Models and Algorithms for Analysis of Large-scale Biological Networks
Organizers: Byung-Jun Yoon*, Xiaoning Qian (USA)
Recent advances in high-throughput techniques for measuring molecular interactions and the development of text-mining techniques for crawling the vast biomedical literature to collect known molecular interactions have led to the emergence of large-scale biological networks. These networks provide valuable resources that enable systematic genome-scale studies of biological interactions, which may unveil important insights on the structure and functional organization of biological networks and lead to the detection of novel network modules. The proposed tutorial consists of two parts. In the first part of the tutorial, we will present probabilistic models for comparative analysis of large-scale biological networks and introduce efficient algorithms for network alignment and network querying. In the second part of the tutorial, we will go over the concept of network clustering and network community detection and present effective computational methods for accurate identification of functional network modules in biological networks.
Deep Audio, Speech & Biosignal Sensing for mHealth & Biomonitoring in Two Nutshells
Organizers: Bjoern Schuller*, Nicholas Cummins (United Kingdom)
With an aging population and ever-increasing health costs, technologies such as smart phones and wearable devices and advances in AI are currently transforming health care. The combination of passive, non-invasive and non-intrusive smart monitoring technologies together with the robust analysis attainable by deep learning means that remote, cost efficient, automated, and objective diagnosis and monitoring of manifold health states is now accessible to the broad public – everywhere and anytime. Speech analysis, for example, has matured to a new form of active and passive remote sensing technology for a broad range of health conditions and wellbeing. Conditions such as depressive disorders, post-traumatic stress, Alzheimer’s and Parkinson’s disease and dementia are detectable, at accuracies approaching clinical utility, via speech analysis. Beyond, earlier pre-diagnosis of epilepsy seizures, or even Rett-Syndrome, Fragile X, and other neurodevelopmental conditions have become feasible. Sleep Apnea, Eating and Alcohol Consumption Behaviour, or Voice Disorder can be continuously monitored, and correctional feedback and aid can now be given to individuals, e.g., on the Autism spectrum “on the go”. This tutorial is an introduction to audio and biosignal sensing “in the wild” for mHealth and will comprise of two main sections: first, an overview of the topic covering, data collection, feature extraction and relevant recent machine learning principles as well as a discussion on the key research problems and challenges including data protection and ethics; second, a hands-on session: attendees will receive training on running open-source established feature extraction and machine learning toolkits, including the state-of-the-art openSMILE and novel openXBOW multisensorial feature extraction software. The tutorial will cater for both established and new researchers to this field. By attending, participants will gain key insights and understandings and be able produce results.
Healthcare Revolution through Wearable Electronics
Organizers: Hammad M. Cheema* (Pakistan)
The first part of the tutorial will introduce wearable electronics domain and the impact it has had in the healthcare realm. Starting from fitness trackers, to smart watches to diagnostic devices, healthcare has witnessed a drastic change in the western countries where ambulatory monitoring is being preferred over in-hospital checkups thus reducing the healthcare costs. In addition, continuous gathering of data is helping physicians and medical scientists to get insight into information which was never available before. The tutorial will also present the basic building blocks which are typically used to develop health centric wearable electronics. The second part of this tutorial will cover Machine Learning and Data analytics techniques for handling wearable sensors. Participants will be provided an overview of how to analyze high-dimensional time-series data using Machine Learning and Pattern Recognition techniques. Special emphasis will be given to feature extraction frontend and participants will learn how to construct and design novel features from data. Different classifiers such as Support Vector Machines and Deep Neural Networks will also be covered. Participants will also be provided hands-on experience on atleast one dataset. Aims and learning objectives: The aims and learning objectives of the tutorial are as follows:
- To share the history of wearable electronics
- To discuss the impact of wearable electronics in healthcare
- To share few case-studies of products/projects in this domain
- To share technological enablers for wearable electronics
- To highlight the future trends in healthcare wearables
- To provide an introduction the basic concepts of machine learning
- To discuss the various parameters and tradeoffs that need to be considered during wearable data analytics
- To provide hands-on exposure to Machine Learning and Data analytic tools for processing wearable sensor data
Structural and Functional Imaging with Optical Coherence Tomography and Optical Coherence Elastography
Organizers: Kirill Larin* (USA)
This tutorial will overview recent advances in development and application of optical imaging techniques for structural and functional imaging and sensing of various transport, developmental, and disease progression in tissues and cells. Special emphasis will be devoted to theory and applications of novel imaging/sensing modality – Optical coherence Tomography (OCT) and Optical Coherence Elastography (OCE). This tutorial will start with basic description of light-tissue interaction including structural and optical models of tissues with single and multiple scattering. It will be shown that light reflection, transmission, scattering, and state of polarization can be effectively controlled by changes of tissue structure and the refractive index of tissue components. Many examples of OCT noninvasive sensing will be provided such as cardiovascular imaging, ophthalmic applications, monitoring of drug diffusion and optical clearing, sensing and quantifying of microbubbles and nanoparticles in tissues and blood, imaging of early embryonic cardiovascular system development, and, the newest hot topic — assessing mechanical properties of tissues and cells. Intended Audience: Engineers, scientists and physicians who are interested in learning optical imaging and spectroscopy, laser methods, instruments design, and application for medical science and clinics will find this course useful.
Smarter Photoacoustic Imaging: Systems, Applications, and Agents
Organizers: Chulhong Kim*, LIANG SONG, Yoshifumi Saijo, JUNJIE YAO (Korea, South)
High-resolution volumetric optical imaging modalities, such as confocal microscopy, two-photon microscopy, and optical coherence tomography, are growing in their importance for biological and medical imaging. However, due to strong light scattering, the penetration depth of optical imaging is limited to the transport mean free path of photons in biological tissues (~1 mm). Photoacoustic imaging, an emerging hybrid modality that can provide strong endogenous and exogenous optical absorption contrasts, has overcome the fundamental depth limitation of optical imaging by maintaining excellent spatio-temporal resolution representative of ultrasound imaging. The image resolution, as well as the maximum imaging depth, is scalable with ultrasonic frequency within the reach of diffuse photons. In biological tissues the imaging depth can be up to a few centimeters. Furthermore, photoacoustic imaging can noninvasively deliver anatomical (i.e., vascular structures, solid tumors and angiogenesis, and internal organs), functional (i.e., total hemoglobin concentration, hemoglobin oxygen saturation, blood flow, pH, and metabolic rate of oxygen consumption), and molecular information from living tissues. For highly sensitive molecular photoacoustic imaging, a valuable tool for personalized medicine, exogenous contrast agents (e.g., organic dyes, metallic and nonmetallic nanoparticles, reporter genes, or fluorescence proteins) with biomarkers are commonly utilized. In this one-day Workshop, the following topics will be discussed; (1) multi-scale photoacoustic imaging systems, (2) pre-clinical imaging of morphology, function, and reporter molecules, (3) clinical applications, and (4) commercial opportunities.
Design Challenge Workshop
Organizers: Bruce Hecht*, Ayesha Khalid (USA)
Emerging challenges and opportunities are enabled through new technologies and design in the field of biomedical engineering. Participants will be challenged to employ and develop some of the most important skills required by leaders: creativity, business acumen, social responsibility, communication, and teamwork. Join us this at the IEEE EMBC Conference 2017, for the Design Challenge Workshop.
Goals: The Workshop is designed to deliver a program that: * Inspires young visionaries from around the world by firing up their competitive spirit;
- Catalyzes innovation
- Cultivates impact through creative solutions to some of the world’s biggest health-related challenges
- Teach young innovators about the power of competitions to solve some of the world’s biggest problems
- Supports the theme of the 39th Annual International Conference of the EMBC to develop Smarter Technologies for a Healthier World.
- The program will activate EMBS’s network to direct their passion, leadership, and entrepreneurial spirit toward solving challenges in biomedical engineering. Participants will be challenged to employ and develop some of the most important skills required by leaders: creativity, business acumen, social responsibility, communication, and teamwork.
Neurophotonics: Imaging and Manipulating the Brain
Organizers: Euiheon Chung*, Jung-Hoon Park (Korea, South)
Recent advances in optical imaging have proven to be a powerful workhorse for new discoveries in neuroscience. Various contrast mechanisms based on light-matter interaction have enabled structural and/or functional neuroimaging with diverse animal models. However, effective study of the brain awaits breakthroughs in many different aspects. For functional imaging at the level of neural circuits, high resolution large field of view imaging at high sampling rates is required. For deep brain imaging, multiple scattering of light becomes the biggest obstacle. For structural imaging of entire brains, aka the connectome, automation of a high resolution, robust sequential imaging system is required. The large data sets that are obtained in these newly developed high throughput systems also require new data storage and processing schemes that were not previously anticipated. This workshop brings together the latest advances in various areas of neurophotonics with a common goal: imaging and manipulating the brain. This workshop will provide a platform to foster sharing and collaboration in this vibrant research arena.
A Hands-on Approach to Neural Connectivity Inference Methods
Organizers: Luiz Antonio Baccala, Koichi Sameshima, Laura Astolfi*
The workshop intends to provide a hands-on learning opportunity. In addition to examining in detail conceptual aspects associated to connectivity estimation from neuroelectrical and hemodynamic data we intend to shorten the learning curve of potential users of connectivity analysis software in regard to inferential aspects and the possible caveats and pitfalls the user may encounter. The full day of activity will consist of short morning overview talks, addressing the basic principles, methodological issues, caveat and pitfalls, and examples of application of the software to real data followed by an afternoon devoted to software use and one-to-one discussion with attendees who will have the opportunity to process workshop examples and their own data by bringing their own laptops. This will enable them to benefit from discussing their results and modelling diagnostics with the workshop experts. Attendees must bring Matlab/Octave if they wish to take part in the practical activities. Information about previous workshop editions together with software and further workshop details will be made available at http://www.lcs.poli.usp.br/~baccala/pdc/handson/.
Current and Future Challenges in Robotics Research for Biomedical Applications
Organizers: Sohee Kim*, Cheol Song
In this workshop, diverse cutting-edge research topics in robotics for medical and biomedical applications will be discussed. Medical robotics is one of the fascinating multidisciplinary fields that combine mechanical, electrical and electronics engineering, computer science, and mechatronics. The goal of this workshop is to bring together the researchers from robotics engineering community and from medical community who are interested in the state-of-the-art research trends and advances in robotics technologies for medical uses. Current research efforts toward next-generation medical robotics technologies will be overviewed by international researchers, in two sessions of macro-scale robotics and micro-scale robotics. In macro-scale, surgical robotics such as ultra-high precision surgical systems and robot-assisted surgery, rehabilitation robots and sport training robotic systems will be presented. In micro-scale, bio-integrated and bio-hybrid robotics such as microrobots, wearable and implantable sensors and systems driven by advances in nano- and micro-electro-mechanical system (MEMS) technologies will be presented, which will be directed toward the integration of biological systems and robotic systems. It will be a unique platform to overview diverse robotics research for medical applications in macro-scale as well as in micro-scale.
Assessment of Consciousness and will Powered Motor Rehabilitation
Organizers: Günter Edlinger*, Christoph Guger
The Brain-Machine/Computer Interface (BCI) research area is a vital and fast expanding field. BCIs have been developed during the last years for people with severe disabilities to improve their quality of life. including tools to help patients with disorders of consciousness (DOC) as well as to improve stroke recovery. Imagine being able to think, hear, and feel – but not to move or communicate. More than 40% of patients diagnosed as vegetative are reclassified as (at least) minimally conscious when assessed by expert teams. A further subset of potentially communicative non-responsive patients might be undetectable through standard clinical testing. One part of the workshop will give an overview over groups that aim to use BCI technology to identify non-responsive patients that might be able to communicate and will introduce state-of-the-art technology for advanced consciousness assessment.
Recently BCI applications have been also used for rehabilitation after stroke based on motor imagery. Several articles have shown that MI-based BCIs can induce neural plasticity and thus serve as an important tool to enhance motor rehabilitation for stroke patients. This part of the workshop is discussing the application of BCI technology for improved stroke recovery by activating the sensorimotor cortex. This activation is translated into control signals for rehabilitation devices like Virtual Reality environments showing moving limbs of avatars, robotic devices attached to the patient’s paralyzed limbs such as exoskeletons, or functional electrical stimulation.
The workshop will discuss necessary prerequisites to successfully perform BCI experiments in non-invasive and invasive ways as well as pros and cons of the various methods. Demonstrations of BCI control will allow to understand the progress of the technology. We will invite audience members to participate in live demonstrations, providing real-world examples of modern BCI performance in field settings.
IEEE Standardization Development for Cuffless and Continuous Blood Pressure Monitoring
Organizers: Carole C. Carey*, Xiao-Rong Ding
Accurate measurement of blood pressure (BP) is necessary for proper diagnosis and precise cardiovascular risk assessment, as well as to gauge the necessity for intervention and to monitor treatment effect. Current standards to evaluate the clinical performance of BP monitors target mostly cuff-based devices that provide a snapshot of BP. These standards include those established by the Association for the Advancement of Medical Instrumentation (AAMI), the British Hypertension Society (BHS), and the European Society of Hypertension (ESH). A recent standard, IEEE Std 1708TM-2014, was published for wearable, cuffless BP measuring devices that have different modes of operation (e.g. to measure short-term, long-term, snapshot, continuous, beat-to-beat). It should be noted that the standard is limited to evaluation of devices that do not use a cuff during measurement. However, issues such as the induction of dynamic change remains to be clarified. On the other hand, the performance assessment of systems that provide continuous measurement tends to be inconclusive, and a comparison of the outcome of different studies is difficult. It is therefore desirable to develop standard for continuous BP monitoring system. In the year immediately after the 200th birth anniversary of Carl Ludwig, we propose this workshop to honor Ludwig as the pioneer of continuous blood pressure recording. The primary goal of this workshop is to provide biomedical engineers with an overview of the recent published IEEE 1708 standard with the discussion on further extension and refinement. This workshop will also attempt to introduce standard to evaluate the performance of continuous BP monitoring systems.
Magnetic Resonance Electrical Impedance Tomography and Electric Properties Tomography: New Insights and Frontiers in Measurement of Tissue Electromagnetic Properties
Organizers: Rosalind Sadleir, Atul S. Minhas (USA)
Tissue electrical conductivity imaging is emerging as a popular and challenging application of MRI techniques. One MR based conductivity imaging technique, Magnetic Resonance Electrical Impedance Tomography (MREIT) has been extensively tested in phantoms and animals. However, the need for external imaging currents has hindered its acceptance in clinical diagnosis applications. Researchers working in this field are now proposing several new approaches with direct clinical significance. Neuromodulation technniques such as transcranial DC or AC stimulation (tDCS/tACS) are naturally suited to MREIT imaging, and its use to image these current flows in the brain has led to new approaches to mechanisms and practice. Conductivity can also be imaged without using current via Electric Properties Tomography (EPT) and Conductivity Tensor Imaging (CTI). Functional neuroimaging may also be possible using MREIT approaches. Development and validation of these techniques in high field scanners requires specialized phantoms suitable for ulta-high magnetic fields. The aim of this workshop is to address the challenges involved in tissue electrical conductivity imaging on these multiple exciting frontiers. The technical focus of the workshop includes numerical methods for MREIT, MRI phase imaging, image data analysis and processing techniques, multi-physics modelling and simulation in MRI, and pulse sequence design for MREIT. The audience will learn how to setup an electrical conductivity imaging experiment, pulse sequences required and data processing methods
The Fast-Changing Landscape of Electroencephalography
Organizers: W. G. Besio (USA)
This workshop is designed to give both novice and experienced electroencephalography (EEG) users a synopsis of the latest innovations in EEG and related areas. 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. We will have talks and demos from four groups on their latest technologies for acquiring and processing the EEG which are making the use of EEG more practical. The target audience of the workshop is the whole community of the IEEE EMBS Society interested in brain research.