Ilaria De Munari

Light on! Real world evaluation of a P300-based brain–computer interface (BCI) for environment control in a smart home

Brain–computer interface (BCI) systems aim to enable interaction with other people and the environment without muscular activation by the exploitation of changes in brain signals due to the execution of cognitive tasks. In this context, the visual P300 potential appears suited to control smart homes through BCI spellers. The aim of this work is to evaluate whether the widely used character-speller is more sustainable than an icon-based one, designed to operate smart home environment or to communicate moods and needs. Nine subjects with neurodegenerative diseases and no BCI experience used both speller types in a real smart home environment. User experience during BCI tasks was evaluated recording concurrent physiological signals. Usability was assessed for each speller type immediately after use. Classification accuracy was lower for the icon-speller, which was also more attention demanding. However, in subjective evaluations, the effect of a real feedback partially counterbalanced the difficulty in BCI use. Practitioner Summary: Since inclusive BCIs require to consider interface sustainability, we evaluated different ergonomic aspects of the interaction of disabled users with a character-speller (goal: word spelling) and an icon-speller (goal: operating a real smart home). We found the first one as more sustainable in terms of accuracy and cognitive effort.

Food Image Recognition Using Very Deep Convolutional Networks

We evaluated the effectiveness in classifying food images of a deep-learning approach based on the specifications of Googles image recognition architecture Inception. The architecture is a deep convolutional neural network (DCNN) having a depth of 54 layers. In this study, we fine-tuned this architecture for classifying food images from three well-known food image datasets: ETH Food-101, UEC FOOD 100, and UEC FOOD 256. On these datasets we achieved, respectively, 88.28%, 81.45%, and 76.17% as top-1 accuracy and 96.88%, 97.27%, and 92.58% as top-5 accuracy. To the best of our knowledge, these results significantly improve the best published results obtained on the same datasets, while requiring less computation power, since the number of parameters and the computational complexity are much smaller than the competitors?. Because of this, even if it is still rather large, the deep network based on this architecture appears to be at least closer to the requirements for mobile systems.

Simple and efficient methods for steady state visual evoked potential detection in BCI embedded system

Brain Computer Interfaces (BCI) can provide severely impaired users with alternative communication paths, by means of interpretation of the users brain activity. Among BCI operating paradigms, SSVEP is largely exploited for its potentially high throughput and reliability. In this paper, two novel SSVEP processing algorithms are presented, focused on calibration-free operation and computational efficiency, targeted for development of BCI embedded modules. A comparison with other popular SSVEP signal processing algorithm (MEC, AMCC, CCA) is also made; results demonstrate the feasibility and effectiveness of the proposed solutions.

A BCI Platform Supporting AAL Applications

Brain Computer Interface (BCI) technology can provide users lacking voluntary muscle control with an augmentative communication channel, based on the interpretation of her/his brain activity. Such technologies, combined with AAL (Ambient Assisted Living) systems, can potentially have a great impact on daily living, extending the scope of the ageing at home paradigm also to individuals affected by severe motor impairments, for whom interacting with the environment is troublesome. In this paper, a low cost BCI development platform is presented; it consists of a customized EEG acquisition unit and a Matlab-based signal processing environment. An application example using SSVEP paradigm is discussed.

A Wireless Sensor Platform for Assistive Technology Applications

In this paper, the development of a prototypal wireless sensor platform is described, aimed at assisting elderly people and people with disabilities in their daily living activities at home. The wireless sensor network is embedded into a more general home control and monitoring network, from which it can borrow remote communication and supervision facilities, enhancing versatility and reliability. A wearable sensor has been developed, capable of smart recognition of abnormal gait and falls. Effective algorithms have been devised, to make the device suitable for low-power hardware implementation. After initial prototyping phases (based on microcontrollers and FPGA) VLSI synthesis has been carried out, to estimate actual silicon area and power consumption. Then, extensions of the approach have been foreseen, accounting for multiple sensor management. An example of a low cost embedded heartbeat monitor is discussed.

An evolutionary approach for standard-cell library reduction

Typically, commercially available standard-cell libraries consist of a large set of items, including cells optimized with respect to speed, area or power consumption. A large number of cells makes the synthesis process and the library maintenance quite demanding. By using a reduced set of properly-selected cells, such efforts can be reduced, without critically affecting performance. This paper introduces an innovative library-reduction strategy, based on a evolutionary algorithm, which allows for selecting an arbitrarily small subset of cells. Library compaction is strictly related to the features of the actual synthesis tool, and is tuned by means of a large set of benchmark circuits, so that it produces results suitable for general-purpose circuit design. Different technologies were accounted for, analyzing dependence of the area, time and power figures on the library cell count. Performance, with respect to full-size library synthesis, do not appreciably degrades, and in several cases actually improves. Synthesis time decreases and library maintenance and characterization tasks can thus be significantly reduced.

Automatic diet monitoring: a review of computer vision and wearable sensor-based methods

Abstract Food intake and eating habits have a significant impact on people’s health. Widespread diseases, such as diabetes and obesity, are directly related to eating habits. Therefore, monitoring diet can be a substantial base for developing methods and services to promote healthy lifestyle and improve personal and national health economy. Studies have demonstrated that manual reporting of food intake is inaccurate and often impractical. Thus, several methods have been proposed to automate the process. This article reviews the most relevant and recent researches on automatic diet monitoring, discussing their strengths and weaknesses. In particular, the article reviews two approaches to this problem, accounting for most of the work in the area. The first approach is based on image analysis and aims at extracting information about food content automatically from food images. The second one relies on wearable sensors and has the detection of eating behaviours as its main goal.

CARDEAGate: Low-cost, ZigBee-based localization and identification for AAL purposes

World population is facing deep demographic changes. A number of societal challenges are to be tackled, and possible support from ICT is sought for. AAL (Ambient Assisted Living) technologies, aimed at fostering independent life of elderly people, are therefore becoming increasingly important. Based on the AAL-system named CARDEA and previously developed at the University of Parma, in this paper a new feature is presented, named CARDEAGate and aimed at providing an inexpensive and scarcely intrusive way for providing user localization and identification information. Such information is needed to implement a number of useful functionalities, and most notably to carry out behavioral analysis in a multi-user context. A simple approach, based on a “wireless barrier” and exploiting ZigBee protocol features is shown to provide reliable monitoring information. Preliminary test results are given, whereas a full characterization is currently under way.

Controlling AAL environments through BCI

A Brain-Computer Interface (BCI) is an alternative/augmentative communication device that can provide users with a different interaction path, based on the interpretation of his/her brain activity. Such technology, applied to Ambient Assisted Living (AAL) contexts, could potentially make the full set of features of such systems accessible to users affected by severe motor impairments, for whom the interaction with the surrounding environment is troublesome. In this paper, a low cost BCI development platform, consisting of a hardware acquisition unit and a Matlab-based prototyping environment is presented. BCI performance assessed by means of an illustrative application example using a 4 class SSVEP paradigm to switch on and off lights. Comparison with other reference methods from literature is also presented.

Thermal stability of AlNi gate AlGaAsGaAs HEMT's

Abstract The study on the instability of gate contacts of Al Ni gate AlGaAs GaAs HEMTs was performed by means of storage tests carried out at three different temperatures: 200°C, 240°C and 300°C. Data from tests as long as 5000 hours were analyzed. At the highest temperature the main failure mode was the reaction between the Al of the gate electrode and the Au of the metallization. At 200°C, 240°C an increase of the barrier height was detected. The activation energy determined and the comparison with the data existing in literature is reported.