Image processing based on a model of the mammalian retina

A first study in order to construct a simple model of the mammalian retina is reported. The basic elements for this model are Optical Programmable Logic Cells, OPLCs, previously employed as a functional element for Optical Computing. The same type of circuit simulates the five types of neurons present in the retina. Different responses are obtained by modifying either internal or external connections. Two types of behaviors are reported: symmetrical and non-symmetrical with respect to light position. Some other higher functions, as the possibility to differentiate between symmetric and non-symmetric light images, are performed by another simulation of the first layers of the visual cortex. The possibility to apply these models to image processing is reported.

Algorithms for colour image processing based on neurological models

373 p. : il., gráf., fot., tablas

Contributions to Ensemble Classifiers with Image Analysis Applications

134 p.

Fixational Instability and Natural Image Statistics: Implications for Early Visual Representations

Under natural viewing conditions small movements of the eye, head, and body prevent the maintenance of a steady direction of gaze. It is known that stimuli tend to fade when they a restabilized on the retina for several seconds. However; it is unclear whether the physiological motion of the retinal image serves a visual purpose during the brief periods of natural visual fixation. This study examines the impact of fixational instability on the statistics of the visua1 input to the retina and on the structure of neural activity in the early visual system. We show that fixational instability introduces a component in the retinal input signals that in the presence of natural images, lacks spatial correlations. This component strongly influences neural activity in a model of the LGN. It decorrelates cell responses even if the contrast sensitivity functions of simulated cells arc not perfectly tuned to counterbalance the power-law spectrum of natural images. A decorrelation of neural activity at the early stages of the visual system has been proposed to be beneficial for discarding statistical redundancies in the input signals. The results of this study suggest that fixational instability might contribute to establishing efficient representations of natural stimuli.

Endothelin-like immunoreactivity and receptor binding in the choroid and retina

This study has examined the localisation and receptor-binding of the endothelins in retina and choroid of human and rat origin. Immunoreactivity to anti-ET1 and anti-ET3 was investigated in trypsin digests, frozen sections and ultrathin sections using immunocytochemistry and immunogold labelling techniques. In addition, receptor binding of 125I-ET1 and 125I-ET3 was visualised and quantified using autoradiography and image analysis. Intense immunoreactivity to anti-ET1 and anti-ET3 was observed in the photoreceptor inner segments and in the outer plexiform layer (OPL) of human and rat retina. Ultrastructural localisation using immunogold labelling confirmed the presence of ET1 and ET3 in the photoreceptor cells. In retinal vascular digests, ET1 was visualised in the arteries, arterioles and at the pre-arteriolar sphincters, however, immunoreactivity to anti-ET3 was absent in the retinal vasculature. Both ETA and ETB-type receptor binding sites to 125I-ET1 and 125I-ET3 were detected in the vascular smooth muscle of choroidal and retinal vessels with the former being predominant. Extravascular binding sites of the ETB-type were found in the ganglion cell layer.

Comparison between laser scanning tomography and computerised image analysis of the optic disc

Aims - To study the interchangeability of the measurements of the optic disc topography obtained by one computerised image analyser and one confocal laser tomographic scanner. Methods - One eye of 28 patients with glaucoma or glaucoma suspects was studied. All cases had simultaneous stereoscopic disc photographs taken with the fundus camera Topcon TRC-SS and optic disc examination with the Heidelberg retina tomograph (HRT) during the same visit. The optic disc photographs were digitised and analysed with the Topcon ImageNet (TI) system. Three variables of the optic disc topography provided by the TI and the HRT were compared - cup volume (CV), rim area (RA), and cup area to disc area ratio (CA/DA). Results - The mean values of CV and RA provided by the TI (0.52 (SD 0.32) mm and 1.58 (0.39) mm , respectively) were greater (p <0.01) than the mean values of CV and RA determined by the HRT (0.32 (0.25) mm , and 1.33 (0.47) mm , respectively). The mean value of CA/DA provided by the TI (0.42 (0.14)) and the HRT (0.42 (0.18)) was similar (p = 0.93). Correlation coefficients between measurements obtained by the two methods ranged from 0.53 to 0.73. Conclusion - There was a significant discrepancy in the measurements of rim area and cup volume of the optic disc obtained by a computerised image analyser and a laser scanning tomograph.

Reproducibility of optic disk topographic measurements with the Topcon ImageNet and the Heidelberg Retina Tomograph

Objective: To compare the reproducibility of optic disk measurements provided by an image analyzer and a scanning laser tomograph. Methods: Ten images of the same eye of 10 normal volunteers were taken with the Heidelberg Retina Tomograph and with the Topcon ImageNet. Intraclass correlation coefficient (ICC) and coefficient of variation (CV) were used to evaluate the reproducibility of the measurements. Results: Eleven parameters were analyzed with the Topcon ImageNet. Six parameters (55%) had ICC greater than 90%. Four parameters (36%) had CV less than 10%. Twelve parameters were evaluated with the Heidelberg Retina Tomograph. Nine parameters (75%) had ICC over 90%. Nine parameters (75%) had CV less than 10%. Conclusion: Both systems provided reproducible data. The optic disk parameters provided by the Heidelberg Retina Tomograph had a better reproducibility than those obtained from the Topcon ImageNet.

Factores asociados a recuperación visual deficiente después de cirugía exitosa para desprendimiento de retina regmatógeno

Introducción: Varias características pueden afectar el pronóstico visual después de resolver quirúrgicamente el desprendimiento de retina. Existen características no observables por el ojo humano por si solo pero si por tomografía óptica coherente que se relacionan con la recuperación visual. Objetivo: Describir las características clínicas y topográfica en los periodos pre y postquirúrgico de ojos que ha sufrido DR regmatógeno con compromiso macular y su relación con la calidad de recuperación visual después de una cirugía considerada exitosa desde el punto de vista anatómico. Materiales y métodos: Estudio descriptivo en el que se comparan algunas características en tres periodos perioeperatorios, uno antes y dos después de cirugía (3 y 6 meses) de 24 ojos con DRregmatógeno y compromiso macular intervenidos mediante retinopexia combinada con vitrectomía pars plana. Resultados: La recuperación visual mejor o igual que logMAR 0,397 (20/50) se dió en 41,7% de ojos y 16,7%. alcanzaron agudeza visual logMAR 0,301 (20/40). Cinco ojos no alcanzaron una ganancia de líneas de visión mayor a cinco. El líquido submacular ausente se observó en la mayoría de ojos que recuperaron más de cinco líneas al igual que aquellos con elipsoide conservado. La regularidad del neuroepitelio y el edema en el periodo posquirúrgico no mostraron comportamientos claros respecto a recuperación visual al igual que la altura del desprendimiento y el número de cuadrantes afectados. Una mejor recuperación visual fue más frecuente en aquellos con menos de cinco semanas de desprendimiento de retina. Conclusiones: El retraso menor a cinco semanas en la resolución del desprendimiento de retina, la conservación del elipsoide y la ausencia de líquido submacular en el periodo postquirúrgico se observó más frecuentemente en ojos con mejor recuperación visual.

Object detecting artificial retina

This paper describes the novel use of agent and cellular neural Hopfield network techniques in the design of a self-contained, object detecting retina. The agents, which are used to detect features within an image, are trained using the Hebbian method which has been modified for the cellular architecture. The success of each agent is communicated with adjacent agents in order to verify the detection of an object. Initial work used the method to process bipolar images. This has now been extended to handle grey scale images. Simulations have demonstrated the success of the method and further work is planned in which the device is to be implemented in hardware.

Aspectos anatômicos do olho e neuroquímicos da retina do mocó (Kerodon rupestris)

The visual system is an important link between the animal and the environment, com profound influences on the habits and lifestyle in various habitats. Adaptive mechanismsto the temporal niche are present in the visual system of many vertebrates, involving changins in ocular dimensios and design, retinal cell distribution and organization of neurochemical circuits related to the retinal resolution or sensitivity. The sensory system of the eye is represented by the retina, whose organization is responsible by receipty, initial analysis, and transmission of the information to the brain. The knowledge of the position of the eyes in the head and the distribution of retinal cells allow to identify adaptive aspects of each species to its visual field, which is characteristic to the ecological niche it occupies. In this research, we study eye anatomical characteristics and retina neurochemical features of the rock cavy (Kerodon rupestris), a tipical Brazilian rodent from the suborder Hystricomorpha, family Caviidae. The rock cavy has lateral eyes well constitute bony orbit and well differentiated extrinsic muscle. The study of the descriptive and morphometric anatomy of the showed mean values of axial diameter 10.7±0,5mm and equatorial diameter 11.6±0.7mm. The pupil is slit shaped and the lens has mean axial diameter 5.4±0.03 mm, corresponding to ~45% of the axial diameter of the eye. The posterior nodal distance and the retinal magnification factor were estimated at 6.74 mm e 118 μm/grau, respectively. Flat mounts were processed for Nissl stain, and the topographic distribution of ganglion cells showed a moderate visual band, just below the optic disc, with higher density in the ventral retina. Retinal vertical sections and flat mounts were processed for immunohistochemistry to visualize tyrosine hydroxilase (TH) and thus two types of TH+ cells were detected. Type 1 cells had strong TH-immunoreactivity, the body cell varied from 120.047 to 269.373 μm2 stratifying in the sublamina 1 of the IPL. Type 2 cells were weakly TH-imunoreactive, had cell body located mostly in the IPL, varying from 54.848 to 177.142 μm2, constituting ~10% of the TH+ cells. Both cell types exhibited similar topographic distribution with higher density found in a horizontal band along of the naso-temporal axis in the dorsal retina. The total population of dopaminergic cells was 2,156±469,4 cells, occupying an average area of 198,164 μm2. The presence of cones and rods was detected by immunohistochemistry in vertical sections and flat mounts. S cones density is around 10 times smaller than L cones, with different degree of spatial organization. Other retinal neuronal populations of the rock cavy were also detected in vertical sections with specific markers. Comparative analysis of the anatomical characteristics of the rock cavy eye 12 suggest that it was designed to acquire higher sensitivity to light, at expense of image sharpness, compatible with a vision at mesopic conditions. Additionally, the distribution of the 2 subtypes of dopaminergic cells in a naso-temporal band in the dorsal retina seems suitable to a gain in sensitivity, coherent with an animal with predominantly crepuscular activity pattern

Quality Assessment and Enhancement Method for high-resolution Fourier-Domain OCT Imaging of Retina Laser Lesions

We present an image quality assessment and enhancement method for high-resolution Fourier-Domain OCT imaging like in sub-threshold retina therapy. A Maximum-Likelihood deconvolution algorithm as well as a histogram-based quality assessment method are evaluated.

Acute onset incomitant image disparity modifies saccadic and vergence eye movements.

New-onset impairment of ocular motility will cause incomitant strabismus, i.e., a gaze-dependent ocular misalignment. This ocular misalignment will cause retinal disparity, that is, a deviation of the spatial position of an image on the retina of both eyes, which is a trigger for a vergence eye movement that results in ocular realignment. If the vergence movement fails, the eyes remain misaligned, resulting in double vision. Adaptive processes to such incomitant vergence stimuli are poorly understood. In this study, we have investigated the physiological oculomotor response of saccadic and vergence eye movements in healthy individuals after shifting gaze from a viewing position without image disparity into a field of view with increased image disparity, thus in conditions mimicking incomitance. Repetitive saccadic eye movements into a visual field with increased stimulus disparity lead to a rapid modification of the oculomotor response: (a) Saccades showed immediate disconjugacy (p < 0.001) resulting in decreased retinal image disparity at the end of a saccade. (b) Vergence kinetics improved over time (p < 0.001). This modified oculomotor response enables a more prompt restoration of ocular alignment in new-onset incomitance.

Automatic assessment of time-resolved OCT images for selective retina therapy

Purpose In recent years, selective retina laser treatment (SRT), a sub-threshold therapy method, avoids widespread damage to all retinal layers by targeting only a few. While these methods facilitate faster healing, their lack of visual feedback during treatment represents a considerable shortcoming as induced lesions remain invisible with conventional imaging and make clinical use challenging. To overcome this, we present a new strategy to provide location-specific and contact-free automatic feedback of SRT laser applications. Methods We leverage time-resolved optical coherence tomography (OCT) to provide informative feedback to clinicians on outcomes of location-specific treatment. By coupling an OCT system to SRT treatment laser, we visualize structural changes in the retinal layers as they occur via time-resolved depth images. We then propose a novel strategy for automatic assessment of such time-resolved OCT images. To achieve this, we introduce novel image features for this task that when combined with standard machine learning classifiers yield excellent treatment outcome classification capabilities. Results Our approach was evaluated on both ex vivo porcine eyes and human patients in a clinical setting, yielding performances above 95 % accuracy for predicting patient treatment outcomes. In addition, we show that accurate outcomes for human patients can be estimated even when our method is trained using only ex vivo porcine data. Conclusion The proposed technique presents a much needed strategy toward noninvasive, safe, reliable, and repeatable SRT applications. These results are encouraging for the broader use of new treatment options for neovascularization-based retinal pathologies.

Confocal analysis of synaptic connectivity of the rod pathway in the rabbit retina

The retina is a specialized neuronal structure that transforms the optical image into electrical signals which are transmitted to the brain via the optic nerve. As part of the strategy to cover a stimulus range as broad as 10 log units, from dim starlight to bright sunlight, retinal circuits are broadly divided into rod and cone pathways, responsible for dark and light-adapted vision, respectively. ^ In this dissertation, confocal microscopy and immunocytochemical methods were combined to study the synaptic connectivity of the rod pathway from the level of individual synapses to whole populations of neurons. The study was focused on synaptic interactions at the rod bipolar terminal. The purpose is to understand the synaptic structure of the dyad synapse made by rod bipolar terminals, including the synaptic components and connections, and their physiological functions in the rod pathway. In addition, some additional components and connections of the rod pathway were also studied in these experiments. The major results can be summarized as following: At the dyad synapse of rod bipolar terminals, three postsynaptic components—processes of All amacrine cells and the varicosities of S1 or S2 amacrine cells express different glutamate receptor subunits, which may underlie the functional diversity of these postsynaptic neurons. A reciprocal feedback system is formed by rod bipolar terminals and S1/S2 amacrine cells. Analysis showed these two wide-field GABA amacrine cells have stereotyped synaptic connections with the appropriate morphology and distribution to perform specific functions. In addition, S1 and S2 cells have different coupling patterns and, in general, there is no coupling between the two types. Besides the classic rod pathway though rod bipolar cells and All amacrine cells, the finding of direct connections between certain types of OFF cone bipolar cells and rods indicates the presence of an alternative rod pathway in the rabbit retina. ^ In summary, this dissertation presents a detailed view of the connection and receptors at rod bipolar terminals. Based on the morphology, distribution and coupling, different functional roles were identified for S1 and S2 amacrine cells. Finally, an alternative to the classic rod pathway was found in the rabbit retina. ^