IJCNN 2017 Special Session on
Machine Learning for Enhancing Biomedical Data Analysis

Aims & Scope The pursuit of precision medicine has led to an explosive growth in the amount of data now available.  Ever-increasing advances in technology and greater levels of granularity have also resulted in an increase in the amount of data and complexity of knowledge surrounding disease.  The increased demand for skilled analysts in systems medicine and engineering has, unfortunately, outpaced the supply.

In spite of these shortcomings, productivity in systems medicine and engineering is nevertheless growing.  Results of these efforts will enable practitioners to improve diagnoses and treatments, and allow health care systems to better manage patients and reduce costs.  Central Data analysis is also crucial to get closer to a real personalized medicine, one of the main goals of modern society in terms of healthcare.

This special session aims at bringing together researchers from the fields of Biomedicine and Machine Learning (ML) in order to exploit the synergies between them, thus taking advantage of the modeling capabilities of ML and the expert knowledge in Biomedicine to make progresses truly relevant to the medical community by focusing on the solution of real problems, whose results hopefully lead to palpable enhancements in clinical routine practice and in increasingly personalized medicine. The proposal of any method useful for medical data analysis, even if it does not fall completely within ML (e.g., biostatistics or biomedical signal processing), is also welcome in this special session. In particular, we welcome papers which present novel algorithms or refined classical methods applied to biomedical problems. These include but are not limited to:

• Proposals of new ML algorithms that outperform previous approaches in clinical problems.
• Practical applications of computational intelligence and ML for mining health-related data.
• Structure finding, including efficient derivation of directed graphs with applications to extract probabilistic graphical relationships between features in biomedical problems.
• Integration of expert clinical knowledge in graphical models.
• Methodologies for fusion of heterogeneous data: clinical tests, subjective assessments, molecular biomarkers, histology, imaging, electrophysiological measurements, etc.
  
• Models of time-to-event data to characterize prognostic outcomes and treatment effects.
• Methodologies for medical decision aid and treatment planning.
• Telemedicine and proposals for a remote healthcare system.

Important Dates
Paper submission: November 15, December 1, 2016
Paper decision notification: January 20, 2017
Camera-ready submission: February 20, 2017

Session Chairs
José D. Martín-Guerrero (Universitat de Valencia, Spain)
Paulo J.G. Lisboa (P.J.Lisboa@ljmu.ac.uk , Liverpool John Moores University, U.K.)
Alfredo Vellido (Universitat Politècnica de Catalunya, Spain)
Azzam F.G. Taktak (University of Liverpool, U.K.)
Leif E. Peterson (Houston Methodist Research Institute, TX, U.S.A.)