Reducing errors in screening-test administration

Abstract

A system for detecting a screening-test error includes a measurement device that measures at least one performance parameter related to at least one screening-test task and a computational device, in communication with the measurement device. The computational device receives the at least one measured performance parameter, calculates at least one performance statistical quantity characterizing the measured performance parameter, and compares the at least one performance statistical quantity to at least one reference statistical quantity associated with an error-free screening test. In accordance with a related embodiment, the system may further include a display device that displays the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity.

Description

RELATED U.S. APPLICATION(S)

[0001] The present application claims priority from U.S. provisional application No. 060/040,435 filed on Mar. 12, 1997, and is a continuation-in-part of U.S. utility application Ser. No. 09/785,673, filed Feb. 16, 2001 which, in turn, is a continuation application of U.S. utility patent application Ser. No. 09/354,488 filed on Jul. 16, 1999, now U.S. Pat. No. 6,190,287, which itself is a divisional application of U.S. utility patent application Ser. No. 09/041,877 filed on Mar. 12, 1998, now U.S. Pat. No. 5,980,429, issued on Nov. 9, 1999, all of which are hereby incorporated herein by reference.

FIELD OF THE INVENTION

[0002] The invention relates to monitoring training programs and the administration of screening tests to improve their cost-effectiveness and maintain quality control. In particular, the invention relates to apparatuses and methods for detecting errors in the administration of screening-tests.

BACKGROUND ART

[0003] A major factor influencing the cost of skill training is the required level of supervision by a skilled practitioner in both detecting a physical condition which may influence the design of a training program and monitoring subject's performance of the training program. This is particularly crucial in today's health care environment in which more efficient use of medical resources is demanded. Allocating a portion of a subject's training to lower paid, less skilled assistants or having a subject train at home without professional supervision significantly reduces the labor costs. These reduced cost training programs increase the risk of inferior training outcomes because the insufficiently supervised trainee is more likely to perform tasks incorrectly, continue doing tasks which are either too simple or too difficult, or perform at a low level of intensity and motivation.

[0004] A second major factor particularly influencing the cost effectiveness of physical rehabilitation and athletic skill training is the motivation of the trainee to acquire the new skills required to enhance performance. A minority of individuals impaired either by injury or disease may have secondary reasons for slowing their recovery. For example, a worker receiving disability payments for a job related injury may prefer to remain on disability rather than have the injury rehabilitated only to return to an unsatisfying job. Similarly, an individual injured in a motor vehicle accident might be motivated to remain impaired in the hopes of winning a financial settlement. Such medical cases frequently occur in conjunction with legal actions in which a judge or jury are forced to make determinations of disability. Objective and accurate information related to a trainee's physical condition and motivation to recover would be valuable in reaching these medical-legal judgments.

[0005] A third major factor influencing the cost effectiveness of physical rehabilitation, but not limited to this type of program, is the quality of management. This is especially important in the realm of larger multi-facility corporate providers of training, operating with numerous practitioners. Traditionally, physical rehabilitation has been a clinical “art” in which the individual practitioner has a substantial degree of independence in determining the specific tasks used in treatment. Issues which must be addressed by quality management include: (1) Are individual practitioners adhering to the corporate standard methods used to treat specific physical disorders? (2) Are individual practitioners and subjects working at the appropriate levels of motivation? (3) Are subjects being adequately challenged during training to produce the best outcome in the shortest possible time? Objective information relative to the motivation, quantity, and quality of exercise tasks performed by subjects would be valuable for the individuals responsible for managing the clinical program.

[0006] A challenge confronted by health care providers is to the ability to extend services to a broader population of individuals while minimizing the escalation of costs. To achieve this goal, providers must deliver their services in as cost-effective manner as possible and must concentrate services where they are likely to effect the greatest benefit. To deliver services efficiently, health care providers must select the most effective evaluation and treatment approaches and predict the likely outcome of the selected approaches relative to the associated costs. For example, the demand for efficient physical rehabilitation services has become especially strong, because demand continues to increase rapidly in proportion to the growing numbers of elderly citizens. As will be discussed further below, elderly citizens often experience chronic medical problems due to fall related injuries.

[0007] A form of service now being implemented by providers is one which is home-based wherein a subject performs a substantial portion of prescribed training exercises at home rather than in a professionally supervised clinical environment. Rehabilitation training performed at clinics tends to be supervised by less skilled clinical aids and assistants rather than by professionally trained physicians and physical therapists. Both home-based and reduced cost clinically based approaches substantially reduce the personnel costs associated with rehabilitation training compared to the cost of services provided by professionally trained personnel. More responsibility is placed on subject motivation to perform the prescribed training exercises. When training exercises are performed in the absence of a professionally trained clinician, there is a substantially greater risk that the subject will be poorly motivated to perform at an optimal level of intensity, will continue to perform simpler, less challenging exercises when they are no longer needed, and will perform exercises incorrectly. All of these factors substantially reduce the efficacy of the rehabilitation training.

[0008] Selection of training exercises likely to yield the best outcome is typically based on information related to the trainee's medical diagnosis, results of functional performance assessments, and the trainee's goals for improvement. The functional performance assessment may involve screening tests that may employ either one or a combination of objective machine based and subjective observationally based methods. The current clinical literature suggests that information related to diagnosis and functional performance contribute useful information to the processes of selecting the most effective training tasks and predicting the progress and outcome of performing said tasks.

[0009] Screening-tests used to detect early stage chronic medical conditions before they become disabling include blood pressure meters, which detect signs of early stage cardiovascular disease, and eye pressure tests, which detect signs of early stage glaucoma. To be cost-effective, screening-tests must be sensitive to physiological changes. Screening-tests should also be accurate, accessible, easily administered to a large number of individuals, and relatively inexpensive to manufacture. When these criteria are met, the cost of screening a large number of individuals may be offset by the savings achieved through reduction in the incidence and/or severity of disabling chronic diseases.

[0010] As noted above, chronic medical problems occurring primarily (but not exclusively) in the elderly population involve falling and fall-related injuries. Such fall-related injuries often lead to the restriction of activities involved in daily living and loss of independent mobility and may require that the subject undergo rehabilitative training. Recent epidemiological studies have found that restriction of activities, and injuries related to falling, are major causes of impairment to functional independence in the over sixty-five year old population. Because people are living longer today, the prevalence of these chronic problems is projected to increase substantially in years to come.

[0011] Further, recent research studies have demonstrated that the risk of fall related restriction of activities and fall related injuries can be identified before disability occurs. If the risk is identified early and treated, the incidence of injury and loss of functional independence can be reduced. According to other recent studies, balance is one of the most important factors influencing fall risk. An American Medical Association review of an article titled “Preventable Medical Injuries in Older Patients” published in the Archives of Physical Medicine provides a succinct summary of this problem. In the current state of the art, both subjective observational and objective technology-based means are available for quantifying an individual's balance function. One well-known example of an observational test with documented ability to detect fall risk is the Berg Balance test. The Berg test requires a clinically trained individual to observe and numerically rate an individual's ability to perform a series of standardized balance and movement tasks. This test has the advantage of requiring no specialized equipment. However, the test requires considerable time to administer, and the results are dependent on the observational skills and experience of a clinically trained administrator.

[0012] In contrast, effective screening-tests for hearing, vision, and blood pressure abnormalities are currently offered to the general public in uncontrolled, non-medical environments such as drug stores and shopping malls. These types of screening-tests are administered by individuals without specialized medical training. However, compared to medical tests administered by highly trained individuals in controlled medical environments, the potential for errors in administration and interpretation of the tests is substantially higher. These screening errors can result in needless worry on the part of the subject if one or more of the errors places a normal subject in an at-risk category. Additionally, if an error in the administration of a test occurs, and the administer fails to detect that the subject is at risk, an opportunity may be lost to help the at-risk individual.

[0013] In screening-tests administered in non-medical environments, data transmission applications are often employed. Many of these data transmission applications require that high volumes of information be transferred from one site to another as rapidly as possible with a minimum of errors. Two devices used to transfer high volumes of information rapidly include the high-speed modem, for transmission of digital data over a phone line; and transmitters on space vehicles (such as satellites), for radio frequency transmission of data over great distances of space. To maximize the efficiency of data transmission, these and many other applications use data compression means. Transmission of high volumes of compressed data, however, increases the risk of errors.

SUMMARY OF THE INVENTION

[0014] In accordance with one embodiment of the invention a system for detecting a screening-test error includes a measurement device that measures at least one performance parameter related to at least one screening-test task and a computational device, in communication with the measurement device. The computational device receives the at least one measured performance parameter, calculates at least one performance statistical quantity characterizing the measured performance parameter, and compares the at least one performance statistical quantity to at least one reference statistical quantity associated with an error-free screening test. In accordance with a related embodiment, the system may further include a display device that displays the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity.

[0015] In accordance with another embodiment of the invention, a system for detecting errors in balance related screening tests includes a force-plate for measuring a quantity related to a stability factor of a balance task performed in trials by a subject under a plurality of distinct sensory conditions and a computation device in communication with the force-plate. The computational device (i) receives the quantity related to the stability factor for each trial, (ii) determines a rank order for the quantities, each quantity for each trial being associated with a rank, and (iii) determines if any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial performed under error-free conditions. In accordance with a related embodiment, the system may also include a display device in communication with the computational device for indicating an instance wherein any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial.

[0016] In accordance with a further embodiment of the invention, a method for detecting a screening-test error includes measuring at least one performance parameter related to at least one screening-test task, calculating at least one performance statistical quantity characterizing the measured performance parameter and comparing the at least one performance statistical quantity to at least one reference statistical quantity associated with an error-free screening test. In accordance with related embodiments, the statistical quantity may represent a value associated with an average or the statistical quantity may represent a value associated with a standard deviation. Additionally, the statistical quantity may represent a value associated with a standard error or the statistical quantity may represent a value associated with a power spectrum. The statistical quantity may further represent a value associated with a root mean square or a value associated with a frequency histogram. The method may also include displaying the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity on a display device.

[0017] In accordance with another related embodiment, (i) the screening-test task may be a balance task, (ii) the at least one performance parameter may be sway deviation, (iii) the at least one performance statistical quantity may correspond to a moving window root mean square value for velocity of the sway deviation, and (iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity may include determining whether the moving window root mean square value deviates from a constant value by a predetermined threshold value. In accordance with a further related embodiment, (i) the screening-test task may be a balance task, (ii) the at least one performance parameter may be vertical force applied to a force plate, (iii) the at least one performance statistical quantity may correspond to a moving window average value for total vertical force applied to the force plate, and (iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity may include determining whether the moving window average value deviates from a constant value by a predetermined threshold value.

[0018] In accordance with an additional related embodiment, (i) the screening-test task may be a balance task, (ii) the at least one performance parameter may be vertical force applied to a force plate, (iii) the at least one performance statistical quantity may correspond to an average of a mathematical derivative of the total vertical force applied to the force plate and (iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity may include determining whether the average deviates from zero by a predetermined threshold value. Similarly, in a further related embodiment, (i) the screening-test task may be a balance task, (ii) the at least one performance parameter may be horizontal force applied to a force plate, (iii) the at least one performance statistical quantity may correspond to an average of a mathematical derivative of the total horizontal force applied to the force plate and (iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity may include determining whether the average deviates from zero by a predetermined threshold value.

[0019] In accordance with another embodiment of the invention, a method for detecting errors in balance related screening tests includes measuring a quantity related to a stability factor of a balance task performed in trials by a subject under a plurality of distinct sensory conditions and obtaining thereby the quantity related to the stability factor for each trial. A rank order for the quantities is determined, each quantity for each trial being associated with a rank, and whether any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial performed under error-free conditions is also determined. In accordance with a related embodiment, the method may also include displaying a number corresponding to the number of times a performance of the balance task by the subject has fallen outside the reference range.

[0020] In accordance with other related embodiments, measuring the quantity related to a stability factor may include following a modified CTSIB protocol and/or determining a rank order for the performance of the plurality of distinct tasks may include determining a rank order according to the level of difficulty of the balance tasks.

[0021] In accordance with yet another embodiment of the invention, a method for detecting a screening test error in an individual trial of a balance task during which sway deviation is measured includes determining a quantity corresponding to a moving window root mean square value for velocity of the sway deviation, the window being short in relation to the duration of the trial but long in relation to the duration of a typical deviation in sway velocity and entering an alarm state when the quantity exceeds a threshold value.

[0022] In accordance with a further embodiment of the invention a method for detecting a screening test error due to malfunctions of at least one vertical force sensing device includes determining a quantity corresponding to a moving window average value for the total vertical force measured by the device, the window being long in relation to the duration of expected spontaneous fluctuations in the total vertical force and entering an alarm state when the quantity deviates from a constant valued by a predetermined threshold value.

[0023] In accordance with another embodiment of the invention, a method for detecting a screening test error due to malfunctions of at least one vertical force sensing device includes calculating an average of a mathematical derivative for the total vertical force measured by the device to determine the rate of change of the total vertical force and determining a quantity corresponding to an average rate of change of the total vertical force over a predetermined period of time. An alarm state is entered when the average deviates from zero by a predetermined threshold value.

[0024] In accordance with a further embodiment of the invention, a method for detecting a screening test error due to malfunctions of at least one horizontal force sensing device includes calculating an average of a mathematical derivative for the total horizontal force measured by the device to determine the rate of change of the total horizontal force and determining a quantity corresponding to an average rate of change of the total horizontal force over a predetermined period of time. An alarm state is entered when the average deviates from zero by a predetermined threshold value.

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] FIG. 1 is a block diagram illustrating a system for detecting errors in the administration and evaluation of screening-tests in accordance with an embodiment of the present invention;

[0026] FIG. 2 is an illustration showing a system that may be used to detect errors in the administration and evaluation of screening-tests directed to balance and coordination tasks in accordance with one embodiment of the invention;

[0027] FIG. 3 is an illustration showing another system that may be used for prescribing an exercise program, defining expected compliance with the program and detecting errors in the administration and evaluation of screening-tests directed to balance and coordination tasks in accordance with one embodiment of the invention;

[0028] FIG. 4 is a flow chart illustrating a method for detecting a screening-test error in accordance with an embodiment of the invention;

[0029] FIG. 5 is a flow chart illustrating a method for detecting errors in balance related screening tests in accordance with another embodiment of the invention;

[0030] FIG. 6 is a flow chart illustrating a method for detecting a screening-test error in an individual trial of a balance task in accordance with another embodiment of the invention;

[0031] FIG. 7 is a flow chart illustrating a method for detecting a screening-test error due to malfunctions of a force sensing device accordance with further embodiment of the invention;

[0032] FIG. 8 is a flow chart illustrating a method for detecting a screening-test error due to malfunctions of a force sensing device accordance with another embodiment of the invention;

[0033] FIG. 9 is a flow chart describing how a prescriber may create and revise an individualized training program; and

[0034] FIG. 10 schematically illustrates an example of a monitoring system in accordance with an embodiment of the invention.

DESCRIPTION OF SPECIFIC EMBODIMENTS

[0035] In accordance with the invention, a computer-based screening device is provided that employs statistical analysis techniques designed to detect potential errors in test administration. The error information can be displayed to a health care practitioner, an operator or a user of the computer upon completion of the test administration, at which time the operator or user can chose whether or not to repeat or to ignore any erroneous test results. Alternatively, the error information can be displayed in a screening-test report to be transmitted by a transmission means to another individual responsible for interpreting the test results. Further, such information may assist a subject or health care practitioner in assessing a physical condition and, consequently, creating a rehabilitation program.

[0036] A screening test should provide data that accurately reflects a subject's capabilities. Thus, errors associated with screening test administration should be minimized or eliminated. There are a number of potential sources of error that can adversely affect the administration of a screening-test and proper interpretation of the screening-test results. If the subject is temporarily distracted or fails to follow instructions during administration of all or a portion of the test, an error may occur. If the subject is improperly instructed and/or misunderstands the test instructions, or initially misunderstands but progressively figures out the instructions during the course of the test administration, an error may occur. Additionally, if the subject is startled or frightened by some aspect of the task and reacts by changing his or her performance, an error may occur. Any of the above three administration problems are common, can alter the subject's ability to perform tasks associated with a screening-test, and thereby produce errors in the screening-test results.

[0037] In the case of screening-tests designed to measure balance control and/or coordination, there are recognized scientific sources for understanding processes of balance control in normal human subjects, in individuals with pathology affecting balance, and in individuals exaggerating symptoms of balance disorder. Among accepted principles of balance control are specific principles from which statistical criteria can be formulated to detect errors in measures of balance function. The following are two easily described examples of balance control principles and their application to detecting errors in test administration.

[0038] 1) The act of balancing is known to be easier when standing with eyes open on a firm surface compared to standing with the eyes closed and/or standing on an irregular or compliant surface. Therefore, individuals should, on a statistical basis, display higher spontaneous sway activity under more difficult task conditions (e.g., standing with eyes closed or standing on an irregular or compliant surface) compared to easier task conditions (e.g., standing with eyes open on a firm surface).

[0039] 2) Standing balance is known to be a relatively continuous control process over an interval of time in which the task conditions are unchanged, so long as the interval is not prolonged to the point of subject fatigue. As a consequence, the statistical properties of spontaneous sway activity will be relatively constant so long as task conditions are constant. Alternatively, the statistical properties of sway activity will change in response to changes in task conditions.

[0040] 3) When the attention of an individual performing a balance task is distracted, the balance performance will temporarily deteriorate or otherwise change. Thus, changes in attention during a test performed under unchanging task conditions will change the statistical properties of the performance measurements.

[0041] In addition to problems in test administration, errors in results can be caused by equipment malfunctions or by improper use of the equipment. The following are some easily understood examples of equipment malfunctions and improper use that can reduce the integrity of the test results.

[0042] 1) As an individual performs a balance test standing erect on a force-plate device with a plurality of force measuring sensors and not grasping for external support, the average sum total of the vertical component of forces measured by the sensors will on the average remain constant and equal to the subject's weight, while the average sum total of the forward-backward and lateral horizontal components of forces measured by the sensors will on the average remain constant and equal to zero. Spontaneous fluctuations in vertical and horizontal forces in one direction can occur if the subject moves the body up or down, bends at the hips, or moves the arms. However, these fluctuations last less than a second or two and are always followed by equal and opposite forces. As the subject moves, the percentage of the total (constant) vertical and horizontal forces measured by each sensor changes. If one of the force measuring sensors fails to accurately measure a vertical or horizontal force component, then the constant vertical and horizontal totals measured by the plurality of sensors will appear to fluctuate over periods of time in excess of any spontaneous fluctuations as the percentages of vertical force change with the subject's movements on the force-plate.

[0043] 2) If the subject is instructed to perform a task on a force-plate surface without grasping for external support, then the sum of the vertical and horizontal plane forces measured by the plurality of sensors will remain constant on the average over time. If the subject steps off the force-plate, grasps a nearby object for additional support, or in any other way receives additional external support, then the total force components (vertical and/or parallel to the force-plate surface) will show systematic changes.

[0044] A number of methods are described in the current art for quantifying performance of a balance task. Balance products manufactured by NeuroCom International, Inc., that measure an individual's performance of balance tasks, use force-plates on which a subject stands to record changes in, for example, the position of the center of force exerted by the feet against the support surface. (Such devices are disclosed in U.S. Pat. No. 5,476,103, U.S. Pat. No. 5,551,445 and U.S. Pat. No. 6,010,465 each of which is incorporated herein, in its entirety, by reference.) An individual's ability to balance over time is quantified by instructing the individual to perform a balance task on the force-plate and then calculating various temporal and statistical quantities related to the motions of the center of force over the duration of a the task. The balance performance of an individual may be quantified in a variety of different ways. Commonly used quantities include the amplitudes of motions, the average velocities of motions, the frequencies of motions, and the standard deviation of motions. Such methods are described in U.S. Pat, Nos. 5,980,429; 5,269,318; 5,052,406; and 4,738,269. Each of the patents herein mentioned is incorporated herein, in its entirety, by reference.

[0045] Another common means for recording signals related to an individual's performance of a balance task is the placement of linear and/or angular motion sensors on one or more parts of the body to measure the motions of that body part. For example, NeuroCom International, Inc. also manufactures a product that uses an inertial device to measure the angular motions of an individual's head as the individual performs a balance task The use of inertial, gravitational, magnetic, optical, and simple spring-based sensors to detect motion of the body have also been described in the prior art.

[0046] Three examples of NeuroCom balance assessment products include the EquiTest™, the Balance Master™, and the VSR™. As above, these three systems exploit force-plate measuring devices, graphic computer-based test administration, computerized data analysis, and graphic displays of results to precisely quantify balance functions. While requiring specialized equipment, tests performed with these products have the advantage over observational test methods that they substantially shorten test times, providing sensitive, objective results which are relatively independent of operator training and experience levels.

[0047] For example, the Balance Master includes a functional performance assessment means for objectively assessing a subject's ability to perform balance and mobility tasks representative of daily living, an analysis means for displaying the subject's performance on the assessment means and for comparing the subject's performance to that of an age-matched normal reference population, a training means comprised of a menu of structured exercise tasks subdivided by activity type and level of difficulty, and an additional assessment means for documenting the subject's compliance with exercises performed with the training means.

[0048] The Balance Master assessment protocols are designed to assess the performance capabilities of the subject on a variety of specific types of balance and mobility tasks. Results of performing each assessment task are is analyzed and displayed in a comprehensive report including graphical summaries of average scores, coefficients of variation, and left-right percentage difference scores where appropriate. Left/Right difference scores are presented for those measures which differentiate between performance of the left and right lower extremity. Each comprehensive report also includes the capability of comparing an individual subject's scores to those of a reference population of individuals. Typically, the reference population will consist of a group of normal individuals age-matched to the subject or a group of subjects with similar diagnoses.

[0049] A number of statistical methods are also described in the art for characterizing a recorded signal related to an individual's performance of a balance task. Statistical methods that may be applied to a recorded signal over the duration of a trial, or to one or more mathematical derivatives of the recorded signal, include but are not limited to the: (1) average, (2) standard deviation, (3) standard error, (4) root mean square, (5) power spectrum, and (6) frequency histogram. Each of the preceding statistical measures can also be calculated over one or more windows of time limited to portions of a trial. One example of this approach is calculating a moving window average.

[0050] In one embodiment of the invention, equipment malfunctions due to failure of one or more of the vertical force sensing devices are detected by the following method. A moving window average is calculated on the total of the vertical forces measured while an individual is standing on a force-plate and performing a balance task. The duration of the time window exceeds the duration of expected spontaneous fluctuations on the total vertical force. One or more of the force sensing devices is determined to have failed when the moving average deviates from a constant value by more than a specified threshold value. The threshold value is specified based on prior knowledge, examples of which include but are not limited to variations in the measurement accuracies among individual vertical force sensing devices, the accuracy changes likely to occur following the failure of a force sensing device, and the accuracy of a moving average calculation Similar methods employing moving averages can be used to detect failures in the devices measuring the horizontal forces.

[0051] In another embodiment of the invention, a mathematical derivative of the total of the vertical forces measured by the force-plate device is calculated to determine the rate of change of the total vertical force. The average of the rate of change is then calculated over the duration of the trial. One or more force sensing devices is determined to have failed during the trial when the average of the rate of change of the total vertical force deviates from a value of zero by a specified threshold value. Specification of the threshold value is based on prior knowledge, examples of which include but are not limited to variations in the measurement accuracies among the individual vertical force sensing devices, signal fluctuations due to mechanical and electrical noise, the characteristics of accuracy changes likely to occur following the failure of a force sensing device, and the accuracy of a moving average calculation To detect failures in the devices measuring horizontal forces, similar averaging operations can be performed on signals related to the rates of change of the horizontal force components.

[0052] The following is one example of a method that can be used to detect errors occurring during individual trials of a standing balance task. A signal related the subject's sway deviations during the trial is recorded and then differentiated to determine the sway velocity. A moving window root mean square quantity is then calculated for the trial. The duration of the moving time window is short compared to the duration of the trial but long compared to the duration of typical deviations in sway velocity. A performance error is determined to have occurred when the moving root mean square quantity deviates from a constant by a specified threshold value. The threshold value is determined empirically by the following procedure. The quantities described above are measured and calculated in a population of individuals known to be performing the same task without error. The threshold value is then set at a value that is greater than the deviations in the moving window root mean square observed in 95% of the error free group population. In addition to the root means square quantity described above, moving window averaging methods of performance error detection can be based on other statistical quantities such as the standard deviation, standard error, power spectrum, and frequency histogram.

[0053] FIG. 1 is a block diagram illustrating a system for detecting errors in the administration of screening-tests in accordance with an embodiment of the present invention. In this embodiment, the system 100 includes a measurement device 101, a computational device 102 in communication with the measurement device and an optional display device 103 in communication with the computational device. The measurement device 101 measures at least one performance parameter related to at least one screening-test task. As noted above, these performance parameters may include, but are not limited to, amplitudes of motions, the average velocities of motions, the frequencies of motions, and the standard deviation of motions.

[0054] The computational device 102 receives the at least one measured performance parameter from the measuring device 101 and calculates at least one performance statistical quantity characterizing the measured performance parameter. The computational device 102 also compares the at least one performance parameter to at least one reference statistical quantity associated with an error-free screening test. The computational device 102 includes a processor and a memory (not shown) and may be programmed to access one or more reference ranges that indicate an error-free performance of each task. The reference range may then be compared to the at least one calculated statistical quantity in order to ascertain whether an error has occurred in the administration of the screening-test. Alternatively, a user or operator may provide the reference ranges, either manually, such as through a keyboard associated with the computational device, or by providing a location wherein the reference range may be found, such as an address, hyperlink, or file.

[0055] The display device 103 may display to the user or operator the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity.

[0056] FIG. 2 is an illustration of a system which may be used to detect errors in the administration of screening-tests directed to balance and coordination tasks in accordance with an embodiment of the invention. This system 200 is illustrative of the VSR system mentioned above and includes a force-plate 201, an optional foam pad 202 for placement on the force-plate 201, a computer 203, an optional visual display device 204, and an optional printer 205. The components of the system 200 may be configured such that the computer 203 performs the tasks associated with the computational device 102 and the visual display device 204 displays the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity. Alternatively, or in addition, the system 200 may correspond to the measuring device 101 of FIG. 1, and it may be configured to communicate with another computational device and display device, either directly, or through a network.

[0057] The force-plate 201 measures a quantity related to a stability factor of a balance task performed by a subject in trials by a subject under a plurality of distinct sensory conditions (such a balance task may be, for example, standing or walking). The foam pad 202 may be used to reduce the accuracy of information, pertaining to balance and orientation, from a subject's somatosensory (proprioceptive, cutaneous, and joint) system. The computer 203 receives the quantity related to the stability factor for each trial and processes the information to provide measurements related to the individual's performance, provides real-time biofeedback, and feeds displays which may be used during the screening-test. When the computer 203 receives the information from the force plate 201, it may determine a rank order for the quantities, wherein each quantity is associated with a rank. The computer 203 may also determine if any of the ranks associated with the given trial has fallen outside a reference range associated with the given trial performed under error-free conditions. Alternatively, another computer, separate from the system 200 and in communication with the system, may be used to make these determinations. The computer that determines whether any of the ranks associated with the given trial has fallen outside the reference range associated with the trial performed under error-free conditions may then generate a display, user interface, or alarm state or condition that will indicate an instance wherein any of the ranks associated with a given one of the trials has fallen outside the reference range associated with the given trial either to the subject or to an operator administering the screening-test. The computer 203 may also generate task instructions to be used by the subject during the screening-test.

[0058] The visual display 204 receives information from the computer 203 and may display any instructions generated by the computer to the subject, any real-time biofeedback information, an alarm condition, and/or the relationship between the quantity related to a stability factor and the reference range upon completion of the screening-test. Alternatively, or in addition, the relation between the quantity related to a stability factor and the reference range may be displayed on a visual display device separate from the system such that an operator or administrator of the screening-test may view the display. In either case, the display device 204 may also display any information related to analysis of the test results, such as whether the results indicated an error in administration of the test and/or equipment malfunction. A printer 205 may receive information from the computer 203 and generate a hard copy report related to the screening-test.

[0059] The system 200 may be used to perform a test protocol consisting of a plurality of tests for each of a plurality of tasks. Among other protocols, the system 200 may perform the modified Clinical Test for Sensory Interaction on Balance (mCTSIB). In accordance with the mCTSIB, the subject maintains a freely standing position on a force-plate for a total of twelve, ten second trials; including three trials each for four increasingly difficult sensory conditions. The four conditions consist of:

[0060] 1) standing on the firm force-plate surface eyes open;

[0061] 2) standing on the firm force-plate surface eyes closed;

[0062] 3) standing on a compliant foam pad placed on top of the force-plate eyes open; and

[0063] 4) standing on a foam pad placed on top of the force-plate eyes closed.

[0064] For each trial, signals from the force-plate 201 are used to calculate, among other quantities, one quantity related to the subject's stability. On completion of the test protocol, the twelve stability scores are ranked, either by the computer 203 or by another computer in communication with the system, in order from the least to the greatest. The rank order for each of the twelve stability scores is then compared to a reference range, which may be accessed by the computer 203 or accessed by another computer as mentioned above. The number of instances in which the rank order of a stability score falls outside of the respective reference range for that sensory condition are summed and may be displayed to the subject via the visual display 204. Alternatively, or additionally, the number of instances in which the rank order falls outside the reference range may be displayed to an operator, health care practitioner or administrator of the screening-test, via another visual display in communication with the system 200. The greater the number of instances of rank order numbers falling outside their respective ranges, the more likely that there were errors in test administration.

[0065] In one embodiment, the rank order reference ranges are set based on the principle that stability decreases as the standing task difficulty increases. In accordance with this method, the rank order range for the three condition one trials is set at 1 through 5; for condition two, the rank order is set at 3 through 7; for condition three, the rank order is set at 6 through 10, and for condition four, the rank order is set at 8 through 12. Widening the reference ranges used in this embodiment would reduce the sensitivity to errors, while narrowing the ranges would increase sensitivity to errors. In a second embodiment, reference ranges are established empirically by testing a plurality of subjects known to be cooperative and proficient using an operator known to be highly proficient. The reference range for condition one would be set to include the rank orders of 95 percent of all condition one trials, the reference range for condition two would be set to include the rank orders of 95 percent of the condition two trials, the reference range for condition three would be set to include the rank orders of 95 percent of all condition three trials, and the reference range for condition four would be set to include the rank order of 95 percent of all condition four trials.

[0066] It should be understood that additional preferred embodiments may employ test protocols that include a plurality of trials conducted on each of a plurality of tasks. For one example, the system 200 may perform the Limits of Stability (LOS) test protocol (described in more detail below) in which a subject performs a total of eight rapid voluntary movements to targets placed at different points on a screen. For each target movement, a plurality of performance scores are calculated. The performance scores calculated may include the reaction time, the velocity of the movement, the distance of the movement, and the accuracy of the movement. By performing the complete LOS test twice, or by performing a portion of the eight targets at least two times each, the requirement of a plurality of trials for each of a plurality of tasks may be met.

[0067] FIG. 3 is an illustration showing another system that may be used in accordance with an embodiment of the invention. The system 300 of FIG. 3, which is illustrative of the Balance Master system mentioned above, includes the following test protocols: walk, quick turn, sit-to-stand, step up/over, bear-bearing squat, rhythmic weight shift, and lunge tests. (Note that each of the systems described herein may be used in accordance with a training program as will be described further below.) All of these tests meet the requirements of a protocol in which a plurality of tasks are performed a plurality of times each.

[0068] The system 300 includes a force-plate 301, optional tools 302 for placement on the force-plate, a computer 303, an optional visual display device 304, and an optional printer 305. The force-plate 301 measures variables of force related to an individual's performance during prescribed seated, standing, and walking assessment and exercise training tasks. The tools 302 may include devices that are available for placement on the force-plate to enable performance of various additional assessment tasks such as step up and step down tasks and sit to stand or stand to sit tasks. Such a system 300 is also described in U.S. Pat. Nos. 5,980,429 and 6,190,287 as well as in U.S. patent application Ser. No. 09/785,673, each of which is hereby incorporated herein by reference.

[0069] As was the case with the embodiment of FIG. 2, when the computer receives the variables of force for each trial from the force plate 301, it determines a rank order for the variables. As above, the computer 303 may also determine if any of the ranks associated with a given on of the trials has fallen outside a reference range associated with a given trial performed under error-free conditions. Again, another computer, separate from the system 300 and in communication with the system, may be used to calculate these qualities. The computer that determine if any of the ranks associated with a given on of the trials has fallen outside a reference range will then generate a display, user interface, or alarm condition or state that will indicate an instance wherein any of the ranks associated with a given one of the trials has fallen outside the reference range associated with the given trial, either to the subject or to an operator administering the screening-test. The computer 303 may also generate instructions to be used by the subject during the screening-test. The visual display 304 and printer 305 function in a similar manner to the visual display 204 and printer 205 of the embodiment of FIG. 2.

[0070] FIG. 4 is a flow chart illustrating a method for detecting a screening-test error. In process 401, at least one performance parameter related to at least one screening-test task is measured. In process 402, at least one performance statistical quantity characterizing the measured performance parameter is calculated 402 and the at least one performance statistical quantity is compared 403 to at least one reference statistical quantity associated with an error-free screening test. Again, the statistical quantity may represent a value associated with an average or the statistical quantity may represent a value associated with a standard deviation. Additionally, the statistical quantity may represent a value associated with a standard error or the statistical quantity may represent a value associated with a power spectrum. The statistical quantity may further represent a value associated with a root mean square or a value associated with a frequency histogram. The method may also include displaying the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity on a display device.

[0071] FIG. 5 is a flow chart illustrating a method for detecting errors in balance related screening tests. A quantity related to a stability factor of a balance task performed in trials by a subject under a plurality of distinct sensory conditions is measured in process 501, obtaining thereby a quantity related to the stability factor for each trial. A rank order for the quantities is determined 502, wherein each quantity for each trial being associated with a rank. Whether any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial performed under error-free conditions is also determined in process 503. In accordance with a related embodiment, the method may also include displaying a number corresponding to the number of times a performance of the balance task by the subject has fallen outside the reference range.

[0072] As noted above, measuring the quantity related to a stability factor may include following a modified CTSIB protocol. Similarly, determining a rank order for the performance of the plurality of distinct tasks may include determining a rank order according to the level of difficulty of the balance tasks.

[0073] FIG. 6 is a flow chart illustrating a method for detecting a screening test error in an individual trial of a balance task during which sway deviation is measured. In accordance with this method, a quantity corresponding to a moving window root mean square value for velocity of the sway deviation is determined 601. The window is short in relation to the duration of the trial but long in relation to the duration of a typical deviation in sway velocity. An alarm state is entered 602 when the quantity exceeds a threshold value.

[0074] FIG. 7 is a flow chart illustrating a method for detecting a screening test error due to malfunctions of at least one vertical force sensing device (or at least one horizontal force sensing device). Here, a quantity corresponding to a moving window average value for the total vertical force (or total horizontal force) measured by the device is determined 701. The window is long in relation to the duration of expected spontaneous fluctuations in the total vertical force (or horizontal force). An alarm state is entered 702 when the quantity deviates from a constant valued by a predetermined threshold value.

[0075] FIG. 8 is a flow chart illustrating an alternative method for detecting a screening test error due to malfunctions of at least force sensing device. Here, an average of a mathematical derivative for the total vertical (or horizontal) force measured by the device is calculated 801 to determine the rate of change and a quantity corresponding to an average rate of change of the total vertical force over a predetermined period of time is be determined in process 802. An alarm state is entered 803 when the average deviates from zero by a predetermined threshold value.

[0076] It should be understood that methods for comparing the statistical properties of measurements to a reference standard other than rank ordering may be used. For one example, multivariate statistics can be used to determine the relations among multiple trials of each of multiple measures. Reference ranges for multivariate statistics can be defined by the empirical method described above. Specifically, multivariate statistics can be calculated for each of a plurality of subjects known to be cooperative and proficient using an operator known to be highly proficient.

[0077] Using specialized systems such as those mentioned above, effective screening-tests may be administered to a large number of people by individuals who have not had specialized medical training. These tests may also be administered in uncontrolled, non-medical environments such as drug stores and shopping malls. To be maximally effective in reducing the incidence of chronic disease and to minimize a screening agent's liability from failing to identify chronic medical conditions, false negative rates (i.e., failing to detect problems) of screening-tests must be as low as possible. In other words, the criteria for judging test outcomes as abnormal must be set broad enough to miss as few at risk individuals as possible. Setting the abnormal criteria too broadly, however, results in large numbers of false positive outcomes that in turn cause needless worry and medical expense.

[0078] When setting the abnormal criteria for a screening-test, both the sensitivity of the test in detecting chronic disease and the potential for errors in test administration and interpretation must be taken into account. As the potential for test error increases, the false positive rate must be increased to maintain an acceptable false negative rate. Hence, methods that reduce the incidence of errors in test administration and interpretation will significantly enhance the effectiveness of a screening-test.

[0079] FIG. 9 illustrates a basic system which a prescriber, health care practitioner or administrator may use when evaluating a subject's physical condition and developing an individualized training program for the subject. The prescriber first determines the initial conditions. An assessment 900 of the present capabilities of the subject is made. This may involve a screening test and the detection of errors associated with the screening test as described above. The current skill level of the individual should be determined and overall performance goals to be achieved within the scope and duration of training should be decided. It may be advantageous to record one or more initial quantity and quality performance measures upon which the subject will build during the program. The specific training program may then be created 901 consisting of one or more tasks defining the expected performance in terms of quality and quantity of executed tasks. Progress toward performance goals is, then, actively monitored 902 by the prescriber by comparing actual task performance with quality and quantity expectations 903. This requires a significant time and monetary commitment by both a professional prescriber and the subject.

[0080] Compliance monitoring of a training program and/or accurate evaluation of a subject's physical condition may be remotely accomplished by a prescriber. FIG. 10 illustrates a specific example of an evaluation and monitoring system. Site 1000 may be the home of a subject 1006 or other location at some distance from a prescriber site 1010. For example, prescriber site 1010 may be a professional office, a health care facility or an educational facility. Tasks are performed by the subject 1006 at training site 1000, while monitoring is occurring at prescriber site 1010. A measuring device 1001 (in this example, a force plate) is in communication with subject 1006 during task performance. Data communication between measuring device 1001 and display unit 1004 or 1014 may be established by use of one or more modems 1011 or other effective linkages. In accordance with a preferred embodiment, measurements made using device 1001 serve to quantify, for example, the accuracy of task performance by the subject 1006 as well as to count the number of tasks performed over a defined time period. A comparator or other means effective in comparing data (such as a processor not shown) compares the measurement and benchmark inputs and is capable of providing these inputs in a format suitable for further calculation. The comparison means may be located at either site 1000 or at site 1010. It may be free-standing and in data communication with device 1001 and unit 1004 or may be an integral component of either device 1001 or of unit 1004. Quality-benchmark data and quantity-benchmark data may be entered and stored in the comparison means or input to the comparison means from a library (not shown) with suitable data communication linkage. Resultant data is further analyzed at site 1010 yielding at least one result. Suitable analysis means (such as computer or controller 1012) may be free-standing and in data communication with unit 1014 or may be an integral component of unit 1014. Results are displayed to the prescriber from display unit 1014. In accordance with a preferred embodiment, a task-display unit 1004, in data communication with device 1001, is provided at site 1000. The task-display unit 1004 provides information to the subject 1006 regarding tasks being performed as part of the training program. Suitable analysis means (such as computer or controller 1013) may also be provided, in data communication with task-display unit 1004.

[0081] In another embodiment of the invention, displayed results are further categorized to provide feedback to the prescriber, health care practitioner or administrator regarding the effectiveness of the individualized training program. Information related to the subject's compliance is compared to information related to the expected performance and the results of the comparison used to help make determinations of training effectiveness based on one or more of the following: (a) the subject's level of motivation and (b) the level of training task difficulty relative to the subject's performance capabilities.

[0082] The subject's motivation and the appropriateness of the prescribed training program is determined based on a comparison of the expected training performance and the subject's actual compliance, with the subject serving as his or her own control. This method includes the following steps: (1) the subject is assessed and quantities related to initial performance capabilities are documented, (2) a training program is prescribed and expected compliance defined relative to the subject's performance capabilities determined in step (1), (3) one or more measurements related to the quantity and accuracy of actual compliance are recorded during execution of the training tasks, and (4) subject motivation, appropriateness of task challenge, and potential for improvement are determined based on algorithms for combining the expected and compliance information as shown in Table 1. 1 TABLE 1 ACCURACY OF QUANTITY OF TRAINING TASKS PERFORMANCE PERFORMED RELATIVE TO EXPECTED RELATIVE TO BELOW EXPECTED EXP. EQUAL EXP. ABOVE. EXP. BELOW EXP. Unmotivated Possibly Re-Instruct Unmotivated EQUAL EXP. Possibly Appropriate Possibly Unchallenged Unmotivated ABOVE EXP. Re-Instruct Possibly Unchallenged Unchallenged

[0083] The subject's motivation and the appropriateness of the selected training tasks are determined by comparing the subject's initial functional performance assessment data with performance, and training compliance data derived from a reference population of individuals. This method includes the following steps: (1) an initial evaluation of the training candidate's deficit from a norm is made, the subject's performance capabilities relative to the performance goals are assessed, and one or more quantities related to the subject's performance capabilities are recorded, (2) a training task is selected and expected compliance defined, based on the compliance achieved by the reference population and, (3) one or more quantities related to the quantity and accuracy of the subject's actual compliance with the training are recorded, and (4) subject motivation and appropriateness of training tasks are determined as shown in Table 2 below. 2 TABLE 2 ACCURACY OF QUANTITY OF TRAINING TASKS PERFORMANCE PERFORMED RELATIVE TO RELATIVE TO REFERENCE POPULATION REFERENCE ABOVE. POPULATION BELOW REF. EQUAL REF. REF. BELOW REF. Too Difficult Possibly Too Difficult Re-Instruct EQUAL REF. Possibly Too Appropriate Possibly Difficult Too Easy ABOVE REF. Re-Instruct Possibly Too Easy Too Easy

[0084] The following comprehensive assessment reports provide objective information to identify the type and severity of balance and mobility problem areas so that a health practitioner, operator or administrator can accurately evaluate a subject's physical condition and prescribe an exercise program which targets identified problem areas at an appropriate level of difficulty.

[0085] 1. Sit To Stand: The subject assumes a comfortable seated position on a backless stool with the feet placed in standardized positions. The subject rises on command to a standing position as quickly and as comfortably as possible and to maintain the erect position for 5 seconds. The task is repeated three times. The difficulty of the task can be adjusted by modifying the height of the stool. Performance measures include the following:

[0086] (1) Weight Transfer Time is the time in seconds required to voluntarily shift the COG forward from the seat to the base of foot support.

[0087] (2) Rising Index documents the maximum vertical force exerted by the legs during the rising phase. This force is expressed as a percentage of the subject's body weight.

[0088] (3) COG Sway documents control of the COG over the base of support during the rising phase of the maneuver and for 5 seconds thereafter. Sway is expressed as mean velocity of COG sway in degrees per second.

[0089] 2. Walk: The subject stands at one end of the forceplate and on command initiates gait to walk from one end of the forceplate to the other as quickly and comfortably as possible, and then terminates gait at the other end of the forceplate. The task is repeated three times. Task difficulty is increased by requiring tandem heel to toe stepping. The following performance measures are calculated:

[0090] (1) Stride Width is the lateral distance in inches between successive steps.

[0091] (2) Stride Length is the longitudinal distance in inches between successive steps.

[0092] (3) Speed is the velocity in feet per second of the forward progression.

[0093] (4) End Sway is the mean velocity in degrees per second of the antero-posterior (AP) component of COG sway after the subject terminates walking.

[0094] 3. Step/Quick Turn: The subject takes two steps forward, executes a 180 degree in-place turn as quickly and comfortably as possible, and then resumes walking in the opposite direction. The maneuver is repeated three times with the subject turning to the right and three times turning to the left. The Comprehensive Report includes the following two performance measures for each turn direction:

[0095] (1) Turn Time quantifies the number of seconds required for the individual to execute the 180 degree in-place turn. Turn time begins when forward progression is arrested and ends when progression in the opposite direction is initiated.

[0096] (2) Turn Sway quantifies the postural stability of the individual during the turn time defined above. Turn Sway is expressed as the average COG sway path length in degrees.

[0097] 4. Step Up/Over: The individual stands in front of a raised platform, steps onto the platform with the leading leg, swings the opposing leg over the platform and down onto the surface on the opposite side. The maneuver is performed as quickly and as comfortably as possible and repeated three times each with left and right lower extremities. The following performance measures are calculated for each extremity:

[0098] (1) Lift-Up Index quantifies the maximum lifting (concentric) force exerted by the leading leg and is expressed as a percentage of the subject's weight.

[0099] (2) Movement Time quantifies the time required to complete the maneuver, beginning with the initial weight shift to the non-stepping (lagging) leg and ending with impact of the lagging leg onto the surface. Time is measured in seconds.

[0100] (3) Impact Index quantifies the maximum vertical impact force as the lagging leg lands on the surface. The force of impact is expressed as a percentage of the subject's weight.

[0101] 5. Forward Lunge: The subject lunges forward as far, quickly, as is comfortably possible three times with each lower extremity. During each lunge, the subject reaches to the maximum distance and, with a smooth continuous movement, then pushes back to the starting position. The report includes the following performance measures for each extremity:

[0102] (1) Distance is the lunge length expressed as a percentage of the subject's height.

[0103] (2) Impact Index is the maximum vertical force exerted by the stepping leg onto the surface during the landing, expressed as a percentage of the subject's body weight.

[0104] (3) Contact Time is the duration in seconds of surface contact with the lunging leg in the forward (or lateral) position.

[0105] (4) Force Impulse is a measure of the total work performed by the lunging leg during the landing and thrust phases of the movement. Force impulse is expressed in units of % body weight (force) multiplied by the time the force is exerted in seconds.

[0106] 6. Weight Bearing/Squat: To accommodate the differing functional levels of subjects, the Weight Bearing/Squat is performed with weight bearing only in the fully erect position, fully erect, 30, and 60 degrees of knee flexion, or fully erect, 30, 60, and 90 degrees of knee flexion.

[0107] 7. Bilateral (Unilateral) Stance: The subject assumes a quiet standing position and is instructed to maintain the position as still as is comfortably possible. Each condition is repeated three times. Bilateral stance (Level I and II) trials last 20-seconds each, while the Level III trials last 10 seconds.

[0108] COG Sway is the mean sway velocity in degrees per second.

[0109] Composite Sway is the mean sway velocity averaged over the six trials.

[0110] COG Alignment is the position of the center of gravity at the start of each trial, expressed as displacement from center in the forward-backward and lateral directions.

[0111] 8. Limits Of Stability: The subject stands with feet in standardized positions, views a cursor display of their COG position, and orients himself so that the COG cursor coincides with a target representing the center of the LOS area. A second target is then placed a predetermined distance from center relative to the LOS boundary. The subject moves the COG cursor on command as quickly and as accurately as possible to the perimeter target and to hold the perimeter position for five seconds. The task is repeated with a total of eight perimeter targets, representing the four cardinal directions (forward, backward, left, right) and the four diagonal directions. Results of the eight target movements are combined into the four cardinal directions by a weighted averaging of the forward-backward and left-right components of the diagonal movement scores into the relevant cardinal direction scores. The report shows the following measures for forward, backward, right, and left target movements:

[0112] (1) Reaction Time is the time in seconds between the command to move and the point at which the subject first initiates movement.

[0113] (2) Movement Velocity is the average speed of movement in degrees per second measured between the time the subject reaches 5% of the End Point distance until 95% of the End Point distance is achieved.

[0114] (3) End Point Excursion is the distance reached upon completion of the first movement to the target. Distance is expressed as a percentage of the theoretical maximum LOS distance. The first movement is completed when progress towards the target ceases.

[0115] (4) Maximum Excursion is the maximum distance achieved during the trial. The maximum excursion may be larger than the End Point excursion if the subject initiates additional movements towards the target following the termination of the first.

[0116] (5) Directional Control is the off-axis (left and/or right of the straight line path to the End Point) distance moved by the subject compared to the on-axis distance moved (the straight line path) to the End Point. Direction control is calculated by subtracting the off-axis distance from the on-axis distance, and expressing the difference as a percentage of the actual on-axis distance. A Direction Control score of 100% indicates a perfect straight line movement from the center to the End Point defined above, while lower percentages indicate larger off-axis movements.

[0117] 9. Rhythmic Weight Shift: The subject stands in-place with feet in standard positions on the forceplate. The subject views a cursor display of their COG position and is instructed to move rhythmically such that the COG cursor moves back and forth the full distance between two boundaries spaced in opposite directions from center and at 50% of the distance to the LOS perimeter. The required rhythm of the back and forth movement is demonstrated to the subject by a pacing target. The subject performs the task either with rhythmic antero-posterior movements or lateral movements between boundaries placed either to the front and back or to the left and right of the center position, respectively.

[0118] To accommodate the differing functional levels of subjects, the Rhythmic Weight Shift assessment is included in all three levels of difficulty:

[0119] (1) Level I (easy) with the pacing targets moving slowly at one cycle every three seconds, (2) Level II (moderate) with the pacing target moving at one cycle every two seconds, and (3) Level III (challenging) with the pacing target moving rapidly at one cycle per second.

[0120] The Rhythmic Weight Shift report includes the following measures:

[0121] (1) On-Axis Velocity is the average speed in degrees per second of the rhythmic movement along the specified movement direction.

[0122] (2) Direction Control is the average velocity of on-axis motion expressed as a percentage of the total (on-axis and off-axis velocity) motion.

[0123] Training protocols associated with the Neurocom International, Inc. systems mentioned above are designed to provide the treating practitioner with a menu of training exercises which are grouped by task type and difficulty level. Because the assessment and training protocols are organized into the same task groups, the assessment results can be readily used by a practitioner to evaluate the subject's physical condition (taking into account possible errors in screening test administration) and to prescribe training tasks of the type and difficulty levels which are likely to achieve functional improvements in performance.

[0124] 1. Seated Activities: Seated exercises begin at the lowest levels with control of trunk alignment over the base of support Intermediate levels exercise the subject's ability to perform weight shifts to specified points within the base of support and to maintain alignment on compliant and movable seated surfaces. The highest levels of seated exercise include trunk control functions necessary for transition from seated to standing activities.

[0125] 2. Weight Shifting Activities: Weight shifting exercises are performed in freely standing positions and emphasize the maintenance of body alignment and active center of gravity control to all locations within the limits of stability area. At the lowest levels of exercise alignment and movement activities are performed on stable, firm surfaces between locations well within the limits of stability. More challenging weight shifting exercises are performed on compliant foam rubber and moving rocker board surfaces, and subjects are required to move to locations at the outermost boundaries of the limits of stability.

[0126] All levels of weight shift training allow the user to emphasize forward, backward, left, or right movement directions.

[0127] 3. Lower Extremity Closed Chain Activities: Lower extremity closed chain activities are performed in freely standing and weight bearing positions and emphasize the weight control, strength, flexibility, and motor control component functions. Activities can be modified to emphasize the left or right extremity and to focus on the ankle, knee, hip, or lower back of the designated side. Lower levels exercise the targeted joint while the subject maintains less than full 100% weight on the involved leg. Higher levels use steps, wedges, and lunge movements to modify the range of joint motion during full weight bearing exercises. Tasks involving controlled COG movements over the base of support at all levels of training exercise lower extremity proprioception and motor control. The challenge is increased at the higher levels by increasing the movement distances and by performing the movements on foam rubber, wedge, and rocker board surfaces.

[0128] 4. Mobility Activities: Mobility activities involve freely moving sit to stand, stepping, turning, and climbing tasks performed over the full extent of the forceplate surface. The tasks emphasize gait stability, step placement and planning, maneuvering over obstacles, and changes in gait direction. The lowest levels of exercise emphasize preparation for and transitions between sitting and standing. Intermediate levels emphasize weight shifts and step initiation, while the highest levels involve narrow based gaits, obstacles, turns, and complex step placement patterns.

[0129] Training exercises and screening protocols involving the above task groups may be divided into five levels of difficulty designed to address the evolving needs of the subject over the course of recovery. Level I training exercises may be designed for the earliest phases of recovery when assessment results indicate that the subject performed well below levels of the reference population. At the other extreme, Level V exercises may be designed for maximizing the training challenge of the highest level subjects and skilled athletes when assessment results indicate that the subject performed well above levels of the reference population. Training Levels II through IV may be designed to progressively increase the challenge in graded steps between the earliest phases of recovery and the maximum challenge.

[0130] Each training sequence provides the following compliance information:

[0131] 1. Training Quantity: The Training Quantity portion of the subject's compliance information file informs the clinician relative to the date(s), exercise task type(s), exercise task difficulty level(s) and repetition of each task performed.

[0132] 2. Training Quality: Depending on the level of challenge selected, a typical exercise task contains one hundred or more individual movement repetitions. For each task performed, the Training Quality portion of the report compares the total number of movement repetitions attempted by the subject to the number of repetitions in which the subject successfully reached the appropriate target goal. For purposes of compliance documentation, a repetition is deemed successfully completed when the subject moves the COG cursor into the designated target zone within the time allotted for the individual movement.

[0133] 3. Training Report: A summary report is generated for a training session consisting of one or more exercise tasks. The graphic portion of the report illustrates the target configuration and COG cursor path for each exercise task. A complete session typically contains up to seven exercise tasks, depending on the difficulty level. The numerical data associated with each exercise includes:

[0134] (1) Exercise Task Duration: The duration of the exercise in seconds.

[0135] (2) Successful Target Repetitions: The number of target repetitions successfully completed.

[0136] (3) Attempted Target Repetitions: The number of target repetitions attempted.

[0137] Included in Neurocom International, Inc. products' assessment protocols is the capability to compare the performance results of the individual subject to that of a reference population of individuals. The graphic presentations of each comprehensive report, areas falling outside the performance range achieved by 95% of the reference population are shaded, whereas areas falling within the 95% range are not shaded. By observing the location of the subject's results on the graph relative to the shaded and unshaded areas, the practitioner can readily determine the performance capabilities of the subject relative to those of the reference population.

[0138] Although the embodiments hereinbefore described are preferred, many modifications and refinements which do not depart from the true spirit and scope of the invention may be conceived by those skilled in the art. It is intended that all such modifications, including but not limited to those set forth above, be covered by the following claims.

Claims

1. A system for detecting a screening-test error, the system comprising:

a measurement device that measures at least one performance parameter related to at least one screening-test task; and

a computational device, in communication with the measurement device, that receives the at least one measured performance parameter, calculates at least one performance statistical quantity characterizing the measured performance parameter, and compares the at least one performance statistical quantity to at least one reference statistical quantity associated with an error-free screening test.

2. A system according to claim 1, further comprising a display device that displays the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity.

3. A system for detecting errors in balance related screening tests, the system comprising:

a force-plate for measuring a quantity related to a stability factor of a balance task performed in trials by a subject under a plurality of distinct sensory conditions; and

a computation device in communication with the force-plate, the computational device

(i) receiving the quantity related to the stability factor for each trial,

(ii) determining a rank order for the quantities, each quantity for each trial being associated with a rank, and

(iii) determining if any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial performed under error-free conditions.

4. A system according to claim 3, further comprising a display device in communication with the computational device for indicating an instance wherein any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial.

5. A method for detecting a screening-test error, the method comprising:

measuring at least one performance parameter related to at least one screening-test task; and

calculating at least one performance statistical quantity characterizing the measured performance parameter; and

comparing the at least one performance statistical quantity to at least one reference statistical quantity associated with an error-free screening test.

6. A method according to claim 5, wherein the statistical quantity represents a value associated with an average.

7. A method according to claim 5, wherein the statistical quantity represents a value associated with a standard deviation.

8. A method according to claim 5, wherein the statistical quantity represents a value associated with a standard error.

9. A method according to claim 5, wherein the statistical quantity represents a value associated with a power spectrum.

10. A method according to claim 5, wherein the statistical quantity represents a value associated with a root mean square.

11. A method according to claim 5, wherein the statistical quantity represents a value associated with a frequency histogram.

12. A method according to claim 5, wherein:

(i) the screening-test task is a balance task;

(ii) the at least one performance parameter is sway deviation;

(iii) the at least one performance statistical quantity corresponds to a moving window root mean square value for velocity of the sway deviation; and

(iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity includes determining whether the moving window root mean square value deviates from a constant value by a predetermined threshold value.

13. A method according to claim 5 wherein:

(i) the screening-test task is a balance task;

(ii) the at least one performance parameter is vertical force applied to a force plate;

(iii) the at least one performance statistical quantity corresponds to a moving window average value for total vertical force applied to the force plate; and

(iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity includes determining whether the moving window average value deviates from a constant value by a predetermined threshold value.

14. A method according to claim 5, wherein:

(i) the screening-test task is a balance task;

(ii) the at least one performance parameter is vertical force applied to a force plate;

(iii) the at least one performance statistical quantity corresponds to an average of a mathematical derivative of the total vertical force applied to the force plate; and

(iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity includes determining whether the average deviates from zero by a predetermined threshold value.

15. A method according to claim 5, wherein:

(i) the screening-test task is a balance task;

(ii) the at least one performance parameter is horizontal force applied to a force plate;

(iii) the at least one performance statistical quantity corresponds to an average of a mathematical derivative of the total horizontal force applied to the force plate; and

(iv) comparing the at least one performance statistical quantity to the at least one reference statistical quantity includes determining whether the average deviates from zero by a predetermined threshold value.

16. A method according to claim 5, further comprising displaying the extent to which the at least one performance statistical quantity differs from the at least one reference statistical quantity on a display device.

17. A method for detecting errors in balance related screening tests, the method comprising:

measuring a quantity related to a stability factor of a balance task performed in trials by a subject under a plurality of distinct sensory conditions;

obtaining thereby the quantity related to the stability factor for each trial;

determining a rank order for the quantities, each quantity for each trial being associated with a rank; and

determining if any of the ranks associated with a given one of the trials has fallen outside a reference range associated with the given trial performed under error-free conditions.

18. A method according to claim 17, further comprising displaying a number corresponding to the number of times a performance of the balance task by the subject has fallen outside the reference range.

19. A method according to claim 17, wherein measuring the quantity related to a stability factor includes following a modified CTSIB protocol.

20. A method according to claim 17, wherein determining a rank order for the performance of the plurality of distinct tasks includes determining a rank order according to the level of difficulty of the balance tasks.

21 A method for detecting a screening test error in an individual trial of a balance task during which sway deviation is measured, the method comprising:

determining a quantity corresponding to a moving window root mean square value for velocity of the sway deviation, the window being short in relation to the duration of the trial but long in relation to the duration of a typical deviation in sway velocity; and

entering an alarm state when the quantity exceeds a threshold value.

22. A method for detecting a screening test error due to malfunctions of at least one vertical force sensing device, the method comprising:

determining a quantity corresponding to a moving window average value for the total vertical force measured by the device, the window being long in relation to the duration of expected spontaneous fluctuations in the total vertical force; and

entering an alarm state when the quantity deviates from a constant valued by a predetermined threshold value.

23. A method for detecting a screening test error due to malfunctions of at least one vertical force sensing device, the method comprising:

calculating an average of a mathematical derivative for the total vertical force measured by the device to determine the rate of change of the total vertical force;

determining a quantity corresponding to an average rate of change of the total vertical force over a predetermined period of time; and

entering an alarm state when the average deviates from zero by a predetermined threshold value.

24. A method for detecting a screening test error due to malfunctions of at least one horizontal force sensing device, the method comprising:

calculating an average of a mathematical derivative for the total horizontal force measured by the device to determine the rate of change of the total horizontal force;

determining a quantity corresponding to an average rate of change of the total horizontal force over a predetermined period of time; and

entering an alarm state when the average deviates from zero by a predetermined threshold value.