Pre-processing visual scenes for retinal prosthesis systems: A comprehensive review

BACKGROUND

Retinal prostheses offer hope for individuals with degenerative retinal diseases by stimulating the remaining retinal cells to partially restore their vision. This review delves into the current advancements in retinal prosthesis technology, with a special emphasis on the pivotal role that image processing and machine learning techniques play in this evolution.

METHODS

We provide a comprehensive analysis of the existing implantable devices and optogenetic strategies, delineating their advantages, limitations, and challenges in addressing complex visual tasks. The review extends to various image processing algorithms and deep learning architectures that have been implemented to enhance the functionality of retinal prosthetic devices. We also illustrate the testing results by demonstrating the clinical trials or using Simulated Prosthetic Vision (SPV) through phosphene simulations, which is a critical aspect of simulating visual perception for retinal prosthesis users.

RESULTS

Our review highlights the significant progress in retinal prosthesis technology, particularly its capacity to augment visual perception among the visually impaired. It discusses the integration between image processing and deep learning, illustrating their impact on individual interactions and navigations within the environment through applying clinical trials and also illustrating the limitations of some techniques to be used with current devices, as some approaches only use simulation even on sighted-normal individuals or rely on qualitative analysis, where some consider realistic perception models and others do not.

CONCLUSION

This interdisciplinary field holds promise for the future of retinal prostheses, with the potential to significantly enhance the quality of life for individuals with retinal prostheses. Future research directions should pivot towards optimizing phosphene simulations for SPV approaches, considering the distorted and confusing nature of phosphene perception, thereby enriching the visual perception provided by these prosthetic devices. This endeavor will not only improve navigational independence but also facilitate a more immersive interaction with the environment.

Towards biologically plausible phosphene simulation for the differentiable optimization of visual cortical prostheses

Blindness affects millions of people around the world. A promising solution to restoring a form of vision for some individuals are cortical visual prostheses, which bypass part of the impaired visual pathway by converting camera input to electrical stimulation of the visual system. The artificially induced visual percept (a pattern of localized light flashes, or 'phosphenes') has limited resolution, and a great portion of the field's research is devoted to optimizing the efficacy, efficiency, and practical usefulness of the encoding of visual information. A commonly exploited method is non-invasive functional evaluation in sighted subjects or with computational models by using simulated prosthetic vision (SPV) pipelines. An important challenge in this approach is to balance enhanced perceptual realism, biologically plausibility, and real-time performance in the simulation of cortical prosthetic vision. We present a biologically plausible, PyTorch-based phosphene simulator that can run in real-time and uses differentiable operations to allow for gradient-based computational optimization of phosphene encoding models. The simulator integrates a wide range of clinical results with neurophysiological evidence in humans and non-human primates. The pipeline includes a model of the retinotopic organization and cortical magnification of the visual cortex. Moreover, the quantitative effects of stimulation parameters and temporal dynamics on phosphene characteristics are incorporated. Our results demonstrate the simulator's suitability for both computational applications such as end-to-end deep learning-based prosthetic vision optimization as well as behavioral experiments. The modular and open-source software provides a flexible simulation framework for computational, clinical, and behavioral neuroscientists working on visual neuroprosthetics.

Self-Supervised Learning for Electroencephalography

Decades of research have shown machine learning superiority in discovering highly nonlinear patterns embedded in electroencephalography (EEG) records compared with conventional statistical techniques. However, even the most advanced machine learning techniques require relatively large, labeled EEG repositories. EEG data collection and labeling are costly. Moreover, combining available datasets to achieve a large data volume is usually infeasible due to inconsistent experimental paradigms across trials. Self-supervised learning (SSL) solves these challenges because it enables learning from EEG records across trials with variable experimental paradigms, even when the trials explore different phenomena. It aggregates multiple EEG repositories to increase accuracy, reduce bias, and mitigate overfitting in machine learning training. In addition, SSL could be employed in situations where there is limited labeled training data, and manual labeling is costly. This article: 1) provides a brief introduction to SSL; 2) describes some SSL techniques employed in recent studies, including EEG; 3) proposes current and potential SSL techniques for future investigations in EEG studies; 4) discusses the cons and pros of different SSL techniques; and 5) proposes holistic implementation tips and potential future directions for EEG SSL practices.

An actor-model framework for visual sensory encoding

A fundamental challenge in neuroengineering is determining a proper artificial input to a sensory system that yields the desired perception. In neuroprosthetics, this process is known as artificial sensory encoding, and it holds a crucial role in prosthetic devices restoring sensory perception in individuals with disabilities. For example, in visual prostheses, one key aspect of artificial image encoding is to downsample images captured by a camera to a size matching the number of inputs and resolution of the prosthesis. Here, we show that downsampling an image using the inherent computation of the retinal network yields better performance compared to learning-free downsampling methods. We have validated a learning-based approach (actor-model framework) that exploits the signal transformation from photoreceptors to retinal ganglion cells measured in explanted mouse retinas. The actor-model framework generates downsampled images eliciting a neuronal response in-silico and ex-vivo with higher neuronal reliability than the one produced by a learning-free approach. During the learning process, the actor network learns to optimize contrast and the kernel's weights. This methodological approach might guide future artificial image encoding strategies for visual prostheses. Ultimately, this framework could be applicable for encoding strategies in other sensory prostheses such as cochlear or limb.

A novel simulation paradigm utilising MRI-derived phosphene maps for cortical prosthetic vision

Objective.We developed a realistic simulation paradigm for cortical prosthetic vision and investigated whether we can improve visual performance using a novel clustering algorithm.Approach.Cortical visual prostheses have been developed to restore sight by stimulating the visual cortex. To investigate the visual experience, previous studies have used uniform phosphene maps, which may not accurately capture generated phosphene map distributions of implant recipients. The current simulation paradigm was based on the Human Connectome Project retinotopy dataset and the placement of implants on the cortices from magnetic resonance imaging scans. Five unique retinotopic maps were derived using this method. To improve performance on these retinotopic maps, we enabled head scanning and a density-based clustering algorithm was then used to relocate centroids of visual stimuli. The impact of these improvements on visual detection performance was tested. Using spatially evenly distributed maps as a control, we recruited ten subjects and evaluated their performance across five sessions on the Berkeley Rudimentary Visual Acuity test and the object recognition task.Main results.Performance on control maps is significantly better than on retinotopic maps in both tasks. Both head scanning and the clustering algorithm showed the potential of improving visual ability across multiple sessions in the object recognition task.Significance.The current paradigm is the first that simulates the experience of cortical prosthetic vision based on brain scans and implant placement, which captures the spatial distribution of phosphenes more realistically. Utilisation of evenly distributed maps may overestimate the performance that visual prosthetics can restore. This simulation paradigm could be used in clinical practice when making plans for where best to implant cortical visual prostheses.

Mixture 2D Convolutions for 3D Medical Image Segmentation

Three-dimensional (3D) medical image segmentation plays a crucial role in medical care applications. Although various two-dimensional (2D) and 3D neural network models have been applied to 3D medical image segmentation and achieved impressive results, a trade-off remains between efficiency and accuracy. To address this issue, a novel mixture convolutional network (MixConvNet) is proposed, in which traditional 2D/3D convolutional blocks are replaced with novel MixConv blocks. In the MixConv block, 3D convolution is decomposed into a mixture of 2D convolutions from different views. Therefore, the MixConv block fully utilizes the advantages of 2D convolution and maintains the learning ability of 3D convolution. It acts as 3D convolutions and thus can process volumetric input directly and learn intra-slice features, which are absent in the traditional 2D convolutional block. By contrast, the proposed MixConv block only contains 2D convolutions; hence, it has significantly fewer trainable parameters and less computation budget than a block containing 3D convolutions. Furthermore, the proposed MixConvNet is pre-trained with small input patches and fine-tuned with large input patches to improve segmentation performance further. In experiments on the Decathlon Heart dataset and Sliver07 dataset, the proposed MixConvNet outperformed the state-of-the-art methods such as UNet3D, VNet, and nnUnet.

Clinical Progress and Optimization of Information Processing in Artificial Visual Prostheses

Visual prostheses, used to assist in restoring functional vision to the visually impaired, convert captured external images into corresponding electrical stimulation patterns that are stimulated by implanted microelectrodes to induce phosphenes and eventually visual perception. Detecting and providing useful visual information to the prosthesis wearer under limited artificial vision has been an important concern in the field of visual prosthesis. Along with the development of prosthetic device design and stimulus encoding methods, researchers have explored the possibility of the application of computer vision by simulating visual perception under prosthetic vision. Effective image processing in computer vision is performed to optimize artificial visual information and improve the ability to restore various important visual functions in implant recipients, allowing them to better achieve their daily demands. This paper first reviews the recent clinical implantation of different types of visual prostheses, summarizes the artificial visual perception of implant recipients, and especially focuses on its irregularities, such as dropout and distorted phosphenes. Then, the important aspects of computer vision in the optimization of visual information processing are reviewed, and the possibilities and shortcomings of these solutions are discussed. Ultimately, the development direction and emphasis issues for improving the performance of visual prosthesis devices are summarized.

Beyond the Cane: Describing Urban Scenes to Blind People for Mobility Tasks

Blind people face difficulties with independent mobility, impacting employment prospects, social inclusion, and quality of life. Given the advancements in computer vision, with more efficient and effective automated information extraction from visual scenes, it is important to determine what information is worth conveying to blind travelers, especially since people have a limited capacity to receive and process sensory information. We aimed to investigate which objects in a street scene are useful to describe and how those objects should be described. Thirteen cane-using participants, five of whom were early blind, took part in two urban walking experiments. In the first experiment, participants were asked to voice their information needs in the form of questions to the experimenter. In the second experiment, participants were asked to score scene descriptions and navigation instructions, provided by the experimenter, in terms of their usefulness. The descriptions included a variety of objects with various annotations per object. Additionally, we asked participants to rank order the objects and the different descriptions per object in terms of priority and explain why the provided information is or is not useful to them. The results reveal differences between early and late blind participants. Late blind participants requested information more frequently and prioritized information about objects’ locations. Our results illustrate how different factors, such as the level of detail, relative position, and what type of information is provided when describing an object, affected the usefulness of scene descriptions. Participants explained how they (indirectly) used information, but they were frequently unable to explain their ratings. The results distinguish between various types of travel information, underscore the importance of featuring these types at multiple levels of abstraction, and highlight gaps in current understanding of travel information needs. Elucidating the information needs of blind travelers is critical for the development of more useful assistive technologies.

End-to-end optimization of prosthetic vision

Neural prosthetics may provide a promising solution to restore visual perception in some forms of blindness. The restored prosthetic percept is rudimentary compared to normal vision and can be optimized with a variety of image preprocessing techniques to maximize relevant information transfer. Extracting the most useful features from a visual scene is a nontrivial task and optimal preprocessing choices strongly depend on the context. Despite rapid advancements in deep learning, research currently faces a difficult challenge in finding a general and automated preprocessing strategy that can be tailored to specific tasks or user requirements. In this paper, we present a novel deep learning approach that explicitly addresses this issue by optimizing the entire process of phosphene generation in an end-to-end fashion. The proposed model is based on a deep auto-encoder architecture and includes a highly adjustable simulation module of prosthetic vision. In computational validation experiments, we show that such an approach is able to automatically find a task-specific stimulation protocol. The results of these proof-of-principle experiments illustrate the potential of end-to-end optimization for prosthetic vision. The presented approach is highly modular and our approach could be extended to automated dynamic optimization of prosthetic vision for everyday tasks, given any specific constraints, accommodating individual requirements of the end-user.

Uncertainty-Guided Voxel-Level Supervised Contrastive Learning for Semi-Supervised Medical Image Segmentation

Semi-supervised learning reduces overfitting and facilitates medical image segmentation by regularizing the learning of limited well-annotated data with the knowledge provided by a large amount of unlabeled data. However, there are many misuses and underutilization of data in conventional semi-supervised methods. On the one hand, the model will deviate from the empirical distribution under the training of numerous unlabeled data. On the other hand, the model treats labeled and unlabeled data differently and does not consider inter-data information. In this paper, a semi-supervised method is proposed to exploit unlabeled data to further narrow the gap between the semi-supervised model and its fully-supervised counterpart. Specifically, the architecture of the proposed method is based on the mean-teacher framework, and the uncertainty estimation module is improved to impose constraints of consistency and guide the selection of feature representation vectors. Notably, a voxel-level supervised contrastive learning module is devised to establish a contrastive relationship between feature representation vectors, whether from labeled or unlabeled data. The supervised manner ensures that the network learns the correct knowledge, and the dense contrastive relationship further extracts information from unlabeled data. The above overcomes data misuse and underutilization in semi-supervised frameworks. Moreover, it favors the feature representation with intra-class compactness and inter-class separability and gains extra performance. Extensive experimental results on the left atrium dataset from Atrial Segmentation Challenge demonstrate that the proposed method has superior performance over the state-of-the-art methods.

Real-world indoor mobility with simulated prosthetic vision: The benefits and feasibility of contour-based scene simplification at different phosphene resolutions

Neuroprosthetic implants are a promising technology for restoring some form of vision in people with visual impairments via electrical neurostimulation in the visual pathway. Although an artificially generated prosthetic percept is relatively limited compared with normal vision, it may provide some elementary perception of the surroundings, re-enabling daily living functionality. For mobility in particular, various studies have investigated the benefits of visual neuroprosthetics in a simulated prosthetic vision paradigm with varying outcomes. The previous literature suggests that scene simplification via image processing, and particularly contour extraction, may potentially improve the mobility performance in a virtual environment. In the current simulation study with sighted participants, we explore both the theoretically attainable benefits of strict scene simplification in an indoor environment by controlling the environmental complexity, as well as the practically achieved improvement with a deep learning-based surface boundary detection implementation compared with traditional edge detection. A simulated electrode resolution of 26 × 26 was found to provide sufficient information for mobility in a simple environment. Our results suggest that, for a lower number of implanted electrodes, the removal of background textures and within-surface gradients may be beneficial in theory. However, the deep learning-based implementation for surface boundary detection did not improve mobility performance in the current study. Furthermore, our findings indicate that, for a greater number of electrodes, the removal of within-surface gradients and background textures may deteriorate, rather than improve, mobility. Therefore, finding a balanced amount of scene simplification requires a careful tradeoff between informativity and interpretability that may depend on the number of implanted electrodes.

Tiled Sparse Coding in Eigenspaces for Image Classification

The automation in the diagnosis of medical images is currently a challenging task. The use of Computer Aided Diagnosis (CAD) systems can be a powerful tool for clinicians, especially in situations when hospitals are overflowed. These tools are usually based on artificial intelligence (AI), a field that has been recently revolutionized by deep learning approaches. blackThese alternatives usually obtain a large performance based on complex solutions, leading to a high computational cost and the need of having large databases. In this work, we propose a classification framework based on sparse coding. Images are blackfirst partitioned into different tiles, and a dictionary is built after applying PCA to these tiles. The original signals are then transformed as a linear combination of the elements of the dictionary. blackThen, they are reconstructed by iteratively deactivating the elements associated with each component. Classification is finally performed employing as features the subsequent reconstruction errors. Performance is evaluated in a real context where distinguishing between four different pathologies: control versus bacterial pneumonia versus viral pneumonia versus COVID-19. blackOur system differentiates between pneumonia patients and controls with an accuracy of 97.74%, whereas in the 4-class context the accuracy is 86.73%. The excellent results and the pioneering use of sparse coding in this scenario evidence that our proposal can assist clinicians when their workload is high.

Detection of Movement Intention in EEG-Based Brain-Computer Interfaces Using Fourier-Based Synchrosqueezing Transform

Patients with motor impairments need caregivers' help to initiate the operation of brain-computer interfaces (BCI). This study aims to identify and characterize movement intention using multichannel electroencephalography (EEG) signals as a means to initiate BCI systems without extra accessories/methodologies. We propose to discriminate the resting and motor imagery (MI) states with high accuracy using Fourier-based synchrosqueezing transform (FSST) as a feature extractor. FSST has been investigated and compared with other popular approaches in 28 healthy subjects for a total of 6657 trials. The accuracy and f-measure values were obtained as 99.8% and 0.99, respectively, when FSST was used as the feature extractor and singular value decomposition (SVD) as the feature selection method and support vector machines as the classifier. Moreover, this study investigated the use of data that contain certain amount of noise without any preprocessing in addition to the clean counterparts. Furthermore, the statistical analysis of EEG channels with the best discrimination (of resting and MI states) characteristics demonstrated that F4-Fz-C3-Cz-C4-Pz channels and several statistical features had statistical significance levels, [Formula: see text], less than 0.05. This study showed that the preparation of the movement can be detected in real-time employing FSST-SVD combination and several channels with minimal pre-processing effort.

Visual percepts evoked with an Intracortical 96-channel microelectrode array inserted in human occipital cortex

BACKGROUND

A long-held dream of scientists is to transfer information directly to the visual cortex of blind individuals, thereby restoring a rudimentary form of sight. However, no clinically available cortical visual prosthesis yet exists.

METHODS

We implanted an intracortical microelectrode array consisting of 96 electrodes in the visual cortex of a 57-year-old person with complete blindness for a six- month period. We measured thresholds and the characteristics of the visual percepts elicited by intracortical microstimulation.

RESULTS

Implantation and subsequent explantation of intracortical microelectrodes were carried out without complications. The mean stimulation threshold for single electrodes was 66.8 ± 36.5 μA. We consistently obtained high-quality recordings from visually deprived neurons and the stimulation parameters remained stable over time. Simultaneous stimulation via multiple electrodes were associated with a significant reduction in thresholds (p<0.001, ANOVA test) and evoked discriminable phosphene percepts, allowing the blind participant to identify some letters and recognize object boundaries. Furthermore, we observed a learning process that helped the subject to recognize complex patterns over time.

CONCLUSIONS

Our results demonstrate the safety and efficacy of chronic intracortical microstimulation via a large number of electrodes in human visual cortex, showing its high potential for restoring functional vision in the blind.

TRIAL REGISTRATION

ClinicalTrials.gov identifier NCT02983370.

FUNDING

Funding was provided by grant RTI2018-098969-B-100 from the Spanish Ministerio de Ciencia Innovación y Universidades, by grant PROMETEO/2019/119 from the Generalitat Valenciana (Spain), by the Bidons Egara Research Chair of the University Miguel Hernández (Spain) and by the John Moran Eye Center of the University of Utah (US).

Hysteresis Modeling in Iron-Dominated Magnets Based on a Multi-Layered Narx Neural Network Approach

A full-fledged neural network modeling, based on a Multi-layered Nonlinear Autoregressive Exogenous Neural Network (NARX) architecture, is proposed for quasi-static and dynamic hysteresis loops, one of the most challenging topics for computational magnetism. This modeling approach overcomes drawbacks in attaining better than percent-level accuracy of classical and recent approaches for accelerator magnets, that combine hybridization of standard hysteretic models and neural network architectures. By means of an incremental procedure, different Deep Neural Network Architectures are selected, fine-tuned and tested in order to predict magnetic hysteresis in the context of electromagnets. Tests and results show that the proposed NARX architecture best fits the measured magnetic field behavior of a reference quadrupole at CERN. In particular, the proposed modeling framework leads to a percent error below 0.02% for the magnetic field prediction, thus outperforming state of the art approaches and paving a very promising way for future real time applications.

Optimization of Model Training Based on Iterative Minimum Covariance Determinant In Motor-Imagery BCI

The common spatial patterns (CSP) algorithm is one of the most frequently used and effective spatial filtering methods for extracting relevant features for use in motor imagery brain-computer interfaces (MI-BCIs). However, the inherent defect of the traditional CSP algorithm is that it is highly sensitive to potential outliers, which adversely affects its performance in practical applications. In this work, we propose a novel feature optimization and outlier detection method for the CSP algorithm. Specifically, we use the minimum covariance determinant (MCD) to detect and remove outliers in the dataset, then we use the Fisher score to evaluate and select features. In addition, in order to prevent the emergence of new outliers, we propose an iterative minimum covariance determinant (IMCD) algorithm. We evaluate our proposed algorithm in terms of iteration times, classification accuracy and feature distribution using two BCI competition datasets. The experimental results show that the average classification performance of our proposed method is 12% and 22.9% higher than that of the traditional CSP method in two datasets ([Formula: see text]), and our proposed method obtains better performance in comparison with other competing methods. The results show that our method improves the performance of MI-BCI systems.

Decision-Feedback Stages Revealed by Hidden Markov Modeling of EEG

Decision response and feedback in gambling are interrelated. Different decisions lead to different ranges of feedback, which in turn influences subsequent decisions. However, the mechanism underlying the continuous decision-feedback process is still left unveiled. To fulfill this gap, we applied the hidden Markov model (HMM) to the gambling electroencephalogram (EEG) data to characterize the dynamics of this process. Furthermore, we explored the differences between distinct decision responses (i.e. choose large or small bets) or distinct feedback (i.e. win or loss outcomes) in corresponding phases. We demonstrated that the processing stages in decision-feedback process including strategy adjustment and visual information processing can be characterized by distinct brain networks. Moreover, time-varying networks showed, after decision response, large bet recruited more resources from right frontal and right center cortices while small bet was more related to the activation of the left frontal lobe. Concerning feedback, networks of win feedback showed a strong right frontal and right center pattern, while an information flow originating from the left frontal lobe to the middle frontal lobe was observed in loss feedback. Taken together, these findings shed light on general principles of natural decision-feedback and may contribute to the design of biologically inspired, participant-independent decision-feedback systems.

Visual cortical prosthesis: an electrical perspective

The electrical stimulation of the visual cortices has the potential to restore vision to blind individuals. Until now, the results of visual cortical prosthetics have been limited as no prosthesis has restored a full working vision but the field has shown a renewed interest these last years, thanks to wireless and technological advances. However, several scientific and technical challenges are still open to achieve the therapeutic benefit expected by these new devices. One of the main challenges is the electrical stimulation of the brain itself. In this review, we analyse the results in electrode-based visual cortical prosthetics from the electrical point of view. We first describe what is known about the electrode-tissue interface and safety of electrical stimulation. Then we focus on the psychophysics of prosthetic vision and the state-of-the-art on the interplay between the electrical stimulation of the visual cortex and the phosphene perception. Lastly, we discuss the challenges and perspectives of visual cortex electrical stimulation and electrode array design to develop the new generation implantable cortical visual prostheses.