Motion Vision: A New Mechanism in the Mammalian Retina

Learning Dendritic-Neuron-Based Motion Detection for RGB Images: A Biomimetic Approach

In this study, we designed a biomimetic artificial visual system (AVS) inspired by biological visual system that can process RGB images. Our approach begins by mimicking the photoreceptor cone cells to simulate the initial input processing followed by a learnable dendritic neuron model to replicate ganglion cells that integrate outputs from bipolar and horizontal cell simulations. To handle multi-channel integration, we utilize a nonlearnable dendritic neuron model to simulate the lateral geniculate nucleus (LGN), which consolidates outputs across color channels, an essential function in biological multi-channel processing. Cross-validation experiments show that AVS demonstrates strong generalization across varied object-background configurations, achieving accuracy where traditional models like EfN-B0, ResNet50, and ConvNeXt typically fall short. Additionally, our results across different training-to-testing data ratios reveal that AVS maintains over 96% test accuracy even with limited training data, underscoring its robustness in low-data scenarios. This demonstrates the practical advantage of the AVS model in applications where large-scale annotated datasets are unavailable or expensive to curate. This AVS model not only advances biologically inspired multi-channel processing but also provides a practical framework for efficient, integrated visual processing in computational models.

A Learning Dendritic Neuron-Based Motion Direction Detective System and Its Application to Grayscale Images

In recent research, dendritic neuron-based models have shown promise in effectively learning and recognizing object motion direction within binary images. Leveraging the dendritic neuron structure and On-Off Response mechanism within the primary cortex, this approach has notably reduced learning time and costs compared to traditional neural networks. This paper advances the existing model by integrating bio-inspired components into a learnable dendritic neuron-based artificial visual system (AVS), specifically incorporating mechanisms from horizontal and bipolar cells. This enhancement enables the model to proficiently identify object motion directions in grayscale images, aligning its threshold with human-like perception. The enhanced model demonstrates superior efficiency in motion direction recognition, requiring less data (90% less than other deep models) and less time for training. Experimental findings highlight the model's remarkable robustness, indicating significant potential for real-world applications. The integration of bio-inspired features not only enhances performance but also opens avenues for further exploration in neural network research. Notably, the application of this model to realistic object recognition yields convincing accuracy at nearly 100%, underscoring its practical utility.

Potential Properties of Natural Nutraceuticals and Antioxidants in Age-Related Eye Disorders

Eye health is crucial, and the onset of diseases can reduce vision and affect the quality of life of patients. The main causes of progressive and irreversible vision loss include various pathologies, such as cataracts, ocular atrophy, corneal opacity, age-related macular degeneration, uncorrected refractive error, posterior capsular opacification, uveitis, glaucoma, diabetic retinopathy, retinal detachment, undetermined disease and other disorders involving oxidative stress and inflammation. The eyes are constantly exposed to the external environment and, for this reason, must be protected from damage from the outside. Many drugs, including cortisonics and antinflammatory drugs have widely been used to counteract eye disorders. However, recent advances have been obtained via supplementation with natural antioxidants and nutraceuticals for patients. In particular, evidence has accumulated that polyphenols (mostly deriving from Citrus Bergamia) represent a reliable source of antioxidants able to counteract oxidative stress accompanying early stages of eye diseases. Luteolin in particular has been found to protect photoreceptors, thereby improving vision in many disease states. Moreover, a consistent anti-inflammatory response was found to occur when curcumin is used alone or in combination with other nutraceuticals. Additionally, Coenzyme Q10 has been demonstrated to produce a consistent effect in reducing ocular pressure, thereby leading to protection in patients undergoing glaucoma. Finally, both grape seed extract, rich in anthocyanosides, and polynsatured fatty acids seem to contribute to the prevention of retinal disorders. Thus, a combination of nutraceuticals and antioxidants may represent the right solution for a multi-action activity in eye protection, in association with current drug therapies, and this will be of potential interest in early stages of eye disorders.

Flexible filtering by neural inputs supports motion computation across states and stimuli

Sensory systems flexibly adapt their processing properties across a wide range of environmental and behavioral conditions. Such variable processing complicates attempts to extract a mechanistic understanding of sensory computations. This is evident in the highly constrained, canonical Drosophila motion detection circuit, where the core computation underlying direction selectivity is still debated despite extensive studies. Here we measured the filtering properties of neural inputs to the OFF motion-detecting T5 cell in Drosophila. We report state- and stimulus-dependent changes in the shape of these signals, which become more biphasic under specific conditions. Summing these inputs within the framework of a connectomic-constrained model of the circuit demonstrates that these shapes are sufficient to explain T5 responses to various motion stimuli. Thus, our stimulus- and state-dependent measurements reconcile motion computation with the anatomy of the circuit. These findings provide a clear example of how a basic circuit supports flexible sensory computation.