Selected contributions submitted to ESGCO 2020 and accepted to be included in the ESGCO 2020 conference proceedings will be invited to submit a full manuscript to the following theme issue to be published in the scientific journal "Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences" entitled:
"Advanced Computation in Cardiovascular Physiology: New Challenges and Opportunities"
consistently with the following summary of issue:
"Recent developments in computational physiology have effectively exploited advanced signal processing and artificial intelligence tools for uncovering characteristic features of physiological and pathological states. While these advanced tools have demonstrated better-than-human diagnostic capabilities, the high complexity of these computational 'black boxes’ may severely limit scientific inference, especially in terms of biological insights on disease mechanisms. This theme issue combines research articles, reviews, and perspective contributions highlighting challenges and opportunities of advanced computational tools for processing comprehensive autonomic nervous system dynamics, with a more specific focus on cardiovascular physiology and pathology. This will include the development and adaptation of complex signal processing methods, multivariate cardiovascular models, multiscale and nonlinear models for central-peripheral dynamics, as well as deep and reinforcement learning algorithms applied to big biomedical datasets. Such a wide panoramic perspective is aimed at fostering the transition from the black-box paradigm to interpretable and personalized clinical models."
-----------------------------------------------------------------------------------------------
Selected contributions submitted to ESGCO 2020 and accepted to be included in the ESGCO 2020 conference proceedings will be invited to submit a full manuscript to the following special issue to be published in the scientific journal "Biomedical Signal Processing and Control" entitled
"Biomedical signal processing and modelling for cardiovascular oscillations"
Main topics of interest to this special issue include, but are not limited to:
- Advances in detection and sensing, and pre-processing for cardiovascular oscillations including heart rate/ blood pressure variability series
- Novel approaches for cardiovascular system modelling
- Advances in time-frequency analysis for cardiovascular oscillations
- Novel avenues in nonlinear and complexity analysis for cardiovascular oscillations
- Multivariate and causal modelling for cardiovascular data
- Heart-centred artificial intelligence approaches for big datasets in physiology and medicine
- Advances in sleep monitoring using cardiovascular oscillations
- Advances in diagnosis and prognosis of pathological states using cardiovascular oscillations
- Real-time processing for cardiovascular oscillations
-----------------------------------------------------------------------------------------------
Selected contributions submitted to ESGCO 2020 and accepted to be included in the ESGCO 2020 conference proceedings will be invited to submit a full manuscript to the following special issue to be published in a "Plos Collection". Invited full manuscript may be submitted to the following scientific journals Plos One, Plos Computational Biology, Plos Medicine.
Title: Methodological and physiological advances in signal processing and modelling for cardiovascular oscillations
Scope: The proposal is intended to collect full research papers generated from select, high- impact and high- profile conference contributions presented at the 11th ESGCO meetings. The proposers of this special issue are also the conference chairs.
Content: Recent developments in computational physiology have effectively exploited advanced signal processing and artificial intelligence tools for uncovering characteristic features of physiological and pathological states. While these advanced tools have demonstrated better-than-human diagnostic capabilities, the high complexity of these computational 'black boxes’ may severely limit scientific inference, especially in terms of biological insights on disease mechanisms. This theme issue combines research articles, reviews, and perspective contributions highlighting challenges and opportunities of advanced computational tools for processing comprehensive autonomic nervous system dynamics, with a more specific focus on cardiovascular physiology and pathology. This will include the development and adaptation of complex signal processing methods, multivariate cardiovascular models, multiscale and nonlinear models for central-peripheral dynamics, as well as deep and reinforcement learning algorithms applied to big biomedical datasets. Such a wide panoramic perspective is aimed at fostering the transition from the black-box paradigm to interpretable and personalized clinical models.
Motivation: Advanced computational tools are changing the landscape of biomedical research. Complex signal processing methods, multivariate nonlinear models, as well as deep and reinforcement learning algorithms are successful examples of mathematical, physical, and engineering sciences which are extremely effective in supporting healthcare. However, while these approaches may achieve better-than-human performance for a disease diagnosis, several high-performing models generate results that are difficult to interpret by clinicians. Retrieving intuitive information which may explain physio-pathological mechanisms, identifying model weaknesses and generalization properties, and extracting biological insights from these computational ‘black boxes’ has often proven extremely arduous.
Recently proposed machine-learning methods rely on the availability of large amounts of high-quality training data, which therefore may not be fully representative of the target patient population because they are affected by various types of bias and noise that are typical of a real-world, clinical environment. Moreover, methodologies for applying deep neural networks to general diagnostics (such as the interpretation of signs and symptoms, past medical history, laboratory results and clinical course) and treatment selection are still under development. Also, while complex and nonlinear advanced signal processing strategies have contributed to the technical understanding and description of physiological signals, their success in developing translational clinical tools has been limited so far.
We posit that a dedicated PLOS issue on these topics would be extremely helpful in focusing transdisciplinary efforts towards the development of advanced computational tools for medicine and healthcare. Cutting-edge research papers, as well as review and perspective articles will describe and encourage novel methods especially targeted to retrieving information that cannot be captured by traditional/existing strategies, and recasting this information into clinically actionable indicators. This timely theme issue will thus allow a significant leap in the development of advanced computational tools embedding both physiological insight and clinical impact.
Impact: The application of advanced computational tools in physiology, especially in the cardiovascular field, is becoming increasingly important as the widespread availability of publicly-available ready-to-use software for processing biomedical data in a blind fashion (i.e., disjoint from any specific technical knowledge), as well as the large amount of data generated by biomedical sensors embedded in daily-use devices (e.g., smartphones and smartwatches) are ever-increasing. In this context, it is of utmost importance to provide guidelines and directions in order to orient the research and industrial communities. To this end, a landmark publication such as the proposed theme issue would critical information on future developments. The theme issue will stimulate a multidisciplinary dialogue on the combination of clinical cardiology and complex data processing methods, also highlighting significant ethical issues on medical management decisions by computational machines. In this way, technicians as well as physicians of the future would be encouraged to develop interpretable computational tools for evaluating patients and supporting their treatment.
,
