Peter Stadler

title: Catching Monsters in RNA Land: Mapping Atypical Transcripts  

Abstract:

Large-scale transcriptomics projects over the last decade have revealed a
plethora of short and long non-coding RNAs. Although biological functions
and molecular mechanisms have been identified for only a small number of
examples, it has become clear that the RNA complement of eukaryotic cells
is high diverse with regard to the size, genomic structure, function, andmechanism. Workflows for transcriptome analysis, however, often
pre-suppose knowlege on gene structures and tends to disregard deviant signals as noise that is to be discarded. The segemehl suite of mapping and downstream analysis tools strives to avoid such potential biases.
It is geared towards mapping with multiple errors and multiple splits in the reads, thus allowing the analysis of editing, complex splicing patterns, and circular transcripts, and transsplicing using a single concise algorithmic approach that works equally well with data from prokaryotes and eukaryotes. Applications range from the discovery of unusual splcing patterns in human and archaeal transcripts to the analysis of macroRNAs, the analysis of dual RNA-seq data of intracelluar parasites to comparative transcriptomics.

Nazim Agoulmine

Title: From small bio-sensors on/in human body to powerful virtual sensors in the Cloud: Opportunities and Challenges in future eHealth.

University of Evry Val d’Essonne, France

Abstract:

Using Information and Communication Technologies (ICT) to improve healthcare quality and reduce costs for ubiquitous healthcare has been recognized and is being implemented as a national priority in major world economies to face the ageing society problem and the cost of healthcare. The aim of this talk is to present how sensor, IoT and Cloud Computing technologies together can help to design new efficient health management systems. Data collected by medical and environmental sensors from individuals in their environment can be transported from using  various communications technologies (WBAN, WPAN, WLAN, WWAN, etc.)  to powerful backend cloud computing datacenters where medical applications are hosted. This model permits to provide healthcare anywhere and anytime virtually approaching patients to doctors. Medical data collected at large scale could also be mined to detect anomalies in large populations or geographic area that can be used by public health policy makers.
Prof. Nazim Agoulmine is a full professor at the University of Evry Val d’Essonne, France since 2000. He is member of the IBISC Research laboratory, co-chairing the COSMO research group. Since 1989, he has been involved in several large research projects in ICT funded by EU and other French agencies. His current research interests are Cloud Computing, Internet of Things, 5G network, eHealth. He is nowadays the chair of the eHealth Technical Committee of the Communication Society. He is the author of more than a 100 publications in the area, several books and book  chapters. He received several awards for his research activities and contributions to the community.

Massimo Mischi

title: Contrast-enhanced imaging for quantitative analysis of angiogenesis

One in four people dies because of cancer. Prostate cancer is the most frequent form of cancer in men; unfortunately, clinical diagnostics is limited to invasive systematic biopsies.

Abstract:

Angiogenesis is associated with cancer growth and development, providing an important opportunity for diagnostic imaging. In particular, dynamic contrast-enhanced ultrasound (DCE-US) is a diagnostic tool that is suitable for analysis of the microvascular bed. Quantitative analysis of cancer angiogenesis is of particular interest for diagnosis and prognostic assessment of cancer aggressiveness. Recently, we successfully proposed contrast-ultrasound dispersion imaging (CUDI) for prostate cancer diagnosis, exploiting the link between intravascular dispersion of ultrasound contrast agents and angiogenic changes in the microvascular architecture. This concept has also been translated to DCE-MRI, introducing new models that account for extravascular contrast leakage.

By the proposed methods, independent estimation of dispersion and velocity components was not possible due to the ambiguity between convection and dispersion. More recently, a new method was proposed that provides a separate estimation of convection and dispersion by full identification of the local dilution system. This provides additional diagnostic information, including blood velocity maps that can be used for additional statistical analysis. By use of tractography, the main stream lines of blood flow can also be reconstructed and rendered. Eventually, combination by machine learning of relevant contrast-transport features derived from DCE-US has been proven to boost even further the accuracy of prostate cancer diagnosis.

Novel quantitative molecular imaging of angiogenesis has additionally been introduced by modeling the transport kinetics of ultrasound contrast agents that are targeted to angiogenic expressions. The developed analytical models enable the assessment of the binding/unbinding kinetics of the agent, showing promise for prostate cancer localization.

In general, the proposed methods for quantitative angiogenesis imaging are not specific to prostate cancer only, and future application to other types of cancer can also be envisaged.