Adriano Tavares

Automatic small bowel tumor diagnosis by using multi-scale wavelet-based analysis in wireless capsule endoscopy images

BackgroundWireless capsule endoscopy has been introduced as an innovative, non-invasive diagnostic technique for evaluation of the gastrointestinal tract, reaching places where conventional endoscopy is unable to. However, the output of this technique is an 8 hours video, whose analysis by the expert physician is very time consuming. Thus, a computer assisted diagnosis tool to help the physicians to evaluate CE exams faster and more accurately is an important technical challenge and an excellent economical opportunity.MethodThe set of features proposed in this paper to code textural information is based on statistical modeling of second order textural measures extracted from co-occurrence matrices. To cope with both joint and marginal non-Gaussianity of second order textural measures, higher order moments are used. These statistical moments are taken from the two-dimensional color-scale feature space, where two different scales are considered. Second and higher order moments of textural measures are computed from the co-occurrence matrices computed from images synthesized by the inverse wavelet transform of the wavelet transform containing only the selected scales for the three color channels. The dimensionality of the data is reduced by using Principal Component Analysis.ResultsThe proposed textural features are then used as the input of a classifier based on artificial neural networks. Classification performances of 93.1% specificity and 93.9% sensitivity are achieved on real data. These promising results open the path towards a deeper study regarding the applicability of this algorithm in computer aided diagnosis systems to assist physicians in their clinical practice.

Aspect-oriented fault tolerance for real-time embedded systems

Real-time embedded systems for safety-critical applications have to introduce fault tolerance mechanisms in order to cope with hardware and software errors. Fault tolerance is usually applied by means of redundancy and diversity. Redundant hardware implies the establishment of a distributed system executing a set of fault tolerance strategies by software, and may also employ some form of diversity, by using different variants or versions for the same processing. This paper describes our approach to introduce fault tolerance in distributed embedded systems applications, using aspect-oriented programming (AOP). A real-time operating system sup-porting middleware thread communication was integrated to a fault tolerant framework. The introduction of fault tolerance in the system is performed by AOP at the application thread level. The advantages of this approach include higher modularization, less efforts for legacy systems evolution and better configurability for testing and product line development. This work has been tested and evaluated successfully in several fault tolerant configurations and presented no significant performance or memory footprint costs.

Classification of endoscopic capsule images by using color wavelet features, higher order statistics and radial basis functions

This paper presents a system to support medical diagnosis and detection of abnormal lesions by processing capsule endoscopic images. Endoscopic images possess rich information expressed by texture. Texture information can be efficiently extracted from medium scales of the wavelet transform. The set of features proposed in this paper to code textural information is named color wavelet covariance (CWC). CWC coefficients are based on the covariances of second order textural measures, an optimum subset of them is proposed. Third and forth order moments are added to cope with distributions that tend to become non-Gaussian, especially in some pathological cases. The proposed approach is supported by a classifier based on radial basis functions procedure for the characterization of the image regions along the video frames. The whole methodology has been applied on real data containing 6 full endoscopic exams and reached 95% specificity and 93% sensitivity.

Path loss exponent analysis in Wireless Sensor Networks: Experimental evaluation

Wireless Sensor Networks are an emerging technology which has been recently adopted in many applications. Due to its wireless nature, the analysis of the radio propagation models plays an important role for performance evaluation in both theoretical and practical aspects. In this regards, path loss exponent is one of the most important parameter which has been considered widely in wireless communications analysis. There are several theoretical evaluations of path loss exponent for wireless sensor networks available in the literature. However there is a lack of experimental evaluation of both path loss exponent and the effect of shadowing. In this paper, three environments (free space, in building and industrial), where wireless sensor nodes are widely deployed, have been chosen in order to evaluate the experimental analysis. Path loss and path loss exponent are measured by means of Received Signal Strength Indicator (RSSI) and based on them, the standard deviation of shadowing effect is also calculated. All the measured parameters are compared with the theoretical analysis available in the literatures.

WECO: A wireless platform for monitoring recycling point spots

There is a growing demand for low cost, very low power and reduced size monitoring systems with wireless communications, to be used in different kinds of industrial environments. In several countries waste separation and recycling is a major issue. Consequently, the number of recycling spots has been steadily increasing. In order to ensure that recycle bins are properly maintained, several monitoring solutions have been proposed. These still have several limitations, such as requiring wires for power and/or communications and not being able to fit in all existing types of bins. This paper presents WECO, a wireless embedded solution for monitoring the level of the bins located in recycling spots. The proposed system automatically alerts a remote central station when a bin reaches a programmable filling level, thus avoiding the need to spot check if the bin is full and ensuring that the recycling spot is kept clean. The developed prototype required hardware-software co-design and aimed to meet the above mentioned requirements, resorting to the IEEE 802.15.4 protocol for wireless communications between all nodes in the network, each based on a System-On-Chip (SoC) CC2530 from Texas Instruments. Due to its wireless nature, the architecture requires a battery for power supplying the nodes, with a life time of at least six years. The filling level readings of each bin in a recycling spot is made using an ultrasonic sensor. The data collected by the monitoring platform is then sent to the remote central station that processes it in order to optimize routes and establish a scheduled collection of the recycling spots.

Towards a lightweight embedded virtualization architecture exploiting ARM TrustZone

Virtualization has been used as the de facto technology to allow multiple operating systems (virtual machines) to run on top of the same hardware platform. In the embedded systems domain, virtualization research has focused on the coexistence of real-time requirements with non-real-time characteristics. However, existent standard software-based virtualization solutions have been shown to negatively impact the overall system, especially in performance, memory footprint and determinism. This work in progress paper presents the implementation of an embedded virtualization architecture through commodity hardware. ARM TrustZone technology is exploited to implement a lightweight virtualization solution with low overhead and high determinism, corroborated by promising preliminary results. Research roadmap is also pointed and discussed.

Application-level fault tolerance in real-time embedded systems

Critical real-time embedded systems need to make use of fault tolerance techniques to cope with operation time errors, either in hardware or software. Fault tolerance is usually applied by means of redundancy and diversity. Redundant hardware implies the establishment of a distributed system executing a set of fault tolerance strategies by software, and may also employ some form of diversity, by using different variants or versions for the same processing. This work proposes and evaluates a fault tolerance framework for supporting the development of dependable applications. This framework is build upon basic operating system services and middleware communications and brings flexible and transparent support for application threads. A case study involving radar filtering is described and the framework advantages and drawbacks are discussed.

Task-Aware Interrupt Controller: Priority Space Unification in Real-Time Systems

In the development of real-time systems, predictability is often hindered by technological factors which break the timing abstractions offered by real time operating systems (RTOSs); namely, the priority space separation between threads and interrupts induces the rate-monotonic problem. Software approaches have tackled this issue, attempting to unify the priority space with varying degrees of success. We present a hardware approach to the problem: unifying the priority space at the interrupt handling subsystem, predictability is greatly enhanced with minimum software modifications. Our solution provides the interrupt controller with awareness of the currently running tasks priority making the solution independent of the used operating system. We show how our approach is minimally intrusive at hardware architecture level and provides benefits beyond the capabilities of previous approaches. Our technique shows a 0.05% run-time overhead if no interrupts occur, and run-time reduction proportional to interrupt rate for rates higher than 5 per s, for a interrupt workload around 0.07 ms.

A customizable and ARINC 653 quasi-compliant hypervisor

This paper presents a novel hypervisor, developed for aerospace applications using an object oriented approach that embodies time and space partitioning (TSP) on a PowerPC (PPC) core embedded in a FPGA, for the NetworkCentric core avionics [1] - an architecture of cooperating components and managed by a real-time operating system, to implement dependable computing and targeting simplicity. To support Integrated Modular Architecture (IMA) [2] partitioned software architectures, the proposed hypervisor adapted to the aerospace application domain the Popek and Goldbergs [3] fidelity, efficiency and resource control virtualization requirements, and extends them with additional ones like timing determinism, reactivity and improved dependability. A distinctive feature of this hypervisor is its I/O device virtualization approach that guarantees real-time performance and small trusted computing base. The object oriented approach will be particularly useful to customize key components of the hypervisor (with different granularity levels) such as partition scheduling and the communications manager using generative programming techniques (Aspect Oriented Programming (AOP) and template meta-programming).

A personal computer based controller for an active power filter

This paper presents the development of controllers used in active power filters, applied to improve power quality in electrical systems. These active filters can be able to compensate for the following current or voltage related problems: short blackouts, current or voltage distortion due to harmonics, current or voltage unbalance in three-phase systems, flicker and momentary over or under voltage. There are several causes for current and voltage distortion, namely, non-linear loads, some type of voltage sources and thunderstorms. These problems cause instant and long term effects on the electrical equipment. The short term effects are imperfections, malfunctioning, interferences and degradation of the performance of devices or equipments. Effects in the long run are, basically, overheating and premature aging of the electric devices. The main goal of the work described in this paper is to develop an active power filter controller based on a personal computer and a standard multifunction data acquisition PCI bus card, due to its relative low cost, high processing capability, versatility and the numerous possibilities offered by such a computer-based system. Four different approaches where tested: using C++ on Microsoft Windows with the manufacturers device driver, using a new device driver for Windows, using Linux with real-time application interface; and using LabVIEW for Windows. Experimental and simulation results of the developed controllers are also presented.