Cardiorespiratory physiology and modeling with clinical applications
Thomas Heldt*, Alberto P Avolio, Nicolas W. Chbat, Brett BuSha, Ricardo Luis Armentano, Leonardo Bocchi, Thomas Penzel, Gaetano Valenza
July 23, 2019 8:30 – 17:30
This full-day workshop is organized and provided by the EMBS Cardiopulmonary Systems Technical Committee and is geared toward undergraduate and early graduate students entering the field of cardiorespiratory systems engineering. We draw from our collective teaching experience and research interests to convey fundamental cardiorespiratory physiology, modeling, and elements of signal processing. The workshop is divided in two sessions. Cardiovascular physiology and modeling are covered in the morning session, and respiratory physiology and modeling in the afternoon. Each session includes lectures with topics ranging from normal physiology, modeling, and some aspects of domain-specific signal processing, highlighting the importance of understanding physiology, for 1) research, deepening the understanding of normal physiology; 2) clinical application, providing some exemplary insights into pathophysiology; and 3) clinical translation through an understanding of physiology-based medical device development and products.
Advances and challenges in the field of emotion recognition and emotion regulation for designing technology-based interventions for affective disorders
Enzo Pasquale Scilingo*, Alberto Greco
July 23, 2019 8:30 – 12:30
Affective disorders (AD) are estimated to be among the highest-ranking causes of mental disease in the upcoming years. Over the last decades, there has been great attention to the impact of technologies on psychotherapeutic practice for the treatment of AD. Understanding emotions and their underlined processes construct the main elements of therapeutic practice, but, apart from solely characterizing the emotional states, the challenge remains on how to capture emotional changes or state emotion regulation. The current personal technologies aim to automatically characterize alternations in emotional experiences through self-reports embedded in mobile technologies or, rarely, using biosensors. The research on digital mental healthcare system tends to propose novel interactive-based ecological assessment and intervention platforms that could be harnessed for emotion regulation. The therapies based on emotion regulation help individuals to have a better insight into their daily life emotional changes and trigger people’s habitual responses through training their emotion regulation strategies. Most of these strategies are gaining benefits from reappraisal, mindfulness and biofeedback techniques. In this workshop, we organize panels to cover the interdisciplinary approach for introducing the strategies of emotion regulation as well as challenges in designing technology-based interventions for affective disorders. We aim to introduce fundamental models of emotion, and advanced signal processing methods to characterize physiological influential theories of emotion regulation and adaptive technology-based emotion regulation strategies. The main objectives of this event are providing better knowledge on developing mental healthcare technologies, understanding how the self-management interactive systems give us insights in emotional experience and covering the state-of-the-art of theories and methods which aim to enhance the affective health and wellbeing.
SimpleITK: A Tool for Biomedical Image Processing, from Cells to Anatomical Structures
Ziv Yaniv*, Bradley Lowekamp, Hans Johnson
July 23, 2019 8:30 – 12:30
SimpleITK is a simplified programming interface to the algorithms and data structures of the Insight Segmentation and Registration Toolkit (ITK). It supports bindings for multiple programming languages including C++, Python, R, Java, C#, Lua, Ruby and TCL. These bindings enable scientists to develop image analysis workflows in the programming language they are most familiar with. The toolkit supports more than 15 different image file formats, provides over 280 image analysis filters, and implements a unified interface to the ITKv4 intensity-based registration framework. The SimpleITK user base is rapidly growing, with more than 100,000 downloads of the Python bindings in the past year. Finally, by combining SimpleITK’s Python bindings with the Jupyter notebook web application one can create an environment which facilitates collaborative and reproduciable development of biomedical image analysis workflows. In this tutorial, we will use a hands-on approach utilizing Jupyter notebooks to explore and experiment with various SimpleITK features in the Python programming language. Participants will follow along using their personal laptops, enabling them to explore the effects of code changes and parameter settings not covered by the instructor. Examples using anatomical and microscopy images will highlight the various capabilities of the toolkit. We will start by introducing the toolkit’s two basic data elements, Images and Transformations. We will then explore the features available in the toolkit’s registration framework including: optimizer selection, the use of linear and deformable transformations, the embedded multi-resolution framework, self-calibrating optimizers and the use of callbacks for registration progress monitoring. Finally, we will show how to use SimpleITK as a tool for image preparation and data augmentation for deep learning via spatial and intensity transformations and its use for segmentation evaluation.
Intellectual Property Strategies for International Protection of Biomedical Products and Technologies
Dorin Panescu*, Theodore Papagiannis, Michael Christensen, Leonard Werner-Jones
July 23, 2019 8:30 – 12:30
Participants will learn about and discuss various considerations related to obtaining and enforcing patents around the world from experienced US and European patent practitioners and from an experienced inventor who has been intimately involved in the international patent process with several companies. The discussion will include an overview of the international PCT application process, the centralized European patent application route, as well as various strategies for U.S. protection. The discussion will also include different options for expediting examination in various countries and considerations for drafting an application in light of anticipated international protection. Participants will be involved in interactive discussion and may be given the opportunity to participate in hands-on activities, including how to strategically evaluate claims and other portions of a patent application for a sample fictitious invention and how to deal with feedback (e.g. rejection letters) from Patent Office Examiners in various regions/countries.
Tentative Outline:
- Overview of patent basics with emphasis on international IP
- Timing of when to file patent applications
- Difference between provisional and non-provisional patent applications and costs
- Different options for international patent protection and associated costs
- Overview of international PCT application process and International Search Reports and Written Opinions
- Overview of centralized European patent application route and validation process after issuance
- Overview of proposed European unitary patent and unified patent court and current status
- Various options to expedite international patent protection
- Different types of patent protection (e.g., invention patents vs. utility models)
- Enforcement in selected key countries
- Overview of a fictitious medical device invention
- Interactive discussion of a sample patent application of the fictitious medical device invention
- Different sections of a patent application
- Considerations for what to include and not include in the application (e.g., how a patent application is different than a scientific article for publication)
- Considerations for drafting an application in light of anticipated international protection
- Interactive exercise regarding claims for fictitious medical device invention
- Do claims sufficiently distinguish over prior art?
- Are claims broad enough to cover alternative designs?
- What types of different claims can be pursued?
- Interactive exercise reviewing an International Search Report/Written Opinion for a PCT Application
- Q&A
Wearable Sensor solutions for integrated mobile EEG/EXG, motion capture & eye tracking in the real and virtual worlds.
Walid Soussou*, Xiaoshan Huang
July 23, 2019 8:30 – 12:30
This tutorial will explore mobile and wearable solutions for biomedical research geared toward real-world and virtual reality environments. We will describe a framework for comprehensive mobile physiological monitoring solutions across a variety of research applications. Presentations will include some basics on the principles of operations of a range of research grade wearable sensors, including: EEG, EOG, EMG, ECG, fNIR, GSR, respiration, temperature motion capture, eye-tracking, facial coding, etc. We will discuss approaches to signal synchronization, and benefits and limitations of using such sensors in real-world settings for naturalistic environments research. We will also present solutions for brain and physiological monitoring in Augmented and Virtual Reality. In addition, time will be allotted for interactive demonstration of the sensors and their enabling platforms. First, we will present a new hybrid system merging high fidelity dry electrode EEG with LED-based fNIR sensors. This fully integrated headset is designed to provide research-grade accuracy and wireless mobility for artifact-free brain monitoring in ambulatory research. We will introduce the separate brain activity monitoring modalities, discuss the significance and utility of multimodal imaging for medical and cognitive research, and present some sample applications. Next, we will present the practical utility of mobile high-density EEG systems for research in real-world settings. We will present several BCI applications of this technology. Finally, we will present a platform for real-time acquisition of multimodal signals and explore the benefits of sensor fusion for research in BCI, neurorehabilitation, neurofeedback, neuroergonomics, neuromarketing, and more. During the hands-on portion of the tutorial, participants will be able to experiment with the various technologies and examine their ease-of-use, signal quality, and practicality, and discuss their benefits and limitations.
Heat-based techniques for cancer removal: a journey through state-of-the-art, research and clinical application
Paola Saccomandi*, Emiliano Schena
July 23, 2019 8:30 – 12:30
This tutorial will focus on cutting-edge research topics in thermal treatments of cancer. Minimally invasive thermal treatments have gained acceptance for clinical application in cancer removal. Laser, Microwave, Radiofrequency, High Intensity Focused Ultrasound allow inducing localized temperature increase into the tumor target, and findings from the medical community to the engineering field improve the procedures day-by-day. The main research directions devoted to solutions aiming at the design of a more selective and safe procedure. This tutorial will expand on the main pillars for successful therapy, considering: i) the techniques used for continuously monitor tissue temperature to have a clear end point for a complete removal of tumor spearing the surrounding healthy tissue, ii) the use of hyperthermal treatment planning tools to predict the effects of the thermal treatment via mathematical model, and to perform patient-specific treatment, and iii) new frontiers methods for improving the treatment selectivity. An outlook to the clinical application of these heat-based techniques will conclude the tutorial. The attendees will gain key insights and understandings about the main solutions for improving thermal treatments for applications in oncology, from the preliminary research to the clinical perspective.
Neuromuscular modeling for evaluating neural prosthesis control and myoelectric decoding
Ning Lan*, Douglas Weber
July 23, 2019 8:30 – 12:30
Control of neural prosthesis such as FES for spinal cord injury and prosthetic hands for amputees has advanced greatly. This is achieved with improved recording and decoding techniques of neural signals, better understanding of principles of neural control in humans, as well as realistic modeling of the neuromuscular system. Implanted micro-electrode arrays are able to provide multiple channels of recordings of neuronal activities at the motor cortex and other areas of the brain. Such neural recordings allowed us to decode motor outputs or motor intentions in patients who lost neural communication with their external limbs due to neurological injuries or amputation. The use of decoded neural signals to control external devices, however, is not straightforward. One promising approach has been to mimic natural modules of neural control of movement in humans, known as muscle synergies. Parallel with these progresses, physiologically realistic models of neuromuscular control are being developed to understand and to evaluate neural control strategies. This workshop will provide a tutorial on the advances in these several aspects of neural prosthesis by a group of distinguished researchers in these fields in the world. It is hoped that this workshop will promote the integration of theoretical, technological and methodological innovations in diverse fields, which may lead to a multi-disciplinary solution of neural prosthesis.
The Fast-Changing Landscape of Electroencephalography
W. G. Besio*
July 23, 2019 8:30 – 12:30
This workshop is designed to give both novice and experienced electroencephalography (EEG) users a synopsis of the latest innovations in EEG and related areas. We will have a keynote talk by Professor Sridhar Sunderam who will discuss neurotechnologies for the analysis and treatment of neural disorders such as epilepsy and spinal cord injury. Next Dr. Han Yuan will give an invited talk on large-scale brain networks. Lastly, we will have talks from four groups on their latest technologies for acquiring and processing the EEG which are making the use of EEG more practical. Finally, we will have hands on demonstrations of the technologies. EEG is the recording of brain electrical activity from the scalp. The EEG measures the difference in potentials between electrodes generated by ionic currents flowing within neurons of the brain. For many years EEG has had limited use due to poor signal quality, low spatial resolution, and non-portability. Even with these limitations EEG is still a standard practice in clinical settings such as diagnosis of epilepsy and for research such as brain computer interfacing. In recent years electrodes, signal acquisition hardware, and signal processing software have undergone major improvements allowing new and improved applications of EEG. These talks will illustrate some of the latest technologies for acquiring EEG. The target audience of the workshop is the whole community of the IEEE EMBS Society interested in brain research.
Science at-home: Breaking with the limitation of today’s electrophysiological studies
Matthias Hohmann*, Vinay Jayaram, Moritz Grosse-Wentrup
July 23, 2019 13:30 – 17:30
Studies that require electrophysiological recordings do not scale well due to the lack of user-friendly systems and complicated hardware. Bringing self-administered electrophysiological systems into the homes of patients and consumers could break with the limitations of small sample sizes, and allow for scalable, longitudinal experiments. In this workshop, speakers from academia as well as industry will present emerging technologies that may be key to scaling electrophysiological studies, and existing use-cases. Workshop presentations will feature: – A smartphone application for unsupervised, large-scale EEG studies – Low-cost, easy-to-use EEG hardware – Longitudinal machine learning in at-home EMG-based control – Low-cost, easy-to-use EMG hardware During the group discussion, attendees will have the opportunity to identify potential features for future versions and evaluate them based on their relevance to the scientific community. Attendees will also discuss challenges in ethics and privacy that emerge with the scaling of electrophysiological studies. The results will be discussed and incorporated into future releases. Our goal is to enable the scientific community to conduct large-scale electrophysiological experiments by actively shaping the development process of the presented technologies together.
The Future of Sleep Health: A data-driven revolution in sleep science and medicine
Luis Fernandez-Luque*, Ignacio Perez Pozuelo, Panagiotis Bamidis, Abdelkader BAGGAG, Raghvendra Mall, Joao Palotti, Michael AUPETIT, Bing Zhai, Wafa Waheeda Syed, Aarti Sathyanarayana, Jaideep Srivastava, Shaharad Taheri
July 23, 2019 13:30 – 17:30
The emergence of mobile and wearable technologies combined with developments in artificial intelligence offers the opportunity to develop new applications to improve sleep quality and help people who suffer from chronic sleep conditions. In addition, large-scale adoption of wearable devices facilitates the capture of immense datasets that can enable research on the interplay between physical activity and sleep. In this workshop, we will discuss the latest insights from leading experts in sleep health, digital health and neuroscience. Further, we will provide hands-on experience in the use of artificial intelligence and signal processing for the analysis of sleep data from wearable devices, as well as describing how to interpret and visualize the results.
Telemedicine and Telemonitoring in AAL Home Environments
Natividad Martinez Madrid*, Ralf Seepold
July 23, 2019 13:30 – 17:30
ICT supports telemedicine to lower health access barriers and to provide better health care. While the potential in Active Assisted Living (AAL) is increasing, it is difficult to evaluate its benefits for user and it requires coordinated actions to launch it. The European Commission’s action plan 2012–2020 provides a roadmap to patient empowerment and healthcare, to link up devices and technologies, and to invest in research towards the personalized medicine of the future. As a quickly developing area in medicine, telemonitoring is a demanding field in research and development. Telemonitoring is an important component of the personalized medicine, where health providers can obtain precise information of outcare or chronic patients to improve diagnose and therapy and also help healthy persons with prevention support. Telemonitoring combines mobile and wearable devices with the personal AAL home environment, a private or (partly) supervised home, most often called ‘smart home’. The focus of this workshop is on new hardware and software solutions specifically designed to be applicable in AAL environments with the objective to empower patients. This workshop presents system-oriented solutions covering wearable and AAL-embedded devices, computer science infrastructure both at the users’ and the medical premises to handle the data and decision support systems to support diagnose and treatment. Major workshop areas:
- IoT platforms for AAL personalized health applications
- Sensors and devices integration
- Open mobile health platforms
- Interoperability through secure cloud solutions
- Design of wearables and biometric measuring devices for AAL
- Embedded hardware for biosignal monitoring
- Mobile ultrasound devices
- AAL-Applications needing multiple monitoring devices
- Sleep monitoring (Non-invasive, long-term, uncontrolled environments)
- Rehabilitation
- Diabetes and other chronic illnesses
- Decision support systems
IMAGE-BASED ESTIMATION OF CARDIOVASCULAR TISSUE MOTION AND ELASTICITY
Spyretta Golemati*, Elisa Konofagou
July 23, 2019 13:30 – 17:30
The estimation of tissue motion and elasticity in the heart and vessels is valuable for characterising cardiovascular status in health and disease. Following the periodic movement of the heart and the resulting blood pressure variations during the cardiac cycle, the cardiovascular tissue performs a complex three-dimensional motion. Motion can be quantified through the calculation of a number of kinematic parameters, including displacements, velocities, and accelerations, as well as strain, which has gained attention for characterising tissue function. Tissue elasticity characterises its mechanical properties and can be assessed through estimations of displacements/strains in combination with pressure indices. To estimate motion from images, different imaging modalities can be used, including ultrasound, magnetic resonance imaging (MRI) and computed tomography (CT). Among these, ultrasound imaging is the most widely used technique for cardiovascular motion estimation, due to its wide availability, easy use, high temporal resolution, and possibility to access various central and peripheral vessels. This workshop aims to present the state-of-the art in image-based estimation of motion and elasticity of the cardiovascular tissue and to highlight novel directions and challenges toward identifying non-invasive markers of cardiovascular risk.
Interpretable & Transparent Deep Learning
Wojciech Samek*, Grégoire Montavon, Klaus-Robert Müller
July 23, 2019 13:30 – 17:30
Deep neural networks (DNNs) are reaching or even exceeding the human level on an increasing number of complex tasks. However, due to their complex non-linear structure, these models are usually applied in a black box manner, i.e., no information is provided about what exactly makes them arrive at their predictions. This lack of transparency can be a major drawback in many practical applications. For instance, in medical diagnosis incorrect predictions can be lethal, thus simple black-box predictions cannot be trusted by default. Instead, the predictions should be made interpretable to a human expert for verification. In the sciences, deep learning algorithms are able to extract complex relations between physical or biological observables. The design of interpretation methods to explain these newly inferred relations can be useful for building scientific hypotheses, or for detecting possible artifacts in the data/model. Also from an engineer’s perspective interpretability is a crucial feature, because it allows to identify the most relevant features and parameters and more generally to understand the strengths and weaknesses of a model. This feedback can be used to improve the structure of the model or speedup training. Recently, the transparency problem has been receiving more attention in the deep learning community. Several methods have been developed to understand what a DNN has learned. Some of this work is dedicated to visualize particular neurons or neuron layers, other work focuses on methods which visualize the impact of particular regions of a given input image. An important question for the practitioner is how to objectively measure the quality of an explanation of the DNN prediction and how to use these explanations for improving the learned model. The tutorial will present recently proposed techniques for interpreting, explaining and visualizing deep models and explore their practical usefulness in biomedical applications.
Tutorial on Computational Modelling of Therapeutic Devices
Socrates Dokos*, JIe Chen
July 23, 2019 13:30 – 17:30
Computational modelling is an important skill for biomedical engineers, playing an increasing role in biological and medical research as well as in the medical device industry. Development of therapeutic devices requires not only extensive bench tests, animal experiments, and clinical trials, but also computational simulations which allow cost-effective investigation of system behavior and device design iteration. Modelling of therapeutic systems may eventually be routinely applied to tailor a range of individual treatments based on patient-specific simulations. In this half-day tutorial, participants will be taken through step by step instructions using COMSOL finite element software to model three therapeutic devices: an external cardiac defibrillator, RF atrial ablation, and a nerve cuff stimulator. We will also teach the use of COMSOL to design a microfluidic device for single-cell trapping with applications to cell sequencing and cell-cell heterogeneity.
July 23, 2019 13:30 – 17:30
The Brain-Computer Interface (BCI) research area is a thriving and rapidly expanding field. BCIs have been developed during the last years for people with severe disabilities to improve their quality of life. However, BCI applications have recently been extended to different research areas, such as rapid functional mapping on the cortical level, virtual reality and rehabilitation & therapy after stroke. The workshop will discuss prerequisites to successfully perform both invasive and non-invasive BCI experiments, and discuss progress in relevant medical domains. Live demonstrations of BCI control will help attendees understand the technology. Many studies over the past decade have shown that ECoG activity in the high gamma band is a reliable indicator of local task-related cortical activity, and could thus complement existing methods for functional mapping. Further highlights of the workshop are new approaches of semiautomatic mapping of the sensorimotor cortex using somatosensory evoked potentials and cortico-cortical evoked potentials for brain mapping in intraoperative scenarios. The workshop will also describe techniques for direct-brain stimulation in Parkinson patients to reduce tremor and to improve treatment opportunities. Attendees will get insights into state-of-the-art hardware and software for BCI research enabling participants to run their own experiments. Advantages and disadvantages of dry and wet biosignal sensors will be discussed as well as differences between invasive and non-invasive BCI applications. Participants will be able to understand the most commonly used methodological approaches (auditory, visual, vibrotactile P300, SSVEP, motor imagery) underlying the design of BCIs and will have the opportunity to operate BCI based medical systems for neurorehabilitation, communication and consciousness assessment. The speakers of this workshop provide a competent mix between neurosurgery, neurology, scientific and technical expertise.
Stress Reduction Methods using Biofeedback and Virtual Reality
Bjoern M Eskofier*, Nicolas Rohleder, Robert Richer, Stefan Gradl
July 23, 2019 13:30 – 17:30
Stress is a hidden epidemic. Its biophysical long-term effects have a large impact on the health and wellbeing of humans in our post-industrialized society. The World Health Organization estimates that mental diseases, including stress-related disorders will be the second leading cause of disabilities by the year 2020. The economic impact of stress is huge, with an estimated loss of about $42 billion for absence and treatment of stress-related illness in the US. Therefore, it is important to understand the underlying mechanisms and identify possibilities/tools to cope with it and prevent negative health outcomes. Therefore, stress-reduction methods should be effective and highly accessible for everyone. State-of-the-art methods for stress coping include biofeedback-guided meditations, breathing exercises, social support by talking to a close person or general conscious relaxation techniques. However, studies suggest that while many of the common stress reduction methods work well for chronic stress, most of them only show limited effects for the coping of acute psychosocial stress where individuals are regularly and repeatedly exposed to potentially stressful situations. Nonetheless, reducing psychosocial stress is of particular importance since it has been shown to be a long-term predictor of negative health outcomes. With the evolution of highly immersive systems in the last years, such as virtual reality (VR), many new promising approaches in different therapeutic fields involving the human psyche have been presented. For that reason, this workshop explores the state-of-the-art in therapeutic biofeedback with or without VR and its future directions.