Role of Machine Learning in Transforming 2013 : ICML 2013 Workshop: Role of Machine Learning in Transforming Healthcare
Call For Papers
The rapid growth of information technology promises to change the practice of medicine as we know it. Large volumes of clinical data are now digitized as part of routine patient care, and clinical decisions are made more accurately and more efficiently than ever before with the growing prevalence of Electronic Medical Record (EMR) systems. In the United States, for instance, EMR adoption has increased dramatically in recent years, driven in part by the recent regulatory mandates and government funding, in particular the HITECH Act in the American Recovery and Reinvestment Act (ARRA). The growth of EMR systems creates the opportunity to extract key, actionable information from the electronic data more robustly and to use it meaningfully (i.e. to reach the Meaningful Use criteria), improving clinical, financial and operational outcomes.
However, both the data and its application within health care are challenging: The data are collected from heterogeneous sources. They are high-dimensional. There is provider bias in the collection process; stake holders (e.g., patient and provider preferences) influence choices made along the way. Acquiring labels is often expensive and non-trivial; for example, experts may disagree on a diagnosis. In addition, for our work to impact health care, it is valuable to deeply understand the context within which it will be deployed.
The purpose of this multi-disciplinary workshop is two-fold:
1. Learning from domain experts from large healthcare organizations (e.g., Kaiser) and senior researchers from related disciplines like operations research, health services, and statistics.
2. Techniques and methodologies machine learning community is using and in process of developing to address these challenges.
We will bring together researchers from machine learning, computational linguistics, medical informatics and large healthcare systems who share an interest in problems of transforming healthcare to meet the growing challenges. The goal of this workshop will be to bridge the gap between the theory of machine learning, natural language processing, and the applications and needs of the healthcare community. We plan to provide a platform for the exchange of ideas, identification of important and challenging applications, and discovery of possible synergies. It is our hope that this will spur vigorous discussions and encourage collaboration between the various disciplines potentially resulting in collaborative projects and grant submissions. We will particularly emphasize the mathematical and engineering aspects of modelling longitudinal patient data for disease progression modelling, financial and clinical outcome analysis etc.
We will try to address many of the topics through both invited and contributed talks. The workshop program will consist of presentations by invited speakers from both healthcare systems and machine learning community, and by authors of extended abstracts submitted to the workshop. In addition, there will be a discussion to identify important problems, applications, and synergies between the the scientific disciplines.
Topics of Interest
We would like to encourage submissions on any of (but not limited to) the following topics:
- Surveillance using large scale health systems data
- Evidence generation from health systems data
- Optimization of care delivery from health systems data
- Computational challenges of learning from observational data
- Population management
- Privacy and security for data with patient health identifiers
- Mining and analysis of clinical data
- Knowledge discovery from electronic health records
- Disease modeling and detection
- Patient monitoring and alerting
- Patient outcome prediction
- Optimization of patient-management workflows
- Design of cost-effective clinical trials
- Methods for personalized medicine and care
- Integration of clinical data sources and domain knowledge
- Integration of phenotypic and genotypic data
- Information Extraction and Retrieval from Clinical Text
- Clinical Ontologies
- Patient Identification
- Patient risk assessment
- Learning from multiple annotators
- Learning with data not missing at random
- Active Learning to reduce expert annotation costs
- Combining Unstructured and Structured Text for inference
Besides the topics above, this year we are specially interested in time series modeling of longitudinal patient data for modeling disease progression, clinical and financial outcomes and also for retrospective studies.
We call for paper contribution of short (2-4 pages) and full (6-8 pages) to the workshop using ICML style (http://icml.cc/2013/wp-content/uploads/2012/12/icml2013stylefiles.tar.gz). The accepted papers will be available from the workshop website's Proceedings page prior to the workshop. Accepted papers will be either presented as a talk or poster. Please indicate your preference for oral or poster presentation. Short papers will only be eligible for poster presentations. Extended versions of some accepted papers will also be invited for inclusion in an edited book on the same topic as the workshop.
Please submit your manuscripts on the workshop EasyChair site (https://www.easychair.org/conferences/?conf=mlhealth13).
Noemie Elhadad, Columbia Univ.
Faisal Farooq, Siemens Healthcare (Co-chair)
Misha Pavel, National Sciences Foundation
Suchi Saria, Johns Hopkins (Co-chair)
Jimeng Sun, IBM Research
Shipeng Yu, Siemens Healthcare