METHODS AND APPARATUS FOR DIAGNOSING DISABILITIES IN A PATIENT

FIELD OF THE INVENTION

The present invention relates to a method and apparatus for diagnosing and tracking disabilities in a patient, and more particularly, in a patient who may suffer progressive decline in both motor and cognitive capabilities, including those suffering from Parkinson's disease, Alzheimer's disease, posterior cortical atrophy and other psychogenic motor illness.

BACKGROUND OF THE INVENTION

Disability management has broad social, economic and political importance. In North America, the cost of disability management has reached tens of billion dollars each year. In California alone, roughly one million workers' compensation cases are filed each year and $3.5 billion were paid in indemnity benefits in 2001. However, many cases of disability are contested because current medico-legal methods utilized to determine the extent of disability are somewhat subjective. The subjective nature of the methods permits fraud, leads to excessive litigation and inequitable settlements where seriously injured workers receive less benefits than those with minor injuries. The process can be inequitable to both claimant and employers and their insurers.

The courts are now requiring that a medical disability due to chronic illness (e.g. AIDS, Parkinson's disease, multiple sclerosis, and diabetes) must be determined objectively on an individual basis. In response to these rulings, various U.S. government agencies have clarified their disability assessment guidelines. They specifically stipulate that medical assessment must show: 1) a disability in comparison to normal average; 2) a record of disability occurrence and any effective mitigating measures (drugs or prosthetics) that have been undertaken and, 3) because abnormal physical findings may be intermittent, their presence over a period of time must be established by a record of ongoing management and evaluation.

For the medical community and insurance industry, ensuring compliance with the new legislative requirements is not a trivial undertaking. First, medical exams are traditionally designed for assessing impairment, not disability, although medical evidence remains the

cornerstone for the determination of disability. A clear conceptual distinction exists between impairment and disability. Physical impairment does not necessarily take away a person's ability to live and work. For a truck driver with Parkinson's disease, tremors are not considered disabling; but may be severely disabling for an eye surgeon. Second, when assessing the degree of disability, objectivity is not assured if functional assessments involve questionnaires, interviews, patient's opinions and subjective scores on an ordinate scale (e.g. mild vs. severe). Furthermore, they are prone to false positives, such as with malingerers and those with hysterical claims of disabilities. Third, for most medical clinics it is nearly impossible to conduct regular disability assessment for each patient and produce a longitudinal record for each case. For patients with chronic illnesses this becomes particularly problematic as motor deficits under these conditions often evolve over many years.

Disability and impairment are sometimes used interchangeably, but they are two distinct concepts. The World Health Organization defines impairment as "any loss or abnormality of psychological, physiological or anatomical structure or function." It defines disability as any restriction or lack of ability to perform an activity in the manner or within the range considered normal for a human being. The American Disability Act uses a two-part definition of disability, (a) a physical or mental impairment that substantially limits (one or more of) the major life activities of an individual; (b) a record of such impairment. Generally speaking, it is assumed that there is a direct correlation between impairment and disability. However, the limitation or severity test is rather stiff, as comparisons must be made to "average" people, and the impairment must be relatively permanent. For example, individuals with moderate difficulty in walking and unable to lift 25 pounds have failed the average- person test, and while those enduring the effects of major surgery and even heart attacks have failed the permanence test.

Current methods that are used to assess disability vary from subjective assessments, such as opinions from patients, relatives and physicians and structured questionnaires, to objective tests such as X-rays, which have a subjective element in their interpretation. The results are ordinal, but not interval. A result that is numerically twice another is not necessarily twice as bad or good. There are also floor and ceiling effects. Examples of questionnaire-based disability ratings are the classic Barthel Index, Functional Independence Measure (proprietary) and Functional Assessment Measure (mainly cognitive disability). The

American Medical Association has a permanent impairment guide (the "AMA Guide"), which is widely adopted as a rating tool. It first generates fractional disability scores for each body part and movement, which are averaged to give a weighted percentage of whole person impairment. The assessment involves comprehensive visual scoring and requires special training. For the regular physician, it is complex and cumbersome to apply on a daily basis in a clinical setting. Visual methods of scoring motor disabilities are not only prone to errors but in fact, they are rarely conducted in most medical clinics because of a lack of trained staff.

Quantifiable disability measures for chronic disabling diseases such as Parkinson's disease, Alzheimer's disease, multiple sclerosis and certain mental illness are problematic, despite the fact that the incidence of these diseases is climbing as the population ages. The methods noted above often become highly unreliable given that the behavioral deficits of these disorders have a long latent period and often fluctuate during the course of treatment. Furthermore, clinical diagnostic tools are predominately subjective, using relative descriptors of the severity of observed deficits. Existing tools can under- or over-estimate the extent of motor disabilities, especially for small and incremental changes that evolve over a long time period.

Therefore, there is a need in the art for objective tests for diagnosing and monitoring disability in a patient, which mitigates the difficulties presented by the prior art.

SUMMARY OF THE INVENTION

The present invention comprises a system for quantitative motor disability assessment. The system is adapted to provide data to determine whether a patient meets the medical, social, or legal definitions of disability and specific examination requirements in a regular clinical setting.

The system comprises a measurement device in the form of small mountable package of inertial sensors and a compact battery of sensorimotor tests capable of capturing the basic capabilities of a patient in terms of movement kinematics, force production and cognitive control capability.

The kinematic and force data may be obtained from the neck, trunk or limbs of a test subject, digitized and transmitted to a central database or processor in real time. Alternatively, the data may be stored and transferred at a later time. The sensorimotor and cognitive tests may be those considered disability-relevant. For example, a set of motor tasks may be considered under US Social Security Guidelines of Disability Evaluation (Blue Book), the contents of which are incorporated herein by reference. Because of the relative simplicity in task design and in the operation of the monitoring device, incremental degradations in a patient's performance in these tasks may be documented during a patient's regular visit to a non-specialist clinic in both urban and rural settings, permitting a longitudinal legal record to capture the evolution of a person's disabilities.

One example to illustrate the current invention would be a patient who has, or is suspected of having, Parkinson's disease (PD). This condition is particularly suitable because patients with PD often display many common motor impairments that evolve over time. As well, there are known cases of legal dispute of disability assessment in the context of this disease.

The kinematic data is processed with diagnostic indicators to report the degree of a person's disability, both currently and longitudinally by comparing patient's data over time.

A loss of brain control over voluntary movement or the appearance of involuntary movement may impair the mobility of one or more body part(s). Obviously, not all impairments will lead to physical disability. Depending upon the individual circumstances and the degree of disease progression, some impairment can affect the quality of daily living more than the ability to work or vice versa. The present invention attempts to measure the relationship between the two parameters. The measure is device-based and relevant to the patient's occupation or daily living activities and user friendly for clinical practitioners including nurses and other health alliance workers.

The present invention operates on the premise that the degree of physical and cognitive disabilities can be evaluated and established based on the degree of degradation of a patient's performance of a basic living-and-working task (LWT). Degradation means the impairment of any one or more of the speed, accuracy, force generation or endurance associated with the completion of an LWT . These parameters are collectively referred to

herein as kinematic data. An LWT may be an ambulatory task or a task which mimics or simulates a household or work related task, or a combination thereof. In one example, the patient whose is a receptionist by profession, has or is suspected to have Parkinson's disease. An LWT design would comprise motions related to responding to telephone ring; picking up the handset; typing on a keyboard; and open and close a file cabinet. The performance degradation will be recorded and its degree or severity assessed against two types of benchmark dataset: the same task performed by and the kinematic data collected from a population of unimpaired and aged matched controls (floor) and by those severely disabled individuals (ceiling), as well as the same task performed by the patient over an extended time period .

Therefore, in one aspect, the invention may comprise a method of diagnosing a disability in a person, or measuring change in a disability in a person, comprising the steps of:

(a) obtaining kinematic data from a patient performing an LWT task; and

(b) processing the kinematic data to obtain at least one objective diagnostic indicator selected from the group consisting of:

i. a quantitative value derived from a measurement of speed, accuracy, acceleration, force generation or endurance during performance of the LWTi. an impairment/disability ratio;

ii. a false claim detection value; and

iii. a subject-object mirror sensing value; and

(c) comparing the diagnostic indicator to a standard or standards, or to a previous diagnostic indicator obtained from the person.

In one embodiment, the LWT task may comprise two or more tasks designed to test different aspects of the actual or suspected disability.

In another aspect, the invention comprises a computer-implemented system for diagnosing a disability in a person, or measuring change in a disability in a person, comprising of:

(a) a plurality of sensors for obtaining kinematic data from a patient performing a LWT;

(b) means for collecting kinematic data from the sensors;

(c) means for utilizing the kinematic data to obtain an objective diagnostic indicator from one or more diagnostic indicators selected from the group consisting of:

i. a quantitative value derived from the measurement of speed, accuracy, acceleration, force generation or endurance during performance of the LWT;

ii. impairment/disability ratio;

iii. false claim detection; and

iv. subject-object mirror sensing; and

(d) means for comparing the diagnostic value to a standard or standards, or to a previous diagnostic value obtained from the person.

The diagnostic indicators, alone or in combination, provide an index with which one may help assess the degree and time course of performance degradation.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described by way of an exemplary embodiment with reference to the accompanying drawings.

Figure 1 is a schematic flowchart of the procedures in one embodiment of the present invention.

Figure 2 is a schematic depiction of a system of the present invention.

Figure 3 shows a plot of thigh tilt measurement of a normal subject and a PD patient in a timed up and go test.

Figure 4 shows foot acceleration, velocity and position measured from a normal subject during a freeze scenario test.

Figure 5 shows foot acceleration, velocity and position measured from a PD patient during a freeze scenario test.

Figure 6 shows a foot tapping power spectrum plot for a normal subject and a PD patient measured during a dual task test.

Figure 7 shows a plot of right thigh tilt during trials of a complete timed up-and-go test for both a normal subject and a PD patient.

Figure 8 shows a plot of right thigh tilt during trials of the standing and initial walking portion of the timed up-and-go test for both normal subject and a PD patient.

Figure 9 shows plots of left foot acceleration and right thigh tilt recorded directly from sensors during a timed up-and-go test, as well as derived plots of foot velocity and foot position.

Figure 10 shows the angular velocity of a hand and a cup in a "mirror sensing" test as well as a power spectrum plot, performed by a person mimicking hand tremors.

Figure 11 shows the angular velocity of a hand tremor by an actor subjected to attention distraction, in an example of false claim detection.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides for a method and system for diagnosing and monitoring physical disability in a patient. When describing the present invention, all terms not defined herein have their common art-recognized meanings. To the extent that the following description is of a specific embodiment or a particular use of the invention, it is intended to be illustrative only, and not limiting of the claimed invention. The following description is intended to cover all alternatives, modifications and equivalents that are included in the spirit and scope of the invention, as defined in the appended claims.

"Kinematic data" means data that is derived from motion of the human body, relative to a fixed point, without reference to the forces causing the motion. As used herein, kinematic data may include "force data" which means data that is derived from acceleration/deceleration associated with the initiation, termination and position changes of an LWT, and also "cognitive data" which means the changes in the timing, force and kinematic performance and the error rate of the performance as a result of changing cognitive conditions such as during divided attention or memory loading. Kinematic data may be processed to provide quantitative measurements such as speed, accuracy, acceleration, force generation or endurance, for the LWT or any portion of the LWT.

An exemplary condition involving motor disability is Parkinson's disease or PD. Gait abnormalities are major disabling symptoms of idiopathic PD. In the early stage of the disease, they are characterized by reduced speed, diminished amplitude of leg movement and shortened stride length. Although other kinematic subcomponents of walking (e.g. cadence, stride duration or double support duration) are also reduced, it is the lost ability to self regulate the stride length that plays a central role in Parkinsonian gait pathology. Indeed, it is known that Levodopa treatment, subthalamic deep brain stimulation and auditory or visual cueing all result in increase in stride length and markedly improved gait function in PD patients. Hence, understanding the neural mechanism and pathophysiology of gait control in PD has important clinical implications.

Kinematic and biomechanical studies of gait control indicate that in order for the body to propagate forward, the central nervous system must not only provide locomotive instructions (i.e. initiate and maintain rhythmic stepping) but also adequate equilibrium control that supports body weight and stability. There are two neural mechanisms for such stability control: reactive and proactive. Reactive control is used for unpredictable upsets to balance, and therefore depends on sensory feedback. Proactive control is broken down into two subtypes. The first is involved in counteracting perturbations caused by the gait movements themselves. The second is an experience-based or feed-forward system that uses internal or external cues to anticipate motion trajectory and predict potential causes of dysequilibrium and implement appropriate avoidance strategies. Impairment of either may cause significant disability. In the case of PD, it is the latter capacity that is compromised.

Without limitation, it is known that other causes of motor disability that may fall within the scope of the claimed invention include Alzheimer's disease, posterior cortical atrophy and psychogenic motor illnesses.

One embodiment of the process of the present invention is illustrated schematically in Figure 1. Initially, a LWT task related to the suspected disability is chosen and the patient performs the task while kinematic data is collected (101). The data are analyzed and an objective diagnostic value is obtained. If the same task has been previously performed by the same patient, the diagnostic value may be compared with previous results (102) to provide a longitudinal PDI (103). The diagnostic value may also be compared with values collected from patients known to suffer the suspected disability (104), as well as from control subjects known not to suffer the suspected disability. In one embodiment, the data may be stored and filed for future use and comparison purposes, and optionally, transmitted to a remote location.

One embodiment of a computer-implemented system of the present invention may comprise a system having an orientation- tracking device including accelerometer sensors (10) that measure the acceleration, both linearly and in angular terms, experienced by the sensor (10) and anything to which the sensor (10) is directly attached. Sensors allow detection, analysis, and recording of physical phenomena that are otherwise difficult to measure by converting the phenomena into more convenient physical measurements such as acceleration or velocity. Acceleration data may be converted into measurements of displacement, velocity, force or other parameters.

The sensors (10) are preferably micro-electro-mechanical systems (MEMS) accelerometers and may also be accompanied by rate gyroscopes so that movements in all 3 dimensions may be recorded. The rate gyros are inertia sensitive and yield digitized signals of instantaneous angular velocity. The sensors may be attached to the foot, leg, hand or trunk, alone or in various combinations, depending on the movement that is being measured. To measure movement in all 6 degrees of freedom (DOF), angular rate and the linear acceleration are measured relative to each of three orthogonal axes.

Preferably, the sensors are small and lightweight so as to have minimal impact on a user's movements. With adequate bandwidth, preferably at least about 40 Hz, the devices may output analog signals (0-5V) to a software package incorporating algorithms developed

to translate each sensor signal into coordinates relative to space. The software package may have a graphical user interface and may also include post analysis subroutines of data mining including various statistical evaluations of peak velocities, movement durations and other available data.

In one embodiment, the sensors (10) communicate with a small handheld computer (12), such as a PDA, which acts as a data logger and may conveniently be worn or carried by the patient. Preferably, the computer (12) has short range wireless connectivity, such as Bluetooth capability, to communicate with a host computer (14), which may be a personal computer such as a desktop, laptop, or another handheld computer. Other suitable wireless protocols will be apparent to those skilled in the art. Alternatively, if the handheld computer (12) has sufficient memory and processing power, data processing and analysis may take place on the handheld (12).

The handheld computer (12) may also communicate wirelessly with a headset (16), such as a common Bluetooth headset used for cellular telephony, so that instructions, sounds, or other auditory signals or responses may be provided to or received from the patient.

The kinematic data obtained from the use of the sensors are then processed on the host computer (14) or remotely to produce one or more diagnostic indicator of the present invention. A diagnostic indicator is a quantitative parameter that helps define the physiological relationship between an impairment and disability.

In one embodiment, a diagnostic indicator of the present invention may be a quantitative measurement derived from kinematic data arising from a LWT. It is very common for physicians to hear a patient complaining about a disability that is specifically related to her or his living or working conditions. Examination of such complaints is difficult in a medical clinic due to a lack of proper tests and the lack of a standardized rating tool. A basic LWT consists of a sequence of natural "acts" which mimic or substitute for everyday tasks the patient may encounter at home or in the workplace.

For Parkinson's disease, the Disability Blue Book indicates that both the inability to ambulate effectively and the inability to perform fine and gross movements effectively are broad categories of impairments that should be considered. A number of specific motor acts

are also described in the Blue Book, which allows one to construct different LWTs and use them as templates for individual patients. For example, an LWT designed to test four basic abilities: posture control (sit down and up), ambulation (linear and rotational), meal preparation and fine/gross movements may consist of the following sequence of actions:

1) sit down on, and then stand up from a chair;

2) make a left (or right) turn and walk along a circular path to a desk;

3) fetch a kitchen pot on the desk and walk toward a cabinet;

4) open the first cabinet drawer;

5) put the pot inside the drawer and close it.

Typically, the patient may be allowed to practice the task once and it will be repeated three times while kinematic data is collected.

In one embodiment, the sensors may be placed on the subject's lower legs, trunk and dominant hand. Compared with measuring ambulation, postural measurement is rather straightforward. Sitting, standing up and turning requires coordinated movements between the limbs and trunk. These movements are routinely used in the Unified Parkinson's Disease Rating Scale (UPDRS). The accelerometers may be placed at the back of the subject at the approximate height of the center of mass. Two other accelerometers may be placed on each foot. The absolute speed and amplitude as well as relative displacement between the sensors will be recorded during locomotion and may be analyzed in both the time and frequency domains. In a preferred embodiment, the sensors and control software are capable of consecutive sampling of time and force of sensor reaction and step initiation. Scoring will be based on statistical deviations of patients' data from average controls, which often have a Gaussian distribution.

In one embodiment, the LWT may comprise a series of tasks, each designed to test a different aspect of the suspected disability. In one embodiment, the LWT test may comprise a test which comprises multiple parts, which may be three parts in one example. In a preferred embodiment, the different parts of a multipart test are known or designed to be targeted at

different populations or subsets within the suspected disability. A first part may test a basic physical task which incorporates a measurement of reaction or motor ability, such as getting up out of a chair upon a randomly generated signal, such as a telephone ring, and walking forward to a designated spot. A second part may test a cognitive aspect with motor ability consequences, and a third part may test simultaneous sensory, cognitive or motor tasks or simultaneous mixed auditory or visual motor tasks. A combination of three different tests of this nature may give greater sensitivity and discrimination to the diagnostic indicators obtained.

In one embodiment, a first task may be referred to as timed up-and-go (TUG) test, where a subject is timed performing the tasks of rising up from a seated position and walking forward. Inability in standing up from a sitting position can be debilitating or even dangerous for those who are left home alone and unable to feed themselves or use the lavatory. Such patients frequently suffer from gait freezing and akinesia and are often prone to fall. In a conventional clinical setting, where a health practitioner gives a start signal to a patient, the patient is often able to anticipate the start signal and perform artificially well on this test. With the present invention, the system may be programmed to provide a random start signal (such as telephone ringing) to the patient via the PDA, thereby minimizing the effects of guessing and voice cueing. The sound signal can also be recorded together with the accelerometer data, allowing reaction time, up time or initiation time to be measured separately and accurately. A linear accelerometer placed on one leg, or a thigh tilt sensor, or both, may provide information about reaction time of the patient's movements and provide numerical data for each of reaction time, up time, and initiation time.

Freezing is a relatively common experience amongst patients with Parkinson's disease. It occurs when a patient stops suddenly while walking, or when beginning to initiate walking, and is unable to move forward again for seconds, or even minutes. Freezing may be triggered by crowded or unfamiliar surroundings, when approaching narrower spaces such as aisles, transitions such as doorways, elevators, or changes in surface appearance or texture. In some cases, freezing may be initiated by a simple line drawn across a floor. Freezing may be exacerbated by tension or nervousness.

Within the present invention, freezing may form part of a multipart test because it is a relatively common symptom of Parkinson's disease and in other diseases in which patients may otherwise appear without symptoms or with mild symptoms on other tests.

When measuring freezing with the system of the present invention, it is preferred to use 3 sensors comprising 2 angular gyros and 1 linear accelerometer. The system may then measure peak accelerations, peak velocities, step interval timing, step frequency, swing foot duration, and stride length.

Another part of a multipart test may comprise dual task measurement. The dual tasks are based on two simultaneous tasks which may bring out a cognitive deficit in attention, which is often part of illness for PD patients and others.

In one embodiment, a patient may be presented with two simultaneous tasks. One task may be a cognitive task and the other a physical task. Preferably, each task will share certain common behavioral component or same neural network resource, which we have found is more sensitive in revealing a disability.

In one example, the first task may be to tap a foot rhythmically on the floor (frequency of tapping is at patient's choice) while at the same time a mental task is performed. In one embodiment, the mental task may be a simple cognitive test such as repeating back sequences of letters or numbers provided verbally to the patient, arithmetical tasks, or a standardized test such as the "Brief Test of Attention" (1997, Psychological Assessment Resources Inc.). A standardized test may have the advantage of a body of literature considering the test, as well as a large database of performance characteristics between age-matched subjects. The system may then measure consistency of tapping frequency and provide a plot of frequency in a power spectrum.

In an alternative dual task embodiment, the patient is asked to walk in a straight line (corridor) while listening for a signal such as a beep, tone or ring from the PDA, where the sounds have random onset times with intervals which may be about 2-5 seconds. In a preferred embodiment, the signal may be a common sound with practical relevance, such as a telephone ring. When the signal starts, the patient is to press a switch as fast as possible to make the ring stop. If no button is pressed, the ring will stop after 5 seconds. The reaction

time to press the switch is recorded along with data from the sensors monitoring the patient's gait and stride. This task should prove difficult to perform for patients with illness or disability, and Parkinson's patients in particular. The system may measure step frequency, stride length, peak linear foot velocity, in relation to the time and speed and the sound of a signal being interrupted.

In one embodiment, the LWT related to the suspected disability is used to measure impairment, which may then be used to determine an impairmentidisability ratio (K ⁾ R). For example, an ambulation function may be used with a patient who has, or is suspected of having Parkinson's disease. An IDR is operationally defined as a numerical value describing the kinematics of an act divided by a clinical disability score. The disability score may be derived from standard and known clinical tests, such as, in the case of Parkinson's disease, the Unified Parkinson's Disease Rating Scale (UPDRS) and will typically include a large subjective factor. In one embodiment, the disability score may be transformed into an ordinal score from 1 to 10.

The impairment value may be obtained from a LWT, such as any of those tasks described above. Such a LWT will yield at least one numerical value, which may be total time to complete the task or some other parameter measured by the test. Then a specific IDR may be calculated, such as a ratio obtained by dividing the impairment value by a disability score. Obviously, an IDR can be obtained from any test resulting in kinematic data.

Since the data sets derived from an ambulation test and disability ratings are known to be negatively correlated and disperse, the actual data plot from a patient population will be distributed in clusters, instead of along a continuum. So a slope factor cannot be reliably derived between impairment and disability.

However, an IDR can be used to mathematically indicate the relationship between impairment and a disability, without involving slope calculation. An IDR is useful because a ceiling effect may be established, and because it permits detection of erroneous subjective disability scores. The ceiling may be derived from impairment measurements from a population of very disabled persons, which will generate an IDR ratio that can be used as an upper limit. Erroneous subjective disability scores will be evidenced by abnormally high or low IDRs. In one embodiment, a patient with an abnormally high IDR value (for example, by

two standard deviations) over the ceiling likely has an incorrect subjective disability score. Conversely, an abnormal low IDR could be the result of an individual exaggerating a disability.

A range of "normal" IDRs may be obtained from a control group known not to suffer from the impairment being assessed, or the suspected disability.

In another example, an objective diagnostic indicator may include the measurement of data from objects being acted upon by the patient, which is termed "mirror sensing" herein. Mirror sensing allows results obtained from different patients and normal subjects to be evaluated based on the same object measurement. It can reduce the standard deviations of the database and enhance the across-population comparability. One linear accelerometer will be placed on each object (pot and drawer). The speed, amplitude and duration of acceleration will be recorded. Cross-correlation techniques such as wavelet analysis may be used to generate various coherence plots for signals acquired from the subject's hand and trunk versus that obtained from the objects. Such plots will provide an objective description of both the intended movement trajectory and the functional outcome of the assessed movement capacity. Since the way that a person picks up a pot or opens a cabinet drawer varies from person to person, the measurement of functional loss based on a common object but not individual motor "habit" may be advantageous.

Attention deprivation or distraction tests may provide a diagnostic value which may assist with false claim detection. Malingering and hysterical claims of disability are common and difficult to deal with clinically. Recent neurophysiological research in animals and humans indicate that intended and unintended behaviors can be reliably differentiated by an attention deprivation test, presumably because intention is highly dependent upon internal and external cues and behavioral context whereas unintended behaviors are not. In one embodiment, false claims of disability may be detected by using the principle that temporary deprivation of attention via a distraction, either totally or partially, will lead to a degradation in intended motor behaviors. Therefore, an "intended" poor performance of any physical test from which kinematic data is derived will exhibit a kinematic or force change in a direction opposite to an unintended performance. For example, an actor trying to maintain the limb in an abnormal position will reinforce the position, even if the actor he/she is required at the

same time to repeat a word presented to him/her alternatively from each ear (dichotic hearing). In contrast, normal individuals or patients with PD will show an opposite change as they may "voluntarily" give priority to the distraction, and shift their attention to the dichotic hearing task. Dichotic hearing is a well known attention shift task, although other attention diversion methods can be used as well.

Examples:

The following examples are included to exemplify the claimed invention and should not be considered limiting of the claimed invention.

In each example involving subject testing, the subject was fitted with a ADXLl 03™ accelerometer (Analog Devices, Inc.) and/or an ADXRS300™ yaw rate gyro (Analog Devices, Inc.) connected to a Dell Axim™ PDA running Microsoft Windows Mobile 5.0™ and National Instruments Lab View 8.2™ PDA module. The data acquisition module was a National Instruments Ni-CF6004™ Data Acquisition compact flash card, having 4 analog inputs (0-5V), a 14 bit A/D converter, with a sampling rate of 200Hz. The PDA supported Bluetooth communication with the host computer, and with a Bluetooth headset including a microphone and an earpiece.

1. Timed Up and Go (TUG)

Both a normal subject and a PD patient subject were seated in a chair and fitted with the system including a headset which played random sounds to the subjects. The subjects were instructed upon hearing a particular sound to get up out of the chair and walk forward. A single accelerometer on the left foot and an accelerometer on right thigh provided information about the subject's movements. As shown in Figure 3, which shows right thigh tilt angle, the interval from the signal to the first peak in the plots, which indicates initial leg movement, represents the reaction time of the subject. The next interval to the next significant peak indicates uptime, or the time it takes to reach a standing position, followed by initiation time, which is the time before the walking motion commences.

As shown in Figure 3, a patient with PD shows significantly longer up time and total time. The increase in total time is largely due to the increase in up time, as both reaction time and initiation time show relatively smaller increases.

In Figure 7, the average TUG total time for a normal subject averaged 7.4 seconds over 4 trials, while the total time average for a PD patient averaged 32.5 seconds over 4 trials. In each trial, the subject stood up from a chair, walked 3 meters, turned around, walked back to the chair and sat down again.

In Figure 8, repeated trials were performed to create a "standing up index" using the right thigh tilt data. The difference in "up time" between normal and PD may again be seen.

More specific data processing was performed on the data recorded from the PD patient. In Figure 9, left foot acceleration and right thigh tilt were recorded, while left foot velocity and position were derived from acceleration.

2. Freezing

A PD patient with known susceptibility to freezing when presented with a line on the floor formed with tape was fitted with 3 sensors comprising 2 angular gyros and 1 linear accelerometer. The patient was asked to walk towards and over the taped line. A person with no known disability was asked to perform the same task. Figure 4 shows foot acceleration, foot velocity and foot position over time for the normal subject. As may be seen, no evidence of freezing occurred as the plots show regular strides.

Figure 5 shows the same plots for the PD patient. In the time interval between about 11 and 19 seconds, the patient's foot position did not change significantly, and demonstrated reduced velocity, although peak acceleration of the foot remained relatively constant.

3. Dual Task

A normal subject and a PD patient were fitted with a single accelerometer on a foot and asked to tap their foot on the floor with a steady beat at a frequency of the patient's choice. At the same time, through the headset, a command set was played to the patient which required the patient to repeat short sequences of letters or numbers, which is a short memory test.

The patient's ability to maintain the same frequency and power amplitude of the tapping motion were recorded. As may be seen in Figure 6, the normal subject was able to maintain a more constant frequency with more forceful tapping that the PD patient.

4. Mirror Sensing

Figure 10 shows an example of mirror sensing of subject-object movements where sensors were placed on the back of the right hand and a normal coffee cup. Concurrent movements of the hand and cup were measured while a subject (unaware of the experiment) mimicked hand tremors for a period of two minutes. The top plot shows the results form small tremors, while the bottom plot shows the result for large movements of the object. The amplitude of cup "shake" is not always correlated with the degree of hand tremors. Both small and large movements of the object can occur without obvious changes in hand tremor power spectrum (shown in the insets). This mismatch may not be the case for a patient who has involuntary tremors.

5. False Claim Detection

Figure 11 shows an example of effect of attention distraction on simulated or intended PD- like hand tremor by an actor. The top trace shows the marked attenuation of tremor amplitude produced by distraction. The bottom traces shows a comparison of the power spectra of tremor before and after distraction using dichotic hearing, emphasizing the magnitude of suppression produced by distraction in a 'faking' subject. Since the unconscious or involuntary tremors under pathological conditions should not attenuate in this way, this simple test may aid discrimination between real and 'faked' disease.

6. Group data analysis

All the data may be processed based on comparison of group variances (ANOVA) with ad hoc post-test for detecting differences between conditions and individual parameters. A common caveat is that large variations in individual performances may give rise to highly dispersed data sets that frequently overcome the statistical power of small data samples. Previous studies on linear gait control noted that the coefficients of variance and regressional

relationship among individual gait parameters sometimes better describe the overall pattern of gait abnormality.

References

The following references may be relevant to the state of the art, and the contents of each are incorporated herein as if reproduced herein in their entirety.

[I] Azulay, J.P., Mesure, S., Amblard, B., BHn, O., Sangla, I. and Pouget, J., Visual control of locomotion in Parkinson's disease, Brain, 122 ( Pt 1) (1999) 111-20.

[3] Blin, O., Ferrandez, A.M., Pailhous, J. and Serratrice, G., Dopa-sensitive and dopa- resistant gait parameters in Parkinson's disease, J Neurol Sci, 103 (1991) 51-4.

[4] Blin, O., Ferrandez, A.M. and Serratrice, G., Quantitative analysis of gait in Parkinson patients: increased variability of stride length, J Neurol Sci, 98 (1990) 91-7.

[5] Brown, P. and Steiger, M., Basal ganglia gait disorders. In T. Brandt, A.M. Bronstein and M.H. Woollacott (Eds.), Clinical disorders of balance, posture and gait, Arnold, London,

1996, pp. 156-67.

[6] Carollo, J.J. and Matthews, D., Strategies for clinical motion analysis based on functional decomposition of the gait cycle, Phys Med Rehabil CUn NAm, 13 (2002) 949-77.

[7] DeLong, M.R., Possible involvement of central pacemakers in clinical disorders of movement, Fed Proc, 37 (1978) 2171-5.

[8] Earhart G, Horak, FB, Melvill Jones G, Block EW, Weber KD, Suchowersky O,

Fletcher WA., Is the Cerebellum Important for Podokinetic Adaptation?, Ann NY Acad Sci,

978 (2003) 978:511-512.

[9] Earhart Gammon M, Melvill Jones G, Horak FB, Block EW, Weber KD, Fletcher

WA. Transfer of podokinetic Adaptation from stepping to hopping. J. Neurophysiol.

87:1142-1144 (2002).

[10] Earhart, G.M., Fletcher, W. A., Horak, F.B., Block, E.W., Weber, K.D., Suchowersky,

O. and Melvill Jones, G., Does the cerebellum play a role in podokinetic adaptation?, Exp

Brain Res, 146 (2002) 538-42.

[I I] Faist, M., Xie, J., Kurz, D., Berger, W., Maurer, C, Pollak, P. and Lucking, C.H., Effect of bilateral subthalamic nucleus stimulation on gait in Parkinson's disease, Brain, 124 (2001) 1590-1600.

[12] Gordon, C.R., Fletcher, W. A., Melvill Jones, G. and Block, E. W., Adaptive plasticity in the control of locomotor trajectory, Exp Brain Res, 102 (1995) 540-5.

[13] Hanakawa, T., Katsumi, Y., Fukuyama, H., Honda, M., Hayashi, T., Kimura, J. and

Shibasaki, H., Mechanisms underlying gait disturbance in Parkinson's disease: a single photon emission computed tomography study, Brain, 122 ( Pt 7) (1999) 1271-82.

[14] Melvill Jones, G.M., Galiana, H.L., Weber, K.D., Fletcher, W.A. and Block, E.W.,

Complex podokinetic (PK) response to post-rotational vestibular stimulation, Arch Ital Biol,

138 (2000) 99-105.

[15] Krystkowiak, P., Blatt, J.-L., Bourriez, J.-L., Duhamel, A., Perina, M., Blond, S.,

Guieu, J.-D., Destee, A. and Defebvre, L., Effects of Subthalamic Nucleus Stimulation and

Levodopa Treatment on Gait Abnormalities in Parkinson Disease, Arch Neurol, 60 (2003) 80-

84.

[16] Kumar, R., Lozano, A.M., Kim, YJ., Hutchison, W.D., Sime, E., Halket, E. and Lang, A.E., Double-blind evaluation of subthalamic nucleus deep brain stimulation in advanced Parkinson's disease, Neurology, 51 (1998) 850-5.

[17] Lang, A.E. and Obeso, J.A., Challenges in Parkinson's disease: restoration of the nigrostriatal dopamine system is not enough, Lancet Neurol, 3 (2004) 309-16.

[20] Melvill Jones G, Fletcher WA, Weber KD, Block EW., Vestibular-Podokinetic Interaction Without Vestibular Perception, Exp Brain Res, (submitted) (2004).

[22] Mesure, S., Darmon, A. and Blin, O., Imbalance of attentional and sensory inputs on gait, Adv Neurol, 87 (2001) 243-50.

[23] Morris, M., Iansek, R., Matyas, T. and Summers, J., Abnormalities in the stride length-cadence relation in parkinsonian gait, Mov Disord, 13 (1998) 61-9. [24] Morris, M.E., Iansek, R., Matyas, T.A. and Summers, J.J., The pathogenesis of gait hypokinesia in Parkinson's disease, Brain, 117 ( Pt 5) (1994) 1169-81. [25] Morris, M.E., Iansek, R., Matyas, T.A. and Summers, JJ., Stride length regulation in Parkinson's disease. Normalization strategies and underlying mechanisms, Brain, 119 ( Pt 2) (1996) 551-68.

[26] Pearson, K.G., Generating the walking gait: role of sensory feedback, Prog Brain Res, 43 (2004) 123-9.

[27] Rubinstein, T.C., Giladi, N. and Hausdorff, J.M., The power of cueing to circumvent dopamine deficits: a review of physical therapy treatment of gait disturbances in Parkinson's disease, Mov Disord, 17 (2002) 1148-60.

[28] Sahyoun, C, Floyer-Lea, A., Johansen-Berg, H. and Matthews, P.M., Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements, Neuroimage, 21 (2004) 568-75.

[29] Shibasaki, H., Fukuyama, H. and Hanakawa, T., Neural control mechanisms for normal versus parkinsonian gait, Prog Brain Res, 143 (2004) 199-205. [30] Thaut, M.H., Mclntosh, G.C., Rice, R.R., Miller, R.A., Rathbun, J. and Brault, J.M., Rhythmic auditory stimulation in gait training for Parkinson's disease patients, Mov Disord, 11 (1996) 193-200. [31] Vaughan, C.L., Theories of bipedal walking: an odyssey, J Biomech, 36 (2003) 513-23

[33] Weber, K.D., Fletcher, W.A., Gordon, C.R., Melvill Jones, G. and Block, E. W., Motor learning in the "podokinetic" system and its role in spatial orientation during locomotion, Exp Brain Res, 120 (1998) 377-85.

[34] Weber, K.D., Fletcher, W.A., Melvill Jones, G. and Block, E.W., Podokinetic after- rotation in patients with compensated unilateral vestibular ablation, Exp Brain Res, 147 (2002) 554-7.