CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit under 35 U.S.C. § 119(e) of United States Provisional Application No. 63/083,652, filed September 25, 2020, which is hereby incorporated by reference in its entirety.

BACKGROUND

[0002] Human electroencephalography (EEG) recordings have been utilized for clinical and research purposes since the 1930s, but much is still unknown about the underlying neuronal dynamics responsible for scalp-recorded electric potential changes as a function of time.

Recently, many lines of investigation in neuroscience and statistical physics have converged to raise the hypothesis that the underlying pattern of neuronal activation which results in EEG trace recordings is nonlinear, with scale-free dynamics, while EEG signals themselves are nonstationary and also show scale-free dynamics. Therefore, traditional statistical methods of EEG analysis (e.g., Fourier Transform, frequency-averaged spectral analysis) may not be the most appropriate means to analyze EEG signals, since these techniques would miss many properties inherent in nonstationary signals with scale-free dynamics.

[0003] Traumatic brain injury (TBI) is a devastating consequence of multifactorial causes of brain damage, which results in considerable morbidity to affected individuals. In 2014, about 2.87 million TBI-related emergency department (ED) visits, hospitalizations, and deaths occurred in the United States. Over the span of eight years (2006-2014), age-adjusted rates of TBI-related ED visits increased by 54%. The peak of incidence of TBI in the general population is in males aged 15 to 24, and it has been estimated that several million people in the U.S. alone may suffer from resultant lifelong cognitive and physical impairment.

[0004] Many serious diagnostic and treatment- specific problems currently limit the ability of clinicians and health care organizations to treat TBI, as there are no single specific biomarkers for neurocognitive impairment or functional capacity limitations associated with TBI. The gold standard for diagnosis and treatment of TBI is a process that remains to be an expensive, intensive battery of neuroimaging, medical, and neuropsychiatric testing to arrive at a diagnosis, and that utilizes a complex diagnostic process to determine impressions of the patient’s level of functional capacity. Additionally, existent screening tools designed to assess for TBI are very nonspecific, and frequently provide positive screens for patients with no history of brain injury. One of the primary areas of cognitive dysfunction that individuals with TBI exhibit post injury, is the executive, which is responsible for important life functions, such as planning and organizing, following through with tasks, analyzing complex information, and modulating emotions. Therefore, there exists an urgent need to develop specific diagnostic tests for executive function deficits in TBI, both to aid in diagnosis and treatment.

[0005] Alzheimer’s disease (AD) is a chronic neurodegenerative disease that usually starts slowly and gradually worsens over time. It is the cause of 60-70% of cases of dementia. The most common early symptom is difficulty in remembering recent events. As the disease advances, symptoms can include problems with language, disorientation (including easily getting lost), mood swings, loss of motivation, not managing self-care, and behavioral issues. As a person’s condition declines, they often withdraw from family and society. [1] Gradually, bodily functions are lost, ultimately leading to death. Although the speed of progression can vary, the typical life expectancy following diagnosis is three to nine years.

[0006] Alzheimer’s disease is usually diagnosed based on the person’s medical history, history from relatives, and behavioral observations. The presence of characteristic neurological and neuropsychological features and the absence of alternative conditions is supportive. Advanced medical imaging with computed tomography (CT) or magnetic resonance imaging (MRI), and with single-photon emission computed tomography (SPECT) or positron emission tomography (PET) can be used to help exclude other cerebral pathology or subtypes of dementia.

[0007] The diagnosis of TBI and AD, as well as other brain-related neurological disorders, is time consuming and expensive.

SUMMARY

[0008] The present technology provides systems and methods that detects and analyze electroencephalography (EEG) signals, which is helpful for making diagnosis of diseases or conditions of the brain. The analysis uses two approaches, multifractal detrended fluctuation analysis (MF-DFA) and information transfer modeling (ITM), separately or in combination. As demonstrated in the experimental examples,

[0009] In accordance with one embodiment of the present disclosure, provided is a computer- implemented method for detecting a disease or condition in the brain of a human subject, comprising recording, with a plurality of electrodes placed on the scalp of the human subject, a period of electroencephalography (EEG) signals from each of the electrodes, extracting, from the EEG signals from each electrode, signal parameters, calculating, from the signal parameters, a multifractal spectrum, and correlating the multifractal spectrum to a reference multifractal spectrum associated with a disease or condition in the brain, thereby identifying the human subject as having the disease or condition.

[0010] Also provided, in another embodiment, is a computer-implemented method for detecting a disease or condition in the brain of a human subject, comprising recording, with a plurality of electrodes placed on the scalp of the human subject, a period of electroencephalography (EEG) signals from each of the electrodes, calculating a series of information transfer constants (K) for the EEG signals using information transfer modeling (ITM), and correlating the calculated information transfer constants to reference information transfer constants associated with a disease or condition in the brain, thereby identifying the human subject as having the disease or condition.

[0011] Yet another embodiment provides a computer- implemented method for detecting a disease or condition in the brain of a human subject, comprising recording, with a plurality of electrodes placed on the scalp of the human subject, a period of electroencephalography (EEG) signals from each of the electrodes, extracting, from the EEG signals from each electrode, signal parameters, calculating, from the signal parameters, a multifractal spectrum, calculating a series of information transfer constants (K) for the EEG signals using information transfer modeling (ITM), and correlating the multifractal spectrum and the information transfer constants (K) to a reference multifractal spectrum and reference information transfer constants associated with a disease or condition in the brain, thereby identifying the human subject as having the disease or condition. [0012] In some embodiments, the disease or condition is selected from the group consisting of traumatic brain injury (TBI), Alzheimer’s disease (AD), mild cognitive impairment (MCI), frontal-lobe lesion, attention deficit hyperactivity disorder, specific learning disability, mood disorder, bipolar disorder, autism spectrum disorders, fetal alcohol syndrome, neuroinflammatory disorder and spina bifida.

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIG. 1 illustrates equipment for recording electroencephalography (EEG) signals from a subject.

[0014] FIG. 2 illustrates a few multifractal spectra from measured EEG signals from subjects having different brain activities or conditions.

[0015] FIG. 3 shows classification tree plots for EEG parameter models with significant correlations with Tower test subscores.

[0016] FIG. 4 shows correlations of FT- and MF-DFA EEG-derived CART model for TPM with actual TPM scores.

[0017] FIG. 5 shows Correlations of FT and ITM EEG-derived CART model for TPM with actual TPM scores.

[0018] FIG. 6 illustrates a computer system that is useful for implementing the technology.

DETAILED DESCRIPTION

[0019] The present disclosure provides an improved methods and systems to make diagnosis of neurological disorders by recording and analyzing EEG signals.

I. Systems for Diagnosing Neurological Disorders

[0020] Quantitative analysis of human electroencephalogram signals (EEG) using currently available technology has not been shown to be clinically useful to help diagnose or treat any human brain disease. As demonstrated in the experimental examples of the instant disclosure, however, applying principles of information transfer modeling (ITM) to the analysis of EEG can allow computer-based, quantitative interpretations to provide concrete, actionable clinical information about underlying brain disease states. For example, computer-based ITM analysis of EEG will provide an accurate assessment of the severity of cognitive impairment in Alzheimer’s disease and traumatic brain injury (TBI).

[0021] ITM works by getting an accurate estimate of the flow of information both within and between EEG leads as a function of both time and spatial distance. These patterns of information flows between and within EEG leads are highly sensitive to the underlying pathology of the brain’s cerebral cortex, and thus provide an accurate assessment of the brain’s functioning.

[0022] Likewise, the examples show that multifractal detrended fluctuation analysis (MF-DFA) is also effective is extracting useful parameters correlating with clinically meaningful data and thus is useful for the diagnosis of AD, TBI, and other neurological disorders.

[0023] In accordance with one embodiment of the present disclosure, therefore, provided is a computer- implemented method for detecting a disease or condition in the brain of a human subject. In some embodiments, the method entails first recording, with a plurality of electrodes placed on the scalp of the human subject, a period of electroencephalography (EEG) signals from each of the electrodes.

[0024] Electroencephalography (EEG) is an electrophysiological monitoring method to record electrical activity of the brain. It is typically noninvasive, with the electrodes placed along the scalp. EEG measures voltage fluctuations resulting from ionic current within the neurons of the brain.

[0025] EEG recording can be obtained by placing electrodes on the scalp (as illustrated in FIG. 1) with a conductive gel or paste, usually after preparing the scalp area by light abrasion to reduce impedance due to dead skin cells. In some embodiments, electrodes are used, each of which is attached to an individual wire. In some embodiments, caps or nets are used to embed the electrodes.

[0026] Electrode locations and names are standardized. For instance, they are specified by the International 10-20 system for most clinical and research applications. This system ensures that the naming of electrodes is consistent across laboratories. In most clinical applications, 19 recording electrodes (plus ground and system reference) are used.

[0027] Each electrode can be connected to one input of a differential amplifier (one amplifier per pair of electrodes). A common system reference electrode can be connected to the other input of each differential amplifier. These amplifiers amplify the voltage between the active electrode and the reference.

[0028] During the recording, a series of activation procedures may be used. These procedures may induce normal or abnormal EEG activity that might not otherwise be seen. These procedures include hyperventilation, photic stimulation (with a strobe light), eye closure, mental activity, sleep and sleep deprivation. During (inpatient) epilepsy monitoring, a patient's typical seizure medications may be withdrawn.

[0029] The digital EEG signal is stored electronically and can be filtered for display. Typical settings for the high-pass filter and a low-pass filter are 0.5-1 Hz and 35-70 Hz respectively. The high-pass filter typically filters out slow artifact, such as electrogalvanic signals and movement artifact, whereas the low-pass filter filters out high-frequency artifacts, such as electromyographic signals. An additional notch filter can be used to remove artifact caused by electrical power lines (60 Hz in the United States and 50 Hz in many other countries).

[0030] For analysis, a specific length of EEG signal may be acquired from each lead. The length may be 5 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, 30 seconds, 40 seconds, 50 seconds, 60 seconds, 90 second or 120 seconds, without limitation. In some embodiments, EEG signals from all leads are used for further analysis. In some embodiments, a subset of selected leads are used. In some embodiments, the EEG signals from one or more of the leads are used as reference (e.g., at a mastoid).

Multifractal detrended fluctuation analysis (MF-DFA)

[0031] In one embodiment, the EEG signals are analyzed with multifractal detrended fluctuation analysis (MF-DFA). Accordingly, one embodiment provides a computer- implemented method for detecting a disease or condition in the brain of a human subject. The method, in some embodiments, entails recording, with a plurality of electrodes placed on the scalp of the human subject, a period of electroencephalography (EEG) signals from each of the electrodes, extracting, from the EEG signals from each electrode, signal parameters, calculating, from the signal parameters, a multifractal spectrum, and correlating the multifractal spectrum to a reference multifractal spectrum associated with a disease or condition in the brain, thereby identifying the human subject as having the disease or condition.

[0032] In some embodiments, a reference multifractal spectrum is not required. Instead, the calculated multifractal spectrum may be fit into a prediction model with suitable parameters.

[0033] The multifractal spectrum can be calculated with multifractal detrended fluctuation analysis (MF-DFA). The signal parameters used for the analysis can include, for instance mean, minimum, maximum Holder exponent, width of Holder exponent spectrum, and combinations thereof.

[0034] Detrended fluctuation analysis (DFA) is a method for determining the statistical selfaffinity of a signal. It can be used for analyzing time series that appear to be long-memory processes or 1/f noise. DFA may be applied to signals whose underlying statistics (such as mean and variance) or dynamics are non- stationary (changing with time). It is related to measures based upon spectral techniques such as autocorrelation and Fourier transform. A detailed description of the DFA method can be found in, e.g., Ihlen EA. Front Physiol. 2012 Jun 4;3 : 141.

[0035] The reference multifractal spectrum and the correlation method may be obtained from a training set. In some examples, the method of Classification and Regression Trees (CART) can be used on the training EEG parameter set and the actual testing EEG parameter set. The method can use output from a Delis-Kaplan Executive Function System (D-KEFS) Tower test subscores to make a classification model with the EEG training data parameters.

[0036] The trained model for each subscore and EEG training data parameters can then be used to predict the subscore from the second EEG test data parameters for each analysis. For correlations between the predicted EEG test data parameters and the actual subscores, Pearson’s moment correlation testing can be performed.

[0037] In some embodiments, the analysis is useful to identify a subject having a brain disease or condition, such as time per move (TPM). Information transfer modeling (ITM)

[0038] In some embodiments, the EGG signals are analyzed with information transfer modeling (ITM), which can be performed both within-lead and between-lead. It is known that any physical system can be considered from the point of view of information transfer, whereby information flows from a source (<?) to a destination (w). From information theory, therefore, the information flow (7) must obey the inequality:

I _u < I _q (1)

[0039] Following this, equations can be derived describing information transfer relationships for several different physical systems. Relevant to EEG, ITM predicts that there will be an information transfer constant K) for each instant such that voltage changes (AV) are proportional to corresponding time intervals (At) :

|AV|~|At| ^K (2)

[0040] EEG is a time series of voltage readings V(t), where t= l,2,...n, (length of series) for each value of t up to n-At , given a time interval At. In this manner, the value of the information transfer constant K) for each instant can be calculated:

[0041] Therefore, each segment of EEG can be characterized by a series of information transfer constant ratios, for different values of the time interval At (i.e, 1, 2, 4, 8, time steps, etc):

[0042] Mean information transfer constants as a function of time interval At can provide an accurate description of sleep staging from a single lead.

[0043] Concerning between-lead ITM analysis, for two EEG leads (e.g., 11 and 12) placed on different parts of the scalp, a measure of between-lead information flow can be defined by the information transfer constant ratio: _ Kll -itcr (5)

Kl2

[0044] For a given time interval At, the between-lead information flow for two leads (11 and 12) can be assessed by the following relation:

_ log (t+At)-V _;i (t)|) iter, At ~ jog (6) (t+At)-V _j2 (t)|)

[0045] In some embodiments, the information transfer constant ratio is used as a function of time interval would be as a mean value over a given length (N) of EEG (e.g., 30 s), with a series of different time intervals (e.g., ranging from 4 ms to 4 s in logarithmic steps).

[0046] In some embodiments, a subset of electrodes covering the majority of the scalp can be selected to keep the number of data points down to a reasonable number amenable to subsequent analysis.

[0047] For ITM, the method of Classification and Regression Trees (CART) can also be used on the training EEG parameter set, which can then be used to assess new or actual patients.

[0048] As demonstrated in the examples, the ITM method can be used to predict the mean accuracy ratio (MAR), first move time (FMT), rule-violations-per-item ratio (RVPI) or time per move (TPM), which are useful for detecting diseases or conditions of the brain.

Combinatory use of MF-DFA and ITM

[0049] As shown in Example 2, after the EEG signals were separately analyzed with ITM and MF-DFA, when the parameters were combined in a machine learning model, even stronger correlation was observed between the EEG data and characteristics of the brain conditions.

[0050] Such characteristics include the achievement (ACH), mean accuracy ratio (MAR), first move time (FMT), rule-violations-per-item ratio (RVPI), time per move (TPM) or total rule violation (TRV). Accordingly, in some embodiments, as a follow-up to the MF-DFA and ITM analyses, a combinatory correlation study can be performed. [0051] The method of any of such embodiments can be used to detect a disease or condition in the brain of a testing subject. Without limitation, the disease or condition includes traumatic brain injury (TBI), Alzheimer’s disease (AD), mild cognitive impairment (MCI), frontal-lobe lesion, attention deficit hyperactivity disorder, specific learning disability, mood disorder, bipolar disorder, autism spectrum disorders, fetal alcohol syndrome, neuroinflammatory disorder and spina bifida.

II. Systems for Implementing the Detections

[0052] FIG. 6 is a block diagram that illustrates a system 600 upon which any embodiments of the EEG acquisition and analysis and related technologies may be implemented. The computer system 600 includes a bus 602 or other communication mechanism for communicating information, one or more hardware processors 604 coupled with bus 602 for processing information. Hardware processor(s) 604 may be, for example, one or more general purpose microprocessors.

[0053] The computer system 600 also includes a main memory 606, such as a random access memory (RAM), cache and/or other dynamic storage devices, coupled to bus 602 for storing information and instructions to be executed by processor 604. Main memory 606 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 604. Such instructions, when stored in storage media accessible to processor 604, render computer system 600 into a special-purpose machine that is customized to perform the operations specified in the instructions.

[0054] The computer system 600 further includes a read only memory (ROM) 608 or other static storage device coupled to bus 602 for storing static information and instructions for processor 604. A storage device 610, such as a magnetic disk, optical disk, or USB thumb drive (Flash drive), etc., is provided and coupled to bus 602 for storing information and instructions. The computer system 600 may be coupled via bus 602 to a display 612, such as a LED or LCD display (or touch screen), for displaying information to a computer user.

[0055] An electroencephalogram (EEG) device 614, including accessories such as electrodes and screens, can be coupled to bus 602 for communicating information and command selections to processor 604. Another input device is cursor control 616, such as a mouse, a trackball, or cursor direction keys for communicating direction information and command selections to processor 604 and for controlling cursor movement on display 612. In some embodiments, the same direction information and command selections as cursor control may be implemented via receiving touches on a touch screen without a cursor. Additional data may be retrieved from the external data storage 618.

[0056] The computer system 600 may include a user interface module to implement a GUI that may be stored in a mass storage device as executable software codes that are executed by the computing device(s). This and other modules may include, by way of example, components, such as software components, object-oriented software components, class components and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables.

[0057] In general, the word “module,” as used herein, refers to logic embodied in hardware or firmware, or to a collection of software instructions, possibly having entry and exit points, written in a programming language, such as, for example, Java, C or C++. A software module may be compiled and linked into an executable program, installed in a dynamic link library, or may be written in an interpreted programming language such as, for example, BASIC, Perl, or Python. It will be appreciated that software modules may be callable from other modules or from themselves, and/or may be invoked in response to detected events or interrupts. Software modules configured for execution on computing devices may be provided on a computer readable medium, such as a compact disc, digital video disc, flash drive, magnetic disc, or any other tangible medium, or as a digital download (and maybe originally stored in a compressed or installable format that requires installation, decompression or decryption prior to execution). Such software code may be stored, partially or fully, on a memory device of the executing computing device, for execution by the computing device. Software instructions may be embedded in firmware, such as an EPROM. It will be further appreciated that hardware modules may be comprised of connected logic units, such as gates and flip-flops, and/or may be comprised of programmable units, such as programmable gate arrays or processors. The modules or computing device functionality described herein are preferably implemented as software modules, but may be represented in hardware or firmware. Generally, the modules described herein refer to logical modules that may be combined with other modules or divided into submodules despite their physical organization or storage.

[0058] The computer system 600 may implement the techniques described herein using customized hard-wired logic, one or more ASICs or FPGAs, firmware and/or program logic which in combination with the computer system causes or programs computer system 600 to be a special-purpose machine. According to one embodiment, the techniques herein are performed by computer system 600 in response to processor(s) 604 executing one or more sequences of one or more instructions contained in main memory 606. Such instructions may be read into main memory 606 from another storage medium, such as storage device 610. Execution of the sequences of instructions contained in main memory 606 causes processor(s) 604 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions.

[0059] The term “non-transitory media,” and similar terms, as used herein refers to any media that store data and/or instructions that cause a machine to operate in a specific fashion. Such non- transitory media may comprise non-volatile media and/or volatile media. Non-volatile media includes, for example, optical or magnetic disks, such as storage device 610. Volatile media includes dynamic memory, such as main memory 606. Common forms of non-transitory media include, for example, a floppy disk, a flexible disk, hard disk, solid state drive, magnetic tape, or any other magnetic data storage medium, a CD-ROM, any other optical data storage medium, any physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, NVRAM, any other memory chip or cartridge, and networked versions of the same.

[0060] Non-transitory media is distinct from but may be used in conjunction with transmission media. Transmission media participates in transferring information between non-transitory media. For example, transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprise bus 602. Transmission media can also take the form of acoustic or light waves, such as those generated during radio-wave and infra-red data communications.

[0061] Various forms of media may be involved in carrying one or more sequences of one or more instructions to processor 604 for execution. For example, the instructions may initially be carried on a magnetic disk or solid-state drive of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a telephone line using a component control. A component control local to computer system 600 can receive the data on the telephone line and use an infra-red transmitter to convert the data to an infra-red signal. An infra-red detector can receive the data carried in the infra-red signal and appropriate circuitry can place the data on bus 602. Bus 602 carries the data to main memory 606, from which processor 604 retrieves and executes the instructions. The instructions received by main memory 606 may retrieve and execute the instructions. The instructions received by main memory 606 may optionally be stored on storage device 610 either before or after execution by processor 604.

[0062] The computer system 600 also includes a communication interface 618 coupled to bus 602. Communication interface 618 provides a two-way data communication coupling to one or more network links that are connected to one or more local networks. For example, communication interface 618 may be an integrated services digital network (ISDN) card, cable component control, satellite component control, or a component control to provide a data communication connection to a corresponding type of telephone line. As another example, communication interface 618 may be a local area network (LAN) card to provide a data communication connection to a compatible LAN (or WAN component to communicated with a WAN). Wireless links may also be implemented. In any such implementation, communication interface 618 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.

[0063] A network link typically provides data communication through one or more networks to other data devices. For example, a network link may provide a connection through local network to a host computer or to data equipment operated by an Internet Service Provider (ISP). The ISP in turn provides data communication services through the world-wide packet data communication network now commonly referred to as the “Internet”. Local network and Internet both use electrical, electromagnetic or optical signals that carry digital data streams. The signals through the various networks and the signals on network link and through communication interface 618, which carry the digital data to and from computer system 600, are example forms of transmission media. [0064] The computer system 600 can send messages and receive data, including program code, through the network(s), network link and communication interface 618. In the Internet example, a server might transmit a requested code for an application program through the Internet, the ISP, the local network and the communication interface 618.

[0065] The received code may be executed by processor 604 as it is received, and/or stored in storage device 610, or other non-volatile storage for later execution. Each of the processes, methods, and algorithms described in the preceding sections may be embodied in, and fully or partially automated by, code modules executed by one or more computer systems or computer processors comprising computer hardware. The processes and algorithms may be implemented partially or wholly in application- specific circuitry.

[0066] The various features and processes described above may be used independently of one another, or may be combined in various ways. All possible combinations and sub-combinations are intended to fall within the scope of this disclosure. In addition, certain method or process blocks may be omitted in some implementations. The methods and processes described herein are also not limited to any particular sequence, and the blocks or states relating thereto can be performed in other sequences that are appropriate. For example, described blocks or states may be performed in an order other than that specifically disclosed, or multiple blocks or states may be combined in a single block or state. The example blocks or states may be performed in serial, in parallel, or in some other manner. Blocks or states may be added to or removed from the disclosed example embodiments. The example systems and components described herein may be configured differently than described. For example, elements may be added to, removed from, or rearranged compared to the disclosed example embodiments.

[0067] Any process descriptions, elements, or blocks in the flow diagrams described herein and/or depicted in the attached figures should be understood as potentially representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process. Alternate implementations are included within the scope of the embodiments described herein in which elements or functions may be deleted, executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those skilled in the art.

[0068] It should be emphasized that many variations and modifications may be made to the above-described embodiments, the elements of which are to be understood as being among other acceptable examples. All such modifications and variations are intended to be included herein within the scope of this disclosure. The foregoing description details certain embodiments of the invention. It will be appreciated, however, that no matter how detailed the foregoing appears in text, the invention can be practiced in many ways. As is also stated above, it should be noted that the use of particular terminology when describing certain features or aspects of the invention should not be taken to imply that the terminology is being re-defined herein to be restricted to including any specific characteristics of the features or aspects of the invention with which that terminology is associated. The scope of the embodiments should, therefore, be construed in accordance with the appended claims and any equivalents thereof.

[0069] The various operations of example methods described herein may be performed, at least partially, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Similarly, the methods described herein may be at least partially processor-implemented, with a particular processor or processors being an example of hardware. For example, at least some of the operations of a method may be performed by one or more processors. Moreover, the one or more processors may also operate to support performance of the relevant operations in a “cloud computing” environment or as a “software as a service” (SaaS). For example, at least some of the operations may be performed by a group of computers (as examples of machines including processors), with these operations being accessible via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., an Application Program Interface (API)).

[0070] The performance of certain of the operations may be distributed among the processors, not only residing within a single machine but deployed across a number of machines. In some example embodiments, the processors may be located in a single geographic location (e.g., within a home environment, an office environment, or a server farm). In other example embodiments, the processors may be distributed across a number of geographic locations. EXAMPLES

Example 1. Information Transfer and Multifractal Analysis of EEG in mild blast-induced TBI

[0071] This example demonstrates that two separate approaches derived from physics of time series analysis, multifractal detrended fluctuation analysis (MF-DFA) and information transfer modeling (ITM), can help diagnose blast-induced TBI.

[0072] Summary of Experimental Design: Participants (n=12) who previously participated in a Greater Los Angeles VA Healthcare System (GLAVAHS) research study of PET scanning as a potential diagnostic tool for TBI were invited back for collection of resting-state EEG (~5 minutes) and cognitive testing via a test not utilized in the original study. Participants eligible for the study had a history of mild blast-induced TBI incurred during their military service within five years of study entry, with no identifiable structural lesions.

[0073] The inclusion criterion for the study was completion of the VA research study “Regional cerebral metabolism in blast-induced mild TBI”. The exclusion criteria were: any indication of suicidal ideation (z.e., as assessed by question 9 on the Beck Depression Inventory, Revised (BDI-II)), or inability to participate in the EEG collection or cognitive testing. Per VA research guidelines, participants were compensated for their participation. All procedures described in this study were approved by the GLAVAHS IRB, and informed consent obtained from all subjects.

Behavioral and Cognitive Measures:

[0074] Beck Depression Inventory, Revised (BDI-II) is a 22-item measure of depressive symptomatology, widely used in psychological research and practice.

[0075] Delis-Kaplan Executive Function System (D-KEFS) - Subtest: Tower Test. The D- KEFS is designed to assess the key components of executive functions believed to be associated primarily with the frontal lobe. The Tower Test measures learning of rules, visuo-spatial planning, inhibition of impulsive responses, perseveration, and the ability to maintain and establish an instruction-based set. The time-per-move ratio score (TPM) was utilized as the primary outcome measure (as deficits in the cognitive ability to plan moves are largely independent of motor function, and are implicated as a sensitive indicator of frontal lobe dysfunction), with secondary outcome measures including total achievement score (ACH), mean first move time score (FMT), move accuracy ratio score (MAR), and rule violations per item ratio score (RVPI), and total rule violations (TRV) score.

[0076] Montreal Cognitive Assessment (MOCA) - is designed to provide an accurate assessment for mild to severe cognitive impairment associated with dementia such as Alzheimer’s disease or moderate to severe brain injury.

EEG Methods:

[0077] EEG Acquisition-. This example collected five minutes of resting-state EEG on participants utilizing a 64-channel EEG cap via the NetStation EEG system (Electrical Geodesics, Eugene, OR). Participants’ EEG activity was continuously recorded during 5 minutes while patients were instructed sit quietly with their eyes closed. EEG data were sampled at 250 Hz with filter settings of 0 to 100 Hz in DC acquisition mode. 64 cap-mounted, equidistant sintered Ag-AgCl electrodes were positioned via manufacturers’ instructions for use. EEG data were processed offline using NetStation EEG software, using a right mastoid reference electrode. However, given time constraints for participant contact and limited equipment availability, many of the recorded leads demonstrated elevated levels of impedance (-40% per participant on average), which likely impaired the overall quality of recorded EEGs. Nonetheless, all leads were utilized for further analysis, as a test of likely “real world” application of the EEG analytic techniques as described.

[0078] EEG Analysis'. This example chose two separate 30s epochs from each subject (one from 90 to 120 s of recording, one from 210 to 240 s of recording) for further analysis, using all 64 leads irrespective of the possibility of movement artifacts or poor impedance. These epochs were used separately for training and testing EEG data in order to respectively train and test the statistical models.

[0079] MF-DFA EEG analysis: 64 lead Digital EEG tracings (30s in each of two epochs as explained above) were analyzed using MF-DFA, using right mastoid as a secondary reference electrode. The following parameters were extracted from each lead: mean, minimum, maximum Holder exponents, and width of Holder exponent spectrum for further analysis.

[0080] ITM EEG analysis: This example performed both within-lead ITE analysis, but further characterized a between-lead EEG ITM analysis herein. Briefly, given appropriate constraints, virtually any physical system can be considered from the point of view of information transfer, whereby information flows from a source (<?) to a destination (w). From information theory, therefore, the information flow (7) must obey the inequality:

I _u < I _q (1)

[0081] This is because by information received cannot exceed information transmitted.

Following this, equations were derived describing information transfer relationships for several different physical systems. Relevant to EEG, ITM predicts that there will be an information transfer constant (K) for each instant such that voltage changes (AV) are proportional to corresponding time intervals (At) :

|AV|~|At| ^K (2)

[0082] EEG is a time series of voltage readings V(t), where t= l,2,...n, (length of series) for each value of t up to n-At , given a time interval At. In this manner, the value of the information transfer constant (K) for each instant can be calculated:

[0083] Therefore, each segment of EEG would be characterized by a series of information transfer constant ratios, for different values of the time interval At (i.e, 1, 2, 4, 8, time steps, etc):

[0084] Mean information transfer constants as a function of time interval At can provide an accurate description of sleep staging from a single lead. [0085] Principles of ITM can also be extended to between-lead information flow as a function of time. For two EEG leads (11 and 12, e.g.) placed on different parts of the scalp, a measure of between-lead information flow can be defined by the information transfer constant ratio:

[0086] This implies that for a given time interval At, the between-lead information flow for two leads (11 and 12) can be assessed by the following relation:

_ log iter, At ~ iog

[0087] In a manner similar to the single-lead ITM above (see equations 2-4), the most meaningful way to utilize the information transfer constant ratio as a function of time interval would be as a mean value over a given length (N) of EEG (e.g., 30 s), with a series of different time intervals (e.g., ranging from 4 ms to 4 s in logarithmic steps).

[0088] Given that the combination of within- and between-lead ITM analysis provides for an extremely large number of possible combinations (i.e., 5 time steps per lead, along with 5 time steps for each between-lead ITM combination >20,000 possible data points for all combinations of 64 leads with only 12 subjects), this would prevent the practical use of data mining techniques. Therefore, this example chose a subset of 10 electrodes covering the majority of the scalp to keep the number of data points down to a reasonable number amenable to subsequent analysis. The leads used were the following: leads 3 and 56 (right frontal area), leads 9 and 19 (left frontal area), 6 and 36 (anterior and posterior midline, respectively), 40 and 46 (right parietal), 26 and 31 (left parietal).

[0089] Spectral Analysis'. This example utilized the R program “spec.pgram” to perform a fast Fourier Transform on the same EEG utilized for MF-DFA and ITM, separately. This example extracted total spectral power for the following bandwidths: alpha (8-14 Hz), beta (16-31 Hz), delta (0.1-4 Hz), gamma (32-50 Hz), and theta (4-7 Hz) for each 30 s segment of each lead. [0090] Statistical Methodology: As above, the two separate 30 s EEG segments (either all 64 leads for MF-DFA and matched FT, or subset of 10 leads for ITM and matched FT analysis) were analyzed by each method listed above, respectively. The method of Classification and Regression Trees (CART) was used on the first (training) EEG parameter set, with a minimum split of 4, using the R program “rpart”, designed to follow CART closely. The rpart model utilized output from the D-KEFS Tower test subscores to make a classification model with the EEG training data parameters (1 ^st set of EEG-derived parameters). The trained rpart model for each subscore and EEG training data parameters was then used to predict the subscore from the second EEG test data parameters for each analysis. For correlations between the predicted EEG test data parameters and the actual subscores, Pearson’s moment correlation testing was performed using R. CART tree plots in FIG. 3 were drawn using the R ‘rpart.plot’ package.

[0091] The plots in FIG. 3 represent CART-derived tree plots of the actual successful correlation models. Each tree branch is described by the lead number, followed by the classifier after a period, then an operator, then the classifying value. For example, “L30.bet < 48e+3” indicates “lead 30 beta power < 4800” as the classifying value in the first panel. FT: alp- alpha; bet- beta power. MF-DFA: avgh- average Holder value; minh- minimum Holder value. ITM: V signifies the lead, with the number after the period indicating the time step At (in units of sampling rate 250 Hz), with the classifier as the value of the mean Information Transfer Constant K at the time step. Classifiers with two leads listed (e.g., “V6V52”) indicates the value of the relevant Information Transfer Constant Ratio KITCR, again with the number after the period indicating the time step in units of the sampling rate.

[0092] Demographic and clinical characteristics of the TBI participants: The subjects had a mean age of 30.8 ± 2.7 years of age. The sample was mostly male (10 subjects), mostly racially White (9 subjects; 2 Black subjects), with a substantial proportion of Hispanic ethnicity (5 subjects), largely representative of the Veteran population in the greater Los Angeles metropolitan region. There was no association between age, gender, race/ethnicity, BDI score, or MOCA score and any Tower test subscores (Table 1). All participants had perfect 30/30 MOCA scores, whereas there was significantly more variability in BDI scores and Tower test subscores amongst participants on the procedure day (Table 1). In addition to a diagnosis of a mild blast- induced TBI, all participants were also independently diagnosed (using the Structured Clinical Interview for Dsm-5 Disorders (Scid-5-cv): Research Version) with post-traumatic stress disorder as a result of their military combat experience. Despite elevated levels of depressive symptoms seen across subjects (BDI score 21 ± 8.9, Table 1), only four subjects carried a previous diagnosis of major depression, and none were taking psychiatric medications at the time of the study.

Table 1: Demographic and clinical characteristics of sample (N=12)

Abbreviations:

ACH: Achievement score;

FMT: First move time;

MAR: mean accuracy ratio;

TPM: Time-per-move ratio;

Listed statistics represent r values (Pearson's correlation for Age, BDI) and F(1 , 10) statistics (ANOVA for Male, White, and Hispanic categories). None of the listed statistics result in statistical significance at the p < 0.05 threshold.

[0093] MF-DFA differs from FT on ability to predict test EEG-derived Tower test scores: The MF-DFA test EEG-derived CART model correlated with actual Tower test TPM scores, whereas the FT CART model did not (Table 2, FIG. 4). However, the FT test EEG-derived CART model did correlate with actual Tower test RVPI scores, whereas the MF-DFA CART model did not (Table 2). Neither FT- nor MF-DFA test EEG-derived CART models correlated with Tower test ACH, FMT, MAR, or TRV subscores (Table 2).

[0094] In the data of FIG. 4, using all 64 leads and the first 30 s set of EEG parameters, a CART model was trained, and then tested on the novel 30 s set. The resulting predicted scores are correlated with the actual TPM score for each participant. A. FT-derived model. B. MF- DFA-derived model. For demonstration purposes only, a small amount of noise was added to the TPM scores.

Table 2: Tower test out-of-sample EEG correlation coefficients with recursive partitioning: MF-DFA vs. FT

ACH- achievement; MAR- mean accuracy ratio;

FMT- first move time; RYPI- rule-violations-per-item ratio;

TPM- time per move; TRY- total rule violations. Corr: Pearson's product-moment correlation value Bold,*: p<0.05.

[0095] CART-derived regression models and brain regions- FT vs. MF-DFA: Regression models derived from CART data are diagrammed schematically in FIG. 3. For the FT-derived CART model for RVPI, lead 2 (lower right frontal), lead 30 (left temporal lobe), and lead 58 (lower right frontal) were included model parameters.

[0096] For the MF-DFA-derived CART model for TPM, lead 1 (lower right frontal), lead 9 (middle left frontal), lead 24 (left temporal), and lead 52 (right temporal) were included model parameters.

[0097] On a reduced EEG dataset, ITM differs from FT on ability to predict test EEG- derived Tower test scores: Utilizing the reduced 10-lead EEG dataset, ITM test EEG-derived CART models correlated with actual Tower test ACH, FMT, MAR, RVPI, and TPM subscores, (Table 3, FIG. 5). By contrast, the FT test reduced lead EEG-derived CART model only correlated with the actual Tower test RVPI scores (Table 3). Neither method’s corresponding EEG-derived CART models correlated with TRV scores (Table 3). [0098] In FIG. 5, using the reduce 10 lead subset and the first 30 s set of EEG parameters, a CART model was trained, and then tested on the novel 30 s set. The resulting predicted scores are correlated with the actual TPM score for each participant. A. FT-derived model. B. ITM- derived model. For demonstration purposes only, a small amount of noise was added to the TPM scores.

Table 3: Tower test out-of-sample EEG correlation coefficients with recursive partitioning: ITE vs. FT

ACH- achievement; MAR- mean accuracy ratio; FMT- first move time;

RYPI- rule-violations-per-item ratio; TPM- time per move;

TRY- total rule violations

Corr: Pearson's product-moment correlation

Using reduced 10-lead EEG dataset.

Bold,*: p<0.05.

[0099] CART-derived regression models and brain regions for the reduced lead set: FT vs. ITM: For the reduced lead FT-derived CART model for RVPI, lead 3 (middle right frontal) and lead 26 (left parietal/temporal) were included model parameters (FIG. 3). For the reduced- lead ITM-derived CART model for ACH, leads 6/56 (middle frontal to left frontal), lead 3 (middle right frontal), lead 31 (left parietal), lead 36 (midline parietal), and lead 40 (right parietal) were included parameters. For the ITM-derived CART model for MAR, lead 6 (middle frontal), leads 26/36 (left temporal to midline parietal), and lead 3 (middle right frontal) were parameters. For the ITM-derived CART model for FMT, leads 19/31 (left frontal to left parietal), lead 3 (middle right frontal), lead 36 (midline parietal), and lead 56 (lateral right frontal) were parameters. For RVPI, lead 3 (middle right frontal), lead 40 (right parietal), and lead 9 (middle left frontal) were CART model parameters. Finally, for TPM, leads 9/56 (middle left frontal to later left frontal), leads 3/6 (middle right frontal to midline frontal), lead 9 (middle left frontal), and lead 36 (midline parietal) were ITM-derived CART model parameters (FIG. 3).

[0100] Differential performance of MF-DFA and FT on the ability to predict Tower test executive function subscores from EEG: EEG-derived parameter models correlated with different aspects of Tower test performance (Table 2, FIG. 4). While TPM has been perhaps the best-studied subscore of the Tower test, RVPI has been shown to be specifically impaired in a small study of patients with focal lateral prefrontal cortex lesions. Therefore, MF-DFA-derived EEG in this study did correlate with the most widely-used measure of executive function in the Tower test, there should certainly continue to be a role for FT, especially with regard to lesions with a propensity to RVPI impairment.

[0101] Differential performance of ITM and FT and on the ability to predict Tower test executive function subscores from EEG: ITM-derived EEG parameters proved to be the most globally correlative with Tower test executive function subscores of the tests examined here, even in the reduced-lead paradigm (FIG. 5, Table 3). Indeed, ITM analysis only failed to correlate with Tower test TRV score amongst all subscores. By comparison, in the same reduced-lead subset, FT correlated only with RVPI (Table 3).

[0102] General utility of MF-DFA, ITM, and FT as diagnostic tools for executive function deficits from EEG: ITM-derived EEG parameters clearly outperformed both MF-DFA and FT in this paradigm, in that in this study they were able to correlate with most Tower test subscores. However, practically speaking, MF-DFA (in the case of TPM), and FT (in the case of RVPI) show promise in correlating with two of the most important subscores. The between-lead ITM analysis first described here represents a considerable improvement over within-lead ITM, as four of the five CART-derived correlation models incorporated between-lead analysis.

Example 2. Combined Use oflnformation Transfer and Multifractal Analysis

[0103] This example tested a further improvement from Example 1. In the new approach, the EEG were separately analyzed with ITM and MF-DFA, and the parameters were then combined in a machine learning model (support vector machine) to find the best predictors. [0104] Correlation coefficients between machine learned models applied to new test EEG vs. actual scores using SVM models.

[0105] The results are shown in Table 4 below.

Table 4. Combined analysis using both ITM and MF-DFA

Bold:* p<0.5, **p<0.001

[0106] As shown, in all cases the best model made from the combined parameter dataset exceeded the predictive ability of the best models made from either MF-DFA or ITM parameters alone.

* * *

[0107] Throughout this specification, plural instances may implement components, operations, or structures described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the individual operations may be performed concurrently, and nothing requires that the operations be performed in the order illustrated. Structures and functionality presented as separate components in example configurations may be implemented as a combined structure or component.

Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.

[0108] Although an overview of the subject matter has been described with reference to specific example embodiments, various modifications and changes may be made to these embodiments without departing from the broader scope of embodiments of the present disclosure. Such embodiments of the subject matter may be referred to herein, individually or collectively, by the term “invention” merely for convenience and without intending to voluntarily limit the scope of this application to any single disclosure or concept if more than one is, in fact, disclosed.

[0109] The embodiments illustrated herein are described in sufficient detail to enable those skilled in the art to practice the teachings disclosed. Other embodiments may be used and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. The Detailed Description, therefore, is not to be taken in a limiting sense, and the scope of various embodiments is defined only by the appended claims, along with the full range of equivalents to which such claims are entitled.

[0110] As used herein, the term “or” may be construed in either an inclusive or exclusive sense. Moreover, plural instances may be provided for resources, operations, or structures described herein as a single instance. Additionally, boundaries between various resources, operations, and data stores are somewhat arbitrary, and particular operations are illustrated in a context of specific illustrative configurations. Other allocations of functionality are envisioned and may fall within a scope of various embodiments of the present disclosure. In general, structures and functionality presented as separate resources in the example configurations may be implemented as a combined structure or resource. Similarly, structures and functionality presented as a single resource may be implemented as separate resources. These and other variations, modifications, additions, and improvements fall within a scope of embodiments of the present disclosure as represented by the appended claims. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.

[0111] Conditional language, such as, among others, “can,” “could,” “might,” or “may,” unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular embodiment. [0112] Although the invention has been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred implementations, it is to be understood that such detail is solely for that purpose and that the invention is not limited to the disclosed implementations, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For example, it is to be understood that the present invention contemplates that, to the extent possible, one or more features of any embodiment can be combined with one or more features of any other embodiment.