Systems and methods for treating retinal degenerative diseases

Abstract

This specification discloses a system and method for treating ocular diseases, such as retinal degenerative diseases. The method for treating these ocular diseases, such as retinal degenerative diseases, includes the step of obtaining an adapted optic neural modulation code, which is adapted to produce a neural modulation response in the visual cortex. The method is further characterized by including the step of outputting the adapted optic neural modulation code to an electronic display of a device visible to the patient.

Description

Technical Field

[0001] The present disclosure relates to systems and methods for treating retinal degenerative diseases, particularly diseases associated with progressive retinal degeneration.

Background Art

[0002] Retinal degenerative diseases are a group of diseases that affect the retina, a thin tissue layer behind the eye. The retina plays a role in capturing visual information and transmitting it to the brain. Currently, treatment options for retinal degenerative diseases are limited, and there are no options to cure or reverse the damage caused by the diseases. Treatment may include medications and, in some cases, surgery. However, these treatments often have only partial effects and may be accompanied by serious side effects. Retinal degenerative diseases can have a significant impact on an individual's quality of life and independence.

[0003] Age-related macular degeneration (AMD) is an example of a visual impairment that causes vision loss without significantly damaging the brain's visual processing centers (e.g., visual fields). AMD is a major cause of blindness in people over 50 years old. This disease affects the macula in its early stages. The macula is a part of the retina and is responsible for central vision. AMD causes irreversible damage to the macula and leads to progressive loss of central vision. However, in many cases, the peripheral vision is preserved in the early stages. AMD is a common cause of blindness and visual impairment and has a significant impact on public health. For example, from the perspective of morbidity, it is a particularly common disease among the elderly. According to the National Eye Institute of the United States, AMD is the leading cause of blindness in Americans over 65 years old, and it is estimated that approximately 11 million people in the country are affected. AMD causes significant visual impairment and functional impairment, affecting daily activities such as reading, driving, and face recognition.

[0004] Age-related macular degeneration (AMD) has a significant economic burden, including medical expenses, rehabilitation costs, and productivity losses. One estimate suggests that the annual direct medical cost of AMD in the United States is approximately $6.7 billion. AMD is associated with increased use of healthcare services, including hospitalization, emergency room visits, and outpatient visits. This strains the healthcare system and increases healthcare costs. Therefore, age-related macular degeneration has a significant impact on public health, affecting millions of people in the United States and worldwide. The disease causes significant disability, economic burden, and increased healthcare utilization, highlighting the need for effective prevention and treatment strategies. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] PCT / US2021 / 049080 [Patent Document 2] PCT / US2022 / 77207 [Patent Document 2] U.S. Provisional Patent Application No. 63 / 324,395 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Approximately 90% of patients with age-related macular degeneration (AMD) have "dry" AMD. In this condition, the macular layer (including photoreceptors and retinal pigment epithelium) gradually thins and atrophies, leading to a decline in function. Currently, there are no approved medications for dry AMD.

[0007] Approximately 10% of all age-related macular degeneration (AMD) cases are "wet" AMD. Typically, patients first develop dry AMD, which then progresses to the wet form. In wet AMD, new blood vessels grow in the choroidal layer behind the retina. This condition is called choroidal neovascularization. These new blood vessels are fragile and leak fluid, lipids, and blood. The leaked material infiltrates the layers of the retina, especially the macular layer, causing scar tissue formation and retinal cell failure. Drugs that inhibit neovascularization can slow the progression of wet AMD, but their use as a treatment method comes with several problems. Drugs such as anti-vascular endothelial growth factor (VEGF) inhibitors have been shown to be effective in suppressing the growth of abnormal blood vessels and reducing fluid leakage in the macula, but they are not effective for all patients. Some patients may have a poor response to medication, or the disease may recur after initial treatment.

[0008] Diabetic retinopathy is another example of visual impairment that causes vision loss without significantly damaging the brain's visual processing centers. Diabetic retinopathy is a complication of diabetes that affects the blood vessels of the retina, leading to damage and vision loss. Currently, there is no effective treatment for diabetic retinopathy, and available treatments aim to manage symptoms and slow disease progression. One of the challenges in treating diabetic retinopathy is that the disease is multifactorial, involving multiple mechanisms contributing to its onset and progression, including oxidative stress, inflammation, and abnormal vascular proliferation. Conventional treatments do not adequately address these mechanisms.

[0009] Retinitis pigmentosa (RP) is a group of genetic disorders in which the photoreceptor cells in the retina, which are responsible for sensing light and transmitting visual information to the brain, gradually deteriorate. This leads to progressive loss of peripheral vision, night blindness, and ultimately, loss of central vision.

[0010] Computer-assisted visual therapy is a treatment option for retinal degenerative diseases, including age-related macular degeneration (AMD) and retinitis pigmentosa. The goal of this therapy is to improve visual function and quality of life by stimulating healthy cells remaining in the retina and retraining the brain to interpret visual information more effectively.

[0011] Visual stimulation therapy is a rehabilitation treatment that uses specific visual stimuli (such as letters, numbers, and magnified images) to improve the visual function of patients with AMD. Its aim is to enhance the activity of remaining healthy retinal cells and compensate for the functional loss caused by AMD. Another approach, pulsed light therapy, stimulates the retina using specific light frequencies and intensities. This therapy is usually performed using specialized equipment that emits high-intensity light pulses at specific frequencies. Another method, electrotherapy, uses electrical stimulation to promote neural plasticity in the visual cortex. This therapy is usually performed using specialized equipment that delivers low-intensity electric currents to the retina. While visual stimulation therapy cannot restore lost vision, it can help patients adapt to vision loss caused by AMD.

[0012] Visual acuity testing is a common method for diagnosing and monitoring degenerative eye diseases such as age-related macular degeneration, diabetic retinopathy, and glaucoma. One representative visual acuity test is the Snerlen visual acuity test, in which the test subject stands at a set distance from a chart or video monitor with letters of various sizes and tries to read the smallest letters possible. Another visual acuity test is the Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity test. This is similar to the Snerlen visual acuity test but is designed to be more precise and sensitive (the chart has rows of letters, with each row consisting of progressively smaller and more difficult-to-read letters). Another visual acuity test specifically for macular degeneration is the Amsler grid test. This involves looking at a grid of straight lines and identifying areas where the lines appear distorted or missing, which may indicate macular damage. Glaucoma patients may undergo a visual field test, which involves identifying flashing lights in the peripheral vision while fixating on a central point. [Means for solving the problem]

[0013] The disclosed embodiments provide digital therapeutics for the treatment of retinal degenerative diseases such as age-related macular degeneration (AMD) and diabetic retinopathy. The disclosed systems and methods substantially train the visual pathway from the eye to the visual cortex.

[0014] For example, AMD is a type of neurodegenerative disease that leads to loss of central vision, resulting in blurred or wavy areas in the central field of vision. Diagnosis of AMD is usually made by dilated pupil examination. The proposed digital therapy is a neurofunctional therapy that focuses on improving the perception of remaining visual abilities, such as peripheral vision, in AMD patients, and aims to activate existing pathways by presenting new visual information. This is a form of digital training that can be performed on any screen, with or without neural feedback. In particular, it is possible to treat functional retina unrelated to lesions at the retinal and cortical levels, and at the cortical level, it is possible to treat the entire functional retina.

[0015] In the disclosed embodiments, visual pathways and cortical processing centers can be trained using empirically designed visual stimuli to improve peripheral vision perception. This includes presenting visual stimuli specifically designed to activate and enhance neural pathways and processing centers responsible for specific areas of the visual field (e.g., peripheral vision). The methods disclosed herein are also applicable to other areas of the cerebral cortex, such as the frontal eye field (FEF), ventromedial parietal lobe (VIP), fusiform gyrus, and parahippocampal gyrus. These cortical areas are involved in the context of visual attention, movement, and other visual perception. Therefore, these methods may not be limited to the peripheral vision, in which case the area of ​​target would be the peripheral part of the lesion site, but not necessarily limited to the peripheral vision. In the disclosed embodiments, stimuli may be designed to take advantage of the binocular nature, that is, the property of the retina of each eye to transmit information to its respective hemisphere. In such cases, a visual impairment of half of the visual field due to a retinal lesion in one eye can be compensated for by enhancing the function of the receptive cortical area through a visual stimulus presented to the other eye.

[0016] In the disclosed embodiments, the development of dynamic neuromodulatory stimulation (e.g., generative video) specifically aims at adaptation to peripheral functions, based at least partially on neurological goals related to occipital lobe processing. Peripheral retraining is determined using eye tracking, and the video generates information that enables improvement of peripheral functions. The system assists in retraining the eye's information acquisition methods, with saccades influencing this retraining. Generative artificial intelligence (AI) can be used to map and optimize peripheral vision functions, particularly visual functions around focal lesions (including color vision and motion perception in the peripheral vision). Because age-related macular degeneration (AMD) leads to the loss of cone cells, the system relies on rod vision and represents a form of neuroenhancement to achieve visual recovery. In the disclosed embodiments, the digital therapy can be individually personalized based on the type of AMD diagnosed to the patient.

[0017] In the disclosed embodiments, the area of ​​the focal lesion is determined based on at least partially measured neural activity, recognizing an active and functional area of ​​the retina. Depending on where the active and functional area is located in the retinal visual field, dynamic visual stimuli can be adapted to achieve a specific neural response in the visual cortex.

[0018] In the disclosed embodiments, the therapeutic approach broadly treats a variety of focal or diffuse lesions by mobilizing existing functional circuits and enhancing the effectiveness of those circuits, thereby resulting in acute and chronic long-term improvements in functional vision, including visual field expansion, improved visual acuity, and enhanced cognitive abilities. For example, the deterioration of vision in glaucoma is progressive, similar to AMD, but the pattern is reversed, usually affecting the periphery first, followed by the central area. Diabetic neuropathy (i.e., retinopathy) itself exhibits a progressive deterioration pattern. In a broad sense, the disclosed embodiments aim to enhance the function of the "processor" or "receiver," i.e., the brain, particularly the thalamus and cerebral cortex, to compensate for the defective "sensors," i.e., the retina and optic nerve.

[0019] The disclosed embodiments can make full use of the implementation of so-called “dandelion chains” and “content lenses” that are overlaid on other screen content, as disclosed, for example, in International Application PCT / US2021 / 049080 (filed September 3, 2021) for “Artificial Intelligence-based Visual Neuromodulation for Therapeutic or Performance Enhancement,” International Application PCT / US2022 / 77207 (filed September 28, 2022) for “System and Method for Generating Spatiotemporal Sensory Codes,” and U.S. Provisional Patent Application 63 / 324,395 (filed March 28, 2022) for “System and Method for Providing Dynamic Neuromodulation Graphics.” These disclosures are incorporated herein by reference.

[0020] In the disclosed embodiments, generative character recognition and object recognition can be used to train a system to read and identify objects such as traffic lights, faces, and locations in the peripheral vision. This training improves general vision and visual acuity by helping the system identify objects that the peripheral vision is not familiar with. The degree of success in visual acuity is assessed using standard measurement techniques and may include specific milestones or thresholds, such as a two-step improvement on an eye chart.

[0021] In the embodiment, the "Dandelion Chain" algorithm is used to generate and optimize optic nerve modulation codes to produce physiological responses that have therapeutic or performance-enhancing effects. These algorithms are implemented in two stages: an "inner loop" that optimizes optic nerve modulation codes through feedback from biomedical sensors to maximize therapeutic effects for individual subjects or groups of subjects, and an "outer loop" that uses various processing techniques to generalize the effects of the optic nerve modulation codes generated by the inner loop to the general user base.

[0022] In an embodiment, to maximize the possibility of discovering consistent responses among subjects, optimization may be performed at the group level. In this case, visual images are simultaneously presented to a group of subjects in the form of a dynamic visual nerve regulation code (e.g., in video format). The biological responses of the group of subjects are aggregated and analyzed in real time to determine which stimulus parameters (i.e., the parameters used to generate the visual nerve regulation code) are associated with the largest responses. The system readjusts and reconstructs the visual parameters to optimize the stimulus, quickly guiding the collective response of the group of subjects in the direction of a larger response. Such group optimization increases the likelihood of inducing a fine-grained graded response range with consistency among subjects. This substantially creates a new homunculus, i.e., a topographical representation of the eye along the postcentral gyrus of the parietal lobe. Thereby, the retinal and functional response data are integrated and can be mapped to the cortex. In other words, instead of simply making local circuit connections to the cortex, the system is performing functional mapping to the cortex.

[0023] The embodiment includes techniques such as transfer learning and ensemble learning using artificial intelligence (AI) such as machine learning models and neural networks (e.g., convolutional neural networks, deep feedforward artificial neural networks, adversarial neural networks) to develop better algorithms and generate generalizable treatment methods. Instead of trying to create a complete model of the brain, in effect, by "connecting" a vast number of users, treatment methods for a large subset of users are identified. Such treatment methods are equivalent to pharmaceuticals in terms of effectiveness in the general population. Therefore, the treatment methods developed in this way can be provided to patients without requiring sensor measurements of individual brain states or brain activities. This approach solves the problem of the generality of treatment methods and brings cost reduction and other efficiencies in the practical aspects of treatment provision.

[0024] Overall, digital therapeutics have great market potential because they provide a non-invasive, cost-effective, and individualized way to treat AMD and other retinal degenerative diseases.

[0025] In the disclosed embodiments, a method of treating an eye disease such as a retinal degenerative disease includes training a neural network (e.g., a generative neural network) that optimizes computer-based vision training in order to treat the retinal degenerative disease by targeting specific neural circuits in the visual cortex. This may include collecting a dataset of patients having different types of retinal degenerative diseases. The dataset may include information characterizing the patients' visual impairments (such as visual acuity, contrast sensitivity, visual field defects, etc.). Further, the method includes implementing a vision training program that presents visual stimuli (such as meaningful symbols, characters, other images, etc.) to the patient via a display device. The visual stimuli may be presented in various forms such as videos, animations, games, etc., or may be presented overlaid on such content. The method further includes constructing a neural network architecture that optimizes the vision training program based on each patient's unique visual impairment. The neural network is trained to identify optimal parameters for each patient, such as the size, spacing, orientation, contrast, luminance, and / or color of the visual stimuli. The neural network may be trained using a dataset of patients having different types of retinal degenerative diseases or may be trained with a focus on a specific disease. The network is trained to identify the relationship between the patients' visual impairments and the optimal vision training parameters. The neural network is verified by tests using a validation dataset of patients with retinal degenerative diseases (or a specific disease) and / or a dataset of patients with normal vision. The performance of the neural network is compared with that of a standard vision training program and its effectiveness is evaluated.

Brief Description of the Drawings

[0026] [Figure 1] A diagram showing an embodiment of a system that generates visual stimuli that produce a physiological response having a therapeutic effect or a performance improvement effect using brain activity data measured while displaying a dynamic visual nerve regulation code to a subject in a target state and a current state. [Figure 2]This flowchart shows one embodiment of a method for generating visual stimuli to produce a physiological response having a therapeutic or performance-enhancing effect, which can be used with the system shown in Figure 1. [Figure 3] This figure shows an embodiment of a system for transmitting visual neural regulatory codes generated in a closed-loop manner using an optimized description space. [Modes for carrying out the invention]

[0027] The following description includes certain details to allow for a full understanding of the various embodiments disclosed. However, those skilled in the relevant art will recognize that some or all of these specific details may be omitted, or that embodiments may be carried out using other methods, components, materials, etc. Furthermore, detailed illustrations and descriptions of well-known structures relating to computer systems, server computers, and / or communication networks have been omitted to avoid unnecessarily obscuring the description of the embodiments.

[0028] Unless otherwise required by context, the word “includes” throughout this specification and the subsequent claims is synonymous with “inclusive” and has an inclusive or open meaning (i.e., does not exclude additional elements or actions of methods not described). The expressions “one embodiment,” “embodiment,” or “specific embodiment” in this specification mean that a particular function, structure, or characteristic described in relation to that embodiment is included in at least one embodiment. Thus, the expressions “in one embodiment,” “in an embodiment,” or “specific embodiment” appearing throughout this specification do not necessarily all refer to the same embodiment. Furthermore, a particular function, structure, or characteristic can be combined in any suitable way in one or more embodiments.

[0029] As used herein and in the appended claims, the singular forms “a,” “an,” and “the” include the plural form unless the context clearly indicates otherwise. Furthermore, the term “or” is generally used to include “and / or” unless the context clearly indicates otherwise. The headings and abstracts of the disclosures described herein are for convenience only and do not constitute an interpretation of the scope or meaning of the embodiments.

[0030] Physiology is a branch of biology that deals with the functions and activities of living organisms or biomolecules (e.g., organs, tissues, cells), and related physical and chemical phenomena. This includes the various organic processes and phenomena of living organisms and their components, as well as specific bodily processes. Therefore, in this specification, the term “physiological” broadly means characteristic or appropriate to the function of an organism (including human physiology). This term encompasses the characteristics and functions of the nervous system, the brain, and all other bodily functions and systems.

[0031] The term “neurophysiology” refers to the physiology of the nervous system. The terms “neural” and the prefix “neural” also refer to the nervous system. In this specification, all these terms and prefixes refer to the physiology of the nervous system and the brain. In some cases, these terms and prefixes are used to refer to a more general physiology, including the nervous system, the brain, and the physiological systems that are physically and functionally related to the nervous system and the brain.

[0032] The visual cortex processes visual information from the retina topographically. That is, different regions of the visual cortex are specialized for processing different areas of the visual field. The part of the visual cortex responsible for peripheral vision processing is located in the posterior region, while the part responsible for central vision processing is located in the anterior region. Visual information is transmitted from the retina through the optic nerve to the lateral geniculate nucleus of the thalamus. From the thalamus, visual information is transmitted to the visual cortex via the optic radiation. Both the thalamus and the visual cortex contain multiple subregions that are geographically separated to allow for the individual processing of visual information from various areas of the visual field.

[0033] Peripheral and central vision are processed differently in the visual cortex. This is because the characteristics of the retinal ganglion cells that project to these areas differ. Peripheral retinal ganglion cells have large receptive fields and are sensitive to changes in contrast and motion. On the other hand, central retinal ganglion cells have smaller receptive fields and are more sensitive to fine details such as edges and textures. The visual cortex processes peripheral and central vision in different ways, using specialized cortical regions that are sensitive to the characteristics of the retinal ganglion cells that project to these areas.

[0034] The retinal circuit is responsible for lower-order visual processing, specifically the initial stages of visual image analysis. It extracts specific spatial and temporal characteristics from raw images obtained from both eyes and transmits them to higher-order visual centers. The rules of this processing are highly plastic. In particular, the retina must adjust its sensitivity to constantly changing lighting conditions. This adaptation allows our vision to remain relatively stable despite the enormous changes in light levels we encounter throughout the day.

[0035] Various aspects of dynamic visual information are distributed and processed across different cortical regions in the primate brain. For example, the middle temporal region (MT) and medial superior temporal region (MST) are specialized areas of the visual cortex that play a crucial role in processing visual information related to motion and depth perception. These regions are located in the dorsal visual pathway and are responsible for processing visual information related to the spatial position and motion of objects.

[0036] The MT region (also known as V5) is a region of the visual cortex specialized in processing visuomotor activity. It receives input from the primary visual cortex (V1) and processes information about the direction, velocity, and trajectory of visual stimuli. The MT region is involved in a variety of visual tasks, including tracking moving objects, perceiving shapes defined by motion, and distinguishing between the movements of different objects.

[0037] The MST region is located adjacent to the MT region and is involved in processing visual information related to the movement of objects in three-dimensional space and depth perception. The MST region receives input from the MT region and integrates information about visual motion, binocular disparity, and visual flow. The MST region is involved in various visual tasks, including depth perception, visual navigation, and self-motion perception. The MT and MST regions are important in processing visual information related to motion and depth perception. Dysfunction of these regions can cause visual impairments such as impaired motion perception and impaired depth perception.

[0038] The middle temporal lobe (MT) and medial superior temporal lobe (MST), located in the dorsal visual pathway, are specialized in visuomotor processing and are important in peripheral visual field processing. The MT and MST regions receive input from the primary visual cortex (V1) and integrate information about the direction, velocity, and trajectory of visual stimuli. In contrast, the part of the visual cortex responsible for central visual field processing is located in the anterior region of the cortex, including the primary visual cortex (V1) and the visual association areas. These regions are specialized in processing detailed information such as edges, texture, and color, and are important in central visual field processing.

[0039] In certain embodiments, simultaneous stimulation of multiple regions in the peripheral retina, along with the use of synchronous and asynchronous harmonization and similar or different image sets, enables retraining and / or overtraining of active neural regions. In some cases, by using combinations of image parameters such as angle, color, motion, brightness, image type (e.g., scene or face), or other components in the peripheral region (i.e., region eccentric to the scotoma (pathology)), the disclosed method can enhance functionality and produce improved function. These parameters are applicable to specific regions, such as the circumferential region formed by the eccentric ring, further enhancing the peripheral visual field's ability to recognize visual details and improving function and cortical interpretation. In a normal physiological state, most scene reconstruction occurs in the fovea (i.e., within 1-2 degrees from the periphery). By training peripheral regions outside the scotoma, circuits that complement the function of the nearby peripheral can be activated, leading to improved function.

[0040] Age-related macular degeneration (AMD) is a disease that affects the macula, a part of the retina responsible for central vision. AMD causes irreversible damage to the macula, leading to loss of central vision. However, in most cases, peripheral vision is not impaired. Inducing accommodation in the visual cortex regions that control peripheral vision can help AMD patients by improving the processing of visual information from healthy peripheral vision.

[0041] Computer-based visual therapy uses specialized software programs and hardware devices to present patients with visual stimuli. These stimuli are designed to target specific areas of the retina and visual cortex and can be customized based on the individual patient's needs and goals. There are several approaches to computer-based visual therapy. Visual acuity training improves visual acuity by training the eyes to recognize and interpret visual information more effectively through a series of exercises and games. Contrast sensitivity training aims to improve the ability to see objects in low-contrast environments, a common impairment in patients with retinal degeneration. Visual field testing training aims to improve the ability to see objects in the peripheral vision, which is particularly important for patients with diseases affecting central vision, such as age-related macular degeneration (AMD). Motion detection training aims to improve the ability to detect and interpret motion, which is a challenging task for patients with retinal degeneration.

[0042] The visual cortex is the part of the brain that processes visual information. It is divided into specialized areas that process different aspects of visual information. The primary visual cortex (V1) is the cortical area that first receives input from the retina. The visual cortex also includes specialized areas that are responsible for processing peripheral vision, such as the middle temporal lobe (MT) and the medial superior temporal lobe (MST).

[0043] In the disclosed embodiments, modulation of the visual cortex using dynamic neural modulation codes modulates the activity of neural circuits within the visual cortex, resulting in changes in visual function. Displaying and viewing dynamic neural modulation codes is a non-invasive method of eliciting a neural modulation response in the brain (e.g., the visual cortex). This therapy targets parts of the visual cortex responsible for peripheral vision processing, such as the MT and MST regions. By modulating the activity of these regions, the techniques disclosed herein improve the processing of visual information from healthy peripheral vision, resulting in improved visual function. Eliciting a modulating effect on peripheral vision visual cortex circuits helps AMD patients optimize, enhance, and / or restore function by improving the processing of visual information from healthy peripheral vision. Thus, driving neural stimulation activity effectively helps retrain parts of the visual cortex for adaptation. This therapy involves modulating the activity of neural circuits in the visual cortex using dynamic neural modulation codes to elicit changes in visual function. This therapy is non-invasive and has the potential to improve the quality of life for AMD patients.

[0044] The disclosed embodiments provide a treatment for age-related macular degeneration that involves modulation of the visual cortex region responsible for peripheral vision. This treatment uses dynamic neural modulation codes to induce modulation of the visual cortex region responsible for peripheral vision. It is a non-invasive technique that stimulates the brain by presenting the user with dynamic neural modulation codes (DNCs). This treatment is designed to activate neural circuits in the visual cortex responsible for processing peripheral vision. The treatment is delivered using a mobile device capable of displaying one or more DNCs, for example, as an overlay on video or other display content.

[0045] Providing dynamic neuromodulation imaging via mobile devices offers several potential advantages as a treatment for age-related macular degeneration (AMD). Mobile devices are widely available and can be used at the patient's home, reducing the need for frequent hospital visits. The treatment is non-invasive and can be easily customized to the patient's needs and preferences.

[0046] Figure 1 shows one embodiment of a system 600 that generates visual stimuli. This system uses brain activity data measured while displaying dynamic optic neural control codes to a subject 605 in a target state and a current state to generate physiological responses that have therapeutic effects, performance-enhancing effects, and effects that promote recovery and rehabilitation. The system 600 is processor-based and may include a networked computer system / server 610, or other types of computer systems with at least one processor and memory / storage. The memory / storage stores processor-executable instructions and data, which, when executed by at least one processor, perform the necessary functions of the system to generate and provide visual stimuli to the user.

[0047] In certain embodiments, the computer system / server 610 is connected to a network 625 to a number of personal electronic devices 630, such as mobile phones and tablets, and to the computer system. Dynamic visual neural control codes are generated based on feedback from one or more users and are used as visual stimuli to produce physiological responses that have therapeutic or performance-enhancing effects as described above.

[0048] System 600 receives a first brain activity dataset measured using a first test apparatus 650, for example, a display 610 and a brain activity measurement device 615, while the subject 605 is in a target state (for example, a state characterized by neurally modulated stimulation of a portion of the visual cortex that controls peripheral vision). The target state is induced by giving the subject 605 a known stimulus or a combination of stimuli. This stimulus may be in the form of a dynamic visual neural modulus code, as described above, and / or various other forms of stimulation, such as visual, image, chemical, or physical.

[0049] Therefore, the initial brain state / activity dataset serves as a baseline for comparison with other measured brain / activity datasets to evaluate the effect of a particular visual stimulus on achieving a desired state. Brain activity data may include, among other things, data obtained from one or more of the following: electroencephalography (EEG), quantitative electroencephalography (qEEG), magnetoencephalography (MEG), single-photon emission computed tomography (SPECT), positron emission tomography (PET), functional magnetic resonance imaging (fMRI), functional near-infrared spectroscopy (fNIRS)—measurements taken while the subject is in a facility equipped to perform such measurements (e.g., a facility equipped with the first test apparatus 650). Various other types of physiological and / or neurological measurements may be used. These types of measurements may be performed in combination with induced target states, as the subject's time in the facility is likely to be limited.

[0050] The system further uses an electronic display 610 to display candidate dynamic visual neural control codes to subject 605 while the subject is in the current state. The current state is different from the target state. For example, in the subject's current state, stimulation of neural circuits in the visual cortex region responsible for central vision is limited, whereas in the target state, neural control stimulation of the visual cortex region responsible for peripheral vision is characteristic. In certain embodiments, candidate dynamic visual neural control codes are at least partially based on a plurality of initial dynamic visual neural control codes. These initial codes are iteratively generated based on feedback data showing the responses of subject groups when one or more initial dynamic visual neural control codes are displayed to multiple subject groups.

[0051] While the system 600 displays candidate dynamic optic neural control codes to the subject 605, it receives a second brain activity dataset measured using a second test apparatus 660, which includes, for example, a display 610 and various types of brain state and / or brain activity measurement devices 615. As above, the brain activity data may include, among other things, data acquired from one or more of the following: electroencephalography (EEG), quantitative electroencephalography (qEEG), magnetoencephalography (MEG), single-photon emission computed tomography (SPECT), positron emission tomography (PET), functional magnetic resonance imaging (fMRI), and functional near-infrared spectroscopy (fNIRS). Brain imaging includes functional imaging (see examples above) and / or structural imaging (e.g., MRI). In certain embodiments, both the first and second brain activity datasets are acquired using the same test apparatus, i.e., the first test apparatus 650 or the second test apparatus 660.

[0052] System 600 analyzes a first brain state / activity dataset (i.e., target state data) and a second brain state / activity dataset to generate at least one parameter indicating the effectiveness of a candidate dynamic visual neural modulation code for subject 605. For example, feedback can be obtained from subject 605 by performing one or more standard visual tests (e.g., visual acuity tests, contrast sensitivity tests, visual field tests) on subject 605 during the target state (i.e., desired state) and the current state. Furthermore, various types of measurement feedback data (i.e., in addition to the image data described above) can be acquired while subject 605 is in the target state and / or the current state. Analysis of such information can provide parameters and / or statistical information indicating the effectiveness of a candidate dynamic visual neural modulation code for the subject.

[0053] Based on at least some parameters and statistics indicating the effectiveness of candidate dynamic optic neural control codes, system 600 either outputs a candidate dynamic optic neural control code as a visual stimulus or performs further iteration. In the latter case, the candidate dynamic optic neural control code is disturbed (i.e., algorithmically modified, adjusted, adapted, randomized, etc.). In certain embodiments, the disturbance of the candidate dynamic optic neural control code is performed using machine learning models, neural networks, convolutional neural networks, deep feedforward artificial neural networks, adversarial neural networks, and / or ensembles of neural networks. The operation of displaying the candidate dynamic optic neural control code to the subject is repeated, and the system receives further brain activity datasets measured during the code display. Analysis is performed again to determine whether to output the candidate dynamic optic neural control code as a visual stimulus or perform further iteration.

[0054] In certain embodiments, the system may generate candidate dynamic visual neural codes from a set of “basic” dynamic visual neural codes. In this case, the system iteratively generates basic dynamic visual neural codes with randomized properties such as texture, color, and geometry. The system obtains and analyzes the neural responses to the basic dynamic visual neural codes. For example, the codes are presented to a subject or group of subjects, and feedback data such as functional magnetic resonance imaging (fMRI) data is obtained. Based at least part of the analysis results of the neural responses to the base dynamic visual neural codes, the system outputs the base dynamic visual neural codes as candidate dynamic visual neural codes, or iterates through one or more of the base dynamic visual neural codes. In certain embodiments, the disturbance of the basic dynamic visual neural codes is performed using one or more of the following: machine learning models, neural networks, convolutional neural networks, deep feedforward artificial neural networks, adversarial neural networks, and ensembles of neural networks.

[0055] In the disclosed embodiments, the treatment protocol using DNC may include patient selection and evaluation, in which case patients with age-related macular degeneration (AMD) without peripheral vision impairment would be eligible candidates for this treatment. The patient's medical history and visual function would be evaluated to determine eligibility for treatment.

[0056] Before treatment, high-resolution diagnostic images, such as functional magnetic resonance imaging (fMRI) scans of the patient's brain accompanied by visual stimulus presentation, may be obtained. fMRI scans are used to map the areas of the visual cortex that control peripheral vision.

[0057] The patient is presented with a DNC (Diagnostic Neural Code) specifically designed to induce a neurally modulated response in the visual cortex region responsible for peripheral vision. For example, the DNC can be displayed on a mobile device or other screen device. The DNC is designed to target specific brain regions identified by fMRI mapping. In one embodiment, for example, peripheral vision testing can be performed on a smartphone screen to determine the degree of central vision loss and identify areas where visual training should be focused. For example, these areas can be identified by having the patient tap on areas that appear clearly to them on the screen.

[0058] In the examples, treatment is provided in multiple sessions per week over several weeks to several months. The frequency and duration of treatment are determined based on the patient's response to the treatment. The patient's visual function is monitored throughout the treatment period using standard visual function tests. The progress of treatment is evaluated based on the degree of improvement in visual function. This treatment is non-invasive and does not require surgical intervention. This treatment is indicated for improving visual function in AMD patients, particularly for improving mobility in the environment.

[0059] From the above perspectives, modulation of the visual cortex using DNC is a promising treatment for age-related macular degeneration. This treatment targets the visual cortical region that controls peripheral vision and is expected to improve the visual function of patients with this disease.

[0060] In this embodiment, dynamic visual stimuli are used to train a patient's peripheral vision, enabling them to perform functions normally handled by the central vision. This training involves presenting dynamic visual stimuli such as DNCs. These stimuli are designed to stimulate neural circuits in the visual cortex that control peripheral vision, prompting the brain to adapt and reorganize to perform tasks normally handled by the central vision.

[0061] Examples may include training in eccentric vision. Age-related macular degeneration (AMD) causes central scotoma, or loss of central vision. Patients with AMD can learn to utilize peripheral vision (also known as eccentric vision) to compensate for the loss of central vision. DNC-based vision therapy provides dynamic neural regulatory codes (DNCs) adapted to stimulate the neural circuits in the visual cortex that control peripheral vision. This can increase the use of eccentric vision and potentially improve the patient's visual function.

[0062] AMD slows down the processing speed of visual information, making rapid information processing difficult. DNC-based visual therapy can provide training that combines dynamic neural regulatory codes (DNCs), which stimulate the visual cortical neural circuits that control peripheral vision, with exercises that improve visual processing speed, such as visual search tasks and reaction time training. AMD has a significant impact on an individual's quality of life. DNC-based visual therapy can provide vision-related quality of life assessments that measure the impact of AMD on an individual's daily activities, such as reading, driving, and social activities.

[0063] DNC-based visual therapy is highly convenient for AMD patients because it can be performed at home. Home-based treatment allows patients to conduct visual training according to their own schedule and in a comfortable environment.

[0064] In one embodiment, the training protocol or method may include patient selection and assessment. This assessment evaluates patients with age-related macular degeneration or other diseases affecting the macula to determine the degree of visual impairment and eligibility for treatment. The method further includes generating dynamic visual stimuli, such as dynamic neural regulatory codes (DNCs), to stimulate neural circuits in the visual cortex that control peripheral vision. The images are adapted to provide the greatest effect in stimulating target areas of the visual cortex, as described later.

[0065] This method may further include peripheral vision training. In this training, the patient is presented with a DNC (Directed Visual Note) while performing tasks that are normally performed using central vision, such as reading, face recognition, and navigating a virtual environment. The stimuli are designed to encourage the patient to use their peripheral vision to perform these tasks. This method may further include progressive adaptation. The training protocol is adapted to gradually increase the complexity and difficulty of the tasks, encouraging the patient to rely more on their peripheral vision to perform them. This method may further include monitoring progress. The patient's visual function is monitored throughout the training period using standard visual function tests. Progress in the training protocol or method is assessed based on the improvement in visual function experienced by the patient.

[0066] Training patients to perform functions normally handled by central vision using dynamic visual stimuli has several potential benefits as a treatment for age-related macular degeneration (AMD). This training promotes neural plasticity (i.e., the functional effectiveness of existing circuits through adaptive training, resulting in enhanced neural function) and adaptation in the visual cortex, preventing atrophy of the visual cortex and encouraging the brain to reorganize to compensate for the loss of central vision. Furthermore, this training can be customized to the individual needs and preferences of each patient.

[0067] Modification of the visual cortex regions responsible for peripheral vision is a promising treatment for age-related macular degeneration (AMD). This treatment uses DNC to modulate neural circuit activity within the visual cortex, leading to changes in visual function. The following are aspects of modifying the visual cortex regions responsible for peripheral vision as a treatment for AMD. One aspect is patient selection and evaluation to identify AMD patients with unimpaired peripheral vision as eligible candidates for this treatment. The patient's medical history and visual function are evaluated to determine treatment eligibility. Another aspect is imaging and mapping of the patient's brain, for example, using functional magnetic resonance imaging (fMRI). A high-resolution MRI scan of the patient's brain is obtained before treatment. The fMRI scan is used to map the portion of the visual cortex responsible for peripheral vision.

[0068] Based on this mapping, DNCs are generated and / or selected that are adapted to produce a neurally modulated response in the visual cortex regions responsible for peripheral vision. The DNCs are adapted based on the neurophysiology of individual patients or used to generate a generalized DNC library based on neurophysiological measurements performed on a population of subjects.

[0069] Treatment is administered over several weeks in multiple sessions per week. The frequency and duration of treatment are determined, at least in part, based on the patient's response to the treatment. The patient's visual function is monitored throughout the treatment period using standard visual function tests. Treatment progress is evaluated based on improvements in visual function. After the initial treatment period, maintenance sessions may be necessary to maintain the treatment effect. The frequency and duration of these sessions are determined, at least in part, based on the patient's response to the initial treatment.

[0070] In this embodiment, modulation of the portion of the visual cortex responsible for peripheral vision may be used in combination with neural feedback to treat AMD. This therapeutic approach involves using a combination of DNC and neural feedback to improve the processing of visual information from healthy peripheral vision. As mentioned above, patient selection and evaluation may be performed, including assessment of the patient's medical history and visual function, to determine eligibility for treatment. Before treatment, it is possible to obtain a high-resolution MRI scan of the patient's brain and map the region of the visual cortex responsible for peripheral vision.

[0071] The DNC (Dynamic Neural Control) is delivered to the patient, for example, via a mobile device screen display, and is specifically designed to induce modulation in the visual cortical areas responsible for peripheral vision. Neural feedback can be used to provide real-time feedback on the patient's neurally modulated responses and visual processing. This feedback can be provided using various techniques, including electroencephalography (EEG), magnetoencephalography (MEG), and functional magnetic resonance imaging (fMRI). Treatment may consist of several sessions per week over several weeks. The frequency and duration of treatment are determined, at least in part, based on the patient's response. Standard visual function tests can be used to monitor the patient's visual function throughout the treatment period. Treatment progress can be evaluated based on the degree of improvement in visual function.

[0072] The combination of DNC administration and neural feedback has several potential advantages as a treatment for AMD. DNC modulates the activity of neural circuits in the visual cortex, leading to changes in visual function. Neural feedback provides real-time feedback on the patient's neurophysiology and visual processing, which can enhance the therapeutic effect. Various imaging techniques such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and magnetoelectroencephalography (MEG) may be used to measure the subject's neural modulatory response to dynamic visual stimuli and optimize the stimulation using this response as feedback. This allows monitoring of the subject's brain activity while viewing dynamic visual stimuli.

[0073] The treatment protocol or method can be described as follows: In the selection and evaluation of subjects, subjects with age-related macular degeneration or other conditions affecting the macula are evaluated to determine the degree of visual impairment and eligibility for treatment. Dynamic visual stimuli (e.g., DNCs) are generated using computer algorithms that produce visual stimuli designed to induce neuromodulatory responses in the neural circuits of the visual cortex that control peripheral vision. DNCs are adapted to maximize their effect in stimulating target areas of the visual cortex. While subjects view the dynamic visual stimuli, neural activity can be monitored using various imaging techniques such as fMRI, EEG, and MEG. These techniques provide real-time feedback on neural activity in the visual cortex, enabling optimization of the dynamic visual stimuli.

[0074] Feedback obtained from imaging diagnostics is used to optimize dynamic visual stimuli. This process may involve repeated adjustments of visual stimuli or improvements in effectiveness through changes in stimulus content. Optimizing dynamic visual stimuli using neural feedback has several potential advantages as a treatment for age-related macular degeneration. The feedback provides real-time information on neural activity in the visual cortex, allowing for the optimization of stimuli to produce the greatest effect. This method can also be customized to the individual needs and preferences of the subject.

[0075] Dynamic visual stimuli, such as dynamic neural modulation codes (DNCs), can be generated and adapted using high-resolution neuroimaging systems such as functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) systems. These systems provide real-time information on neural activity in the visual cortex, enabling the creation and optimization of visual stimuli tailored to the individual needs and preferences of the subject. DNCs are optimized to maximize the effect of stimulating neural modulation responses in target areas of the visual cortex. During DNC generation (e.g., in video format) and use, the subject's neural activity can be monitored using high-resolution neuroimaging systems such as fMRI and MEG, allowing for adaptation to specific applications and performance optimization.

[0076] As mentioned above, age-related macular degeneration (AMD) is a disease that affects the macula, the part of the retina responsible for central vision. This disease causes irreversible damage to the macula, leading to loss of central vision. By combining DNC-based visual therapy with medication, it is possible to slow the progression of the disease and improve visual function. Drug therapy is usually used to slow the progression of wet-type AMD and maintain visual function. Drugs commonly used to treat AMD are anti-vascular endothelial growth factor (VEGF) inhibitors such as bevacizumab, ranibizumab, and aflavelcept. These drugs are injected intraocularly and can suppress the proliferation of abnormal blood vessels and fluid leakage in the macula that cause wet-type AMD.

[0077] DNC-based visual therapy is used to improve the processing of visual information in healthy peripheral vision. This therapy involves providing a dynamic neural regulatory code (DNC) adapted to stimulate the neural circuits in the visual cortex that control peripheral vision. This therapy can improve visual function in healthy peripheral vision and compensate for central vision loss due to disease.

[0078] A combination of DNC-based vision therapy and medication can help slow the progression of AMD and improve visual function. The medication reduces abnormal vascular growth and fluid leakage in the macula, while DNC-based vision therapy improves the processing of visual information in healthy peripheral vision, which can compensate for the loss of central vision caused by the disease.

[0079] In addition to drugs aimed at suppressing progressive retinal degeneration, another embodiment includes a combination of drugs that enhance visual cortical function and visual DNC-based visual therapy. For example, local circuits in the visual cortex involved in peripheral vision utilize ACh neurotransmission via muscarinic acetylcholine (ACh) receptors and can be enhanced by the use of acetylcholinesterase inhibitors (AChIs) such as donepezil or muscarinic agonists such as arecoline. Combining such pharmacological interventions with DNC-based neurophysiological interventions to enhance attention to peripheral vision is envisioned as an effective method for improving residual visual acuity and compensating for lost vision. Furthermore, other therapeutic combinations of DNC-based visual therapy with drugs that act on glutamate and gamma-aminobutyric acid (GABA) neurotransmission are also envisioned.

[0080] The disclosed embodiments provide a non-invasive and cost-effective method for treating AMD. This method involves stimulating a neuromodulatory response in the visual cortex and improving visual function using dynamic optic neuromodulatory stimulation generated by the screen of a digital device. This may include an assessment of visual function before the initiation of treatment. This assessment evaluates the patient's visual function using standard tests such as visual acuity tests, contrast sensitivity tests, and visual field tests. These test results are used to determine the severity of the disease and monitor the progress of treatment.

[0081] Based at least partly on an assessment of visual function, a customized optic neuromodulatory stimulation program is designed for each patient. This program is designed to stimulate neuromodulatory responses in specific areas of the visual cortex (e.g., areas related to peripheral vision) by displaying the DNC on the screen of a digital device (e.g., the user's mobile device).

[0082] In the embodiment, the patient is instructed to perform a visceral neuromodulatory stimulation program for a predetermined time (usually 20-30 minutes per session) daily. The patient performs the visceral neuromodulatory stimulation program using a digital device such as a tablet or smartphone. This includes displaying one or more DNCs on the device screen, either alone or superimposed on other display materials. The patient's visual function is regularly monitored during treatment to evaluate the effectiveness of the visceral neuromodulatory stimulation program. Visual function tests are repeated periodically to monitor the progress of treatment.

[0083] Among other advantages, the embodiments disclosed herein provide a non-invasive and cost-effective method for treating age-related macular degeneration (AMD). This method is easy to implement and does not require the use of invasive procedures. It is convenient for patients as it can be performed at home. Furthermore, it can be customized to each patient, providing personalized treatment.

[0084] Dynamic neural control images, such as videos, can be used to induce targeted neural control responses in a patient's visual cortex. This is achieved by using mobile devices such as smartphones and tablets to present visual neural control stimuli designed to stimulate specific areas of the visual cortex responsible for peripheral vision. Dynamic neural control images are created using computer algorithms that generate visual neural control stimuli designed to stimulate neural circuits in the visual cortex responsible for peripheral vision. The images are adapted to maximize their effect in stimulating the target area of ​​the visual cortex.

[0085] Dynamic neuromodulatory images are presented to the patient using a mobile device such as a smartphone or tablet. The patient views the images at predetermined times and frequencies determined by the treatment protocol or method. The patient's visual function is monitored throughout treatment using standard visual function tests. Treatment progress is evaluated based, at least in part, on improvements in visual function.

[0086] Dynamic neuromodulatory imaging can be presented to patients in various formats, including videos, animations, and games. Visual stimuli are designed to activate neural circuits in the visual cortex that control peripheral vision, promoting plasticity (i.e., the functional effectiveness of existing circuits through adaptive training, resulting in enhanced neural function) and adaptation in these circuits.

[0087] Figure 2 shows one embodiment of method 900 for generating and providing to a user visual stimuli that induce a physiological response having a therapeutic or performance-enhancing effect. The disclosed method can be used in a system such as the one shown in Figure 1 above.

[0088] Method 900 includes step 920 of displaying candidate dynamic visual neural modulation codes to a subject (using an electronic display) while the subject is in a current state, which is distinct from a target state characterized by a neural modulation stimulation of a target area of ​​the visual cortex. Method 900 further includes step 930 of receiving a first brain activity dataset measured while the subject is being displayed the candidate dynamic visual neural modulation codes. Method 900 further includes step 940 of analyzing a second brain activity dataset measured while the subject is in a target state and the first brain activity dataset to generate at least one parameter indicating the effectiveness of the candidate dynamic visual neural modulation codes for the subject.

[0089] Based at least partially on at least one parameter indicating the effectiveness of a candidate dynamic visual neural control code, the method further performs one of the following (950): (i) output the candidate dynamic visual neural control code as a visual stimulus (970); (ii) disturb the candidate dynamic visual neural control code, repeatedly display the candidate dynamic visual neural control code to a subject, receive a first brain activity dataset measured during the display of the candidate dynamic visual neural control code to the subject, and analyze the second brain activity dataset and the first brain activity dataset (960).

[0090] Figure 3 shows an embodiment of a system 300 that delivers optic nerve-coordinate codes generated in a closed-loop manner using an optimized description space. The system 300 includes electronic devices such as mobile devices (e.g., cell phones and tablets) and virtual reality headsets (referred to herein as user devices 310). Patients view the optic nerve-coordinate codes on a user device such as a smartphone or tablet by using an app or by streaming them from a website. In the embodiment, the app or web-based software enables the therapeutic optic nerve-coordinate codes to be integrated (e.g., overlaid) with content displayed on the screen (e.g., a website displayed in a browser, the user interface of an app, or the user interface of the device itself) without interfering with the normal use of the content. Thus, the disclosed embodiment provides a dynamic lens or filter-like function between the displayed content and the viewer.

[0091] In the embodiment, the system can be adapted to personalize the visual neural modulation codes using sensors and data from a user device (e.g., a smartphone). For example, the user device may provide eye-tracking and pupil dilation measurement using the user device's camera. Furthermore, the user device can present the patient with vision-related tests and questionnaires developed using artificial intelligence, and automatically personalize the visual neural modulation codes and exposure time to obtain the optimal therapeutic effect.

[0092] The user device 310 comprises at least one processor 315 and memory 1420 (e.g., random access memory, read-only memory, flash memory, etc.). Memory 320 contains a non-temporary medium readable by the processor and adapted to store instructions that the processor can execute. When executed by the processor 315, these instructions cause the processor 315 to perform a method of delivering visceral neural control codes. The user device 310 has an electronic display 325 adapted to display images rendered and output by the processor 315.

[0093] The user device 310 also has a network interface 330, which is implemented as a hardware and / or software-based component and includes wireless network communication capabilities (e.g., Wi-Fi or a cellular network). The network interface 330 is used to acquire one or more adaptive optic nerve modulatory codes adapted to produce a physiological response having a therapeutic or performance-enhancing effect 335. In some cases, the optic nerve modulatory codes may be acquired in advance and stored in the memory 320 of the user device 310.

[0094] In embodiments, the acquisition of adapted visual neural control codes (e.g., via network interface 330) may involve communication over a network (e.g., wireless network 340). This communication takes place with a server 345, which is configured as a computing platform having one or more processors and memory for storing data and program instructions executed by the group of processors. (Internal components of the server are not shown). The server 345, like the user device 310, includes a network interface. This can be implemented as hardware and / or software-based components such as a network interface controller or card (NIC), a local area network (LAN) adapter, or a physical network interface. In embodiments, the server 345 may provide a user interface for operating and controlling the acquisition of visual neural control codes.

[0095] The processor 315 outputs visual neural control codes to the display 325 that are adapted to produce a physiological response having a therapeutic or performance-enhancing effect in the user 335 viewing the display 325. The visual neural control codes may be generated by any of the methods disclosed herein. In this way, the visual neural control codes are presented to the user 335, and a therapeutic or performance-enhancing effect is realized. When outputting adapted visual neural control codes to the display 325 of the user device 310, each displayed visual neural control code, or a sequence of visual neural control codes (i.e., visual neural control codes displayed in a predetermined order), may be displayed for a predetermined time. These functions substantially provide the ability to set individually prescribed “dosages” for each user in a manner similar to that of prescription drugs. In embodiments, the determined display time of the adapted visual neural control codes may be adapted based on user feedback data indicating the user 335’s response. In some embodiments, the output of the adapted visual neural code may include overlaying the visual neural code onto displayable content (e.g., the displayable output of an application running on a user device, the displayable output of a browser running on user device 310, the user interface of user device 310, etc.).

[0096] The user device 310 has a short-range wireless communication interface 350, such as Bluetooth (trademark registered). This is for communicating with devices in the vicinity of the user device 310 (for example, a sensor (e.g., 360) that measures the physiological response of subject 335 while a visual neural modulation code is presented to subject 335). In embodiments, the sensors may include components of the user device 310 itself. These may acquire feedback data, for example, by tracking eye movements or receiving input to displayed prompts.

[0097] As described above, an application or web-based software running on the user device 310 enables the integration (e.g., overlay) of therapeutic optic nerve modal codes with content displayed on the screen (e.g., a website displayed in a browser, the user interface of an application, or the user interface of the device itself). This allows therapeutic optic nerve modal codes to be provided without interfering with the normal use of the content. In one embodiment, the user device 310 presents a combination of displayable content and adaptive optic nerve modal codes on the display 325. This allows the user to receive treatment with adaptive optic nerve modal codes while simultaneously viewing displayable content such as application output or web pages displayed in a web browser. This method reduces the burden on the user because the treatment is performed while the user is paying attention to the normal functions of the device 310. Furthermore, since this method can be integrated into existing devices, users can receive treatment without purchasing dedicated hardware, i.e., hardware devices specifically designed for treatment.

[0098] In the disclosed embodiments, the dynamic neural modal code may take the form of semantic content such as letters, numbers, or other types of graphic images. In this case, the semantic content may be adapted in a manner similar to that of adaptations of non-graphic and / or non-semantic images, as disclosed herein. Furthermore, semantic images may be used as components of visual acuity tests and visual acuity training. This includes training a neural network (e.g., a generative counter-network (GAN)) with visual acuity training data to adapt the visual acuity training to achieve more effective and targeted results. The training is based on neural goals, as described above with respect to the dynamic neural modal code (DNC) and / or behavioral feedback such as the subject's responses.

[0099] More generally, neurological goals of semantic-based visual training include neural activation time and intensity, as measured by fMRI. In embodiments, GANs can be trained to optimize activation rate, encoded neural location, and behavioral response. In embodiments, target neural responses can be generated using a combination of DNCs and semantic images. In some cases, content such as DNCs in video format can be overlaid on other screen content using the “content lens” implementation described later. Semantic and other figurative visual stimuli, used alone or in combination with DNC-based stimuli, can lead to general sensitization training and excitation of visual processing, improving overall vision. These methods can be adapted to specific types and severity of visual disorders, at least partially based on measured neural responses. In fact, in certain embodiments, these methods are independent of the type and severity level of the disorder. For example, instead of relatively simple visual training such as tracking a dot displayed on a screen, a trained neural network can present combinations of geometric shapes, letters, numbers, symbols, and / or symbols to maximize the effectiveness of visual training. This can be combined with non-semantic approaches using DNCs generated by the processes disclosed herein. This approach to visual acuity training and testing is more effective because neural networks optimize the output of visual acuity tests (e.g., letters, dots, numbers) and improve visual acuity training. This can lead to improved and accelerated treatment. Therefore, evaluations of treatment effectiveness based on visual acuity testing (e.g., evaluations by the U.S. Food and Drug Administration (FDA)) will reflect these improved capabilities. The neural network also optimizes parameters related to visual dynamics, such as the speed, rotation, and color of the presented semantic content (e.g., letters). The presentation of semantic content controlled by these dynamic parameters yields better therapeutic effects than conventional visual acuity tests that are not adapted to specific neurological and / or behavioral goals. After acquiring or generating the adapted visual neural modulation codes as described above, the adapted visual neural modulation codes are combined with displayable content to form one or more dynamic neural modulation composite images.The combination of adapted visual neural modulation codes and displayable content includes the implementation of image overlays using techniques such as pixel addition, multiplication blending, screen blending, and alpha blending. Various other image overlay techniques are also available. Specific overlay techniques can be selected based on a subjective evaluation of the appearance of the dynamically synthesized image (e.g., clarity, brightness, contrast, etc.). Alternatively, overlay techniques can be selected by comparing the effects of the resulting dynamically neural modulation composite image on the subject.

[0100] Alpha blending, an example of image overlay technology, is the process of combining one image with a background to represent partial or complete transparency. For each image element (i.e., pixel), color information is stored, such as a combination of red, green, and blue. Each pixel is also associated with a numerical value α, which varies from 0 to 1. This is called the "alpha channel." A pixel with a value of 0 is completely transparent, and the color of the underlying pixel is visible. A pixel with a value of 1 is completely opaque.

[0101] The existence of an alpha channel allows us to represent image compositing operations using compositing operations. For example, given two images A and B, the most common compositing operation is to combine the images so that A is displayed in the foreground and B in the background. This is expressed as AoverB. Specifically, the over operator can be implemented by applying the following formula to each pixel. α_0 = α_a + α_b (1 - α_a) C_0=(C_a α_a+C_b α_b (1-α_a)) / α_0 Here, C0, Ca, and Cb represent the color components of the pixels in the resulting image, image A, and image B, respectively, and are applied individually to each color channel (red / green / blue). Also, α_0, α_a, and α_b are the alpha values ​​of each pixel. In "premultiplied alpha," the RGB components are multiplied by the corresponding alpha values. This represents the emission (alpha values ​​represent occlusion) of the object or pixel. In this case, the color components are as follows: C_0 = C_a + C_b (1 - α_a) The composite image is output to the display 325 by the processor 315. Displayable content includes displayable outputs of applications, browsers, and / or user interfaces of user devices. Each dynamic neural modulation composite image is displayed for a predetermined time period. This time can be adapted based on user feedback data (e.g., feedback data indicating the user's neurological and / or physiological responses).

[0102] In one embodiment, the acquisition or generation of adapted visual neural code, and the coupling of adapted visual neural code with displayable content, may be performed, at least in part, by the graphics processing unit (GPU) (not shown) of the user device 310. This allows the processor 315 of the user device 310 to operate without being burdened with additional processing tasks.

[0103] The user device 310 can acquire user feedback data, such as feedback data indicating the user's neurological and / or physiological responses, during the process of outputting a dynamic neural modulation composite image to the electronic display 325. User feedback data can be acquired, for example, by tracking eye movements using components of the user device 310 and / or by receiving input to displayed prompts. Various other types of components can be used to measure various types of user feedback data. In some cases, user feedback data can be acquired by receiving data from wearable neural sensors.

[0104] In one embodiment, the output from a sensor measuring the user's eye movements can be received while the user device 310 is outputting a visual neural control code. For example, the front camera of the user device 310 can be used as an eye movement tracking sensor. In this case, the processor 315 runs software that analyzes images and / or videos captured by the front camera to determine the position of the user's eyes and track their movement. Other types of sensors and measurement techniques can also be used to perform these functions. In some cases, the facial recognition hardware and software components of the user device 310 perform or assist in performing eye movement tracking.

[0105] Based on the analysis results of sensors that measure the user's eye movements, the user's visual focal point on the electronic display 325 can be identified. Based on the identified visual focal point of the user on the electronic display 325, values ​​for a series of adaptive rendering parameters can be calculated. In this case, the adaptive rendering parameters effectively shift one or more primary reference points of the displayed visual neural control codes to match the user's visual focal point, thereby ensuring that the user's attention is focused on the most effective parts. For example, the reference points of the displayed visual neural control codes can be shifted to match the user's visual focal point as they move across the screen while reading the viewable content.

[0106] The embodiments described above involve the use of non-representational (i.e., abstract, non-semantic, and / or non-representational) visual stimuli, such as the illustrated visual neural control codes. These offer advantages over concrete content. Non-representational visual stimuli can be placed under strict experimental control for the purpose of stimulus optimization. Under AI guidance, specific features (shape, color, duration, motion, frequency, hue, etc.) can be expressed as parameters and gradually readjusted and reconstructed frame by frame and pixel by pixel to induce a desired biological response. Unlike photographs of people or landscapes, non-representational visual stimuli are free from cultural and linguistic biases, making them more versatile as a global therapy. Furthermore, when non-representational images are combined as composite images, they interfere less with the displayable content.

[0107] To activate specific target regions of the visual cortex, neural selectivity can be tested using the vast hypothetical space of generative deep neural networks, without making assumptions about features or semantic categories. Using genetic algorithms, stimuli that maximize neural firing and / or feedback data indicating the response of a user or subject group during stimulus presentation can be searched within this space. This leads to the evolution of composite images with complex combinations of shape, color, and texture. Sometimes they resemble animals or familiar people, and sometimes they reveal novel patterns that do not fit into clear semantic categories.

[0108] In this embodiment, by using a combination of a pre-trained deep generative neural network and a genetic algorithm, feedback data representing neural responses during stimulus presentation, as well as user or subject group responses, can guide the evolution of the synthetic image. By training with a large number of images, the generative adversarial network acquires the ability to model the statistical features of natural images, rather than merely memorizing training data. This represents a vast and general image space constrained only by the statistical properties of natural images. This provides an efficient space for executing the genetic algorithm because the brain also learns from real-world images, and therefore the images preferred by the brain are likely to follow the statistical features of natural images.

[0109] Aspects of the teaching methods and techniques of the present invention may be embodied in the form of a system, a computer program product, or a method. Similarly, aspects of the teaching methods and techniques of the present invention may be embodied in hardware, software, or a combination of both. Aspects of the teaching methods and techniques may be embodied in the form of computer-readable program code embodied on a computer-readable medium, or as a computer program product embodied (e.g., stored, transmitted, etc.) on one or more computer-readable media.

[0110] Computer-readable media may be computer-readable storage media and / or computer-readable transmission media. Computer-readable storage media (or non-temporary computer-readable media) may be, for example, electronic, optical, magnetic, electromagnetic, infrared, or semiconductor systems, apparatus, devices, or any combination thereof. Computer-readable transmission media may include carrier waves, transmitted signals, etc. Computer-readable transmission media can transmit instructions between components of a single computer system and / or between multiple separate computer systems.

[0111] The computer program code in embodiments of the present invention may be written in any suitable programming language. The program code may be executed on a single computer or on multiple computers and / or processors. The computer may include a processing unit capable of communicating with a computer-available medium, where the computer-available medium includes a set of instructions, and the processing unit is designed to execute that set of instructions.

[0112] The detailed description above illustrates various embodiments of the apparatus and / or process using block diagrams, schematic diagrams, and examples. To the extent that these block diagrams, schematic diagrams, and examples include one or more functions and / or operations, those skilled in the art will understand that each function and / or operation in these block diagrams, flowcharts, or examples can be implemented individually and / or collectively by a wide range of hardware, software, firmware, or substantially any combination thereof. Those skilled in the art will recognize that many of the methods or algorithms described herein may employ additional operations, omit some operations, and / or perform operations in an order different from that specified. The various embodiments described above can be combined to provide further embodiments.

[0113] Based on the detailed description above, these and other modifications can be incorporated into the implementation. In general, the terms used in the following claims should not be limited to any specific embodiment disclosed in the specification and claims, but should be interpreted to include all possible embodiments, including equivalents of the full scope of such claims. Thus, the claims are not limited by the content of the disclosure.

[0114] This application claims priority to U.S. Provisional Patent Application No. 63 / 490,235, filed on 14 March 2023, which is incorporated herein by reference in its entirety.

Claims

1. The steps include obtaining one or more adaptive visual neural modal codes that are adapted to produce a neural modal response in the visual cortex, The steps include outputting the adaptive visual nerve modulatory code to an electronic display visible to the patient in order to improve the patient's eye disease, A method that includes this.

2. The method according to claim 1, wherein one or more adaptive visual neural modulatory codes are adapted to improve the processing of visual information in a target neural region.

3. The method according to claim 1 or 2, wherein the one or more adaptive optic neural modal codes are adapted to improve the processing of visual information in an intact peripheral field of vision in order to compensate for loss of central vision.

4. The method according to any one of claims 1 to 3, further comprising administering drug therapy to slow the progression of the disease and / or improve visual function.

5. The method according to claim 4, wherein the drug therapy is designed to reduce the growth of abnormal blood vessels and fluid leakage in the macula.

6. The steps include: analyzing a first brain activity dataset and generating at least one parameter that indicates the effectiveness of a candidate dynamic visual neural regulatory code for the subject; Based at least partially on at least one parameter that demonstrates the effectiveness of the candidate dynamic visual neural control code, (i) Output the candidate dynamic visual nerve modulation code as an adaptive dynamic visual nerve modulation code. or (ii) Disturbing the candidate dynamic optic neural control code, presenting the disturbed candidate dynamic optic neural control code to the subject, receiving a second brain activity dataset measured while the disturbed candidate dynamic optic neural control code is presented to the subject, iteratively repeating the analysis using the second brain activity dataset, and generating at least one parameter indicating the effectiveness of the candidate dynamic optic neural control code for the subject. A step that performs one of the following, The method according to any one of claims 1 to 5, wherein one or more adaptive visual neural control codes are generated by the method.

7. The method according to claim 6, wherein the disturbance is obtained from a machine learning model trained to target specific neural circuits in the visual cortex to treat retinal degenerative diseases, which may be a further candidate for dynamic visual neural regulatory codes.

8. The method according to claim 7, wherein the training of the machine learning model comprises displaying dynamic visual neural regulatory codes to subjects having different types of retinal degenerative disease and measuring the neural responses from the subjects during the presentation.

9. The steps include implementing drug therapy to reduce the growth of abnormal blood vessels and / or fluid leakage in the macula, The steps include providing a visual therapy based on dynamic neural modalities to improve the processing of visual information in healthy peripheral vision, The step of performing visual therapy based on the dynamic neural control code in combination with the drug therapy, Treatment methods for eye diseases, including those mentioned above.

10. The drug therapy according to claim 9, comprising an anti-vascular endothelial growth factor (VEGF) agent.

11. The method according to claim 9 or 10, wherein the visual therapy based on the dynamic neural modulus code is provided by presenting a dynamic neural modulus code adapted to stimulate neural circuits in the visual cortex that control peripheral vision.

12. The method according to claim 11, wherein the dynamic neural control code is presented on the screen of a digital device.

13. The method according to claim 12, wherein the digital device is a tablet or a smartphone.

14. The method according to claim 11, wherein the dynamic neural modulatory code is customized for each patient based at least in part on an assessment of the patient's visual function.

15. The method according to claim 14, wherein the evaluation of the visual function includes performing one or more standard tests, including a visual acuity test, a contrast sensitivity test, and a visual field test.

16. The method according to claim 11, wherein the dynamic neural modulatory code is presented to the patient for a predetermined period of time.

17. The method according to claim 11, wherein the dynamic neural modulatory code is presented to the patient in a format selected from the group consisting of videos, animations, and games.

18. A system for providing patients with visual therapy based on dynamic neural modulation codes, A digital device configured to present the aforementioned dynamic neural modulatory code and adapted to stimulate the neural circuits of the visual cortex responsible for peripheral vision, A memory device for storing a visual therapy program based on the aforementioned dynamic neural control code, A processor that runs a visual therapy program based on a customized dynamic neural modulation code, A system equipped with this feature.

19. The system according to claim 18, wherein the digital device is a tablet or a smartphone.

20. The system according to claim 18 or 19, wherein the visual therapy program based on the customized dynamic neural modulatory code is at least in part based on an assessment of the patient's visual function.

21. The system according to claim 20, wherein the evaluation of the visual function includes performing one or more standard tests, including a visual acuity test, a contrast sensitivity test, and a visual field test.

22. The system according to any one of claims 18 to 21, wherein a visual therapy program based on the customized dynamic neural modulatory code is presented to the patient for a predetermined period of time.

23. The system according to any one of claims 18 to 22, wherein the dynamic neural modulatory code is presented to the patient in a format selected from the group consisting of videos, animations, and games.

24. The system according to any one of claims 18 to 23, further comprising a monitoring device configured to monitor the patient's visual function throughout the treatment period.

25. The steps include providing a dynamic neural regulatory code (DNC) adapted to stimulate a target neural circuit in the visual cortex, To improve the processing of visual information in target neural circuits of the visual cortex, the steps include presenting the dynamic neural modulatory code on the screen of a digital device, Methods for treating eye diseases, including [specific example].

26. The method according to claim 25, wherein the visual therapy based on the dynamic neural modulatory code improves the processing of visual information in a healthy part of the visual field and compensates for vision loss in another part of the visual field.

27. The method according to claim 25 or 26, further comprising the step of combining visual therapy based on the dynamic neural modulatory codes with drug therapy in order to slow the progression of retinal degeneration.

28. The method according to claim 27, wherein the drug therapy involves administering a drug that reduces the growth of abnormal blood vessels and leakage of body fluids in the macula.

29. A method for generating adaptive dynamic visual neural control codes for visual therapy based on dynamic neural control codes, The steps include presenting candidate dynamic visual neural control codes to a subject using an electronic display while the subject is in a state different from the target state characterized by neural control stimulation of a target area of ​​the visual cortex, The steps include receiving a first brain activity dataset measured while the subject is shown the candidate dynamic visual neural regulatory code, The steps include analyzing a second brain activity dataset measured while the subject is in a target state and the first brain activity dataset to generate at least one parameter that indicates the effectiveness of the candidate dynamic visual neural regulatory code for the subject, Based on at least one parameter that demonstrates the effectiveness of the candidate dynamic visual neural control code, (i) Outputting the candidate dynamic visual nerve conditioning code as an adaptive dynamic visual nerve conditioning code, (ii) Disturbing the candidate dynamic visual neural regulatory code, repeatedly presenting the candidate dynamic visual neural regulatory code to the subject, receiving a first brain activity dataset measured during the presentation of the candidate dynamic visual neural regulatory code to the subject, and analyzing the second brain activity dataset and the first brain activity dataset. A step that performs one of the following, A method that includes this.

30. The method according to claim 29, wherein the disturbance is obtained from a machine learning model trained to target specific neural circuits in the visual cortex to treat retinal degenerative diseases, which may be a further candidate for dynamic visual neural regulatory codes.

31. The training of the aforementioned machine learning model is A step of displaying dynamic visual neural regulatory codes to subjects with different types of retinal degenerative disease, The steps include measuring the neural response from the subject during the aforementioned display, The method according to claim 30, including the method described in claim 30.

32. Acquisition step of obtaining one or more visual neural modal codes adapted to generate a neural modal response having a therapeutic effect or a performance-enhancing effect, The steps include generating one or more adapted visual neural control codes by carrying out the method of claim 29 and outputting them to an electronic display of a user device visible to the user, Includes, A method for providing a visual nerve modulatory code adapted to generate a physiological response having a therapeutic effect or a performance-enhancing effect.

33. The acquisition step described above is: The steps of receiving one or more adapted visual neural control codes via a network, or The method according to claim 32, further comprising the step of obtaining one or more adapted visual neural modal codes from the memory of the user device.

34. The method according to claim 32 or 33, wherein, in outputting the one or more adaptive optic nerve modulatory codes to the electronic display, each of the one or more adaptive optic nerve modulatory codes is displayed for a predetermined period, and the predetermined period is adapted based on user feedback data indicating the user's response.

35. The method according to any one of claims 32 to 34, wherein the output of the one or more adaptive visual nerve modulatory codes to the electronic display comprises combining the one or more adaptive visual nerve modulatory codes with display content.

36. The method according to claim 35, wherein the displayed content includes at least one of the display output of an application, the display output of a browser, and the user interface of a user device.

37. A computer-readable storage medium comprising instructions configured, when executed by a processor, to cause the processor to perform the method according to any one of claims 1 to 17 and 25 to 36.

Images