Wireless electromagnetic powering of implantable medical devices is a diverse research area, with goals including replacing percutaneous wires, miniaturizing and extending the lifetime of implanted devices, enabling wireless communication, and biosensing, all while maximizing safety and efficiency of wireless power transfer. Many challenges in wireless transcutaneous powering are associated with tissue as an electromagnetic transmission medium. Tissue is lossy and variable, and safety is a concern due to absorption of electromagnetic energy in high-water-content tissue. The purpose of this overview is to summarize reported variability of tissue properties, particularly in the context of electromagnetic safety, with a focus on models of tissue that can represent variability in the design and evaluation of systems for wireless transcutaneous power transfer.

The design of effective transcutaneous systems demands the consideration of inevitable variations in tissue characteristics, which vary across body areas, among individuals, and over time. The purpose of this paper was to design and evaluate several printed antenna topologies for ultrahigh frequency (UHF) transcutaneous power transfer to implantable medical devices, and to investigate the effects of variations in tissue properties on dipole and loop topologies. Here, we show that a loop antenna topology provides the greatest achievable gain with the smallest implanted antenna, while a dipole system provides higher impedance for conjugate matching and the ability to increase gain with a larger external antenna. In comparison to the dipole system, the loop system exhibits greater sensitivity to changes in tissue structure and properties in terms of power gain, but provides higher gain when the separation is on the order of the smaller antenna dimension. The dipole system was shown to provide higher gain than the loop system at greater implant depths for the same implanted antenna area, and was less sensitive to variations in tissue properties and structure in terms of power gain at all investigated implant depths. The results show the potential of easily-fabricated, low-cost printed antenna topologies for UHF transcutaneous power, and the importance of environmental considerations in choosing the antenna topology.

The impressive evolution of neural networks and deep learning techniques during the last few years has offered new incomparable routes to solve many complex problems. Moreover, the fact that neural networks are structured and supervised has made it possible to perform automatic parameter tuning that guarantees convergence to the best expressive model for the problem assessed. In this work, we investigated the use of recurrent neural networks (RNNs) to solve the sequential sparse recovery problem through unfolding the iterative soft thresholding algorithm (ISTA) into a stacked RNN. Specifically, we examined the performance of the unsupervised iterative algorithm and the supervised network for a purely compressive sampling reconstruction problem of time-frequency representations. Our results demonstrated that the trained stacked neural network outperforms the iterative algorithm in the quality of the reconstructed data and points to several future directions to improve the performance.

In this paper we considered windows used for local vertex spectrum analysis of graph signals. In addition to a review of the convolution based windowing method, two methods based on the vertex neighborhood are presented. They are based on the graph path lengths. In the first one the number of edges in a path determine window size, while the edge weights are taken into account in the second method. Signal localization is performed by using these window functions. Windowing methods are used for signal local vertex spectrum calculation with a test signal. Norm one based concentration measure is used for comparison.

Deep learning, a relatively new branch of machine learning, has been investigated for use in a variety of biomedical applications. Deep learning algorithms have been used to analyze different physiological signals and gain a better understanding of human physiology for automated diagnosis of abnormal conditions. In this paper, we provide an overview of deep learning approaches with a focus on deep belief networks in electroencephalography applications. We investigate the state-of-the-art algorithms for deep belief networks and then cover the application of these algorithms and their performances in electroencephalographic applications. We covered various applications of electroencephalography in medicine, including emotion recognition, sleep stage classification, and seizure detection, in order to understand how deep learning algorithms could be modified to better suit the tasks desired. This review is intended to provide researchers with a broad overview of the currently existing deep belief network methodology for electroencephalography signals, as well as to highlight potential challenges for future research.

Currently, brain and social networks are examples of new data types that are massively acquired and disseminated [1]. These networks typically consist of vertices (nodes) and edges (connections between nodes). Usually, information is conveyed through the strength of connection among nodes, but in recent years, it has been discovered that valuable information may also be conveyed in signals that occur on each vertex. However, traditional signal processing often does not offer reliable tools and algorithms to analyze such new data types. This is especially true for cases where networks (e.g., the strength of connections), or signals on vertices, have properties that change over the network.

Importance Gait performance is affected by neurodegeneration in aging and has the potential to be used as a clinical marker for progression from mild cognitive impairment (MCI) to dementia. A dual-task gait test evaluating the cognitive-motor interface may predict dementia progression in older adults with MCI. Objective To determine whether a dual-task gait test is associated with incident dementia in MCI. Design, Setting, and Participants The Gait and Brain Study is an ongoing prospective cohort study of community-dwelling older adults that enrolled 112 older adults with MCI. Participants were followed up for 6 years, with biannual visits including neurologic, cognitive, and gait assessments. Data were collected from July 2007 to March 2016. Main Outcomes and Measures Incident all-cause dementia was the main outcome measure, and single- and dual-task gait velocity and dual-task gait costs were the independent variables. A neuropsychological test battery was used to assess cognition. Gait velocity was recorded under single-task and 3 separate dual-task conditions using an electronic walkway. Dual-task gait cost was defined as the percentage change between single- and dual-task gait velocities: ([single-task gait velocity – dual-task gait velocity]/ single-task gait velocity) × 100. Cox proportional hazard models were used to estimate the association between risk of progression to dementia and the independent variables, adjusted for age, sex, education, comorbidities, and cognition. Results Among 112 study participants with MCI, mean (SD) age was 76.6 (6.9) years, 55 were women (49.1%), and 27 progressed to dementia (24.1%), with an incidence rate of 121 per 1000 person-years. Slow single-task gait velocity (<0.8 m/second) was not associated with progression to dementia (hazard ratio [HR], 3.41; 95% CI, 0.99-11.71; P = .05) while high dual-task gait cost while counting backward (HR, 3.79; 95% CI, 1.57-9.15; P = .003) and naming animals (HR, 2.41; 95% CI, 1.04-5.59; P = .04) were associated with dementia progression (incidence rate, 155 per 1000 person-years). The models remained robust after adjusting by baseline cognition except for dual-task gait cost when dichotomized. Conclusions and Relevance Dual-task gait is associated with progression to dementia in patients with MCI. Dual-task gait testing is easy to administer and may be used by clinicians to decide further biomarker testing, preventive strategies, and follow-up planning in patients with MCI. Trial Registration clinicaltrials.gov: NCT03020381.

A biomarker is defined as a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. In this talk, I will present my efforts to develop computational biomarkers that can characterize temporal and spatial signatures (i.e., the unique patterns of moment-to-moment changes of physiologic variables under normal or pathologic conditions) and their relationship to other variables. Specifically, I will elaborate my efforts to develop computational biomarkers for detecting swallowing difficulties, gait changes and handwriting changes. These computational biomarkers are obtained by mining large data sets in order to characterize changes in the considered functional outcomes under various conditions.

A method for a resistive circuit analysis based on graph spectral decompositions is proposed. It is shown that the Laplacian matrix can be used in order to calculate node potentials. Based on the Laplacian eigenvalues and eigenvectors it is possible to decompose a complex resistive circuit into smaller, weakly connected sub-circuits.