Pablo Sergio Mandolesi

Registering eye movements during reading in Alzheimer’s disease: Difficulties in predicting upcoming words

Reading requires the fine integration of attention, ocular movements, word identification, and language comprehension, among other cognitive parameters. Several of the associated cognitive processes such as working memory and semantic memory are known to be impaired by Alzheimer’s disease (AD). This study analyzes eye movement behavior of 18 patients with probable AD and 40 age-matched controls during Spanish sentence reading. Controls focused mainly on word properties and considered syntactic and semantic structures. At the same time, controls’ knowledge and prediction about sentence meaning and grammatical structure are quite evident when we consider some aspects of visual exploration, such as word skipping, and forward saccades. By contrast, in the AD group, the predictability effect of the upcoming word was absent, visual exploration was less focused, fixations were much longer, and outgoing saccade amplitudes were smaller than those in controls. The altered visual exploration and the absence of a contextual predictability effect might be related to impairments in working memory and long-term memory retrieval functions. These eye movement measures demonstrate considerable sensitivity with respect to evaluating cognitive processes in Alzheimer’s disease. They could provide a user-friendly marker of early disease symptoms and of its posterior progression.

Eye Movement Alterations During Reading in Patients With Early Alzheimer Disease

PURPOSE Eye movements follow a reproducible pattern during normal reading. Each eye movement ends up in a fixation point, which allows the brain to process the incoming information and to program the following saccade. Alzheimer disease (AD) produces eye movement abnormalities and disturbances in reading. In this work, we investigated whether eye movement alterations during reading might be already present at very early stages of the disease. METHODS Twenty female and male adult patients with the diagnosis of probable AD and 20 age-matched individuals with no evidence of cognitive decline participated in the study. Participants were seated in front of a 20-inch LCD monitor and single sentences were presented on it. Eye movements were recorded with an eye tracker, with a sampling rate of 1000 Hz and an eye position resolution of 20 arc seconds. RESULTS Analysis of eye movements during reading revealed that patients with early AD decreased the amount of words with only one fixation, increased their total number of first- and second-pass fixations, the amount of saccade regressions and the number of words skipped, compared with healthy individuals (controls). They also reduced the size of outgoing saccades, simultaneously increasing fixation duration. CONCLUSIONS The present study shows that patients with mild AD evidenced marked alterations in eye movement behavior during reading, even at early stages of the disease. Hence, evaluation of eye movement behavior during reading might provide a useful tool for a more precise early diagnosis of AD and for dynamical monitoring of the pathology.

Lack of contextual-word predictability during reading in patients with mild Alzheimer disease

In the present work we analyzed the effect of contextual word predictability on the eye movement behavior of patients with mild Alzheimer disease (AD) compared to age-matched controls, by using the eyetracking technique and lineal mixed models. Twenty AD patients and 40 age-matched controls participated in the study. We first evaluated gaze duration during reading low and highly predictable sentences. AD patients showed an increase in gaze duration, compared to controls, both in sentences of low or high predictability. In controls, highly predictable sentences led to shorter gaze durations; by contrary, AD patients showed similar gaze durations in both types of sentences. Similarly, gaze duration in controls was affected by the cloze predictability of word N and N+1, whereas it was the same in AD patients. In contrast, the effects of word frequency and word length were similar in controls and AD patients. Our results imply that contextual-word predictability, whose processing is proposed to require memory retrieval, facilitated reading behavior in healthy subjects, but this facilitation was lost in early AD patients. This loss might reveal impairments in brain areas such as those corresponding to working memory, memory retrieval, and semantic memory functions that are already present at early stages of AD. In contrast, word frequency and length processing might require less complex mechanisms, which were still retained by AD patients. To the best of our knowledge, this is the first study measuring how patients with early AD process well-defined words embedded in sentences of high and low predictability. Evaluation of the resulting changes in eye movement behavior might provide a useful tool for a more precise early diagnosis of AD.

Hybrid sensor network and fusion algorithm for sound source localization

In this paper, we present a network composed of two-different types of nodes for the task of sound source localization. One of the nodes is a low-power A-scale (dbA) logarithmic sound pressure level sensing (SPLS) unit, and the other is a bearing estimator sensing (BES) unit based on a cross-correlation approach. The sensor boards are mounted on MICADOT nodes, therefore, they are wireless connected. An information fusion algorithm collects the information provided by five SPLS units and one BES unit and provides an estimation of the location of the target. Field results collected from the experiment are shown.

Nonlinear analog controller for a buck converter: theory and experimental results

The design of a switching power supply (buck DC-DC converter) with regulated output voltage for a wide range of current consumption is considered. The control objective is to regulate the output voltage despite the existence of current load variations with bounded time derivative. A variable-structure technique is used to design a continuous time controller. A discrete and an analog version of the same controller are compared. Laboratory experiences are performed to obtain conclusions about the real performance of the proposed controller with satisfactory results.

SCDVP: A Simplicial CNN Digital Visual Processor

In this work we present a programmable and reconfigurable single instruction multiple data (SIMD) visual processor based on the S-CNN architecture, namely, the Simplicial CNN Digital Visual Processor (SCDVP), oriented to high-performance low-level image processing. The cells in the array have a selectable neighborhood configuration and several registers, which provide the chip with extended spatial and temporal processing capabilities, in particular optical flow. A prototype 64 × 64 cell chip with two program memories and a column adder was fabricated in a 90 nm technology, which running at 133 MHz delivers 105.5 GOPS. The calculation at the cell level is performed with time coded signals and the program memory is located outside the array. This produces a very efficient realization in terms of area: 53.8 GOPS per mm2, which outperforms all results reported so far. We show that even after normalization, to account for technology scaling, the proposed architecture is the most efficient among all reported digital processors. Computation performance to power ratio also exceeds all previous results with 817.8 GOPS/W. Experimental results of the working chip are reported.

Joint sampling-time error and channel skew calibration of time-interleaved ADC in multichannel fiber optic receivers

A joint sampling-time error and channel skew background calibration technique for time interleaved analog to digital converters (TI-ADC) is presented. The technique is aimed at applications in dual-polarization QPSK/QAM receivers for coherent optical communications at high data rates (e.g., 40Gb/s and beyond). Unlike previous proposals, the calibration algorithm introduced here is used to jointly compensate for sampling-time and channel skew errors. Estimates of the gradient of the mean squared error (MSE) or the bit error rate (BER) with respect to the sampling phases of the different signal lanes and interleaves are computed and used to iteratively minimize a cost function (i.e., MSE or BER). Computer simulations demonstrate the excellent behavior of the proposed compensation technique. The calibration algorithm can be implemented with minimal hardware requirements and with a slow clock. This allows power dissipation in a CMOS VLSI implementation to be minimized.

Carbon nanotube integration with a CMOS process.

This work shows the integration of a sensor based on carbon nanotubes using CMOS technology. A chip sensor (CS) was designed and manufactured using a 0.30 μm CMOS process, leaving a free window on the passivation layer that allowed the deposition of SWCNTs over the electrodes. We successfully investigated with the CS the effect of humidity and temperature on the electrical transport properties of SWCNTs. The possibility of a large scale integration of SWCNTs with CMOS process opens a new route in the design of more efficient, low cost sensors with high reproducibility in their manufacture.

Field test results for low power bearing estimator sensor nodes

The paper describes experimental results of low power sensor nodes designed to perform bearing estimation. The nodes are intended to form a wireless sensor network able to locate an audio source. Two different nodes are tested: one is based on a cross-correlation derivative integrated circuit (IC), and the other on a gradient flow IC. Implementation details and experimental results of both systems working in a natural environment are presented.

A 6-bit 2GS/s CMOS time-interleaved ADC for analysis of mixed-signal calibration techniques

A 6-bit 2-GS/s time interleaved (TI) successive approximation register (SAR) analog-to-digital converter (ADC) is designed and fabricated in a 0.13 μm CMOS process. The architecture uses 8 time-interleaved track-and-hold amplifiers (THA), and 16 SAR ADCs. The chip includes (i) a programmable delay cell array to adjust the interleaved sampling phase, and (ii) a 12 Gbps low voltage differential signaling (LVDS) interface. These blocks make the fabricated ADC an excellent platform to evaluate mixed-signal calibration techniques, which are of great interest for application in high-speed optical systems. Measurements of the fabricated ADC show 33.9 dB of peak signal-to-noise-and-distortion ratio (SNDR) and 192 mW of power consumption at 1.2 V.