Abstract: A method for preferential system identification (PSID) for modeling neural dynamics of a brain includes extracting behaviorally relevant latent brain states via a projection of future behavior onto past neural activity. The method further includes identifying, based on the extracted latent brain states, model parameters for a linear state space dynamic model of neural activity. An extension of PSID includes accounting for an effect of external inputs to the brain by using oblique projections along the external inputs instead of orthogonal projections. PSID may be extended to apply to any primary signals and secondary signals in place of the neural activity and the behavior. In some embodiments, the extracting and the identifying may be performed using a numerical optimization of recurrent neural networks (RNN) instead of the projection.

Abstract: A sensor (e.g., an optical sensor) that may be implanted within a living animal (e.g., a human) and may be used to measure an analyte (e.g., glucose or oxygen) in a medium (e.g., interstitial fluid, blood, or intraperitoneal fluid) within the animal. The sensor may include a sensor housing, an analyte indicator covering at least a portion of the sensor housing, and a multiple metal protective system including multiple metals incorporated in and/or in close proximity to a surface of the analyte indicator that reduce deterioration of the analyte indicator.

Abstract: Provided is a system for predicting disease based on biosignal data and medical knowledge base convergence. The system includes a first system unit configured to receive, from a user terminal, biosignal data collected from at least one sensor for sensing a biosignal, and calculate a disease score from the biosignal data based on a pre-trained prediction model, a second system unit configured to provide medical knowledge data for the first system unit, analyze a query input from the user terminal to provide a corresponding response, and augment the medical knowledge data based on the query and response; and a unified distributed repository that includes a database for enqueuing the biosignal data, a manager database for storing additional information of a user, and a medical knowledge base for storing predetermined medical knowledge data.

Abstract: The present disclosure relates to software used in planning the correction of bone deformities preoperatively or postoperatively, and in particular relates to virtually manipulating rings and struts of an external fixation frame in order to plan the steps for making a desired correction to two or more bone portions of a patient. The software can be used prior to surgery, allowing a user to virtually define a bone deformity, and virtually add and manipulate fixation rings and struts to the bone deformity. Based on the virtual manipulations, a correction plan can be generated that describes length adjustments that should be made to the plurality of model struts over a period of time to correct the bone deformity. The software can also be used after surgical fixation of the fixation frame and struts to the deformed bone.

Abstract: A computing platform configured to receive and process an on-demand real-time patient-specific data analysis order is provided. The computing platform can receive an order, determine the viability of the order, and then perform the desired analysis based on parameters provided within the order. As part of the analysis, the computing platform can mine one or more data sources to collect data relevant to the ordered diagnostic. Once the data is collected, the computing platform can analyze the data according to one or more pre-programmed algorithms. The selection of which algorithms to apply to the data set can be determined by the type of on-demand real-time patient-specific data analysis ordered. The on-demand real-time patient-specific data analysis in some examples can be ordered using an external ordering user interface.

Abstract: The breastfeeding device to prevent baby suffocation includes a main unit and a remote unit. The main unit is mounted on a silicon pad or patch having an adhesive layer for attaching the main unit to one of the mother's breasts. The main unit includes a microcontroller having a sensor, such as a pressure sensor, connected thereto, either directly on the same printed circuit board or mounted external to the main unit and connected to the microcontroller by a wire or cable. The main unit has a transmitter mounted thereon for sending a signal to the remote unit when the sensor detects too much pressure being applied to the baby's mouth and nose. The remote unit has a receiver for receiving the signal, and a microcontroller for switching on an audible alarm when such a signal is received.

Abstract: A method (100) and system (1) for diagnosing reflux disease in an individual (3). The method (100) comprises: administering (110) an oral dose (17) of tracer in to the individual; capturing (120), with a single photon emission computed tomography (SPECT) scanner (5), data representative of multiple image frames (21) of the individual in areas of interest (23) that include the pharynx (25) and oesophagus (27), wherein the data representative of multiple image frames (21) are captured over time after administration of the oral dose; and determining (130) one or more trend(s) in tracer activity over time in the areas of interest (23) based on the data representative of multiple image frames (21), wherein the trend(s) in tracer activity over time is indicative of a reflux disease.

Abstract: An Internet of Thing (IoT) device includes a body with a processor, a camera and a wireless transceiver coupled to the processor.

Abstract: Various embodiments of the present disclosure are directed to model feature classification, analysis, and updating for analyzing one or more risk determination model(s). Embodiments include an improved apparatus configured to generate an actionable feature data object subset of a risk model feature set for a risk determination model, for example using an actionable determination model. The apparatus may provide the actionable feature data object subset for rendering to an interface of a client device. The apparatus may additionally or alternatively be configured to utilize user feedback, provided either directly or identified by the apparatus based on user interactions, to update the actionable determination model. The apparatus may additionally or alternatively be configured to maintain and utilize linked claim data object(s) for use in rendering a linked claim scores analysis interface that provides additional insight regarding the risk level of a particular entity.

Abstract: Systems and methods for reducing acute incident risks for dementia patients are provided. Benefits of the systems and methods include reducing healthcare costs, improving patient and caregiver outcomes and reducing caregiver burden. A server system analyzes data related to a dementia patient and the corresponding caregivers (e.g., family member, paid caregiver, physician) and calculates an acute incident risk. Based on the acute incident risk, the server system determines patient therapies to reduce the acute incident risk and also identifies caregiver education to improve the caregiver belief state, which reduces the acute incident risk. The system periodically re-calculates the acute incident risk and identifies patient therapies and caregiver education and motivation to further reduce or maintain the acute incident risk.

Abstract: A method includes receiving location data of a monitoring device when carried by a user and receiving motion data of the monitoring device. The motion data is associated with a time of occurrence and the location data. The method includes processing the received motion data to identify a group of the motion data having a substantially common characteristic and processing the location data for the group of the motion data. The group of motion data by way of processing the location data provides an activity identifier. The motion data includes metric data that identifies characteristics of the motion data. The method includes transferring the activity identifier and the characteristics of the motion data to a screen of a device for display. The activity identifier being a graphical user interface that receives an input for rendering more or less of the characteristics of the motion data.

Abstract: A breath analysis device into which a user exhales a breath sample is capable of venting an initial portion of the breath sample from the device, and routing a second portion of the breath sample into a disposable cartridge containing an interactant. The device may include a sensor, such as a pressure sensor, for detecting the initiation of exhalation, and may include a controller that switches a valve during the exhalation process to route a desired portion of the breath sample into the cartridge. After the exhalation process, an LED/photodiode arrangement, or another type of optical sensor, may be used to measure a color change produced by a chemical reaction in the cartridge, to thereby measure a concentration of a ketone or other analyte in the breath sample.

Abstract: One embodiment provides a method, including: receiving a selection to access interaction data associated with a target; receiving a designation of an interaction parameter, wherein the designation of the interaction parameter adjusts a defining standard for a direct interaction and an indirect interaction for the interaction data; determining, using a processor and based on the interaction parameter, whether each of the interactions from the interaction data is associated with a direct interaction or an indirect interaction; and generating an interaction visualization from the interactions, wherein the interaction visualization indicates interactions of the target with the at least one entity and wherein the interaction visualization visually distinguishes direct interactions from indirect interactions. Other aspects are described and claimed.

Abstract: An adaptive method of cleaning structural health monitoring (SHM) data based on local outlier factor is provided, including following steps: step 1: dividing SHM data to be analyzed into a series of data segments by using adaptive sliding window (ASW) technology; step 2: extracting time-domain statistical factors and frequency-domain statistical factors of each of the data segments to refine data information, thereby forming objects for study; step 3: calculating an outlier degree of each of the objects by using a weighted multiscale local outlier factor (WMLOF) based on feature factors; and step 4: detecting anomalies in SHM data by comparing a WMLOF value with a threshold value. The adaptive method can improve data quality effectively.

Abstract: Data characterizing an individual is received. Thereafter, one or more variables are extracted from the data so that, using a predictive model populated with the extracted variables, a likelihood of the individual adhering to a treatment regimen can be determined. The predictive model is trained on historical treatment regimen adherence data empirically derived from a plurality of subjects. Subsequently, data characterizing the determined likelihood of adherence can be promoted.

Abstract: Embodiments provide systems, methods, and apparatus for reducing the impact of user-entered data errors in a data management system (DMS) such as for diabetes. Embodiments include storing user-entered data received from a user into a primary DMS database; storing secondary data received from a source other than the user into a secondary tracking database; associating the secondary data with one or more events described by the user-entered data; cross-checking the user-entered data against the associated secondary data; evaluating user-entered data based on the cross-checking results; presenting for review evaluated user-entered data; treating user-entered data in the primary DMS database based on review results; and determining a diabetes management plan based on the treated user-entered data. Numerous other aspects are provided.

Abstract: The present disclosure relates to methods, systems, and non-transitory computer-readable media for determining causal contributions of dimension values to anomalous data based on causal effects of such dimension values on the occurrence of other dimension values from interventions performed in a causal graph. For example, the disclosed systems can identify an anomalous dimension value that reflects a threshold change in value between an anomalous time period and a reference time period. The disclosed systems can determine causal effects by traversing a causal network representing dependencies between different dimensions associated with the dimension values. Based on the causal effects, the disclosed systems can determine causal contributions of particular dimension values on the anomalous dimension value. Further, the disclosed systems can generate a causal-contribution ranking of the particular dimension values based on the determined causal contributions.

Abstract: The present invention relates to a sensor device (100, 101, 102, 103) for sensing physiological information of a subject. To efficiently avoid or reduce ambient interference, the sensor device comprises a communication unit (7) for receiving, from an external source, context information indicating one or more of a modulation mode, settings of an interference reduction algorithm, and spectral information indicating an ambient light spectrum, said spectral information allowing the sensor device to determine the modulation mode and/or settings of an interference reduction algorithm, one or more light emitters (1, 2) for emitting modulated light onto the tissue of the subject, and a light detector (4) for detecting light which is transmitted through the tissue or/and which is reflected from the tissue.

Abstract: A method includes collecting a set of user information, determining a user current performance by: determining a sleep score or debt, measuring the user's previous physical activity, heart rate, or body temperature, calculating weights for the sleep score and the sleep debt, the measured previous physical activity, heart rate, or body temperature, calculating a weighted sum of the weighted sleep score, the weighted sleep debt, the weighted user's previous physical activity, weighted heart rate, or weighted body temperature, setting a user target level of performance, measuring a circadian rhythm and a user's duration of sleep or sleep cycle, comparing the circadian rhythm, the duration of sleep or sleep cycle, the current performance level, and current time zone information over days to determine a separate user performance level, comparing the target level of performance to the separate user performance level, and providing an alert and feedback user related to the comparison.

Abstract: The present embodiments provide systems and methods for, among others, tracking sensor insertion locations in a continuous analyte monitoring system. Data gathered from sensor sessions can be used in different ways, such as providing a user with a suggested rotation of insertion locations, correlating data from a given sensor session with sensor accuracy and/or sensor session length, and providing a user with a suggested next insertion location based upon past sensor accuracy and/or sensor session length at that location.

Abstract: A method for calibrating measurements of human physiological state applied on an electronic device couples the device to at least one uncalibrated measurement device, worn or placed. Readings of a physiological state are collected by the uncalibrated measurement device and the readings are input into a model for calibration. A sub-module corresponding to the type of reading, for blood pressure or heart rate for example, is called up from the calibration model and a regression algorithm is applied to the calibration model to generate a calibrated reading, each sub-module in the model corresponding to one type of reading. The method also outputs calibrated physiological measurements by the called-up sub-module. A related electronic device and a non-transitory storage medium are also disclosed.

Abstract: An aspect provides a method, including receiving an indication that a tag associated with a real time location system has been detected at a pre-determined location, the detected tag being associated with a patient in a hospital; and assigning a status of the patient based on detection of the tag; and updating a bed availability indication based on detection of the tag. Other aspects are described and claimed.

Abstract: Method and System for providing a medical drug recommendation. Time and location data is read from received client data. Weather data is retrieved based on the time and location data and the weather data is stored in a user data set. A medical drug recommendation is computed from a user data set that comprises migraine event data by using a statistical model. Furthermore, a server is prepared to provide medical drug recommendations, wherein a statistical model for delivering a medical drug recommendation is adjusted by providing user data as input data to a statistical model, the user data sets comprising migraine events.

Abstract: A glove comprising a set of electrocardiogram leads, the set of electrocardiogram leads including: (a) a plurality of precordial electrocardiogram leads; (b) a pair of aVL and aVR electrocardiogram leads; and (c) an aVF electrocardiogram lead; wherein the glove records a 12-lead electrocardiogram by pressing the glove against skin of a user.

Abstract: A severity calculator analyzes patient information including a patient avatar and a patient state and along with a patient trajectory calculator, computes a predicted outcome which is used to administer treatment. The severity calculator systems and computer program products are also disclosed.

Abstract: A robotically-assisted surgical device assists robotic surgery with a surgical robot including a robot arm, and includes a processing unit and a display unit. The processing unit acquires 3D data of a subject; acquires kinematic information of the robot arm; acquires information of a surgical procedure for operating; acquires position planning information for a plurality of ports to be pierced on a body surface of the subject; acquires measurement information obtained by measuring a position of a first port pierced on the body surface among the plurality of ports; determines a position of at least one of remaining ports other than the first port, based on the measurement information of the position of the first port, the information of the surgical procedure, the kinematic information, and the 3D data; and causes the display unit to display information indicating the determined position of the remaining port.

Abstract: Aspects of the present disclosure are directed to devices, systems, and methods for optimized integration of a human operator with a machine for safe and efficient operation. Accordingly, aspects of the present disclosure are directed to systems, methods, and devices which evaluate and determine a cognitive state of an operator, and allocate tasks to either the machine and/or operator based on the cognitive state of the operator, among other factors.

Abstract: A system, method, and computer-readable medium are disclosed for performing an entity behavior cataloging operation. The entity behavior cataloging operation includes: identifying a security related activity, the security related activity being based upon an observable from an electronic data source; analyzing the security related activity, the analyzing identifying an event of analytic utility associated with the security related activity; generating entity behavior catalog data based upon the event of analytic utility associated with the security related activity; and, storing the entity behavior catalog data within an entity behavior catalog, the entity behavior catalog providing an inventory of entity behaviors for use when performing a security operation.

Abstract: A method and a system for determining the maximum heart rate, called HRmax of a user of in a freely performed physical exercise and using an apparatus with software and memory. An intensity model takes account of, in addition to heart rate, respiration rate and/or kinetics information.

Abstract: Described are systems and methods for generating sleep scores based at least in part on the visual appearance of hypnograms. The disclosed implementations generate an embedding vector representative of the visual appearance of a hypnogram and utilize that embedding to determine a sleep score of the sleep session represented by the hypnogram. For example, a deep neural network, such as a convolutional neural network (“CNN”), may be trained with input data that includes hypnograms that have been labeled according to their sleep quality, or sleep score. The trained network may then be provided with an unlabeled hypnogram, generate an embedding representative of that hypnogram, and determine a sleep score corresponding to that hypnogram.

Abstract: A system has a body with a base having a top surface with a receiving region to receive the bottom of a single foot. Among other things, the base may be in the form of an open or closed platform with a plurality of temperature sensors in communication with the top surface of the receiving region. The plurality of temperature sensors are within the receiving region and configured to activate after receipt of a stimulus applied to one or both the platform and the plurality of temperature sensors. A comparator is configured to form a temperature range as a function of the temperature value distribution and compare a percentage of the range size of the temperature distribution to a threshold value. An output produces ulcer information indicating the emergence of an ulcer or pre-ulcer when the percentage of the range size equals or exceeds the threshold value.

Abstract: Analytics dashboards for critical event management systems that include artificial-intelligence (AI) functionalities, and related software. AI functionalities disclosed include pattern recognition and predictive modelling. One or more pattern-recognition algorithms can be used, for example, to identified patterns or other groupings within stored critical events, which can then be used to improve response performance and/or to inform the generation of predictive models. One or more predictive-modeling algorithms can be used to generate one or more predictive models that can then be used, for example, to make predictions about newly arriving critical events that can then be used, among other things, to provide optimal response performance and allow users to efficiently and effectively manage responses critical events. These and other features are described in detail.

Abstract: A system for using artificial intelligence to analyze user activity data, the system comprising a computing device configured to receive from a user, at least a biological extraction and at least a user activity datum, determine a current user location, generate a diagnostic output as a function of the biological extraction, wherein the diagnostic output comprises a condition of the user, retrieve, from a fingerprint database, at least a datum of user fingerprint data, identify a plurality of compatible elements at the current user location as a function of the condition of the user, select at least a compatible element as a function of the fingerprint data, and present, via a graphical user interface, the at least a compatible element to a user device.

Abstract: Systems, methods and computer-readable media are provided for identification of patients or family member having genetic disease or probable genetic disease. During or after registration of a patient, parents, grandparents, or siblings of the patient are identified. If it is determined that one of the patient or the parents, grandparents, or siblings of the patient has been assigned with a diagnosis indicating a genetic disease, an alert for genetic disease or probable genetic disease for the patient or family member of the patient is provided. A clinician is then prompted to confirm or rule out the patient or family member inheriting the disease.

Abstract: A system and method for generating a master realtor script to aid marketing and sales of new development properties with manually entered parameters or parameters gathered through machine learning. The system includes a VR headset that is in communication with a server via a communication network. The server includes a database containing preferred parameters of a given user for a real estate option that are either manually entered or generated by a machine learning module. This may include, number of rooms, color, housing type, preferred furniture, family members, house layout, and other parameters. The system will then predict user preferences based on the predetermined parameters and assign a weight to each different type of parameters. Afterwards, a master realtor script is averaged and provided in real time via a virtual reality headset based on the weighted parameters.

Abstract: Embodiments of the disclosure provide systems and methods for generating a diagnosis report based on a medical image of a patient. The system includes a communication interface configured to receive the medical image acquired by an image acquisition device. The system further includes at least one processor. The at least one processor is configured to detect a medical condition based on the medical image and automatically generate text information describing the medical condition. The at least one processor is further configured to construct the diagnosis report, where the diagnosis report includes at least one image view showing the medical condition and a report view including the text information describing the medical condition. The system also includes a display configured to display the diagnosis report.

Abstract: The present invention relates to a visualized virtual agent configured to provide a visual response to a user, wherein the visualized virtual agent is further configured to provide an artificial physiological color change response to a user on the basis of one or more user-specific parameter, wherein the visualized virtual agent is configured to acquire user-specific data for at least one user-specific parameter using at least one sensor, the one or more user-specific parameter comprising activity data of the user and/or physio-psychological data of the user, and wherein the artificial physiological color change response is provided in response to monitored changes of the acquired user-specific data for the at least one user-specific parameter of the activity data of the user and/or the physio-psychological data of the user, the artificial physiological color change response comprising at least in part a color change of the visualized virtual agent.

Abstract: A DI recovery management system generates a plurality of ranked claimant records and recovery scores. A predictive machine learning model inputs disability income claim data and disability income claimant data into an event history model utilizing discrete-time survival analysis in conjunction with a gradient boosting machine learning model. The claim termination event is one of a plurality of preselected recovery events that indicate that a claimant has achieved return-to-work capacity. Claimant data used in modeling includes diagnosis data representative of workplace disability duration guidelines. The predictive machine learning model is continually trained using updated disability income claims data. The training procedure transforms claimant records extracted from a DI claims database into a longitudinal format that includes multiple person-year records corresponding to each claimant record.

Abstract: Systems, devices, and methods are provided that enable the revision of RF command handling software stored in ROM, and that enable to supplementation of RF command handling software stored in ROM. Examples of the systems, devices, and methods make use of a lookup data structure stored within writable non-volatile memory.

Abstract: The present systems and methods provide a mechanism to assess viewer behavior, features of stimuli, and the interaction between viewer behavior and stimuli. The systems and methods described herein for quantifying blink response and blink inhibition provide moment-by-moments measurements of viewer engagement by measuring what is or is not engaging enough to warrant viewers' inhibition of blinking. The present disclosure describes measures of visual scanning, eye movements, blink data, and blink timing data to derive a measure of how engaged a person is with what he or she is looking at. Blink-related data as a measure of viewer engagement provides a mechanism for determining the most engaging spatial and temporal aspects of a stimulus.

Abstract: Methods and apparatus for encrypting and decrypting data for wearable devices that are not based on authentication techniques, such as login/password or handshaking, are provided. A computing device receives a message. The message includes encrypted data and a cryptographic reference. The encrypted data includes physiological data of a wearer of the wearable device. The cryptographic reference includes a reference to a first cryptographic technique. The computing device determines the first cryptographic technique based on the reference to the first cryptographic technique. The computing device determines a cryptographic key. The computing device decrypts the encrypted data using the first cryptographic technique and the cryptographic key to obtain decrypted data. The computing device stores the decrypted data.

Abstract: A server in communication with a first device capable of receiving input from a camera attached to the first device, and a second device capable of displaying output, is provided. The server is configured to receive, from the first device, data relating to characteristics of the camera and characteristics of an image obtained by the camera of a screen of the second device; determine a screen pixel length of the screen; and provide instructions to the second device to display on the screen an object at a desired visual angle in response to the data received from the second device and the screen pixel length, while the first device is positioned at an arbitrary distance from the screen.

Abstract: The invention relates to mental health and using technology that can analyze and interpret hormones from drops of sweat and also a peer to peer feedback system to predict mental health issues (such as depression) in an individual. The invention further relates to the monitoring of sweat and feedback and interpreting the two in tandem to identify symptoms of mental health illnesses in an individual over a period of time (to predict and prevent that certain mental health disease). Furthermore, the records of the sweat analysis and peer to peer feedback will be utilized to provide smart suggestions and track the mental health of an individual.

Abstract: A system for use with a user interface system, a first data provider, a second data provider and a display device. The system comprises a data input component, a fishbone diagram generating component, an output component, a user interface component, and a data trend generating component. The data trend generating component can generate data trend image data based on a trends diagram instruction. The output component can further output the data trend image data to the display device to be displayed so as to include a first graphical function and a second graphical function, wherein the first graphical function corresponds to first patient laboratory test data, and wherein the second graphical function corresponds to second patient laboratory test data.

Abstract: The present invention provides an alarm system for monitoring a physical disorder of a user during an immersive virtual reality (VR) experience. The alarm system includes a physical information monitoring module, a control module, a display module, an alarm module, and a power module. The control module is configured to receive the human body physical sign information transmitted by the physical information monitoring module and feed back the information to the display module and the alarm module. When a user wearing a VR device has an adverse physical response during an experience and use, the alarm system of the present invention can quickly transmit a corresponding warning and prompt according to a detected abnormal physical sign. If the user has an adverse physical response and cannot carry out self-help, a remote call for help can be quickly made through software, thereby providing maximum protection for the user.

Abstract: Methods and systems determine risks of deterioration of hospitalized or other monitored or cared-for patients, for example in a treatment facility such as a hospital or under home-health care. In embodiments, a warning or other instruction is issued to medical professionals to alert them that certain patients have moderate or high risk of transfer to a higher level of care or should be monitored more frequently. A medical professional can accept alerts regarding prediction of deterioration, causing a prophylactic transfer or increased monitoring, or a transfer or monitoring order can occur automatically. Data relating to all patients in a unit of a medical facility can be viewed including warnings relating to risk of transfer or deterioration, so that a medical facility can intervene prior to an event such as a cardiac event and/or plan to accommodate patients at higher levels of care or monitoring.

Abstract: Methods and systems may incorporate voice interaction and other audio interaction to facilitate access to prescription related information and processes. Particularly, voice/audio interactions may be utilized to achieve authentication to access prescription-related information and action capabilities. Additionally, voice/audio interactions may be utilized in performance of processes such as obtaining prescription refills and receiving reminders to consume prescription products.

Abstract: Systems and methods for a comprehensive online healthcare platform are provided that increase the efficiency of the medication selection process. In one embodiment, a method comprises responsive to a query from the user via the user device, receiving a list of one or more medications identified from a storage device and receiving medication information about each identified medication in the list of the one or more medications from the storage device. The method further includes, calculating a user personalized grade for each of the identified medications based on the medication information and the user information, transmitting the user personalized grade and medication information to the user device and displaying simultaneously on the user device, the user personalized grade, cost factor, prescription experience data, and secondary technical effects of each identified medication.

Abstract: Described herein are software platforms (or “platforms”), systems, devices, and methods for providing effective healthcare to a patient while increasing the efficiency of the healthcare provider. A platform provides effective healthcare to a patient through, but not limited to, monitoring of the physical parameters and therapeutic objectives of the patient. In addition, a platform is further configured to provide a communication link between a patient and a healthcare provider. The platform may be implemented by a system comprising one or more computing devices that interact with or run one or more applications of the platform.

Abstract: The invention disclosed herein is for an in vitro cell-based assay for predicting the conversion from mild cognitive impairment to Alzheimer's disease in a patient who has received a diagnosis of MCI. The method comprises the following steps: a) culturing human hippocampal progenitor cells in a culture medium comprising serum, obtained from said patient, during a period of proliferation of said progenitor cells; b) subsequently culturing said hippocampal progenitor cells in a culture medium comprising serum, obtained from said patient, during a period of differentiation of said progenitor cells; c) determining the level of proliferation of said cultured progenitor cells; d) determining the average cell count of said cultured progenitor cells; and e) monitoring apoptotic cell death after differentiation of the proliferated hippocampal progenitor cells, wherein the outcomes of each of (c) to (e) are applied to a statistical analysis, the result of which is predictive of conversion from MCI to AD in the patient.