Systems and methods for cognitive training and monitoring

Abstract

A system and method for analyzing user feedback in response to a cognitive training program, comprising: training at least one machine learning algorithm using a predefined data set to predict a training success rate, wherein the predefined data set includes previously received user feedback for users with known characteristics; receiving new user feedback; and determining a prediction of the training success rate using at least one machine learning algorithm based on the received new user feedback.

Description

Field of the Invention

[0001] The present invention relates to cognitive training. More specifically, the present invention relates to systems and methods for monitoring and analyzing user feedback in response to a cognitive training program. Background of the Invention

[0003] People with cognitive problems, or those seeking to improve their cognitive skills, sometimes use cognitive training programs to improve their cognitive health and train their memory, similar to physical training in a gym. For example, a person can use memory cards, solve crosswords, or sit in front of a computer screen and perform various tasks designed to improve cognitive abilities such as memory, calculation, vocabulary, etc.

[0004] The main problem with self-administered cognitive training programs (e.g., without any professional supervision) is user adherence and / or engagement with the training program. The time of use of a weekly, e.g., repetitive program (where the user gradually loses interest in the program) may decrease, and the training results decrease accordingly. For any training to be effective, one needs to train continuously for a long time. Although most people usually start training with high motivation to improve their cognitive abilities, the average user does not complete the training, and over time, they may stop or significantly reduce the amount of training sessions. Usually, this situation may occur due to monotonous training, difficulty in understanding how these activities relate to daily needs, and / or due to loss of interest. Therefore, current training programs do not produce significant cognitive effects when self-administered.

[0005] Overview

[0006] Accordingly, in some embodiments of the present invention, there is provided a method for analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm using a predefined data set to predict a training success rate, where the predefined data set may include previously received user feedback for users with known characteristics; receiving, by the processor, new user feedback; and determining, by the processor, a prediction of the training success rate using at least one machine learning algorithm based on the received new user feedback. In certain embodiments, at least one machine learning algorithm may be trained with reinforcement learning.

[0007] In some embodiments, a behavior pattern may be determined from the user feedback. In some embodiments, at least one machine learning algorithm may be implemented on a recurrent neural network with long short-term memory units. In some embodiments, a training churn rate may be predicted. In some embodiments, the received feedback may be monitored for at least one of timing, training session length, training session success rate, attention stability, freeze period, location, training platform, and number of breaks in a training session.

[0008] In some embodiments, user feedback can be classified to determine a user profile from a list of predefined profiles, wherein the prediction of the training success rate determined can also be based on the determined user profile. In some embodiments, the user profile can be determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, medical condition, and marital status. In some embodiments, the user profile can be determined based on clustering of the received feedback and based on at least one user characteristic.

[0009] In some embodiments, a user can be monitored with at least one electroencephalogram (EEG) sensor, wherein a cognitive training program can be changed based on the measured EEG signals. In some embodiments, a user's eye movements can be monitored with at least one imager to determine the user's attention. In some embodiments, a behavior pattern can be determined based on user feedback, and an alert can be issued when the determined behavior pattern exceeds a predefined threshold.

[0010] Accordingly, in some embodiments of the present invention, there is provided a system for performing cognitive analysis of user feedback in response to a cognitive training program, the system comprising: a database that includes a dataset of previously received user feedback for users with known characteristics; and a processor coupled to the database and configured to: train at least one machine learning algorithm with the dataset to predict a training success rate, receive new user feedback, and based on the received new user feedback, determine a prediction of the training success rate using at least one machine learning algorithm. In some embodiments, at least one machine learning algorithm can be trained with reinforcement learning.

[0011] In some embodiments, the processor can determine a behavior pattern based on user feedback. In some embodiments, the processor can classify user feedback to determine a user profile from a list of predefined profiles, wherein the prediction of the training success rate using at least one machine learning algorithm can also be based on the determined user profile. In some embodiments, the processor can predict a training dropout rate using at least one machine learning algorithm. In some embodiments, the processor can monitor the received feedback for at least one of timing, training session length, training session success rate, attention stability, freeze period, and number of breaks in a training session.

[0012] In some embodiments, at least one machine learning algorithm may be implemented on a recurrent neural network with long short-term memory units. In some embodiments, a user profile may be determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, and marital status. In some embodiments, a user profile may be determined based on clustering of received feedback. In some embodiments, at least one electroencephalogram (EEG) sensor may be coupled to a processor, where the processor may utilize the at least one EEG sensor to monitor a user, and where a user profile may be determined based on measured EEG signals.

[0013] In some embodiments, at least one imager may be coupled to a processor, and where the processor may utilize the at least one imager to monitor eye movements of a user. In some embodiments, the processor may determine a behavior pattern based on user feedback and issue an alert when the determined behavior pattern exceeds a predefined threshold.

[0014] Thus, according to some embodiments of the present invention, there is provided a method for cognitive training, including: determining, by a processor in response to a cognitive training program, a behavior pattern from received user feedback; and correcting, by the processor based on the determined behavior pattern, the cognitive training program using at least one machine learning algorithm so as to improve cognitive training. In some embodiments, at least one machine learning algorithm may be trained using previously received user feedback for users with known characteristics.

[0015] Thus, according to some embodiments of the present invention, there is provided a method for analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm using a predefined data set to predict a training success rate, where the predefined data set may include previously received user feedback for users with known characteristics; and updating, by the processor, training variables based on the prediction of the training success rate using at least one machine learning algorithm. In some embodiments, new user feedback may be received, and the training variables may be re-updated based on the received new user feedback.

[0016] Thus, according to some embodiments of the present invention, there is provided a method for analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm using a predefined data set to predict a training dropout rate, where the predefined data set may include previously received user feedback for users with known characteristics; receiving, by the processor, new user feedback; and determining, by the processor based on the received new user feedback, a prediction of the training dropout rate using at least one machine learning algorithm.

[0017] Accordingly, in some embodiments of the present invention, there is provided a method for analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm using a predefined data set to predict cognitive decline, where the predefined data set may include previously received user feedback for users with known characteristics; receiving, by the processor, new user feedback; and determining, by the processor, a prediction of cognitive decline using at least one machine learning algorithm based on the received new user feedback. Brief Description of the Drawings

[0019] The subject matter of the present invention is particularly pointed out and distinctly claimed at the end of the specification. However, the organization and method of operation of the present invention, as well as its objectives, features, and advantages, will be best understood by reference to the following detailed description when read in conjunction with the accompanying Figure 1 drawings, in which:

[0020] Figure 1 FIG. 1 shows a block diagram of an example computing device in accordance with some embodiments of the present invention;

[0021] Figures 2A - 2E FIG. 2 shows a block diagram of a system for cognitive analysis of user feedback in response to a cognitive training program in accordance with some embodiments of the present invention;

[0022] Figure 3 FIG. 3 shows a block diagram of a system for cognitive data collection in accordance with some embodiments of the present invention;

[0023] Figures 4A - 4B FIG. 4 shows a flowchart of a method for cognitive analysis of user feedback in response to a cognitive training program in accordance with some embodiments of the present invention;

[0024] Figure 5 FIG. 5 shows a flowchart of a method for analyzing user feedback to determine a training dropout rate in response to a cognitive training program in accordance with some embodiments of the present invention; and

[0025] Figure 6 FIG. 6 shows a flowchart of a method for analyzing user feedback to determine cognitive decline in response to a cognitive training program in accordance with some embodiments of the present invention.

[0026] It will be understood that, for simplicity and clarity of illustration, the elements shown in the figures are not necessarily drawn to scale. For example, the dimensions of some elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated in multiple figures to indicate corresponding or analogous elements.

[0027] Detailed Description of the Invention

[0028] In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, those skilled in the art will understand that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, and components, modules, units, and / or circuits have not been described in detail so as not to obscure aspects of the present invention. Some features or elements described with respect to one embodiment may be combined with features or elements described with respect to other embodiments. For the sake of clarity, the same or similar features or elements may not be discussed repeatedly.

[0029] Although embodiments of the present invention are not limited in this regard, discussions using terms such as, for example, "processing", "computing", "calculating", "determining", "establishing", "analyzing", "checking", etc. may refer to operations and / or processes of a computer, computing platform, computing system, or other electronic computing device that manipulates and / or transforms data represented as a physical (e.g., electronic) quantity within a register and / or memory of the computer into other data similarly represented as a physical quantity within a register and / or memory of the computer or other non-transitory storage medium that may store instructions for performing the operations and / or processes. Although embodiments of the present invention are not limited in this regard, as used herein, the terms "plurality" and "aplurality" may include, for example, "multiple" or "two or more". Throughout the specification, the terms "plurality" or "a plurality" may be used to describe two or more components, devices, elements, units, parameters, etc. The term "set", as used herein, may include one or more items. Unless explicitly stated, method embodiments described herein are not limited to a particular order or sequence. Additionally, some or some of the elements of the method embodiments described may occur or be performed synchronously, at the same point in time, or simultaneously.

[0030] Reference Figure 1, which shows a schematic block diagram of an example computing device according to some embodiments of the present invention. The computing device 100 may include a controller or processor 105 (e.g., a central processing unit processor (CPU), a chip, or any suitable computing or computational device), an operating system 115, a memory 120, executable code 125, a storage device 130, an input device 135 (e.g., a keyboard or a touch screen), and an output device 140 (e.g., a display), a communication unit 145 (e.g., a cellular transmitter or a modem, a Wi-Fi communication unit, etc.) for communicating with remote devices via a communication network (such as, for example, the Internet). The controller 105 may be configured to execute program code to perform the operations described herein. The systems described herein may include one or more computing devices 100, e.g., serving as various devices and / or components as Figure 2A shown. For example, the system 200 may be or may include the computing device 100 or its components.

[0031] The operating system 115 may be or may include any code segment (such as, a code segment similar to the executable code 125 described herein) that is designed and / or configured to perform tasks involving coordinating, scheduling, arbitrating, supervising, controlling, or otherwise managing the operations of the computing device 100 (e.g., scheduling the execution of software programs or enabling software programs or other modules or units to communicate).

[0032] The memory 120 may be or may include, for example, random access memory (RAM), read-only memory (ROM), dynamic RAM (DRAM), synchronous DRAM (SD-RAM), double data rate (DDR) memory chips, flash memory, volatile memory, non-volatile memory, cache memory, buffers, short-term memory units, long-term memory units, or other suitable memory units or storage device units. The memory 120 may be or may include multiple potentially different memory units. The memory 120 may be a computer or processor non-transitory readable medium, or a computer non-transitory storage medium, such as RAM.

[0033] The executable code 125 may be any executable code, such as an application, a program, a process, a task, or a script. The executable code 125 may be executed by the controller 105 under the control of the operating system 115. For example, the executable code 125 may be a software application that executes the methods further described herein. Although, for clarity, Figure 1 the executable code 125 shows a single item in, the system according to embodiments of the present invention may include multiple executable code segments similar to the executable code 125, which may be stored in the memory 120 and cause the controller 105 to execute the methods described herein.

[0034] The storage device 130 can be or can include, for example, a hard disk drive, a Universal Serial Bus (USB) device, or other suitable removable and / or fixed storage device units. Additionally, in some embodiments, some of the components shown in Figure 1 can be omitted. For example, the memory 120 can be a non-volatile memory having the storage capacity of the storage device 130. Thus, although shown as a separate component, the storage device 130 can be embedded or included in the memory 120.

[0035] The input device 135 can be or can include a keyboard, a touch screen or touchpad, one or more sensors, or any other or additional suitable input device. Any suitable number of input devices 135 can be operatively connected to the computing device 100. The output device 140 can include one or more displays or monitors and / or any other suitable output device. Any suitable number of output devices 140 can be operatively connected to the computing device 100. As shown by blocks 135 and 140, any applicable input / output (I / O) device can be connected to the computing device 100. For example, a wired or wireless network interface card (NIC), a Universal Serial Bus (USB) device, or an external hard drive can be included in the input device 135 and / or the output device 140.

[0036] Embodiments of the present invention can include articles such as computer or processor non-transitory readable media or computer or processor non-temporary storage media, such as, for example, a memory, a disk drive, or a USB flash drive that encodes, includes, or stores instructions (e.g., computer-executable instructions) that, when executed by a processor or controller, perform the methods disclosed herein. For example, the article can include a storage medium such as the memory 120, computer-executable instructions such as the executable code 125, and a controller such as the controller 105. Such non-transitory computer-readable media can be, for example, a memory, a disk drive, or a USB flash drive that encodes, includes, or stores instructions, such as computer-executable instructions, that, when executed by a processor or controller, perform the methods disclosed herein. The storage medium can include, but is not limited to, any type of disk, including semiconductor devices such as read-only memory (ROM) and / or random access memory (RAM), flash memory, electrically erasable programmable read-only memory (EEPROM), or any type of medium suitable for storing electronic instructions, including programmable storage devices. For example, in some embodiments, the memory 120 is a non-transitory machine-readable medium.

[0037] A system according to an embodiment of the present invention may include components such as, but not limited to, multiple central processing units (CPUs) or any other suitable general-purpose or specific processors or controllers (e.g., a controller similar to controller 105), multiple input units, multiple output units, multiple memory units, and multiple storage device units. The system may additionally include other suitable hardware components and / or software components. In some embodiments, the system may include or may be, for example, a personal computer, a desktop computer, a laptop computer, a workstation, a server computer, a network device, or any other suitable computing device.

[0038] According to some embodiments, there are provided systems and methods for personalizing a computerized cognitive training program configured to support the cognitive health of adult users, particularly memory function.

[0039] Now refer to Figure 2A , which shows a block diagram of a system 200 for performing cognitive analysis of user feedback in response to a cognitive training program 210 according to some embodiments. In Figure 2A , the direction of the arrows indicates the direction of information flow, and the dashed elements indicate software and / or algorithms.

[0040] System 200 may include at least one processor 201 (such as Figure 1 the controller 105 shown), for example, a processor in a mobile device and / or a PC on which the cognitive training program 210 may be implemented. Processor 201 may be coupled to a database 202 (such as Figure 1 the storage device system 130 shown), which includes a dataset of previously received user feedback 203 for users with known characteristics. For example, users with known characteristics such as age, gender, medical, and / or mental conditions may provide feedback 203 to the cognitive training program 210 (e.g., during the initial calibration phase described in detail herein).

[0041] During training, the processor 201 can collect information related to one or more of the training time of day, the length of the training session, the platform used by the user (e.g., PC, tablet, smartphone), and / or the training location (e.g., at home or in a public place) in order to improve the cognitive training program 210 for future training of that user. In some embodiments, the processor 201 can also collect information from the user feedback 203, such as the user's response time in different scenarios presented to the user (e.g., in a game) and / or the type of answers (e.g., correct / wrong / missed answers), and / or the target location on the display, and / or the type of input (e.g., using a keyboard or touch screen), and / or the number of times the user takes a break. In addition to collecting information related to the type of answers, the processor 201 can also collect information related to at least one of the following: success rate, attention stability, attention lapse, spatial attention, longest streak (e.g., the number of consecutive correct answers in a game), learning curve, sleep quality, and / or mood (e.g., determined based on the questions in the training session or directly from a dedicated device such as a smartwatch or other sensors). In some embodiments, the information collected by the processor 201 can be stored in the database 202 as user feedback 203.

[0042] According to some embodiments, the cognitive training program 210 can be modified based on cognitive functions identified with a low success rate (e.g., in a game, exercise, etc.), and / or a high standard deviation of response time (e.g., crossing a certain predefined threshold), and / or a specific type of error (e.g., location vs. correct identification), etc. The cognitive functions being trained can include memory components such as visual perception, binding of features and objects, organizing information, semantic networks, attention (which is crucial for the memory process), etc. The user's attention can be trained to focus attention, attention orientation, selective attention, visual-spatial attention, sustained attention, executive attention, and / or attention control (including, for example, divided attention and inhibition), etc. In some embodiments, the processor 201 can monitor the user feedback 203 to determine the training attention based on the overall response time and the identified attention lapse (e.g., the deviation of the response time) and attention stability (e.g., the magnitude of the overall standard deviation of the response time).

[0043] The processor 201 may execute at least one machine learning algorithm 204 (e.g., using deep learning with a deep neural network) to train using a dataset of user feedback 203 and predict the user's training success rate 205. The at least one machine learning algorithm 204 may be trained using data previously collected on the user while operating according to predefined rules. In some embodiments, the at least one machine learning algorithm 204 may be trained using supervised training of a machine learning (e.g., using a neural network) with a computer network. The supervised training may include training using a labeled dataset (e.g., taking the example of training a human user with a system), or may include training under the supervision of a human operator who labels samples to teach the network. In some embodiments, once a predetermined amount of new data is collected, the at least one machine learning algorithm 204 may be activated (e.g., semi-automatically) for a predetermined period of time for re-training of the algorithm. In some embodiments, the at least one machine learning algorithm 204 may be implemented on a recurrent neural network (RNN) having, for example, long short-term memory (LSTM) units. An RNN is a class of artificial neural network where the connections between nodes form a directed graph along a time series. Different from a feedforward neural network, an RNN with an LSTM structure may have feedback connections to process an input sequence using its internal state (memory).

[0044] In some embodiments, the processor 201 may predict the training success rate 205 (e.g., using supervised learning), where similar users may be identified, for example, from a continuously updated dataset of long-term users based on similar content (e.g., gender, location, age, education, etc.) and / or based on similar behaviors (e.g., success regarding their training history). Thus, in some embodiments, the training success rate 205 of a particular user may be calculated and / or predicted by the processor 201 based on the success of similar users for a particular training session.

[0045] In some embodiments, the cognitive training program 210 may receive input (e.g., directly and / or via the processor 201) from the at least one machine learning algorithm 204, the input including, for example, which exercises and / or levels and / or variables to use in order to receive the expected training success rate 205. In some embodiments, the processor 201 may continuously receive user feedback 203 during and / or after training with information on the training progress made according to the provided instructions in order to compare the training results with the predicted training success rate 205 in order to continuously improve the at least one machine learning algorithm 204.

[0046] In some embodiments, the processor 201 may use at least one machine learning algorithm 204 (e.g., execute an algorithm to obtain a result) to generate suggestions that increase (or decrease) the training success rate 205. During training, the processor 201 may modify the cognitive training program 210 based on the calculated training success rate 205. For example, if the calculated training success rate 205 is below a predefined threshold, it may be easier to modify the cognitive training program 205 for the user. In some embodiments, the processor 201 may generate statistical data (e.g., charts) to be displayed to the user to reflect which variables affect their cognitive abilities, such as training time (e.g., time of day or day of the week), sleep quality (e.g., optimal sleep time for maximum concentration), and / or whether there are differences in these variables during training sessions. In some embodiments, at least some of the features may be determined by the processor 201 (e.g., success rate or training dropout rate) by different machine learning algorithms.

[0047] Now refer to Figure 2B and Figure 2C , which respectively show block diagrams of another systems 230 and 240 for determining a behavior pattern 206 in response to a cognitive training program 210 according to some embodiments. Figure 2B and Figure 2C Some of the elements in Figure 2A may be the same as or similar to the elements shown in

[0048] According to some embodiments, the processor 201 may, for example, use a machine learning algorithm 204 to determine a user's behavior pattern 206 (e.g., a pattern of behavior and / or feedback, such as reaction time and / or correct answers that may reflect cognitive and / or motor abilities) from user feedback 203. In some embodiments, the behavior pattern 206 may be a cognitive behavior pattern.

[0049] The processor 201 may determine a prediction of the training success rate 205 based on the determined user behavior pattern 206 and / or based on user characteristics, using at least one machine learning algorithm 204. In some embodiments, when the determined user behavior pattern 206 exceeds a predefined threshold, the processor 201 may issue an alert. For example, the machine learning algorithm 204 may compare the user's behavior during training (e.g., from user feedback 203) with an initial state (e.g., from user profile 207) to determine whether the user's behavior pattern 206 meets or exceeds a predefined threshold (such as, for example, determining that the training success rate has decreased by 40%). In some embodiments, the information collected in the alert program may be fed back into the user's behavior pattern 206 algorithm. For example, a sliding window technique may be used in conjunction with supervised learning (e.g., RNN) or unsupervised learning by finding differences between previous windows. For both cases, a distance function between subsequences may be defined and used to calculate the distance to the previous window as input for several anomaly detection methods.

[0050] For example, the machine learning algorithm 204 may be configured to implement a predefined training success rate 205 (e.g., 80%) for each user, where the training success rate 205 is measured, for example, during a training session to keep the user training for a relatively long time (e.g., if a decreased success rate also indicates a decrease in training persistence), and / or when a new level of training (e.g., in a game) is successfully completed, the training success rate 205 is measured. When a decrease in the success rate 205 is identified, for example, due to a decrease in training time, the machine learning algorithm 204 may be configured to implement a lower training success rate 205 in order to keep the user training.

[0051] In some embodiments, a set of actions for direct interaction with the user may be defined based on the behavior pattern 206 of a specific user in order to further increase the training success rate 205 and / or if the user's behavior has changed significantly. User feedback data 203 may be collected using reinforcement and / or supervised learning regarding the impact of different actions on the user's performance. Based on the collected user feedback data 203, predefined rules and / or machine learning algorithms may impose training and / or parameters to be presented accordingly. For each such action (e.g., initiating a phone call to the user, providing educational materials, etc.), the impact on the training success rate 205 may be measured in order to understand which actions increase the training success rate 205. In some embodiments, once the system 200 and / or system 230 and / or system 240 learns the response to an action, at least one machine learning algorithm 204 may predict for each user which actions may be needed and at what time during training to apply them accordingly.

[0052] In some embodiments, the processor 201 may classify the user feedback 203 (and / or the processor 201 may instruct an algorithm to perform the classification) to determine at least one user profile 207, including age, gender, etc., from a list of predetermined profiles (e.g., stored in the database 202); for example, the at least one profile may be determined by at least one machine learning algorithm 204. In some embodiments, the determined at least one user profile may be Figures 2A - 2E used by at least one of the systems 200, 230, 240, 260, and 270 shown. In some embodiments, the prediction of the user behavior pattern 206 may also be based on the determined at least one user profile 207. In some embodiments, the at least one user profile 207 may also or alternatively be determined based on at least one user characteristic, such as gender, age, education, location, language, occupation, current employment status, medical status, and / or marital status. The at least one user profile 207 may also or alternatively be determined based on a clustering of the received feedback.

[0053] Now referring to Figure 2D , which shows a block diagram of another system 260 for determining a training attrition rate 264 in response to a cognitive training program 210, according to some embodiments. Figure 2D Some of the elements in Figure 2A may be the same as or similar to the elements shown in

[0054] According to some embodiments, the processor 201 may use a machine learning algorithm 204 to predict the training attrition rate 264. The user training attrition rate may be defined using different levels of ongoing participation (e.g., based on the training duration and / or the time of day of training).

[0055] In some embodiments, data on ongoing usage and / or user participation may be collected (e.g., if the user closes the account or stops training) in order to use the machine learning algorithm 204 to predict the user's training habits and / or the expected training attrition rate 264. In some embodiments, the machine learning algorithm 204 may receive as input data of other users previously identified as having reduced engagement and / or stopped training, for comparison with the new training data of the collected users, in order to predict the expected training attrition rate 264, for example, also based on the user profile 207.

[0056] According to some embodiments, the processor 201 can detect changes and / or anomalies in user behavior, for example, by using unsupervised learning. The processor 201 can monitor user performance and define a profile for each user, with each profile having an expected behavior. In some embodiments, non-trivial or significant deviations from the expected behavior (e.g., predefined prior to training) detected from a new training session can be flagged to issue an alert, e.g., initiate contact with the user (e.g., call the user to try and understand the reason for the anomaly, or whether there is a medical issue or significant change, such as grief, etc.).

[0057] In some embodiments, the measurement of user behavior can be performed at predetermined intervals, e.g., once a month, as an objective measurement or assessment to see if there are any changes in cognitive ability.

[0058] In some embodiments, the machine learning algorithm 204 can receive data for the behavior pattern 206 of the user ( Figure 2C shown therein) as input to determine a change in behavior, e.g., determine the training dropout rate 264.

[0059] Now refer to Figure 2E , which shows a block diagram of another system 270 according to some embodiments for determining cognitive decline 274 in response to a cognitive training program 210. Figure 2E Some of the elements in Figure 2A may be the same as or similar to the elements shown in

[0060] In some embodiments, the processor 201 can predict and / or detect cognitive decline 274 associated with mild cognitive impairment (MCI) using the machine learning algorithm 204. MCI can result in a significant and measurable decline in cognitive abilities including memory and thinking skills (judgment, correct decision-making, etc.). People with MCI have an increased risk of developing Alzheimer's disease or other types of dementia.

[0061] In some embodiments, if users with different MCI stages are initially flagged, e.g., through a diagnosis by an external medical institution, at least one machine learning algorithm 204 can learn the behavior patterns of these users, e.g., in order to later identify similar patterns in users not flagged at a certain stage of MCI. Thus, the system 200 and / or the system 270 can be used for MCI prediction. In some embodiments, multi-label time series can be used with a minority class prediction algorithm (for new users), e.g., in conjunction with an attention mechanism or LSTM. Oversampling or generative adversarial network (GAN) mechanisms can be used to enhance the example set. In some embodiments, unsupervised detection can be used in conjunction with algorithm-based clustering, such as local outlier factor (LOF), kernel density estimation (KDE), or K-Means, to identify the MCI level of the user.

[0062] In some embodiments, the machine learning algorithm 204 may receive data of other users previously identified as having cognitive decline (e.g., having MCI or dementia) as input for comparison with the newly collected training data of the user, e.g., in order to predict the expected cognitive decline 274, which may also be based on the user profile 207, for example.

[0063] In some embodiments, the machine learning algorithm 204 may receive data on the behavior pattern 206 of the user (as Figure 2C shown) as input in order to determine changes in behavior, e.g., to determine cognitive decline 274.

[0064] Now referring Figure 3 , Figure 3 FIG. shows a block diagram of a system 300 for cognitive data collection according to some embodiments. In some embodiments, the system 300 may also include some or all elements of the system 200 (such as the processor 201 and the database 202), where elements of the system 300 are added in order to collect cognitive data from the user 30.

[0065] In some embodiments, the system 300 may include at least one electroencephalogram (EEG) sensor 301 coupled to the processor 201 to measure EEG signals, and the processor 201 is configured to monitor the cognitive signals of the user 30 using the at least one EEG sensor 301. In some embodiments, the user profile 207 may also be determined based on the measured EEG signals. For example, the user 30 may wear a head-mounted device with at least one EEG sensor 301 to collect measurements on brain waves and specific activities using commercial EEG channels (1 - 16) or clinical electroencephalogram channels (16 - 64) depending on private use or clinician permission, respectively. In some embodiments, the determined training success rate 205 may be refined based on data collected by the at least one EEG sensor 301.

[0066] In some embodiments, the EEG sensor 301 may be similarly used to provide neurofeedback, following different brain waves (e.g., α, β, θ) and their relationships, e.g., to find correlations between brain waves in response to a cognitive training program 210. In some embodiments, the measured signals and specific responses to the measured waves and / or wave relationship thresholds may be integrated into the training session (e.g., integrated into a game). In some embodiments, the training may also include an option for a dual task that has neurofeedback based on brain wave activity (e.g., based on the α level or θ or β level) and a predetermined level and / or measure maintained during training.

[0067] In some embodiments, system 300 may include, for example, at least one imager 302 coupled to processor 201, which is configured to monitor the eye movements of user 30 with respect to the content displayed on display 310 by the at least one imager 302, and thus determine the concentration and / or attention of user 30 during training. In some embodiments, the determined training success rate 205 may be refined, for example, based on data collected by the at least one imager 302.

[0068] In some embodiments, the at least one imager 302 may track eye movements and / or pupil size using, for example, the camera of a computerized device such as a tablet, smartphone, etc. (such as Figure 1 computing device 100 as shown) or via a clinical eye tracker. Eye tracking data may be collected during training and with respect to the states presented in training display 310. In some embodiments, processor 201 may analyze the collected eye tracking data to identify saccades, fixations, pupil size, etc. related to the content presented on display 310 to determine attention quality, attention measurements, and memory measurements.

[0069] In some embodiments, at least some of the training sessions may be performed in a virtual reality environment. For example, a single wearable device (e.g., a head-mounted device) may include EEG sensor 301 and / or imager 302 and / or virtual reality imaging displayed by the head-mounted device to combine biofeedback with pulse rate and sweating monitoring.

[0070] In some embodiments, processor 201 may analyze data collected from external devices such as EEG sensor 301 and / or imager 302, as well as other external devices used by the user such as activity trackers, smartwatches, smartphones, clinical data, and test results, to improve cognitive training program 210 and accordingly increase training success rate 205. The additional data collected may relate to sleep quality, daily activities, location (e.g., using GPS data), stability (e.g., hand stability when holding a device), and / or emotional state (e.g., based on voice and speech recognition, and / or based on nutrition, medications, etc.). In some embodiments, at least one machine learning algorithm 204 may be used to compute the collected data, for example, to provide more accurate personalized training, personal recommendations, and / or cognitive markers.

[0071] According to some embodiments, the processor 201 may calculate a training success rate 205 and / or a general training progress based on the percentage of correct answers and / or response time and based on spatial attention detection, e.g., to measure how the user is distracted in the surrounding space. Thus, in some embodiments, the training may include targets in different regions of the display 310, and the received responses between these regions (e.g., registered as user feedback 203) may be compared, e.g., to create a spatial attention map and locate (within the display area) the "ignored" regions. These regions may be marked and trained to improve the user's spatial attention.

[0072] Now refer to Figure 4A , which shows a flowchart of a method for analyzing user feedback 203 in response to a cognitive training program 210 according to some embodiments.

[0073] At least one machine learning algorithm 204 may be trained in step 401 (e.g., by the processor 201) using a predefined dataset to predict the training success rate 205, where the predefined dataset may include previously received user feedback for users with known characteristics. New user feedback may be received in step 402, and a prediction of the training success rate 205 may be determined in step 403 based on the received new user feedback, using at least one machine learning algorithm 204 (e.g., by the processor 201). In some embodiments, at least one machine learning algorithm 204 may be trained with reinforcement learning. In some embodiments, a transfer learning algorithm may be used to establish a prediction model for a cognitively declining person using data from healthy people.

[0074] Now refer to Figure 4B , which shows a flowchart of a method for analyzing user feedback 203 in response to a cognitive training program 210 according to some embodiments. In some embodiments, in step 404, at least one machine learning algorithm 204 may be trained (e.g., by the processor 201) using a predefined dataset to predict the training success rate 205, where the predefined dataset may include previously received user feedback for users with known characteristics. The training set and / or training variables may be determined (e.g., using a predefined threshold) to meet the prediction of the training success rate 205, where the training variables may be updated in step 405 according to the prediction of the training success rate using at least one machine learning algorithm 204. When new user feedback is received in step 406, the training variables may be updated again in step 405.

[0075] According to some embodiments, at least one machine learning algorithm 204 may be trained (e.g., by processor 201) using a predefined dataset to, for example, determine a user's behavior pattern 206, where the predefined dataset may include previously received user feedback for users with known characteristics and behavior patterns of other users previously calculated. In some embodiments, new user feedback may be received, and the at least one machine learning algorithm 204 may be used (e.g., by processor 201) to perform a comparison of the user's behavior pattern 206 with the newly received data to identify anomalies in the behavior pattern 206.

[0076] According to some embodiments, at least one machine learning algorithm 204 may be trained (e.g., by processor 201) using a predefined dataset to, for example, determine a user's behavior pattern 206, where the predefined dataset may include previously received user feedback for users with known characteristics and behavior patterns of other users previously calculated. The at least one machine learning algorithm 204 may be trained to predict the user's behavior in other situations accordingly, such as predicting the behavior of a specific user in a stressed state.

[0077] Now refer to Figure 5 , which shows a flowchart of a method for analyzing user feedback 203 to determine a training dropout rate 264 in response to a cognitive training program 210 according to some embodiments.

[0078] In some embodiments, at least one machine learning algorithm 204 may be trained in step 501 (e.g., by processor 201) using a predefined dataset to predict the training dropout rate 264, where the predefined dataset may include previously received user feedback for users with known characteristics (e.g., user feedback regarding the number of training sessions, training frequency, training time / date, etc.). New user feedback may be received in step 502, and the at least one machine learning algorithm 204 may be used in step 503 (e.g., by processor 201) to determine a prediction of the training dropout rate 264 based on the received new user feedback.

[0079] Now refer to Figure 6 , which shows a flowchart of a method for analyzing user feedback 203 to determine cognitive decline 274 in response to a cognitive training program 210 according to some embodiments.

[0080] In some embodiments, at least one machine learning algorithm 204 may be trained in step 601 (e.g., by processor 201) using a predefined dataset to label possible cognitive decline and / or predict cognitive decline 274, where the predefined dataset may include previously received user feedback for users with known characteristics. The predefined dataset may include previously computed patterns that characterize different cognitive deterioration states, for example, based on previously received user feedback for users with known characteristics including a clinical diagnosis of cognition. New user feedback may be received in step 602, and a prediction of cognitive decline 274 may be determined in step 603 based on the received new user feedback (e.g., by processor 201) using at least one machine learning algorithm 204.

[0081] Although certain features of the present invention have been illustrated and described herein, many modifications, alternatives, variations, and equivalents thereof will be apparent to those skilled in the art. Accordingly, it is to be understood that the appended claims are intended to cover all such modifications and changes that fall within the true spirit of the present invention.

[0082] Various embodiments have been presented. Of course, each of these embodiments may include features of the other embodiments presented, and embodiments not specifically described may include various features described herein.

Claims

1. A method for analyzing user feedback in response to a cognitive training program, the method comprising: A processor trains at least one machine learning algorithm using a predefined dataset to predict the training success rate of a specific user, where the predefined dataset includes previously received user feedback for users with known characteristics, and the user feedback represents user responses to the displayed scenarios; The processor receives new user feedback, which represents a new user response to the displayed scenario; The processor determines a prediction of the training success rate for the new user using the at least one machine learning algorithm based on the received new user feedback; The processor modifies the cognitive training program according to the predicted training success rate of the new user; The processor retrains the at least one machine learning algorithm using reinforcement learning with the received new user feedback to improve the prediction of the training success rate; In response to the modified cognitive training program, the processor determines the behavior pattern of the new user; The processor determines a decrease in the training success rate based on the determined behavior pattern; and When the decrease in the training success rate exceeds a predefined threshold, the processor issues an alarm.

2. The method according to claim 1, further comprising determining a behavior pattern from the user feedback by the processor.

3. The method according to claim 1, wherein, The at least one machine learning algorithm is implemented on a recurrent neural network with long short-term memory units.

4. The method according to claim 1, further comprising predicting a training dropout rate by the processor.

5. The method according to claim 1, further comprising monitoring the received feedback by the processor for at least one of timing, training session length, training session success rate, attention stability, freeze period, location, training platform, and number of breaks in a training session.

6. The method according to claim 1, further comprising classifying the user feedback by the processor to determine a user profile from a list of predefined profiles, wherein, The predicted training success rate determined is also based on the determined user profile.

7. The method according to claim 6, wherein, The user profile is also determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, medical status, and marital status.

8. The method according to claim 7, wherein, The user profile is also determined based on clustering of the received feedback and based on the at least one user characteristic.

9. The method according to claim 1, further comprising monitoring the user by the processor using at least one electroencephalogram (EEG) sensor, wherein, The cognitive training program is changed based on measured EEG signals.

10. The method according to claim 1, further comprising monitoring eye movements of the user by the processor using at least one imager to determine the user's attention.

11. The method according to claim 1, further comprising: Continuously receive the new user feedback; Based on the new user feedback, compare the training result with the predicted training success rate; Improve the at least one machine learning algorithm based on the comparison.

12. The method according to claim 11, further comprising: Obtain a set of actions that directly interact with the new user; And Train the at least one machine learning algorithm using reinforcement learning by: Collecting new user feedback data using information about the impact of actions in the set of actions on the user performance; Based on the collected user feedback data, applying cognitive training parameters to be presented; For each action in the set of actions, measure the impact on the training success rate to understand which actions improve the training success rate; And Train the at least one machine learning algorithm to predict which actions may be needed to improve the training success rate.

13. A system for analyzing user feedback in response to a cognitive training program, the system comprising: A database that includes a dataset of previously received user feedback for users with known characteristics, where the user feedback represents user responses to the displayed scenarios; And A processor coupled to the database and configured to: Train at least one machine learning algorithm using the dataset to predict the training success rate of a specific user; Receive new user feedback, which represents a new user response to the displayed scenario; And Based on the newly received user feedback, determine a prediction of the training success rate for the new user using the at least one machine learning algorithm; Modify the cognitive training program according to the predicted training success rate of the new user; Use reinforcement learning to retrain the at least one machine learning algorithm using the received new user feedback to improve the prediction of the training success rate; Determine the behavior pattern of the new user in response to the modified cognitive training program; Determine the decrease in the training success rate based on the determined behavior pattern; And When the determined decrease in the training success rate exceeds a predefined threshold, issue an alert.

14. The system according to claim 13, wherein, The processor is further configured to determine a behavior pattern from the user feedback.

15. The system according to claim 13, wherein, The processor is further configured to classify the user feedback to determine a user profile from a list of predefined profiles, wherein the prediction of the training success rate using the at least one machine learning algorithm is also based on the determined user profile.

16. The system according to claim 15, wherein, The processor is further configured to determine the user profile based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, and marital status.

17. The system according to claim 15, wherein, The processor is further configured to determine the user profile based on the clustering of the received feedback.

18. The system according to claim 15, further comprising at least one electroencephalogram (EEG) sensor coupled to the processor, wherein, The processor is further configured to monitor the user using the at least one electroencephalogram (EEG) sensor, and wherein the processor is further configured to determine the user profile based on the measured EEG signals.

19. The system according to claim 13, wherein, The at least one machine learning algorithm is implemented on a recurrent neural network with long short-term memory units.

20. The system according to claim 13, wherein, The processor is further configured to predict the training dropout rate using the at least one machine learning algorithm.

21. The system according to claim 13, wherein, The processor is further configured to monitor the received feedback for at least one of timing, training session length, training session success rate, attention stability, freeze period, and number of breaks in a training session.

22. The system according to claim 13, further comprising at least one imager coupled to the processor, and wherein, The processor is further configured to monitor the eye movement of the user using the at least one imager.

Images