Abstract- Some of most compelling application domains of the IoT and Swarm concepts relate to how humans interact with the world around it and the cyberworld beyond. While the proliferation of communication and data processing devices has profoundly altered our interaction patterns, little has been changed in the way we process inputs (sensory) and outputs (actuation). The combination of IoT (Swarms) and wearable devices offers the potential for changing all of this, opening the door for true human augmentation. The epitome of this would be a direct interface to the human brain.
Yet, making sense of the plethora of information received from the often noisy sensors and making reliable decisions within very tight latency bounds (< 10 ms) typically demands huge computational workloads to be performed in wearable form factors at extreme energy efficiency. In this presentation, we will make the case why alternative non-Von Neumann computational paradigms and architectures may be the right choice for these cognitive processing tasks. Even more, we will focus on a computational model called Hyper-Dimensional Computing (HDC), and illustrate with concrete examples of why this approach may be the right one in a post-Moore data-driven arena.
Abstract- Neural interfaces are making it possible to record and stimulate brain activity at high spatiotemporal resolution, at the level of single neurons and across neural networks. In parallel, brain-inspired nano- and microelectronic devices and circuits are being developed that emulate functional properties of biological neurons and networks. Memristors, in particular, are promising candidates to emulate synapses in terms of transmission and signal processing capability. We show first evidence that memristors can be used to compress information of signals from biological neurons as recorded by high-resolution multielectrode arrays and discuss the perspective that these devices will serve as synaptic-like bioelectronic links between biological neurons and artificial counterparts in advanced brain-chip interfaces.
Abstract- Recently, combination of genetic engineering and optical technology enables to measure and control biological functions with light. Fluorescent protein such as GFP can be used as an optical tag of a specific molecule, and photoactive protein such as ChR2 can be applied for optical manipulation of biological functions.
This presentation introduces some kinds of implantable optical devices for measuring and controlling biological functions in the brain of a freely-moving rodent. Future direction is addressed for achieving bidirectional optical communication with brain.
Authors- Georges Gielen, Jorge Marin, Elisa Sacco – KU Leuven Abstract-
Sensors are increasingly used in many emerging application areas such as internet of things and autonomous driving. Integrated sensor interfaces offer the benefit of low cost and low power consumption.
However, besides the nominal performance, also the drift as a function of temperature, supply and other variations is a key limitation to the precision of sensor interfaces.
This paper presents drift mitigation techniques that can be applied to integrated sensor interfaces.
Abstract- Organic bio-electronics represents one of the most exciting directions in printable electronics, promising to deliver new technologies for healthcare and human well‐being. Among the others, organic field-effect transistors have been proven to work as highly performing sensors. Selectivity is achieved by integrating a layer of functional biological recognition elements, directly coupled with an electronic interface. The devices were shown to reach detection limits down to the picomolar (10-12 M) range with highly repeatable responses (within few percentage of standard deviation) even for hundreds of reiterated measurements.
In this lecture recent developments in the field of organic and printable electronics implemented to probe biological interfaces will be discussed highlighting the importance of the interplay among disciplines such as organic electronics, analytical chemistry and biochemistry to reach a comprehensive understanding of the underpinning phenomena. It will also be shown that applications can lead to label-free electronic biosensors with unprecedented detection limits and selectivity. Notably, the extremely good sensing performance level can be rationalized by quantifying electrostatic and capacitance contributions characterizing the surface confined biological recognition elements interacting with their affinity ligands. Examples of the detection of clinical relevant biomarkers will be provided too.
Abstract- Information technology has had profound impacts on our lives. The problem is that, so far, technology has required our explicit attention to provide services. This limits the scenarios in which it can or we would like it to take action. On the other hand, perceptive systems aim to understand our activities and intentions to proactively, collaboratively, and adaptively provide services. This requires systems to form projections of the world, but also construct models for how to respond.
This talk starts by looking at how deploying large numbers of form-fitting sensors, which are explicitly associated with the physical objects we interact with (including each other), can provide contextually-relevant and structured data for enabling the construction of such models. Then, a possible platform technology for creating such sensors is examined, namely Large-Area Electronics (LAE). The challenges of realizing full systems from this are explored. In particular, perceptive systems present demanding functional requirements, but, through emerging algorithms from statistical signal processing and machine learning, also open up new opportunities for addressing technological limitations. Several LAE systems for human monitoring are presented, demonstrating the potentials.
Abstract- Imec has launched a virtual personal health coach research program which combines wearable technology and data sciences for providing new coaching methodologies towards managing an active lifestyle, managing stress, and smoking cessation, with the ambition to have a big impact on preventive health and reduce the future incidence of chronic disease.
Chronic diseases currently account for 70 to 85% of all healthcare costs in US, EU and OECD countries. Nearly half of these chronic diseases are linked to lifestyle and behavior and are therefore in principle preventable. At the same time we are creating a next generation of chronic patients when looking at behavioral statistics of adolescents and adults. A third of all adolescents and half of all adults are not sufficiently active (in terms of aerobic exercise and physical activity). Nearly three quarters does not meet the recommended targets for muscle-strengthening physical activity. We consume too much sodium (which increases our risk of hypertension). Nearly 25% (40%) of adults said they ate vegetables (fruit) less than once a day. Approximately 19% of all adults still smokes cigarettes and e-cigarettes are on the rise. Half of all alcohol-related deaths are due to binge drinking. Unhealthy behavior leads to high blood pressure and high LDL cholesterol which are risk factors for heart disease and stroke. A major effort should therefore be devoted to preventive health and focus on maintaining or restoring a healthy lifestyle. Consumer wearables are already aiming to address our behavior, most notably fitness, and to some extent diet. However, since changing behavior is so difficult, generic smartphone apps providing averaged advice are not well accepted by users and have limited positive outcomes. If wearables were to learn our habits, and were able to capture triggers towards bad behavior, they could provide the right positive advice at the right time. Doing so, they could have the same impact and high-quality result as a personal coach who learns our behavior, what motivates us and gives tailored personalized advice.
Abstract- Deep convolutional neural networks are being regarded today as an extremely effective and flexible approach for extracting actionable, high-level information from the wealth of raw data produced by a wide variety of sensory data sources. CNNs are however computationally demanding: today they typically run on GPU-accelerated compute servers or high-end embedded platforms. Industry and academia are racing to bring CNN inference (first) and training (next) within ever tighter power envelopes, while at the same time meeting real-time requirements. Recent results, including our PULP and ORIGAMI chips, demonstrate there is plenty of room at the bottom: pj/OP (GOPS/mW) computational efficiency, needed for deploying CNNs in the mobile/wearable scenario, is within reach. However, this is not enough: 1000x energy efficiency improvement, within a mW power envelope and with low-cost CMOS processes, is required for deploying CNNs in the most demanding CPS scenarios. The fj/OP milestone will require heterogeneous (3D) integration with ultra-efficient die-to-die communication, mixed-signal pre-processing, event-based approximate computing, while still meeting real-time requirements.
Abstract- Today, CMOS Temperature sensors are predominantly based on parasitic bipolar junction transistors (BJTs). This is because such sensors can achieve high accuracy (< 0.1C error) after a single room-temperature calibration. Although resistor-based temperature sensors can achieve higher resolution and energy-efficiency, they usually require multi-point calibration to reach similar levels of accuracy. In a recent breakthrough, we have discovered that temperature sensors based on silicided poly resistors are an exception to this rule. Two temperature sensor architectures will be presented that use such resistors to achieve good accuracy (<0.2C) after a one or two-point calibration, as well as state-of-the-art energy-efficiency and resolution.
Abstract- The High-Luminosity program of the LHC poses unprecedented challenges for the tracking systems of ATLAS and CMS in terms of radiation tolerance and instantaneous particle rates.
Thanks to use of advanced technologies and novel solutions, the future trackers are expected to meet those requirements, along with improved tracking performance and important additional functionalities.
In parallel with the development of the next trackers, R&D is continuing to prepare the ground for further novelties beyond the HL-LHC era.