Beyond basic research, biology in general and biomedicine in particular are increasingly and rapidly becoming data-based sciences, an evolution driven by technological advances in image and signal non-invasive data acquisition (perfectly exemplified by the 2014 Nobel Prize in Chemistry for the development of super-resolved fluorescence microscopy), or high-throughput genomics, to name just a few. In the Biomedical field, the large amount of data generated from a wide range of devices and patients is creating challenging scenarios for researchers, related to storing, processing and even just transferring information in its electronic form, all these compounded by privacy and anonymity legal issues. The situation is not different in healthcare, where electronic health records are becoming commonplace and new possibilities such as remote home monitoring, or wearable medical devices are likely to make an impact as part as the ambitious p-Health, or 4-P (Predictive, personalized, preventive, participatory) paradigm for medicine. Data-based healthcare finds a paramount example in the current Institute for Systems Biology (ISB, Seattle) "Hundred Person Wellness Project"(*), a pilot in which 100 healthy individuals are intensively monitored on a daily basis. New data requirements require new approaches to data analysis, some of the most interesting ones are currently stemming from the Computational Intelligence (CI) and Machine Learning (ML) community. This session is particularly interested in the proposal of novel CI and ML approaches to problems in the biomedical and healthcare domains, with a non-exclusive focus on methods that overcome the "black-box syndrome" by making models interpretable and thus fulfil the usability requirements of most real medical applications.
Topics that are of interest to this session include (but are not necessarily limited to):
- Novel applications of existing CI and ML methods to biomedicine and healthcare
- Novel CI and ML techniques for biomedicine and healthcare.
- CI and ML-based methods to improve model interpretability in biomedicalproblems, including data/model visualization techniques.
- Novel CI and ML techniques for dealing with non-structured and heterogeneousdata formats.
- Development of user-friendly interactive exploratory interfaces and subjectspecificmodels.
- Alfredo Vellido, Computer Science Department, Universitat Politècnica de Catalunya BarcelonaTECH, Spain
- Paulo J.G. Lisboa and Sandra Ortega-Martorell, School of Computing and Mathematical Sciences, Liverpool John Moores University, United Kingdom
- José D. Martín, Intelligent Data Analysis Laboratory (IDAL), Department of Electronic Engineering, University of Valencia, Spain