A Real-time clustering system for spatio-temporal signals from network of neurons

Over past few years, the studies of cultured neuronal networks have opened up avenues for understanding the ion channels, receptor molecules, and synaptic plasticity that may form the basis of learning and memory. The hippocampal neurons from rats are dissociated and cultured on a surface containing a grid of 64 electrodes. The signals from these 64 electrodes are acquired using a fast data acquisition system MED64 (Alpha MED Sciences, Japan) at a sampling rate of 20 K samples with a precision of 16-bits per sample. A few minutes of acquired data runs in to a few hundreds of Mega Bytes. The data processing for the neural analysis is highly compute-intensive because the volume of data is huge. The major processing requirements are noise removal, pattern recovery, pattern matching, clustering and so on. In order to interface a neuronal colony to a physical world, these computations need to be performed in real-time. A single processor such as a desk top computer may not be adequate to meet this computational requirements. Parallel computing is a method used to satisfy the real-time computational requirements of a neuronal system that interacts with an external world while increasing the flexibility and scalability of the application. In this work, we developed a parallel neuronal system using a multi-node Digital Signal processing system. With 8 processors, the system is able to compute and map incoming signals segmented over a period of 200 ms in to an action in a trained cluster system in real time.

3-D GPU Based Real Time Diffuse Optical Tomographic System

3-Dimensional Diffuse Optical Tomographic (3-D DOT) image reconstruction algorithm is computationally complex and requires excessive matrix computations and thus hampers reconstruction in real time. In this paper, we present near real time 3D DOT image reconstruction that is based on Broyden approach for updating Jacobian matrix. The Broyden method simplifies the algorithm by avoiding re-computation of the Jacobian matrix in each iteration. We have developed CPU and heterogeneous CPU/GPU code for 3D DOT image reconstruction in C and MatLab programming platform. We have used Compute Unified Device Architecture (CUDA) programming framework and CUDA linear algebra library (CULA) to utilize the massively parallel computational power of GPUs (NVIDIA Tesla K20c). The computation time achieved for C program based implementation for a CPU/GPU system for 3 planes measurement and FEM mesh size of 19172 tetrahedral elements is 806 milliseconds for an iteration.

An Implementation of Partitioned Scheduling Scheme for Hard Real-Time Tasks in Multicore Linux with Fair Share for Linux Tasks

The correctness of a hard real-time system depends its ability to meet all its deadlines. Existing real-time systems use either a pure real-time scheduler or a real-time scheduler embedded as a real-time scheduling class in the scheduler of an operating system (OS). Existing implementations of schedulers in multicore systems that support real-time and non-real-time tasks, permit the execution of non-real-time tasks in all the cores with priorities lower than those of real-time tasks, but interrupts and softirqs associated with these non-real-time tasks can execute in any core with priorities higher than those of real-time tasks. As a result, the execution overhead of real-time tasks is quite large in these systems, which, in turn, affects their runtime. In order that the hard real-time tasks can be executed in such systems with minimal interference from other Linux tasks, we propose, in this paper, an integrated scheduler architecture, called SchedISA, which aims to considerably reduce the execution overhead of real-time tasks in these systems. In order to test the efficacy of the proposed scheduler, we implemented partitioned earliest deadline first (P-EDF) scheduling algorithm in SchedISA on Linux kernel, version 3.8, and conducted experiments on Intel core i7 processor with eight logical cores. We compared the execution overhead of real-time tasks in the above implementation of SchedISA with that in SCHED_DEADLINE's P-EDF implementation, which concurrently executes real-time and non-real-time tasks in Linux OS in all the cores. The experimental results show that the execution overhead of real-time tasks in the above implementation of SchedISA is considerably less than that in SCHED_DEADLINE. We believe that, with further refinement of SchedISA, the execution overhead of real-time tasks in SchedISA can be reduced to a predictable maximum, making it suitable for scheduling hard real-time tasks without affecting the CPU share of Linux tasks.

Optimización del Lead Time de vehículos mediante la implementación de una solución RTLS (Real Time Location System)

A lo largo de este documento, se va a explicar la implantación del proyecto que he realizado basado en la localización de vehículos en la fábrica de Mercedes Benz España situada en Vitoria-Gasteiz. Durante la realización de este proyecto, se han llevado a cabo diversos estudios con el fin de conseguir la correcta implantación de las tecnologías empleadas. Se han realizado diferentes alternativas de posicionamiento de los componentes y diversas pruebas para comprobar el correcto funcionamiento de la solución. La solución del proyecto se realizará en distintas fases. La primera de ellas tratará sobre el estudio en una determinada zona de la fábrica, más concretamente la denominada “Área Técnica”, en esta zona se encuentran los vehículos que sufren algún retoque una vez están montados, esta zona se utilizará como piloto para una vez finalizado y comprobado su éxito ampliar la solución al resto de zonas. Previamente a mi incorporación se realizó un estudio para la colocación de los elementos necesarios en esta zona y se ha visto las posibilidades y beneficios que aportaría el control de los vehículos dentro de la fábrica. La siguiente fase será implantar la solución en el resto de las áreas que se encuentran dentro de la fábrica de Vitoria-Gasteiz así como la instalación de unos dispositivos que estarán ubicados en las puertas. Estos ayudarán a mejorar la ubicación de los vehículos ya que podremos conocer si los vehículos se encuentran dentro o fuera de la fábrica. Finalmente se ha realizado la integración de la solución en los sistemas actuales que utilizan en la fábrica para la gestión de los vehículos durante su ciclo de vida.

Heterogeneous reconfigurable system for adaptive particle filters in real-time applications

This paper presents a heterogeneous reconfigurable system for real-time applications applying particle filters. The system consists of an FPGA and a multi-threaded CPU. We propose a method to adapt the number of particles dynamically and utilise the run-time reconfigurability of the FPGA for reduced power and energy consumption. An application is developed which involves simultaneous mobile robot localisation and people tracking. It shows that the proposed adaptive particle filter can reduce up to 99% of computation time. Using run-time reconfiguration, we achieve 34% reduction in idle power and save 26-34% of system energy. Our proposed system is up to 7.39 times faster and 3.65 times more energy efficient than the Intel Xeon X5650 CPU with 12 threads, and 1.3 times faster and 2.13 times more energy efficient than an NVIDIA Tesla C2070 GPU. © 2013 Springer-Verlag.

Real-time far distance micro-vibration measurement using an external cavity semiconductor laser interferometer with a feedback control system

An external cavity semiconductor laser interferometer used to measure far distance micro-vibration in real time is proposed. In the interferometer, a single longitudinal mode and excellent coherent characteristic grating external cavity semiconductor laser is constructed and acted as a light source and a phase compensator. Its coherent length exceeds 200 meters. The angle between normal and incidence beam of the far object is allowed to change in definite range during the measurement with this interferometer, and this makes the far distance interference measurement easier and more convenient. Also, it is not required to keep the amplitudes of the first and second harmonic components equal, and then the dynamic range is increased. A feedback control system is used to compensate the phase disturbance between the two interference beams introduced by environmental vibration.

A tripartite microbial reporter gene system for real-time assays of soil nutrient status

Plant-derived carbon is the substrate which drives the rate of microbial assimilation and turnover of nutrients, in particular N and P, within the rhizosphere. To develop a better understanding of rhizosphere dynamics, a tripartite reporter gene system has been developed. We used three lux-marked Pseudomonas fluorescens strains to report on soil (1) assimilable carbon, (2) N-status, and (3) P-status. In vivo studies using soil water, spiked with C, N and P to simulate rhizosphere conditions, showed that the tripartite reporter system can provide real-time assessment of carbon and nutrient status. Good quantitative agreement for bioluminescence output between reference material and soil water samples was found for the C and P reporters. With regard to soil nitrate, the minimum bioavailable concentration was found to be greater than that analytically detectable in soil water. This is the first time that bioavailable soil C, N and P have been quantified using a tripartite reporter gene system.

An Extended Event Reasoning Framework for Decision Support under Uncertainty

To provide in-time reactions to a large volume of surveil- lance data, uncertainty-enabled event reasoning frameworks for CCTV and sensor based intelligent surveillance system have been integrated to model and infer events of interest. However, most of the existing works do not consider decision making under uncertainty which is important for surveillance operators. In this paper, we extend an event reasoning framework for decision support, which enables our framework to predict, rank and alarm threats from multiple heterogeneous sources.

An Adaptive Machine Learning Decision System for Flexible Predictive Maintenance

Process monitoring and Predictive Maintenance (PdM) are gaining increasing attention in most manufacturing environments as a means of reducing maintenance related costs and downtime. This is especially true in industries that are data intensive such as semiconductor manufacturing. In this paper an adaptive PdM based flexible maintenance scheduling decision support system, which pays particular attention to associated opportunity and risk costs, is presented. The proposed system, which employs Machine Learning and regularized regression methods, exploits new information as it becomes available from newly processed components to refine remaining useful life estimates and associated costs and risks. The system has been validated on a real industrial dataset related to an Ion Beam Etching process for semiconductor manufacturing.

Rapid quantification of microRNAs in plasma using a fast real time PCR system

The ability to rapidly detect circulating small RNAs, in particular microRNAs (miRNAs), would further increase their already established potential as biomarkers in a range of conditions. One rate-limiting factor is the time taken to perform quantitative real time PCR amplification. We therefore evaluated the ability of a novel thermal cycler to perform this step in less than 10 minutes. Quantitative PCR was performed on an xxpress® thermal cycler (BJS Biotechnologies, Perivale, UK), which employs a resistive heating system and forced air cooling to achieve thermal ramp rates of 10 °C/s, and a conventional peltier-controlled LightCycler 480 system (Roche, Basel, Switzerland) ramping at 4.8 °C/s. The threshold cycle (Ct) for detection of 18S rDNA from a standard genomic DNA sample was significantly more variable across the block (F-test, p=2.4x10-25) for the xxpress (20.01±0.47SD) than the LightCycler (19.87±0.04SD). RNA was extracted from human plasma, reverse transcribed and a panel of miRNAs amplified and detected using SYBR green (Kapa Biosystems, Wilmington, Ma, USA). The sensitivity of both systems was broadly comparable and both detected a panel of miRNAs reliably and indicated similar relative abundances. The xxpress thermal cycler facilitates rapid qPCR detection of small RNAs and brings point-of care diagnostics based upon circulating miRNAs a step closer to reality.

Rapid detection of health-care-associated bloodstream infection in critical care using multipathogen real-time polymerase chain reaction technology: a diagnostic accuracy study and systematic review

Background: There is growing interest in the potential utility of real-time polymerase chain reaction (PCR) in diagnosing bloodstream infection by detecting pathogen deoxyribonucleic acid (DNA) in blood samples within a few hours. SeptiFast (Roche Diagnostics GmBH, Mannheim, Germany) is a multipathogen probe-based system targeting ribosomal DNA sequences of bacteria and fungi. It detects and identifies the commonest pathogens causing bloodstream infection. As background to this study, we report a systematic review of Phase III diagnostic accuracy studies of SeptiFast, which reveals uncertainty about its likely clinical utility based on widespread evidence of deficiencies in study design and reporting with a high risk of bias. 

Objective: Determine the accuracy of SeptiFast real-time PCR for the detection of health-care-associated bloodstream infection, against standard microbiological culture. 

Design: Prospective multicentre Phase III clinical diagnostic accuracy study using the standards for the reporting of diagnostic accuracy studies criteria. 

Setting: Critical care departments within NHS hospitals in the north-west of England. 

Participants: Adult patients requiring blood culture (BC) when developing new signs of systemic inflammation. 

Main outcome measures: SeptiFast real-time PCR results at species/genus level compared with microbiological culture in association with independent adjudication of infection. Metrics of diagnostic accuracy were derived including sensitivity, specificity, likelihood ratios and predictive values, with their 95% confidence intervals (CIs). Latent class analysis was used to explore the diagnostic performance of culture as a reference standard. 

Results: Of 1006 new patient episodes of systemic inflammation in 853 patients, 922 (92%) met the inclusion criteria and provided sufficient information for analysis. Index test assay failure occurred on 69 (7%) occasions. Adult patients had been exposed to a median of 8 days (interquartile range 4–16 days) of hospital care, had high levels of organ support activities and recent antibiotic exposure. SeptiFast real-time PCR, when compared with culture-proven bloodstream infection at species/genus level, had better specificity (85.8%, 95% CI 83.3% to 88.1%) than sensitivity (50%, 95% CI 39.1% to 60.8%). When compared with pooled diagnostic metrics derived from our systematic review, our clinical study revealed lower test accuracy of SeptiFast real-time PCR, mainly as a result of low diagnostic sensitivity. There was a low prevalence of BC-proven pathogens in these patients (9.2%, 95% CI 7.4% to 11.2%) such that the post-test probabilities of both a positive (26.3%, 95% CI 19.8% to 33.7%) and a negative SeptiFast test (5.6%, 95% CI 4.1% to 7.4%) indicate the potential limitations of this technology in the diagnosis of bloodstream infection. However, latent class analysis indicates that BC has a low sensitivity, questioning its relevance as a reference test in this setting. Using this analysis approach, the sensitivity of the SeptiFast test was low but also appeared significantly better than BC. Blood samples identified as positive by either culture or SeptiFast real-time PCR were associated with a high probability (> 95%) of infection, indicating higher diagnostic rule-in utility than was apparent using conventional analyses of diagnostic accuracy. 

Conclusion: SeptiFast real-time PCR on blood samples may have rapid rule-in utility for the diagnosis of health-care-associated bloodstream infection but the lack of sensitivity is a significant limiting factor. Innovations aimed at improved diagnostic sensitivity of real-time PCR in this setting are urgently required. Future work recommendations include technology developments to improve the efficiency of pathogen DNA extraction and the capacity to detect a much broader range of pathogens and drug resistance genes and the application of new statistical approaches able to more reliably assess test performance in situation where the reference standard (e.g. blood culture in the setting of high antimicrobial use) is prone to error.