WO2000001303A1

WO2000001303A1 - Apparatus for assessment of grip function

Info

Publication number: WO2000001303A1
Application number: PCT/GB1999/002115
Authority: WO
Inventors: Paul Mcarthur
Original assignee: The University Of Sheffield
Priority date: 1998-07-02
Filing date: 1999-07-02
Publication date: 2000-01-13
Also published as: GB2339025A; AU4633799A; GB9814226D0

Abstract

Apparatus for use in grip analysis comprising a glove having one or more grip transducers disposed thereon. The transducer(s) are actuable by a wearer of the glove to provide an indication of the degree of actuation.

Description

APPARATUS FOR ASSESSMENT OF GRIP FUNCTION

This invention relates to the field of apparatus for the assessment of grip function.

The calculation of the forces acting at a joint is complex. Mathematical models have taken into account forces generated by the tendons acting across the proximal interphalangeal joint (PIPJ) during static grip activities, but owing to the number of unknown variables featured in the equations used, a number of assumptions and modifications of the model used in the calculation have to be made. For example, modifications to the numerical models have lead to the reduction of the small joints of the finger to simple one-dimensional hinge joints. The calculation of tendon forces has been made using the principle of Newton's First Law of motion.

For static grip, all forces in the finger must balance, due to the isometric nature of the activity. In principle, owing to the nature in which the grip force has conventionally been measured - using systems designed to provide a single force reading - this has meant that a single force balances all those generated by the complexity of tendons governing the grip activity.

Commercially-available grip analysis systems provide data allowing the investigator to monitor the grip of an individual or a group of subjects. Various types of grip analysis system are known and each of these is discussed briefly below:

Mano βtric systems comprise an air-filled deformable chamber connected to a pressure monitor, either analogue or mercury column. These systems are cheap and comfortable to use but they have potential for significant error; differing original chamber volumes and pressures combined with variable grip techniques generate inconsistent results. Furthermore, the deformation of a chamber can only give a single reading and hence the analysis of individual digits during a mass grip activity is not possible with this type of system.

The Jamar Dynamometer is a hydraulic dynamometer which is currently recognised as the standard for cynametric analysis. It has an adjustable handle for use with a wide variety of patients, but it is heavy and cumbersome for patients with a weak grip and small hands. The data obtained from a Jamar is only available from a mass grip activity; individual digits are not assessed, so that pinch and key assessment requires a separate pinch meter. A single strain gauge system produced by MEI (see Helliwell et al, Annals of Rheumatic Disease 1987 Vol 46, 203-8) does allow mass grip, pinch and key grip analysis, but not assessment of individual digits during grip.

Higher generation dynamometers incorporate individual digit analysis during mass grip action. These systems incorporate a pressure sensor for each finger across a handle that can be adjusted for differing hand sizes.

An example of such a device is the KK Digits Grip (a trademark) which is described in US patent r.o 5,317,916. This device has a series of strain gauges arranged in series to allow the measurement of each individual finger contribution during mass grip. A PC is required to utilise the software provided to analyse the signals generated by the strain gauges .

Systems such as the NK Digits Grip (trademark) illustrate the contribution of individual fingers during mass grip but do not yield data on the role of the thumb. Furthermore, for pinch, key and torque analysis, further instrumentation is required.

There is thus a need for a grip analysis system which is able to provide data on the roles of each individual finger and the thumb during different types of grip activity.

It is therefore an object of the present invention to alleviate at least some of the problems of the above-described prior art.

According to a first aspect of the present invention there is provided apparatus for use in grip analysis comprising a glove having one or more grip transducers disposed thereon, said transducer (s) being actuable by a wearer of said glove to provide an indication of the degree of actuation.

Preferably, the or each transducer is a piezoelectric transducer .

Preferably, the or each piezoelectric transducer comprises a three layer structure in which a first (outer) layer comprises a brass electrode plate, a second (middle) layer comprises poly (vinylidene) -trifluoroethylene and a third (outer) layer comprises double-sided circuit board material.

Preferably, a plurality of transducer pockets are provided about the glove for locating transducers therein.

Preferably, the transducers are located in at least the following positions:

(a) centred between the two joint interphalangeal joint creases on each finger;

(b) centred substantially halfway between the distal joint crease and the fingertip on the each finger; and

(c) in the position corresponding to (b) on the thumb.

Alternatively, the transducers are located in at least the following positions: (a) centred between the two joint interphalangeal joint creases on each finger; (b) centred substantially halfway between the distal joint crease and the fingertip on the index and middle fingers; and

(c) in the positions corresponding to both (a) and (b) on the thumb .

Preferably, two or more transducers are arranged in a single transducer pocket .

Preferably, two or more transducers are arranged on different joints .

Preferably, the glove further comprises an insulating layer arranged to electrically isolate a corresponding transducer, in use, from a wearer of the glove. Ideally, said insulating layer comprises printed circuit board lacquer.

It will be appreciated that the scope of the present invention is intended to include a glove for use in grip analysis substantially as described herein with reference to and as illustrated by any appropriate combination of the accompanying drawings .

According to a second aspect of the invention, there is provided a method of analysing grip function comprising the steps of:

(a) providing a glove substantially as described above;

(b) waiting for the pyroelectric effect of the wearer's skin warming the transducers to stabilise; (c) asking the wearer to perform the desired gripping task;

(d) amplifying the output charge from the transducers;

(e) interpreting the amplified output charges to provide the results of the grip analysis .

It will be appreciated that the scope of the present invention is intended to include a method of analysing grip function substantially as described herein with reference to and as illustrated by any appropriate combination cf the accompanying drawings .

Preferred embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

Figure 1 is a schematic representation of the palm side of a glove embodying one aspect of the present invention;

Figure 2 is a schematic representation of the dorsal side of the glove shown in Figure 1 ;

Figure 3 is a side view, partly in cross-section, and drawn to an enlarged scale of the index finger and thumb portions of the glove of Figure 1, showing the transducer pockets;

Figure 4 is cross-sectional view of one finger of the glove and wearer's finger;

Figure 5 is a schematic view of the palm side of an index or middle finger of the glove, showing the position of the transducers;

Figure 6 is a schematic view of the palm side of a ring or little finger of the glove, showing the position of the transducers ;

Figure 7 is a schematic view of the palm side of a thumb of the glove, showing the position of the transducers;

Figure 8 shows the results of a comparison cf grip measured by the NK Digits Grip ("DG"); Figure 9 shows the results of a comparison of grip measured by the DG and Gait Scan ("GS");

Figure 10 shows the results of a comparison of tasks 1 and 2 measured by GS;

Figure 11 illustrates how the tip-pulp of the ring and little fingers would contribute to grip as measured by the DG, but not as measured by the glove owing to the lack of transducer;

Figure 12 shows the relationship between the pulps of each finger (and hence the transducers) to the strain gauge of the DG;

Figure 13 illustrates the forces acting at the transducer- strain gauge interface for the index and middle fingers; and

Figure 14 illustrates the small force contributions of soft tissues overlapping the transducers.

Referring to Figure 1, the glove 1 comprises a standard size 7.5 cotton glove which has been modified to house the transducers. Cotton is used as it is an easily available and easily-worked fabric from which close-fitting glove can be made. Furthermore, use of cotton will reduce any static charge which would be generated by man-made fabrics that could potentially interfere with the charge generated by the transducers .

It will, however, be appreciated that the glove could be fabricated from other materials, for example Lycra 25467 (as manufactured by Spentex BCA Ltd) .

A glove according to the present invention is manufactured as follows. With a volunteer wearing the glove, the joint creases are marked on the palm side. Two small, square pockets 3 of sufficient size to house a transducer snugly are stitched onto each finger region on the palm surface 2 of the glove .

The first of these pockets 3A is located centrally between the two joint interphalangeal joint creases as this is the region subjects use to produce maximum force during the grip activity considered m the study comparing the glove to the NK Digits Grip device ( see below) .

The second pocket 3B is located halfway between the distal joint crease and the fingertip; this provides for both pulp- pinch assessment and assessment of the contribution of the fingertip pulp to the overall grip force.

Two transducer-receiving pockets are also stitched to the palm side of the thumb, in positions corresponding to those sited on the fingers. Alternatively (and as shown in Figure 1) only one pocket 3A is stitched to the palm side of the thumb.

Figure 2 shows the dorsal side of the glove 1 from which it can be seen that Velcro straps 4 ensure a close fit of the transducers on the palm side of the glove to the fingers and thumb of the wearer. The dorsal side of the glove is provided with a zip fastener 5 which facilitates the putting on and taking off of the glove and elasticated straps 6 ensure a close fit of the glove to the wearer's hand, regardless of hand size.

The cables 7 connecting each transducer to a multi-pin connector (not shown) leave the fingers on the ulnar side and leave the thumb on the radial side, as shown in Figure 2. The cables 7 are then taken over the dorsum of the hand and relayed via a charge amplifier unit to a PC.

Although the embodiment described herein utilises the Gait Scan software of Neville 1991, it is envisaged that other software packages could equally be used, for example Matlab.

As shown in Figures 3 and 5 to 7 , the transducers 8 are placed in eight pockets - two on the thumb, index and middle fingers and one in the proximal pocket of each of the ring and little fingers. In an alternative form (shown in Figures 1 and 2), the transducers 8 are placed in the pockets as follows - one in the distal pocket on the thumb and two on each finger.

Figure 4 shows a cross-sectional view of one finger of the glove 1 with the wearer's finger inside. This shows how the cable 7 leaves the transducer 8 on the radial side.

The selection of a suitable transducer needs to take into account the following desired properties.

The transducer must :

1. Be small enough to be used on the finger with minimal interference to hand function.

2. Respond rapidly to changes in load.

3. Stabilise quickly when loaded and unleaded i.e. display minimal hysteresis.

4. Respond accurately within the required load range (approx. 0-200N) .

A piezoelectric [PE] transducer has been developed for use in an in-sole gait assessment system - Gait Scan [GS] (AJ Neville, University of Kent and Canterbury, PhD thesis 1991) . This transducer is 10mm x 10mm x 2.8mm and comprises three layers. It will be appreciated that the exact dimensions of the transducers may change as development continues.

The active piezoelectric portion is a 500 micron thick middle layer of poly (vinylidene) -trifluoroethylene [p (VdF-TrFE) ] . The two outer layers are electrode plates, one of brass and the other of double-sided circuit board.

The transducer, through the properties of p(VdF-TrFE) produces a small charge output when compressed ,the piezoelectric effect) . The charge generated is of the order of pico- Coulombs (pC) and therefore requires amplification before it can be interpreted. The transducer is insulated against external charge by printed circuit board lacquer.

The PE transducer produces a very linear response to loading in excess of 100N.

One negative effect of the PE transducer is the pyroelectric effect. As the transducer changes temperature, it generates charge . Thus a transducer warmed from room temperature to skin contact temperature will generate a charge which may be interpreted as a force reading.

However, the transducer is dynamic i.e. will only provide a charge output during a change in either the force applied or the temperature Therefore once the temperature has stabilised, no charge will be generated which is attributable to the pyroelectric effect.

The Gait Scan software allows for visualisation of the transducer output during the temperature stabilisation period.

The size, accuracy of load bearing and negligible hysteresis make the PE transducer suitable for use in a glove according to the present invention. Furthermore it is advantageous that the dynamic nature of the PE transducer allows the detection of an oscillating force i.e. a glove according to the present invention can detect vibrating force. A Fourier transform of the signal detected when grasping a tuning fork shows peaks at 50Hz (mains supply used to amplify the signal) , 128Hz (tuning fork) and 256Hz (second harmonic of the tuning fork) . The system could be set to detect frequencies of the order of megahertz .

A pilot study was conducted in order to compare the performance of a glove according to the present invention with grip analysis using the known NK Digits Grip (hereafter referred to as "DG") device.

The Pilot Study

Three male volunteers were recruited to the study. All were right handed with size 7.5 hands. Each volunteer carried out a series of tasks. Firstly to obtain a measure of grip strength the volunteer performed three cycles of grip using the DG. (The DG software requests four grips per cycle. After each cycle three of the four grips are selected for averaging - in the case of this study the three highest values were selected) . No time limited was imposed on the subject to carry out the tests .

With a baseline established, the next stage of the investigation was to compare the grip recorded by a glove according to the present invention (hereafter referred to as "the glove^I with the grip recorded by the DG.

The GS software is set for the analysis of gait, therefore compensations had to be made. The software was set to allow a period of twenty seconds during which the subject had to perform four maximum grips per cycle . This corresponded with the four grips per cycle of the DG system.

Once the pyroelectric effect of the skin warming the transducers had passed (usually one minute) the subject, wearing the glove performed a series of five cycles gripping the DG, with an inter-cycle rest period of two minutes. The glove was continually worn throughout this test, (thereby minimising temperature variation in the transducers) . This will be referred to as Task 1.

A second test, Task 2 was performed similar to that of the first to assess the repeatability of transducer location within an individual subject. The same series of five grip cycles were performed but after each cycle the glove was removed for one minute. After one minute the glove was replaced allowing a further minute for the transducers to warm up (avoiding a pyroelectric contribution to the overall charge output during the grip activity.)

One volunteer carried a third test. This was simply to perform the series of five grip cycles wearing the cotton glove with the transducers removed. This will later be referred to as Task 3.

Results

The baseline for each volunteer was determined by performing five grip cycles using the DG. The results are given in Table 1.

Table 1

Volunteer Mean Max Grip in Newtons Standard Deviation

- 478 . 7 10 . 9

2 305 . 2 32 . 8

3 297 . 1 26 . 2

Subject 1 performed Task 3 the results of which are compared to the baseline grip and are illustrated in Figure 8. Table 2

Student t-Test Total Index Middle Ring Li ttle (p -values)

Baseline cf . 0, .66462 0.23831 0.81062 0.9813 0.92275

Glove - No 9 4 5 93 4

Transducers

Baseline cf. 0. .32954 0.33400 0.4509 0.36441

Glove - With 7 5 05 1

Transducers

Transducers cf . 0. .00227 0.42805 0.0065- 0.0112 0.01112

None 3 4 3 4 5

Table 2 gives the p - values for the comparison of the baseline data with that of the data obtained from the two tasks performed by subject 1. Although there is a noticeable difference in the total grip strengths recorded when the DG is gripped without the glove, with the glove but no transducers and with the glove and the transducers the (Figure 8) difference is not statistically different (Table 2) . The same is true of the individual finger contributions. However there is a statistically significant difference between the recordings obtained whilst wearing the glove with and without the transducers, with the exception of the index finger. This would imply a different pattern of contribution by the fingers to the total grip strength when the transducers are incorporated into the glove. Table 3

Total Index Middle Ring Li ttle Grip

Correlation 0.825834 0.795993 0.7972=3 0.5325 -0.2159 coefficients 52

Student t- 0.722241 0.81092 0.7215=3 0.8966 0.08229

Test p- 64 9 values

Each subject performed Task 1 and the ~ean grip strength measured by the DG and GS systems were calculated and are represented in Figure 9. Table 3 gives the correlation coefficients (Cc) and student t-Test p-values of the comparison of both data sets .

The appearance of Figure 9 indicates similarity between the data derived from the DG and that from the GS systems during Task 1. The Cc for total grip and the contributions made by the index and middle finger support this observation, but a lower Cc value for the ring finger and a negative value for the little finger indicate a difference in the forces measured. When considering the statistical significance of the differences between the measurements made by the DG compared to the GS, none is apparent 0.06 < p > 0.96 (Student t-Test) .

Task 2

Task 2 was devised to assess the repeatability of grip measurement using the GS system when removing and replacing the glove .

Figure 10 illustrates the difference between the grips measured by the GS system during Tasks 1 and 2. The subjects grasped the DG handle as a proformer. Table 5 lists the comparative values of the data using the Student t-Test.

Table 5 : Task 1 cf . Task 2

Task Mean (n=3) sd increase T-Test

Total 1 257.0221 61.26845

Total 2 291.7567 63.19892 113.51 0.0384

Index 1 89.40725 25.21805

Index 2 99.48338 26.83314 111.27 0.133

Middle 1 72.80234 28.3681

Middle 2 96.7088 33.0789 132.84 0.002

Ring 1 65.80316 12.52852

Ring 2 65.00872 12.30673 98.79 0.841

Little 1 29.0093 9.456607

Little 2 30.5558 9.926717 105.33 0.706

The total grip strength measured during Task 2 (repeated donning of the glove) is greater than Task 1, 291.8N +\-61.3 257. ON +\- 63.2 a 113% increase in grip strength. (p=0.038). However individual finger contributions are more difficult to interpret .

The index and little fingers show a statistically significant increase in the force measured during Task 2. The mean ring finger measurements indicate a small reduction in the measured force but this is not of statistical significance. When the middle finger is analysed a statistically significant increase in force measurements is seen, from 72.8N +\- 28.4 during Task 1 to 96.7N +\- 33.1 during Task 2 (p=0.002) . So far the data presented has concerned the contribution of individual fingers during tasks designed to compare the glove of the present invention (which uses the Gait Scan (GS) system) to the DG. The glove of the present invention uses eight PE transducers situated on the palmar surface of the fingers and thumb. The tasks carried out did not produce readings from the thumb transducers as they did not come into contact with the grasped proformer. The handle of the grasped DG rested firmly in the palm and against the base of the thumb, against the thenar muscles.

The index and middle fingers had two transducers situated on the palmar surface over the tip and interphalangeal (IP) pulps . Table 6 gives the percentage values the two transducers made to the total force, measured by each of the two fingers.

Table 6 : The contribution of each transducer - represented as percentage of the total force measured by each finger.

Finger Tip -pulp Transducer IP -pulp Transducer

Index 22.59 77.41

(sd +\- 5.42) sd +\- 5.42)

Middle 23.50 76.50

(sd +\- 7.11) sd +\- 7.11)

The results of the various tasks utilising various combinations of the glove and the DG are discussed below.

Task 1: Gripping the DG with the Glove worn continually

Comparison of the data recorded by the DG and the glove (GS system) indicated good correlation between the two systems (Cc

0.83 for total grip strength with no statistical significance p=0.73). Similarly high correlation coefficients are calculated for the index and middle fingers with no statistical significance apparent (Cc = 0.E0 and 0.80, p=0.81 and 0.72 respectively). However, although there is no statistical significant difference between the results for the ring finger (p=0.9), the correlation coefficient (0.53) is less than for the index and middle fingers. Furthermore when the data for the little finger are compared, a negative correlation is seen and the p-value is only 3.08.

To explain the probable reason for the discrepancy in the results, particularly in the ring and little fingers, we must consider the arrangement of the transducers .

The grip data recorded for the index and middle fingers was the sum of the force measured by two transducers; for the ring and little fingers only one transducer recorded data. Figure 11 illustrates how the tip-pulp of the ring and little fingers would contribute to the grip measured by the DG but not to that measured by the glove due to the lack of transducer. Furthermore the ergonomics of the DG handle, essentially a cylindrical rod with all strain gauges set at constant distance from the palm causes the pulps cf each finger and therefore transducer to have a different relationship to the strain gauge (Figure 12) .

When the forces measured by the transducers and the DG differences are apparent. Figure 13 illustrates the forces acting at the transducer-strain gauge interface for the index and middle fingers .

The force measured by the transducers are Ftp and Fip representing the tip-pulp (TP) and interphalangeal pulp (IP) transducers. The DG strain gauge force (Fdg) is given as a reaction against the force applied by the finger. Assuming the transducer is always tangential to the strain gauges convex surface the forces Fip and Ftp will be measured directly by the transducer. However the strain gauge can only measure the vertical components of the two applied forces, Ftpv and Fipv respectively.

Therefore : Ftpv + Fipv = Fdg

But: Ftpv < Ftp

Fipv < Fip

So one might expect that the force measured by the DG strain gauge would be less than the sum of the forces measured by the transducers .

Statistically there is no difference between the two groups but data represented in Figure 9 suggests the mean values obtained by the transducers are marginally less than those obtained from the DG. One possible explanation for this discrepancy is the size of the transducer. For the ring and little fingers there was only one transducer at the interphalangeal pulp with none at the tip-pulp. The mean values measured for these fingers were less than those measured by the DG system with smaller correlation coefficients and p-values, particularly in the little finger, approaching statistical significance.

In Figure 11 the point is made that the tip-pulp still contributes to the total grip generated - and is measured by the DG but not the glove. It would be safe to assume that in the index and middle fingers some contribution is made by parts of the tip and interphalangeal pulps that are not covered by the transducers of the glove. The pulps of the finger adapt to the surface contours they are grasping so some soft tissue overlap the edges of the transducer during maximal grip may occur contributing to force transmission, forces that would not be measured by the transducer (Figure 14) .

Task 2 : Constant wearing versus re-donning of the glove between test cycles When each subject carried out task 2 it was apparent from the data that although graphical representation of the data suggested an increase in the mean for total grip (Figure 10) the difference was not significant (p=0.04). However when analysis of individual fingers was studied the middle finger data for re-donning of the glove was significantly greater (p=0.002). The significance of this is not clear but may be clarified with further study of an increased population.

Task 3: Gripping of the DG without the glove, with the glove but no transducers and with the glove with the transducers

Once more, graphical representation of the results of this study (only one subject) suggests grip strength was reduced with the consecutive addition of the cotton glove and then the glove with the transducers. However there was no statistical significance between the date when compared to the baseline data. Significance did exist between the data obtained from gripping the DG with glove alone and when the transducers were added to the glove - with the exception of the index finger.

The explanation of this reference can be made to Bechtol ' s paper of 1954 (Journal of Bone and Joint Surgery Vol. 36A: 820-24) in which he refers to the position of the hand during grip. The paper reports grip strength variation relative to the position of the handle . In effect by introducing the transducers as an interface between the finger and the DG we have increased the handle position albeit only by 2.3mm. This would account for the grip strength changes observed.

From the above discussion, it can be seen that the glove has been validated against the DG - there was no statistical significance between the data derived from the DG and that derived from the glove of the present invention using the GS system.

Further developments include the possibility of two transducers being placed on the ring and little fingers in the same orientation as those on the index and riddle fingers (cf . Figures 5-7) . This would increase force detection and probably improve the correlation coefficient of total grip strength when compared to the DG. The contribution to the total finger grip of the tip-pulp transducer during maximum grip was 22.6% (sd +\- 5.4) and 23.5% (sd +\- 7.1) for the index and middle fingers respectively during task 1.

It is envisaged that the glove can be validated against other instrumentation (for example a hydraulic pinch meter) to evaluate other tasks such as pinch and key grip, for which the thumb tip-pulp is essential. The DG cannot perform this type of analysis. Such analysis may be performed with only one transducer located on the thumb of the glove - this will be located in the tip-pulp position.

Using PE transducers which can detect force in two directions (or even a tri-axial PE transducer) , the glove can be developed to assess shear force and torque - for example as used in the unscrewing of a jam jar lid. In this way, the glove of the present invention provides a convenient way of assessing numerous tasks of different types in a single session; tasks which conventionally would need to be assessed by a number of different pieces of apparatus.

The glove according to the present invention offers a number of advantages over the known prior art . For example :

1. The contribution of the thumb to different types of grip activity can be assessed.

2. The different types of grip activity which previously needed to be assessed with different pieces of equipment can now be assessed with the glove alone .

3. The glove allows scope for a wide variety of grip analyses which were not previously achievable, for example in the field of sports grips (tennis racquets, golf clubs etc.) or steering devices (car steering wheels, motorcycle handlebars, aircraft controls etc.).

4. Since the glove can cope with rapid changes in grip force, the glove can be used to detect and analyse vibration and thus improve ergonomic understanding and therefore design of, for example, tool handles, steering wheels etc.

5. Since the glove can have more than one transducer on each digit, data can be provided in relation to each segment of each digit, rather than providing a single reading for each (whole) digit, as with the DG.

Claims

1. Apparatus for use in grip analysis comprising a glove having one or more grip transducers disposed thereon, said transducer (s) being actuable by a wearer of said glove to provide an indication of the degree of actuation.

2. Apparatus as claimed in claim 1 wherein the or each transducer is a piezoelectric transducer.

3. Apparatus as claimed in claim 2 wherein the or each piezoelectric transducer comprises a three layer structure in which a first (outer) layer comprises a brass electrode plate, a second (middle) layer comprises poly (vinylidene) - trifluoroethylene and a third (outer) layer comprises double- sided circuit board material .

4. Apparatus as claimed in any of the preceding claims wherein a plurality of transducer pockets are provided about the glove for locating transducers therein.

5. Apparatus as claimed in any of the preceding claims wherein the transducers are located in at least the following positions : (a) centred between the two joint interphalangeal joint creases on each finger;

(c) in the position corresponding to (b) on the thumb.

6. Apparatus as claimed in any of the preceding claims wherein the transducers are located in at least the following positions :

(a) centred between the two joint interphalangeal joint creases on each finger;

(b) centred substantially halfway between the distal joint crease and the fingertip on the index and middle fingers; and

(c) in the positions corresponding to both (a) and (b) on the thumb.

7. Apparatus as claimed in any of the preceding claims wherein two or more transducers are arranged in a single transducer pocket .

8. Apparatus as claimed in any of the preceding claims wherein two or more transducers are arranged on different joints.

9. Apparatus as claimed in any of the preceding claims wherein the glove further comprises an insulating layer arranged to electrically isolate a corresponding transducer, in use, from a wearer of the glove .

10. Apparatus as claimed in claim 9 wherein said insulating layer comprises printed circuit board lacquer.

11. Apparatus for use in grip analysis comprising a glove substantially as described herein with reference to and as illustrated by any appropriate combination of the accompanying drawings .

12. A method of analysing grip function comprising the steps of:

(a) providing a glove substantially as described in any of the preceding claims; (b) waiting for the pyroelectric effect of the wearer's skin warming the transducers to stabilise;

(c) asking the wearer to perform the desired gripping task;

(d) amplifying the output charge from the transducers; (e) interpreting the amplified output charges to provide the results cf the grip analysis.

13. A method of analysing grip function substantially as described herein with reference to and as illustrated by any appropriate combination of the accompanying drawings .