, uncertainties in original hypotheses, the passage of time and accumulating costs make medical discovery a complex process. An example of such a discovery is metabolic syndrome. The concept of metabolic syndrome evolved over 90 years to reach our current point of understanding. It is now known that the syndrome occurs as a cluster of metabolic and medical disorders, including obesity, impaired control of blood glucose, high levels of fat in the blood, and high blood pressure. The hope of knowledge discovery in healthcare data is to expedite such discoveries.

Artificial intelligence and machine learning approaches hold the potential to reveal not readily apparent, hidden information in biological and medical healthcare datasets. The results of such discoveries can aid the development of novel diagnostic and prognostic tests, inform descriptive, predictive and prescriptive analytics and guide hypothesis generation. By combining advances in algorithmic and computational approaches together with perspectives from medical care professionals the hope of this workshop is to ensure advancements result in positive impact and relevance to the healthcare community.

CALL FOR PAPERS

We invite submissions in the following categories:

Long papers (6 pages + 1 page references): Long papers should present original research work and be no longer than seven pages in total: six pages for the main text of the paper (including all figures but excluding references), and one additional page for references.

Short papers (3 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 four pages in total: three pages for the main text of the paper (including all figures but excluding references), and one additional page for references.

Both long and short papers must be formatted according to IJCAI guidelines and submitted electronically through easychair:

https://easychair.org/conferences/?conf=ijcai16kdhealth

RELEVANT TOPIC AREAS

We welcome contributions in areas that include, but are not limited to, the following:

· Mathematical model development in biology and medicine

· Handling large healthcare datasets: dealing with missing values and non-uniformly sampled data

· Integration of multi-level data in healthcare (e.g. behavioral data, diagnoses, vitals, radiology imaging, Doctor's notes, phenotype, and different omics data)

· Biomedical data collection efforts

· Clinical decision support systems

· Detecting and extracting hidden information from healthcare data

· Mathematical modeling of disease interaction and progression

· Development of novel diagnostic and prognostic tests utilizing quantitative data analysis

· Extracting causal relationships from healthcare data

· Novel visualization techniques in biomedicine

· Nonlinear dynamics in medicine

· Deep learning approaches in healthcare

· Active and transfer learning in healthcare

· Applications of probabilistic analysis in medicine

· Physiological data analysis

· Predictive and prescriptive analytics using available patient data

· Classification of pathologic mental and physical health states