Estudi GPGPU dedicat al processament de neuroimatges

La tecnologia GPGPU permet paral∙lelitzar càlculs executant operacions aritmètiques en els múltiples processadors de que disposen els xips gràfics. S'ha fet servir l'entorn de desenvolupament CUDA de la companyia NVIDIA, que actualment és la solució GPGPU més avançada del mercat. L'algorisme de neuroimatge implementat pertany a un estudi VBM desenvolupat amb l'eina SPM. Es tracta concretament del procés de segmentació d'imatges de ressonància magnètica cerebrals, en els diferents teixits dels quals es composa el cervell: matèria blanca, matèria grisa i líquid cefaloraquidi. S'han implementat models en els llenguatges Matlab, C i CUDA, i s'ha fet un estudi comparatiu per plataformes hardware diferents.

Development of a tool for the construction of "ground truth" for complex color images with text

Aquesta memoria resumeix el treball de final de carrera d’Enginyeria Superior d’Informàtica. Explicarà les principals raons que han motivat el projecte així com exemples que il·lustren l’aplicació resultant. En aquest cas el software intentarà resoldre la actual necessitat que hi ha de tenir dades de Ground Truth per als algoritmes de segmentació de text per imatges de color complexes. Tots els procesos seran explicats en els diferents capítols partint de la definició del problema, la planificació, els requeriments i el disseny fins a completar la il·lustració dels resultats del programa i les dades de Ground Truth resultants.

The Part-time pay penalty in a segmented labor market

This paper is the first to examine the implications of switching to PT work for women's subsequent earnings trajectories, distinguishing by their type of contract: permanent or fixedterm. Using a rich longitudinal Spanish data set from Social Security records of over 76,000 prime-aged women strongly attached to the Spanish labor market, we find that PT work aggravates the segmentation of the labor market insofar there is a PT pay penalty and this penalty is larger and more persistent in the case of women with fixed-term contracts. The paper discusses problems arising in empirical estimation (including a problem not discussed in the literature up to now: the differential measurement error of the LHS variable by PT status), and how to address them. It concludes with policy implications relevant for Continental Europe and its dual structure of employment protection.

Integración de políticas medio ambientales: Transformaciones institucionales en España

La Integración de la Políticas Ambientales (IPA) consiste en la incorporación de los objetivos ambientales en todos los estadios de elaboración e implementación de las políticas no ambientales (o sectoriales), con un reconocimiento específico de dicho objetivo como principio director de la programación y la puesta en práctica. Se complementa con la previsión de las consecuencias ambientales mediante una evaluación de conjunto de la política con objeto de minimizar las contradicciones entre las políticas ambientales y no ambientales dando prioridad a las primeras. La naturaleza transversal de la IPA no se ajusta a la forma tradicional de gobernanza jerárquica basada en la autoridad del Estado, la diferenciación sectorial, y los instrumentos típicos de dirección y control. Por ello, no es extraño que la IPA se encuentre a menudo asociada a los llamados "nuevos modos de gobernanza". Esta tesis propone analizar empíricamente el estado de la cuestión en España con respecto a la Integración de la política ambiental , tanto a nivel estatal como autonómico (Cataluña, País Vasco), yendo más allá de los compromisos políticos a favor de un mayor grado de integración de los factores ambientales en las demás políticas. En particular, los ámbitos de investigación seleccionados para esta finalidad son el de las políticas de mitigación de las causas y los efectos del cambio climático, y las políticas de desarrollo sostenible. En términos de ‘variable independiente’ La hipótesis que se quiere comprobar es que la acción llevada a cabo por la Unión Europea ha generado profundos cambios en las estructuras de las instituciones mencionadas y en los procesos de elaboración y implementación de determinadas políticas (energía y transporte) así como en las relaciones entre sectores de la administración pública y entre el sector público y el privado.

Intraobserver and interobserver variability of ascending aorta diameter as assessed with ECG-gated MDCT: automatic versus manual measurements

Purpose: Recently morphometric measurements of the ascending aorta have been done with ECG-gated MDCT to help the development of future endovascular therapies (TCT) [1]. However, the variability of these measurements remains unknown. It will be interesting to know the impact of CAD (computer aided diagnosis) with automated segmentation of the vessel and automatic measurements of diameter on the management of ascending aorta aneurysms. Methods and Materials: Thirty patients referred for ECG-gated CT thoracic angiography (64-row CT scanner) were evaluated. Measurements of the maximum and minimum ascending aorta diameters were obtained automatically with a commercially available CAD and semi-manually by two observers separately. The CAD algorithms segment the iv-enhanced lumen of the ascending aorta into perpendicular planes along the centreline. The CAD then determines the largest and the smallest diameters. Both observers repeated the automatic measurements and the semimanual measurements during a different session at least one month after the first measurements. The Bland and Altman method was used to study the inter/intraobserver variability. A Wilcoxon signed-rank test was also used to analyse differences between observers. Results: Interobserver variability for semi-manual measurements between the first and second observers was between 1.2 to 1.0 mm for maximal and minimal diameter, respectively. Intraobserver variability of each observer ranged from 0.8 to 1.2 mm, the lowest variability being produced by the more experienced observer. CAD variability could be as low as 0.3 mm, showing that it can perform better than human observers. However, when used in nonoptimal conditions (streak artefacts from contrast in the superior vena cava or weak lumen enhancement), CAD has a variability that can be as high as 0.9 mm, reaching variability of semi-manual measurements. Furthermore, there were significant differences between both observers for maximal and minimal diameter measurements (p<0.001). There was also a significant difference between the first observer and CAD for maximal diameter measurements with the former underestimating the diameter compared to the latter (p<0.001). As for minimal diameters, they were higher when measured by the second observer than when measured by CAD (p<0.001). Neither the difference of mean minimal diameter between the first observer and CAD nor the difference of mean maximal diameter between the second observer and CAD was significant (p=0.20 and 0.06, respectively). Conclusion: CAD algorithms can lessen the variability of diameter measurements in the follow-up of ascending aorta aneurysms. Nevertheless, in non-optimal conditions, it may be necessary to correct manually the measurements. Improvements of the algorithms will help to avoid such a situation.

The Body Action and Posture Coding System (BAP): Development and Reliability

Several methods are available for coding body movement in nonverbal behavior research, but there is no consensus on a reliable coding system that can be used for the study of emotion expression. Adopting an integrative approach, we developed a new method, the Body Action and Posture (BAP) coding system, for the time-aligned micro description of body movement on an anatomical level (different articulations of body parts), a form level (direction and orientation of movement), and a functional level (communicative and self-regulatory functions). We applied the system to a new corpus of acted emotion portrayals, examined its comprehensiveness and demonstrated intercoder reliability at three levels: a) occurrence, b) temporal precision and c) segmentation. We discuss issues for further validation and propose some research applications.

Récit(s) en fragments. Logiques sérielle et mondaine dans Little Nemo in Slumberland de Winsor McCay

Cette étude examine dans une perspective narratologique le fonctionnement de la sérialité dans les planches de la bande dessinée Little Nemo in Slumberland, qui sont tiraillées entre l'autonomisation de l'épisode et l'inscription dans un récit plus vaste. Les implications du rythme hebdomadaire de parution y sont envisagées du point de vue des niveaux de temporalité ainsi que des principes de segmentation en épisodes et de liage entre ceux-ci. Des seuils sont repérés dans cette série dont le traitement varie sur le plan de la narrativité et des relations instaurées entre le texte des récitatifs et les images. La diversité de ces régimes fait la richesse de cette oeuvre, et montre comment McCay a su exploiter les possibilités de son moyen d'expression, notamment en privéligiant pour ses représentations oniriques et spectaculaires une logique mondaine qui, comme on l'explique ici, n'est pas sans parenté avec les univers actuels de jeux vidéo.

Sobreeducación y Sobrecualificación en los Universitarios Catalanes. Una perspectiva de género

Partiendo de una muestra compuesta por más de 19.000 titulados universitarios encuestados en 2008 y 2011, 3 años y medio después de haber finalizado sus estudios, obtenemos que la incidencia de la sobreeducación femenina está por debajo de la masculina. Con la sobrecualificación los resultados son similares. Las estimaciones de las ecuaciones salariales indican que el desajuste educativo y competencial generan una importante reducción de los ingresos para ambos sexos. La penalización a la sobreeducación es mayor para las tituladas. En cambio, se observa que la pérdida de ingresos causada por la sobrecualificación en las competencias de gestión es mayor para los hombres. El efecto de la sobrecualificación en sus niveles más elevados llega a equilibrar el impacto más favorable de la sobreeducación masculina. Los resultados obtenidos no se pueden explicar a través de la segmentación del mercado laboral de los graduados, ni tampoco a partir de la preferencia femenina por condiciones laborales a las cuales otorgan atributos positivos. En cambio, no es posible rechazar la existencia de discriminación hacia las egresadas. Tampoco es posible descartar que las asimetrías en el impacto de los desajustes estén causadas por una concepción distinta del fenómeno de la sobreeducación por parte de hombres y mujeres.

Sound Texture Modelling

In the PhD thesis “Sound Texture Modeling” we deal with statistical modelling or textural sounds like water, wind, rain, etc. For synthesis and classification. Our initial model is based on a wavelet tree signal decomposition and the modeling of the resulting sequence by means of a parametric probabilistic model, that can be situated within the family of models trainable via expectation maximization (hidden Markov tree model ). Our model is able to capture key characteristics of the source textures (water, rain, fire, applause, crowd chatter ), and faithfully reproduces some of the sound classes. In terms of a more general taxonomy of natural events proposed by Graver, we worked on models for natural event classification and segmentation. While the event labels comprise physical interactions between materials that do not have textural propierties in their enterity, those segmentation models can help in identifying textural portions of an audio recording useful for analysis and resynthesis. Following our work on concatenative synthesis of musical instruments, we have developed a pattern-based synthesis system, that allows to sonically explore a database of units by means of their representation in a perceptual feature space. Concatenative syntyhesis with “molecules” built from sparse atomic representations also allows capture low-level correlations in perceptual audio features, while facilitating the manipulation of textural sounds based on their physical and perceptual properties. We have approached the problem of sound texture modelling for synthesis from different directions, namely a low-level signal-theoretic point of view through a wavelet transform, and a more high-level point of view driven by perceptual audio features in the concatenative synthesis setting. The developed framework provides unified approach to the high-quality resynthesis of natural texture sounds. Our research is embedded within the Metaverse 1 European project (2008-2011), where our models are contributting as low level building blocks within a semi-automated soundscape generation system.

Segmentació d'imatge de ressonància magnètica de pit utilitzant atlas probabilístic

La localització d'òrgans és un tòpic important en l'àmbit de la imatge mèdica per l'ajuda del tractament i diagnosi del càncer. Un exemple es pot trobar en la cal•libració de models farmacoquinètics. Aquesta pot ésser realitzada utilitzant un teixit de referència, on, per exemple en imatges de ressonància magnètica de pit, una correcta segmentació del múscul pectoral és necessària per a la detecció de signes de malignitat. Els mètodes de segmentació basat en atlas han estat altament avaluats en imatge de ressonància magnètica de cervell, obtenint resultats satisfactoris. En aquest projecte, en col•laboració amb el el Diagnostic Image Analysis Group de la Radboud University Nijmegen Medical Centre i la supervisió del Dr. N.Karssemeijer, es presenta la primera aproximació d'un mètode de segmentació basat en atlas per segmentar els diferents teixits visibles en imatges de ressonància magnètica (T1) del pit femení. L'atlas consisteix en 5 estructures (teixit greixòs, teixit dens, cor, pulmons i múscul pectoral) i ha estat utilitzat en un algorisme de segmentació Bayesià per tal de delinear les esmentades estructures. A més a més, s'ha dut a terme una comparació entre un mètode de registre global i un de local, utilitzats tant en la construcció de l'atlas com en la fase de segmentació, essent el primer el que ha presentat millors resultats en termes d'eficiència i precisió. Per a l'avaluació, s'ha dut a terme una comparació visual i numèrica entre les segmentacions obtingudes i les realitzades manualment pels experts col•laboradors. Pel que fa a la numèrica, s'ha emprat el coeficient de similitud de Dice ( mesura que dóna valors entre 0 i 1, on 0 significa no similitud i 1 similitud màxima) i s'ha obtingut una mitjana general de 0.8. Aquest resultat confirma la validesa del mètode presentat per a la segmentació d'imatges de ressonància magnètica del pit.

Early white matter alterations in men predisposed to FXTAS revealed by MT imaging.

Introduction: The Fragile X - associated Tremor Ataxia Syndrome (FXTAS) is a recently described, and under-diagnosed, late onset (≈ 60y) neurodegenerative disorder affecting male carriers of a premutation in the Fragile X Mental Retardation 1 (FMR1) gene. The premutation is an CGG (Cytosine-Guanine-Guanine) expansion (55 to 200 CGG repeats) in the proximal region of the FMR1 gene. Patients with FXTAS primarily present with cerebellar ataxia and intention tremor. Neuroradiological features of FXTAS include prominent white matter disease in the periventricular, subcortical, middle cerebellar peduncles and deep white matter of the cerebellum on T2-weighted or FLAIR MR imaging (Jacquemmont 2007, Loesch 2007, Brunberg 2002, Cohen 2006). We hypothesize that a significant white matter alteration is present in younger individuals many years prior to clinical symptoms and/or the presence of visible lesions on conventional MR sequences and might be detectable by magnetization transfer (MT) imaging. Methods: Eleven asymptomatic premutation carriers (mean age = 55 years) and seven intra-familial controls participated to the study. A standardized neurological examination was performed on all participants and a neuropsychological evaluation was carried out before MR scanning performed on a 3T Siemens Trio. The protocol included a sagittal T1-weighted 3D gradient-echo sequence (MPRAGE, 160 slices, 1 mm^3 isotropic voxels) and a gradient-echo MTI (FA 30, TE 15, matrix size 256*256, pixel size 1*1 mm, 36 slices (thickness 2mm), MT pulse duration 7.68 ms, FA 500, frequency offset 1.5 kHz). MTI was performed by acquiring consecutively two set of images; first with and then without the MT saturation pulse. MT images were coregistered to the T1 acquisition. The MTR for every intracranial voxel was calculated as follows: MTR = (M0 - MS)/M0*100%, creating a MTR map for each subject. As first analysis, the whole white matter (WM) was used to mask the MTR image in order to create an histogram of the MTR distribution in the whole tissue class over the two groups examined. Then, for each subject, we performed a segmentation and parcellation of the brain by means of Freesurfer software, starting from the high resolution T1-weighted anatomical acquisition. Cortical parcellations was used to assign a label to the underlying white matter by the construction of a Voronoi diagram in the WM voxels of the MR volume based on distance to the nearest cortical parcellation label. This procedure allowed us to subdivide the cerebral WM in 78 ROIs according to the cortical parcellation (see example in Fig 1). The cerebellum, by the same procedure, was subdivided in 5 ROIs (2 per each hemisphere and one corresponding to the brainstem). For each subject, we calculated the mean value of MTR within each ROI and averaged over controls and patients. Significant differences between the two groups were tested using a two sample T-test (p<0.01). Results: Neurological examination showed that no patient met the clinical criteria of Fragile X Tremor and Ataxia Syndrome yet. Nonetheless, premutation carriers showed some subtle neurological signs of the disorder. In fact, premutation carriers showed a significant increase of tremor (CRST, T-test p=0.007) and increase of ataxia (ICARS, p=0.004) when compared to controls. The neuropsychological evaluation was normal in both groups. To obtain general characterizations of myelination for each subject and premutation carriers, we first computed the distribution of MTR values across the total white matter volume and averaged for each group. We tested the equality of the two distributions with the non parametric Kolmogorov-Smirnov test and we rejected the null-hypothesis at a p=0.03 (fig. 2). As expected, when comparing the asymptomatic permutation carriers with control subjects, the peak value and peak position of the MTR values within the whole WM were decreased and the width of the distribution curve was increased (p<0.01). These three changes point to an alteration of the global myelin status of the premutation carriers. Subsequently, to analyze the regional myelination and white matter integrity of the same group, we performed a ROI analysis of MTR data. The ROI-based analysis showed a decrease of mean MTR value in premutation carriers compared to controls in bilateral orbito-frontal and inferior frontal WM, entorhinal and cingulum regions and cerebellum (Fig 3). The detection of these differences in these regions failed with other conventional MR techniques. Conclusions: These preliminary data confirm that in premutation carriers, there are indeed alterations in "normal appearing white matter" (NAWM) and these alterations are visible with the MT technique. These results indicate that MT imaging may be a relevant approach to detect both global and local alterations within NAWM in "asymptomatic" carriers of premutations in the Fragile X Mental Retardation 1 (FMR1) gene. The sensitivity of MT in the detection of these alterations might point towards a specific physiopathological mechanism linked to an underlying myelin disorder. ROI-based analyses show that the frontal, parahippocampal and cerebellar regions are already significantly affected before the onset of symptoms. A larger sample will allow us to determine the minimum CGG expansion and age associated with these subclinical white matter alterations.

Connectome alterations in schizophrenia

Introduction : DTI has proven to be an exquisite biomarker of tissue microstructure integrity. This technique has been successfully applied to schizophrenia in showing that fractional anisotropy (FA, a marker of white matter integrity) is diminished in several areas of the brain (Kyriakopoulos M et al (2008)). New ways of representing diffusion data emerged recently and achieved to create structural connectivity maps in healthy brains (Hagmann P et al. (2008)). These maps have the capacity to study alterations over the entire brain at the connection and network level. This is of high interest in complex disconnection diseases like schizophrenia. We report on the specific network alterations of schizophrenic patients. Methods : 13 patients with chronic schizophrenia were recruited from in-patient, day treatment, out-patient clinics. Comparison subjects were recruited and group-matched to patients on age, sex, handedness, and parental social economic-status. This study was approved by the local IRB and subjects had to give informed written consent. They were scanned with a 3T clinical MRI scanner. DTI and high-resolution anatomical T1w imaging were performed during the same session. The path from diffusion MRI to a multi-resolution structural connection matrices of the entire brain is a five steps process that was performed in a similar way as described in Hagmann P et al. (2008). (1) DTI and T1w MRI of the brain, (2) segmentation of white and gray matter, (3) white matter tractography, (4) segmentation of the cortex into 242 ROIs of equal surface area covering the entire cortex (Fig 1), (5) the connection network was constructed by measuring for each ROI to ROI connection the related average FA along the corresponding tract. Results : For every connection between 2 ROIs of the network we tested the hypothesis H0: "average FA along fiber pathway is larger or equal in patients than in controls". H0 was rejected for connections where average FA in a connection was significantly lower in patients than in controls. Threshold p-value was 0.01 corrected for multiple comparisons with false discovery rate. We identified consistently that temporal, occipito-temporal, precuneo-temporal as well as frontal inferior and precuneo-cingulate connections were altered (Fig 2: significant connections in yellow). This is in agreement with the known literature, which showed across several studies that FA is diminished in several areas of the brain. More precisely, abnormalities were reported in the prefrontal and temporal white matter and to some extent also in the parietal and occipital regions. The alterations reported in the literature specifically included the corpus callosum, the arcuate fasciculus and the cingulum bundle, which was the case here as well. In addition small world indexes are significantly reduced in patients (p<0.01) (Fig. 3). Conclusions : Using connectome mapping to characterize differences in structural connectivity between healthy and diseased subjects we were able to show widespread connectional alterations in schizophrenia patients and systematic small worldness decrease, which is a marker of network desorganization. More generally, we described a method that has the capacity to sensitively identify structure alterations in complex disconnection syndromes where lesions are widespread throughout the connectional network.

Segmentación interactiva multi-clase de personas

La segmentació de persones es molt difícil a causa de la variabilitat de les diferents condicions, com la postura que aquestes adoptin, color del fons, etc. Per realitzar aquesta segmentació existeixen diferents tècniques, que a partir d'una imatge ens retornen un etiquetat indicant els diferents objectes presents a la imatge. El propòsit d'aquest projecte és realitzar una comparativa de les tècniques recents que permeten fer segmentació multietiqueta i que son semiautomàtiques, en termes de segmentació de persones. A partir d'un etiquetatge inicial idèntic per a tots els mètodes utilitzats, s'ha realitzat una anàlisi d'aquests, avaluant els seus resultats sobre unes dades publiques, analitzant 2 punts: el nivell de interacció i l'eficiència.

Ascending aorta measurements as assessed by ECG-gated multi-detector computed tomography : a pilot study to etablish normative values for transcatheters therapies

Rapport de synthèse : Mesures de l'aorte ascendante par scanner synchronisé au rythme cardiaque: une étude pilote pour établir des valeurs normatives dans le cadre des futures thérapies par transcathéter. Objectif : L'objectif de cette étude est d'établir les valeurs morphométriques normatives de l'aorte ascendante à l'aide de l'angiographie par scanner synchronisé au rythme cardiaque, afin d'aider au développement des futurs traitements par transcathéter. Matériels et méthodes : Chez soixante-dix-sept patients (âgé de 22 à 83 ans, âge moyen: 54,7 ans), une angiographie par scanner synchronisé au rythme cardiaque a été réalisée pour évaluation des vaisseaux coronaires. Les examens ont été revus afin d'étudier l'anatomie de la chambre de chasse du ventricule gauche jusqu'au tronc brachio-céphalique droit. A l'aide de programmes de reconstructions multiplanaires et de segmentation automatique, différents diamètres et longueurs considérés comme importants pour les futurs traitements par transcathéter ont été mesurés. Les valeurs sont exprimées en moyennes, médianes, maximums, minimums, écart-types et en coefficients de variation. Les variations de diamètre de l'aorte ascendante durant le cycle cardiaque ont été aussi considérées. Résultats : Le diamètre moyen de la chambre de chasse du ventricule gauche était de 20.3+/-3.4 mm. Au niveau du sinus coronaire de l'aorte, il était de 34.2+/-4.1 mm et au niveau de la jonction sinotubulaire il était de 29.7+/-3.4 mm. Le diamètre moyen de l'aorte ascendante était de 32.7+/-3.8 mm. Le coefficient de variation de ces mesures variait de 12 à 17%. La distance moyenne entre l'insertion proximale des valvules aortiques et le départ du tronc brachio-céphalique droit était de 92.6+/-11.8 mm. La distance moyenne entre l'insertion proximale des valvules aortiques et l'origine de l'artère coronaire proximale était de 12.1+/-3.7 mm avec un coefficient de variation de 31%. La distance moyenne entre les deux ostia coronaires était de 7.2+/-3.1 mm avec un coefficient de variation de 43%. La longueur moyenne du petit arc de l'aorte ascendante entre l'artère coronaire gauche et le tronc brachio-céphalique droit était de 52.9+/-9.5 mm. La longueur moyenne de la continuité fibreuse entre la valve aortique et la valvule mitrale antérieure était de 14.6+/-3.3 mm avec un coefficient de variation de 23%. L'aire moyenne de la valve aortique était de 582.0+/-131.9 mm2. La variation du diamètre antéro-postérieur et transverse de l'aorte ascendante était respectivement de 8.4% et de 7.3%. Conclusion Il existe d'importantes variations inter-individuelles dans les mesures de l'aorte ascendante avec cependant des variations intra-individuelles faibles durant le cycle cardiaque. De ce fait, une approche personnalisée pour chaque patient est recommandée dans la confection des futures endoprothèses de l'aorte ascendante. Le scanner synchronisé au rythme cardiaque jouera un rôle prépondérant dans le bilan préthérapeutique. Abstract : The aim of this study was to provide an insight into normative values of the ascending aorta in regards to novel endovascular procedures using ECG-gated multi-detector CT angiography. Seventy-seven adult patients without ascending aortic abnormalities were evaluated. Measurements at relevant levels of the aortic root and ascending aorta were obtained. Diameter variations of the ascending aorta during cardiac cycle were also considered. Mean diameters (mm) were as follows: LV outflow tract 20.3+/-3.4, coronary sinus 34.2+/-4.1, sinotubular junction 29.7+-3.4 and mid ascending aorta 32.7+/-3.8 with coefficients of variation (CV) ranging from 12 to 17%. Mean distances (mm) were: from the plane passing through the proximal insertions of the aortic valve cusps to the right brachio-cephalic artery (BCA) 92.6111.8, from the plane passing through the proximal insertions of the aortic valve cusps to the proximal coronary ostium 12.1+/-3.7, and between both coronary ostia 7.2+/-3.1, minimal arc of the ascending aorta from left coronary ostium to right BCA 52.9 X9.5, and the fibrous continuity between the aortic valve and the anterior leaflet of the mitral valve 14.óf3.3, CV 13-43%. Mean aortic valve area was 582+-131.9 mm2. The variations of the antero-posterior and transverse diameters of the ascending aorta during the cardiac cycle were 8.4% and 7.3%, respectively. Results showed large inter-individual variations in diameters and distances but with limited intra-individual variations during the cardiac cycle. A personalized approach for planning endovascular devices must be considered.

Early processing in the human lateral occipital complex is highly responsive to illusory contours but not to salient regions.

Human electrophysiological studies support a model whereby sensitivity to so-called illusory contour stimuli is first seen within the lateral occipital complex. A challenge to this model posits that the lateral occipital complex is a general site for crude region-based segmentation, based on findings of equivalent hemodynamic activations in the lateral occipital complex to illusory contour and so-called salient region stimuli, a stimulus class that lacks the classic bounding contours of illusory contours. Using high-density electrical mapping of visual evoked potentials, we show that early lateral occipital cortex activity is substantially stronger to illusory contour than to salient region stimuli, whereas later lateral occipital complex activity is stronger to salient region than to illusory contour stimuli. Our results suggest that equivalent hemodynamic activity to illusory contour and salient region stimuli probably reflects temporally integrated responses, a result of the poor temporal resolution of hemodynamic imaging. The temporal precision of visual evoked potentials is critical for establishing viable models of completion processes and visual scene analysis. We propose that crude spatial segmentation analyses, which are insensitive to illusory contours, occur first within dorsal visual regions, not the lateral occipital complex, and that initial illusory contour sensitivity is a function of the lateral occipital complex.