Several recent papers underline methodological points that limit the validity of published results in life science and especially in neurosciences. At least three main points are emphasized that lead to invalidated findings: the endemic low statistical power of the published studies due to the small size of the population involved, data analysis and reporting are often selective and biased, and studies are rarely replicated then false discoveries or solutions persist. To overcome the poor reliability of research finding, several actions should be promoted including large cohort studies, data sharing and data re-analysis. Large-scale online databases may contribute to the definition of a “collective mind” facilitating open collaborative work or “crowd science”. Although technology alone cannot change scientist’s practices, technical solutions should be identified and implemented allowing effective population imaging and data sharing without increasing the burden on researchers.
This research topic (RT), cross-listed between the speciality section Computer Image Analysis of Frontiers in ICT and Frontiers in Neuroinformatics, is devoted to the methodological aspects and existing solutions to support the constitution and the management of large cohorts and facilitate data sharing between distributed data repositories in life science domain. Moreover, specific links to dedicated software and hardware infrastructures should be developed for the sharing and execution of image processing workflows making easier replication and comparison of data analysis procedures. This RT emphasizes the main conceptual and technical challenges posed by population imaging at a large scale: in brief, do we need standards for data model as provided by domain and application ontologies? How to interoperate between distributed repositories? Which levels of quality control (manual or automatic) and data provenance are required? Are big data centers preferable to federated distributed systems? What are the current solutions for high performance computing for in vivo imaging large repositories? Should we develop generic or tailored to usages solutions? Are cloud-computing clusters better solutions than grids or crowd computing? Is image processing pipelines sharing different of data sharing?
Submitted papers should be related to methodological issues for population imaging including data management and processing of large bio-imaging databases. Topics of interest include, but are not limited to, the following conceptual and technical approaches for:
• New paradigms and techniques for handling large image datasets:
o Data structures, domain and application ontologies,
o Federated databases, interoperability of data repositories,
o Data quality control,
o Data provenance,
o Data curation
• Image processing of large datasets:
o Data-intensive computing, parallel algorithms,
o Pipelines composition,
o Information fusion,
o Statistical techniques, semantic queries, data mining, machine learning and
meta-analysis,
o Data protection methods
• Infrastructures for facilitating data and software sharing, reused and re-analysis:
o Environments to support data-intensive computing,
o High computing access facilitation,
o Image computing in the cloud,
o High performance computing,
o Distributed storage systems
• Applications of population imaging
o Case studies using specific platforms (pros and cons, ...)
o Needs and requirements for specific multi-centre studies.
o Ethical considerations
Several recent papers underline methodological points that limit the validity of published results in life science and especially in neurosciences. At least three main points are emphasized that lead to invalidated findings: the endemic low statistical power of the published studies due to the small size of the population involved, data analysis and reporting are often selective and biased, and studies are rarely replicated then false discoveries or solutions persist. To overcome the poor reliability of research finding, several actions should be promoted including large cohort studies, data sharing and data re-analysis. Large-scale online databases may contribute to the definition of a “collective mind” facilitating open collaborative work or “crowd science”. Although technology alone cannot change scientist’s practices, technical solutions should be identified and implemented allowing effective population imaging and data sharing without increasing the burden on researchers.
This research topic (RT), cross-listed between the speciality section Computer Image Analysis of Frontiers in ICT and Frontiers in Neuroinformatics, is devoted to the methodological aspects and existing solutions to support the constitution and the management of large cohorts and facilitate data sharing between distributed data repositories in life science domain. Moreover, specific links to dedicated software and hardware infrastructures should be developed for the sharing and execution of image processing workflows making easier replication and comparison of data analysis procedures. This RT emphasizes the main conceptual and technical challenges posed by population imaging at a large scale: in brief, do we need standards for data model as provided by domain and application ontologies? How to interoperate between distributed repositories? Which levels of quality control (manual or automatic) and data provenance are required? Are big data centers preferable to federated distributed systems? What are the current solutions for high performance computing for in vivo imaging large repositories? Should we develop generic or tailored to usages solutions? Are cloud-computing clusters better solutions than grids or crowd computing? Is image processing pipelines sharing different of data sharing?
Submitted papers should be related to methodological issues for population imaging including data management and processing of large bio-imaging databases. Topics of interest include, but are not limited to, the following conceptual and technical approaches for:
• New paradigms and techniques for handling large image datasets:
o Data structures, domain and application ontologies,
o Federated databases, interoperability of data repositories,
o Data quality control,
o Data provenance,
o Data curation
• Image processing of large datasets:
o Data-intensive computing, parallel algorithms,
o Pipelines composition,
o Information fusion,
o Statistical techniques, semantic queries, data mining, machine learning and
meta-analysis,
o Data protection methods
• Infrastructures for facilitating data and software sharing, reused and re-analysis:
o Environments to support data-intensive computing,
o High computing access facilitation,
o Image computing in the cloud,
o High performance computing,
o Distributed storage systems
• Applications of population imaging
o Case studies using specific platforms (pros and cons, ...)
o Needs and requirements for specific multi-centre studies.
o Ethical considerations