FIELD OF THE INVENTION The present invention relates to apparatuses and methods for the measurement, the monitoring and/or the analysis of motor control.

BACKGROUND OF THE INVENTION It is often important, in psychology and neuroscience, to determine the response time to various stimuli to thereby analyse the neuromuscular activity.

A simple method has been used over the years to determine the reaction time to a stimulus. The patient is confronted with a visual or auditory stimulus and has to react as quickly as possible by depressing an actuator such as a switch, for example. Of course, many variations of this method have also been used to allow the study of tasks of various complexities to try to understand neuromuscular mechanisms

associated to the task at hand. It is even possible to diagnose various disorders in medicine, psychology and kinesiology.

Figure 1 of the appended drawings, which is labelle "Prior Art", illustrates the simple method described hereinabove. A series of stimuli 10 are generated to get the attention of the patient. After a random duration delay (see arrow 12), the real stimulus 14 is generated. The system then evaluates the reaction time 16 between the real stimulus 14 and the patient's response 18 to the stimulus. The reaction time 16 varies according to various criteria, such as, for example, the type of stimulus generated, the complexity of the task to be performed by the user to respond to the stimulus, the patient's aptitudes. The analysis of these variations helps the physician to understand the brain and neuromuscular activity of the patient.

The physician, knowing the complexity of the task to be performed, may analyse the reaction time of the patient.

A drawback with this conventional method is that only the reaction time is known, and that all the conclusions of the physician are based on this single piece of information.

OBJECTS OF THE INVENTION An object of the present invention is therefore to provide an improved apparatus and method for the measurement, the monitoring

and/or the analysis of motor control ability, performance, training, rehabilitation and/or disorder.

SUMMARY OF THE INVENTION More specifically, in accordance with the present invention, there is provided an apparatus for the measurement, the monitoring and/or the analysis of motor control ability, performance, training, rehabilitation and/or disorders of a patient, the apparatus comprising: a controller; a stimuli generator connected to the controller to be controlled thereby; an input device connected to the controller to supply raw data thereto; the input device measuring motor control signals from a predetermined motor control task done by the patient upon generation of a stimulus by the stimuli generator; wherein the controller is configured for a) acquiring raw data from the input device, b) fitting this raw data onto a predetermined curve following a predetermined equation and c) extracting the values of the parameters from the predetermined equation.

According to another aspect of the present invention, there is provided a method for the measurement, the monitoring and/or the analysis of motor control ability, performance, training, rehabilitation and/or disorders of a patient, the method comprising:

supplying a controller; supplying a stimuli generator connected to the controller to be controlled thereby; generating a stimulus to indicate to the patient that the test has begun; supplying an input device connected to the controller to supply raw data thereto; measuring motor control parameters of a predetermined task done by the patient upon said generation of a stimulus by the stimuli generator; transferring time indexed motor control parameters from the input device to the controller; fitting the time indexed motor control parameters onto a predetermined curve following a predetermined equation; and extracting the values of the parameters from the predetermined equation.

It is to be noted that even though the terms"patient"and "physician"are used throughout the present disclosure, this is done for concision purposes since the methods and apparatuses described herein as being part of the present invention can be used for other purposes than therapeutic as will be better understood by the foregoing disclosure.

Other objects, advantages and features of the present invention will become more apparent upon reading of the following non- restrictive description of preferred embodiments thereof, given by way of example only with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS In the appended drawings: Figure 1, which is labeled"Prior Art", is a schematic representation of the information gathered using conventional methods and apparatus; Figure 2 is a typical curve of the velocity vs. time for a simple quick movement; Figure 3 is a schematic block diagram of an apparatus according to an embodiment of the present invention; Figure 4 is a flowchart illustrating a method according to an embodiment of the present invention; Figure 5 is a timeline for the analysis of the reaction according to the present invention; Figure 6 is an example of acquired data for a healthy patient; Figure 7 is an example of acquired data for a patient having the Alzheimer disease; and

Figure 8 is an example of the decomposition of a complex movement into a sequence of elementary strokes.

DESCRIPTION OF THE PREFERRED EMBODIMENT It has been found that simple rapid human movements may be represented by a velocity function that can be described by a Delta-Lognormal equation as illustrated in Figure 2.

According to the kinematics theory, simple human movements can be described in the velocity domain as the response of a synergistic action of an agonist and an antagonist neuromuscular network (Plamondon R., (1993a)"The Generation of Rapid Human Movements : Part I: A-log-normal Law", EPM/RT-93/4; Plamondon R., (1995a)"A Kinematic Theory of Rapid Human Movements. Part I: Movement Representation and Generation", Biological Cybernetics, 72: 297-307.). Each network is composed of a large set of coupled neuromuscular subsystems that react to an input command D1 (for the agonist) and D2 (for the antagonist) with an impulse response that can be described by a lognormal function (Plamondon R., (1993a)"The Generation of Rapid Human Movements: Part I: A"log-normal Law", EPM/RT-93/4; Plamondon R., (1995a)"A Kinematic Theory of Rapid Human Movements. Part): Movement Representation and Generation", Biological Cybernetics, 72: 297-307.).

Each lognormal impulse response A (t; to, j2) can be characterized by three parameters: the starting time to, the parameter p which reflets the logtime delay, and CYj2 which reflets the logresponse time of the neuromuscular system (Plamondon R., (1993a)"The Generation of Rapid Human Movements: Part): A~log-normal Law", EPM/RT-93/4; Plamondon R., (1995a)"A Kinematic Theory of Rapid Human Movements. Part): Movement Representation and Generation", Biological Cybernetics, 72: 297-307.).

The resulting curvilinear velocity V (t) of a single movement is then described by subtracting the weighted impulse response of the antagonist network from the agonist one, which is called a delta-lognormal response (see Equations 1 and 2).

AA (f)=D,A(,/o-,')-DA(,/)(1) where Equation 1 results in velocity profiles that can have one, two or even three peaks (Plamondon R., (1995a)"A Kinematic Theory of Rapid Human Movements. Part l: Movement Representation and Generation", Biological Cybernetics, 72: 297-307; Plamondon, R., (1998) "Kinematic Theory of Rapid Human Movements: Part III : Kinetic Outcomes", Biological Cybernetics, 78: 133-145; Plamondon R., Alimi

M. A., (1997)"Speed/Accuracy Tradeoffs in Target Directed Movements", Behavioral and Brain Science, 20: 279-349; Plamondon R., Guerfali W., (1998)"The Generation of Handwriting with Delta-Lognormal Synergies", Biological Cybernetics, 78: 119-132.).

The vectorial delta-lognormal model (Plamondon R., (1995)"A Delta-Lognormal Model for Handwriting Generation", Proc. 7th Biennal Conference of the International Graphonomics Society, 126-127.; Guerfali W., Plamondon R., (1995)"The Delta Lognormal Theory for the Generation and Modelling of Cursive Characters", Proc. 3 rd Int. Conf. on Document Analysis and Recognition: 495-498.; Guerfali W., (1996) "Modèle delta lognormal vectoriel pour I'analyse du movement et la generation de l'écriture manuscrite", Ph. D. dissertation, École Polytechnique de Montreal.) describes single two-dimensional movements as a velocity vector, the magnitude of which follows a delta-lognormal response. Apart from the seven parameters of the delta-lognormal function, each velocity vector is also characterized in the space domain by three static parameters which globally reflect the geometric properties of the set of muscles and joints used in a particular movement: the starting point Po, the starting direction 60 and the global curvature Co (Plamondon R., Guerfali W., (1998)"The Generation of Handwriting with Delta- Lognormal Synergies", Biological Cybernetics, 78: 119-132.). The curvature is considered to be positive if the movement is clockwise, and negative otherwise.

An extraction technique to yield the seven variables to, D" D2,,,2,, and a2 is described in (Guerfali W., (1996)"Modele Detta

Lognormal Vectoriel pour I'Analyse du Mouvement et la Génération de l'Écriture Manuscrite", Ph. D. dissertation, École Polytechnique de Montreal", Guerfali W., Plamondon, R., (1998)"A New Method for the Analysis of Simple and Complex Human Movements", Journal of Neuroscience Methods, 82: 35-45) and is therefore included herein by reference.

For simple movements, the extraction of the set of parameters that best fit a delta-lognormal velocity profile can be achieved in two steps. First, estimate the parameters that describe the kinematics of the movement D"D2,, u1, >2, o 2 and to. Second, estimate the static parameters that describe the geometric properties of the movement in the 2D plane, Po, Co and eo. Seeking the optimal set of kinematic parameters that best fits the movement observed requires the use of robust optimization approaches which ensure algorithm convergence. Nonlinear regression technique can be used. To ensure convergence, to reduce the search space and to minimize computation time, a graphical method is used to find an approximation of the optimal solution prior to nonlinear optimization (Guerfali W., Plamondon R., (1994)"Robust Parameter Extraction Techniques for the Delta LogNormal Model", Vision Interface, 218-225). To estimate the geometric properties of a single movement, various approaches can be used, and both graphical and analytical methods can be employed to estimate the overall curvature Co of a single stroke and its starting angle 60 and starting point Po.

The analysis of complex movements, or multiple-stroke movements, such as fluent handwriting requires decomposing a fluent

movement into a multiple independent strokes. The analysis of complex movements can thus be divided into five steps. First, estimate the minimal number of strokes (or delta-lognormal curves which compose the curvilinear velocity) that can generate the observed movement. Second, estimate the kinematic parameters of each single stroke (i): D, (i),D2(i), µ1(i), µ2(i), #1(i), #2(i) and to (i), while avoiding as much as possible the time-overlap effects. Third, estimate the static parameters that describe the geometric properties of each movement in the 2D plane, P0(i), C0(i) and eio (i,. Fourth, optimize the parameters estimated with nonlinear regression techniques for all the successive vectorial delta-lognormal functions. Fifth, make geometrical corrections of parameters Po, Co and Oo to minimize the error between the reproduced shape and the observed shape and restart step 4 until satisfactory conditions, (which can be thresholds on error or time consumption) are reached.

Figure 3 of the appended drawings illustrates an embodiment of an apparatus 20 according to the present invention. This apparatus is designed to detect, record and analyse the movements of the patient's hand. The apparatus 20 includes a controller in the form of a computer 22 running a software program as will be described hereinbelow, a stimuli generator in the form of a sound or light generator 24 connected to the computer 22 and an input device in the form of a digitizing tablet 26.

The purpose of the input device is to detect and measure the raw data, in the form of motor control signals, from a predetermined motor control task as will be understood by the foregoing description.

The software program running on the computer 22 has many functions, such as, for example, to precisely control the timing of the output of the sound or light generator 24, to control the acquisition of the data from the digitizing tablet 26, to store the raw data onto a computer readable medium, to compute displacement, velocity, acceleration or force signals from raw data, and fitting these signals onto the Delta Lognormal curve, to extract the seven variable values from the raw data and to display the results.

Figure 4 of the appended drawings is a flowchart schematically illustrating a possible method used to operate the apparatus 20. This possible method of operation will now be described.

In the first step 30, the operator starts the system.

Then, in step 32, the system proceeds to its initialisation.

This process includes, for example, the loading of the appropriate computer program, the determination of the nature of the test to be performed, the test of the sound or light generator 24 and of the digitizing tablet 26. Of course, the initialisation step may include a digitizing tablet calibration substep.

The loaded computer program then begins the test sequence by giving instructions to the patient (step 34). Some of these instructions are general and apply to every tests and others are dependent on the nature of the test to be done.

After the instructions have been given to the patient, the computer 22 controls the sound or light generator 24 to emit a series of "get ready"signals (step 34) to get the attention of the patient. Of course, the"get ready"signal could take other forms and could use another type of stimulus as the actual test stimulus. For example, the"get ready"signal could be a recorded voice or a visual stimulus.

In step 38, the computer generates a pause having a random duration after the"get ready"signal.

Once this delay is over, the computer controls the sound or light generator 24 (step 40) so as to emit the stimulus signal.

Simultaneously, from this moment, the time indexed X, Y position of the stylus onto the digitizing tablet is continuously recorded by the computer 22 during the entire duration of the test. Thus, the time indexed X, Y position of the stylus onto the digitizing tablet is the motor control signal measured by the input device.

When the patient movement is over, the computer 22 stores (step 42) the time indexed X, Y positions onto the computer hard disk with other information pertaining to the nature of the test performed, for subsequent analysis.

As will readily be apparent to one skilled in the art, since both the stimuli generator and the input device are controlled by the controller, the controller may easily time index all the X, Y positions, forces of velocity detected by the input device.

The raw data of the time indexed X, Y positions is used to determine the velocity of the stylus at all time. In step 44, these calculated values are fitted onto the velocity vs. time model illustrated in Figure 2. In this step, the computer determines if the movement is a simple movement or a complex movement with or without superposition.

Description of this feature is described in (Guerfali W., Plamondon, R., (1998)"A New Method for the Analysis of Simple and Complex Human Movements", Journal of Neuroscience Methods, 82: 35-45) which is hereby included herein by reference. In general, if a complex movement is detected, this movement is decomposed according to a vector superimposition of Delta Lognormal functions (see for example Equation 3 hereinbelow for the speed domain where #(i)(t-t0(i) follows a Delta- lognormal profile as discussed hereinabove with respect to Equations 1 and 2). The number n being an indication of the complexity of the movement.

It is to be noted that if the movement is a complex movement, the raw data will be fitted onto more than one Delta-Lognormal curves. Figure 8 of the appended drawings shows how a compiex movement may be represented by separate overlapping Delta-Lognormal curves. More specifically, Figure 8 illustrate how a complex movement 70 can be decomposed into elementary strokes. In Figure 8, four strokes 72,74,76 and 78 (shown in dashed lines) can be represented by respective Delta-lognormal curves from which the parameters can be

extracted. The parameters, the shape, the timing sequence and the amplitude of the strokes can then characterize and distinguish particular behavior. This decomposition of complex movements can be generalized and used for unidimensional, bi-dimensional or tri-dimensional complex movements.

The next step 46 consists of the extraction of the value of the seven parameters (to, D"D2,, u"p2, C1 and C2) of the Delta Lognormal equation using the fitted data, as briefly described hereinabove. Of course, these values may be optimized using non-liner regression techniques.

The extracted values may then be displayed (step 48) for example, on the monitor of the computer 22. They may also be saved along with the raw data and printed if desired.

Optionally, in step 50, the acquired velocity vs. time curve and the theoretical velocity vs. time model are displayed, for example on the monitor of the computer 22, for visual comparison of these two curves by the operator.

As will easily be understood by one skilled in the art, steps 44 to 50 may be executed anytime after steps 30-42 since the raw data is stored onto the computer hard disk.

Turning now to Figure 5 of the appended drawings, a generic example of the data acquired by the apparatus 20 will be described.

As discussed hereinabove with respect to Figure 1, a series of stimuli 52 are generated to get the attention of the patient. After a random delay (see arrow 54), the real stimulus 56 is generated, as discussed hereinabove. The reaction time (see arrow 58) is calculated from the real stimulus 56 to the moment a predetermined velocity is reached. The points 60 on the velocity vs. time curve 62 are calculated from the acquired position values. The curve 62 is extrapolated from these calculated values.

Figure 6 of the appended drawings illustrates actual data from a test done with the apparatus of Figure 3 on a healthy patient. As can be seen from this figure, the stimulus is represented by a vertical line 64 and the curve 66 has a similar shape to the theoretical shape illustrated in Figure 2.

On the other hand, Figure 7 of the appended drawings illustrates the same test performed by a patient suffering from the Alzheimer disease. As is visually apparent from this figure, the curve 68 is different in shape from the theoretical curve of Figure 2. It shows various speed variations. It has been found that this velocity behaviour is common to many Alzheimer patients.

It is to be noted that while the description hereinabove is concerned with the velocity of a stylus (not shown) as it moves onto the digitizing tablet, other types of motor control parameters such as the displacement, the acceleration, the strength or the pressure could also be used to yield similar results, depending on the type of tasks to be performed by the patient when the stimulus is heard. It is to be noted that the type of input device is advantageously chosen to adequately detect and quantify the relevant motor control parameters of the type of task to be performed. For example, strain gauges could be used to acquire strength data and cameras or other electromagnetic systems could be used to acquire two or three-dimensional movements.

Similarly, while the stimuli generator is described hereinabove as being designed to generate audible stimuli, other types of stimuli generator could also be used to generate, for example, visual, olfactory, tactile, or gustatory stimuli.

As will easily be understood by one skilled in the art, the apparatus and method discussed hereinabove have been given as example only of what can be done according to the present invention. It has been found that the present invention can be adapted to yield significant results in the following fields and/or situations, for example: -Diagnostic and follow-up of patients suffering from dystrophy; -Diagnostic and follow-up of patients suffering from neuro-degenerative illness; -Evaluation of the effects of drugs;

-Evaluation of the effects of medications; -Evaluation of the effects of alcohol; -Evaluation of sports activities (physical capacity, performances and rehabilitation); -Helping the design of training material; -Evaluation of the response time and of latency time; -Study of the Speed-Accuracy trade-off; -Evaluation of the reflexes and response time (for example, in the evaluation of the condition of patients for insurance companies) -Measure of performances (flight simulation, army, aerospace industry) -Evaluation of computer interface ergonomics; -Stress evaluation; -Lie detection; -Evaluation of the behaviour of a patient faced with multiple stimuli (for example, arcade games); and -General studyofhuman and animal behaviour.

Having listed various general applications of the apparatus and method of the present application, we will now briefly describe examples of specific applications.

Alcohol level detector An apparatus similar to the one illustrated in Figure 3 is packaged in a single small package (not shown) having a digitizing tablet top surface. The controller is initially programmed with the"normal" response of the user of the apparatus which is stored in a memory of the

controller. When in use, the apparatus asks the user to do a particular task, for example, when the stimuli is heard or seen, make a rapid movement from one corner to the other corner of the tablet. The apparatus then extracts the parameters values as discussed hereinabove and compares them with the stored values to determine if they are similar or if the user is not in a"normal"state. Of course, such an apparatus could be linked to the ignition system of a vehicle.

Golf training An apparatus provided with a camera system as an input device and where the controller is provided with the variable values of particular movements of a professional golfer could be used to indicate to a user the level of similarity of the movements of the professional golfer and the movements of the user. Of course, such a system could be used for other sports such as for example, hockey, baseball, training or biking.

Injury rehabilitation In this application example, strain sensors are mounted to the movable parts of a conventional gym training machine. The controller of the apparatus is configured to store the results into its memory and automatically compare the most recent results with previous results to indicate the level of progression of the user. Of course, not only rehabilitation but also training could be done on such an equipped gym training machine.

Personal identification An apparatus similar to the alcohol level detector may also be used to identify or characterize a user or a class of users.

Although the present invention has been described hereinabove by way of preferred embodiments thereof, it can be modified, without departing from the spirit and nature of the subject invention as defined in the appended claims.