SCREENING TEST FOR THE DETECTION OF DEEP VEIN THROMBOSIS

Description

FIELD OF THE INVENTION

[0001] The present invention relates to systems and apparatus for assisting healthcare workers in the screening and detection of abnormalities and pathological states in soft tissue, and has particular application at least, but not confined to, the screening and detection of deep vein thrombosis (DVT) of the lower limbs.BACKGROUND OF THE INVENTION

[0002] Deep vein thrombosis (DVT) is a significant complication in all surgical and medical wards, as well as in other aspects of community life. It is a condition that has significant implications for pulmonary embolism, and possible death. There are also dangers of continuing morbidity in the legs, and also in the lungs, from the presence of venous thromboembolism.

[0003] Currently, there is no recognised clinical assessment for DVT that has an accuracy greater than 60 percent, and many patients with DVT have no overt clinical findings.

[0004] In cases of suspected DVT, it is necessary to undertake specific investigations for confirmation. These investigations are expensive, and include techniques such as Doppler ultrasonography which requires significant equipment and expertise. Tests for pulmonary embolism may also be necessary, involving methods such as computed tomography (CT) or chest scanning using a radioactive marker. Blood tests may also be performed. As will be appreciated, a number of these tests are invasive and / or uncomfortable for the patient. Many of these tests cannot be performed at the bedside.

[0005] A particular concern is that many patients develop DVT with minimal or no significant changes or symptoms (e.g. pain and swelling), and as a result many patients are sent home early following surgery without any assessment of the possibility of DVT. Deaths from complications of venous thromboembolism have been known to result in such cases.

[0006] There is, accordingly, a pressing need for a reproducible, objective investigation that can be performed by a trained technician, a trained clinician, or other healthcare worker. Ideally, assessment of the possible presence of DVT should be able to be conducted simply, with high reliability, at relatively low cost, and without the requirement for invasive or uncomfortable procedures. The present invention seeks to address these requirements.

[0007] US 2006 / 0085047 A1 discloses a method for automatically controlling single channel neuromuscular electrical stimulation of the plantar muscle, in response to the sensing of motion of the foot or leg: to reduce accommodation of the stimulated plantar muscle and attendant reduction of contractions, which when undiminished increase blood flow for the prevention of DVT and pulmonary embolism; to turn off the stimulation during walking or running to prevent slips or falls; and to reduce power consumption of the unit that provides the stimulation.

[0008] US 2009 / 0005834 A1 discloses a device adapted for mitigating the formation of a blood clot. The device comprises: a sensor adapted to identify muscle movement associated with blood circulation in a vein and to produce a signal indicating the movement; and a controller adapted to monitor the sensor signal and to activate an actuator upon the controller determining that less than a defined number of movements have been sensed over a defined period of time. The actuator is adapted to produce a stimulating action of a given set of parameters urging the user to move his / her feet if, for example, the predefined number of dorsiflexions is not met.

[0009] US 2008 / 103397 A1 is a further document important for understanding the background of the present invention. It discloses a DVT diagnosis system in the form of a wearable leg pressure sleeve. It discloses applying mechanical stimulate and comparing the detected pressure signal accordingly. It fails at least to disclose analysing the detected signal to the degree as claimed.

[0010] For convenience of description, where the term 'muscle' is employed in this specification (including in the appended claims), this should be understood to encompass a single muscle, or one or more muscles comprising a muscle group.

[0011] Embodiments of the invention take advantage of the observation that, upon clinical examination, in certain pathological states there is a change in the calf muscles that may be identified in a response to a percussive stimulus applied to the tissues. In DVT, it has been observed that the normal mobility of the calf is reduced. It has thus been observed that tapping the calf, which under normal circumstances has considerable mobility like the bounce of jelly, results in a more 'dough-like' response in the presence of DVT. It has been found, however, that skill and experience are required in order to master the use of this technique for the detection of DVT. It is reasonably anticipated that similar effects will also occur in comparable pathological states of other soft tissues.

[0012] Embodiments of the invention further take advantage of the fact that disorders such as DVT are normally present in only one comparable muscle of the subject. Thus, for example, if a DVT is present in the right calf muscle, the response of this muscle to a mechanical stimulus should be different from the response of the subject's left calf muscle, in which DVT is unlikely to be present.

[0013] Embodiments of the present invention therefore advantageously employ a consistent, reproducible, objective analysis technique, comprising determining and comparing first and second parameter sets of corresponding pairs of muscles of the subject, in order to detect significant differences indicative of the possible presence of DVT or other abnormality.

[0014] Accordingly, in embodiments of the invention the first and second muscles are corresponding muscles of respective left and right limbs of the subject. The first and second muscles are left and right calf muscles of the subject.

[0015] Embodiments of the invention decompose the oscillatory mechanical responses of the first and second muscles, such that the first and second parameter sets each comprise at least one frequency parameter and at least one corresponding damping parameter characterising the oscillatory mechanical response. Typically, the first and second parameter sets each comprise frequency parameters and corresponding damping parameters of two or more oscillatory components of the mechanical responses of the first and second muscles of the subject.

[0016] According to embodiments of the invention, a possible presence of DVT in the subject is assessed based on a difference between the frequency parameters and / or the damping parameters of the first and second parameter sets. For example, a lower frequency response and / or more rapid damping in one of the first and second muscles, as compared with the other, is indicative of the possible presence of DVT in the subject. The frequency response may typically be a resonant frequency, while damping may be measured as a damping factor.

[0017] The apparatus comprises a signal processor. For the purpose of comparing the first parameter set with the second parameter set, the signal processor may be configured to present information relating to the first and second parameter sets on a display, for evaluation and further diagnostic consideration, by an operator such as a trained technician, trained clinician, or other healthcare worker.

[0018] In other embodiments, further statistical analysis of the first and second parameter sets, obtained over multiple responses of the first and second muscles of the subject to multiple mechanical stimuli, may be performed in order to assess the statistical and / or clinical significance of differences between the first and second parameter sets, and to provide a specific indication, such as a likelihood or probability of the presence of an abnormality such as DVT in the subject.

[0019] According to embodiments of the invention, the system may comprise a motion sensor that is adapted to be fixed, in use, to a muscle of the subject. The motion sensor is configured to transmit a signal representing motion of the sensor. The motion sensor may comprise an accelerometer. Advantageously, the accelerometer comprises a multi-axis accelerometer, such as a two-axis or three-axis accelerometer.

[0020] The apparatus comprises a receiver. The receiver is configured to receive a signal from a transmitter associated with the motion sensor. In some embodiments, a communications channel between the transmitter and the receiver comprises a wired connection.

[0021] In other embodiments, the communications channel between the transmitter and the receiver comprises a wireless connection. Advantageously, the use of a wireless connection between the motion sensor and the signal processor avoids the possibility that the presence of physical connections may influence the mechanical response of the first and second muscles resulting from the applied mechanical stimuli.

[0022] In various embodiments, the signal communicated from the transmitter to the receiver may be an analog signal, which may be sampled and digitised at the receiver, or may be a digital signal, having been sampled and digitised at the motion sensor.

[0023] Further features, benefits and advantages of embodiments of the invention will be apparent from the following description, which is provided by way of example only, and should not be considered as limiting of the scope of the invention as defined in the claims appended hereto.BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Embodiments of the invention will now be described with reference to the accompanying drawings, in which like reference numerals indicate like features, and wherein: Figure 1 is a schematic diagram of a system embodying the invention; Figure 2 is a schematic diagram of a motion sensor suitable for use in the system of Figure 1; Figure 3 is a diagram illustrating position and support of a leg of a subject arranged for testing by the system of Figure 1; Figure 4 is a graph of a motion sensor data set representing an exemplary mechanical response obtained according to an embodiment of the invention; Figure 5 is a graph illustrating decomposition of a data set into three decaying sinusoidal components according to an embodiment of the invention; Figure 6 is a chart illustrating an illustrative statistical analysis of frequency parameters obtained in accordance with an embodiment of the invention; and Figure 7 is a chart illustrating an illustrative statistical analysis of damping parameters obtained in accordance with an embodiment of the invention. DETAILED DESCRIPTION OF EMBODIMENTS

[0025] Figure 1 is a schematic diagram of a system 100 embodying the invention. The system 100 captures data sets obtained by applying mechanical stimuli to muscles of a subject. Shown schematically in Figure 1 is a cross-section through the lower leg 102 of a subject, comprising bones 104 (i.e. fibula and tibia), muscles 106 (i.e. peroneals, tibialis posterior, gastrocnemius, and / or soleus muscles). Collectively, the various muscles making up the back portion of the lower leg are called 'the calf'.

[0026] In accordance with the exemplary system 100, one or more motion sensors 108 are fixed to the exterior of the limb, e.g. at the rear and / or side of the calf. In some configurations, a single motion sensor, such as a multi-axis accelerometer, may be sufficient. It may be advantageous to provide multiple motion sensors to enable data from each to be combined and / or selection of signals from a sensor providing the strongest or cleanest signal.

[0027] The system 100 further includes signal processing apparatus 110, which is configured to receive and analyse information captured by the motion sensor, or sensors, 108. As shown in Figure 1, the signal processor 110 comprises a combination of hardware and software, in which the hardware and / or the software are configured to embody various features of the invention. The signal processing apparatus 110 may comprise a standard hardware configuration, such as a personal computer, a smartphone or other portable device, or any other suitable computing and communications platform. Alternatively, the signal processing apparatus 110 may comprise custom or semi-custom hardware, including programmable components such as microprocessors, or programmable-logic devices such as a field programmable gate array (FPGA). Where programmable-logic devices are used, these may be configured to include functional blocks implementing features of discrete devices, such as microprocessors, memory devices, and custom analog and / or digital components. As will be appreciated by electronics system designers and engineers, various hardware and / or software implementation options are available, falling within the scope of the present invention.

[0028] The exemplary signal processor 110, as illustrated in Figure 1, comprises a microprocessor component or functional block 112. The microprocessor 112 is interfaced to, or otherwise operably associated with, a non-volatile memory / storage device 114. The non-volatile storage 114 may be a hard disk drive (e.g. if the signal processor 110 is implemented using a personal computer), or may include a solid-state non-volatile memory, such as read-only memory (ROM), flash memory, or the like. The microprocessor 112 is also interfaced to volatile storage 116, such as random access memory (RAM) which contains program instructions and / or transient data relating to the operation of the signal processor 110.

[0029] In a conventional configuration, the non-volatile storage device 114 maintains program and data content relevant to the normal operation of the signal processing apparatus 110. For example, if the apparatus 110 is implemented using a personal computer, smartphone, or the like, the storage device 114 may contain operating system programs and data, as well as other executable application software necessary to the intended functions of the signal processing apparatus 110. The storage device 114 may also contain program instructions which, when executed by the microprocessor 112, instruct the apparatus 110 to perform operations in accordance with an exemplary configuration, for assessing a subject at risk of DVT. In operation, instructions and data held in non-volatile storage 114 may be transferred to volatile memory 116 as required.

[0030] The microprocessor 112 is also operably associated with a communications interface 118 in a conventional manner. The communications interface 118 enables communication between the signal processor 110 and the one or more motion sensors 108. Communications between the signal processing apparatus 110 and each motion sensor 108 may be via a wired connection, such as a Universal Serial Bus (USB) connection, or may be via a wireless connection, such as a Bluetooth, Bluetooth Low Energy (BLE), Wi-Fi, or other wireless communications channel. In the exemplary embodiment of the system 100, communication is conducted via a wireless BLE channel.

[0031] In use, the volatile storage 116 includes a corresponding body 120 of program instructions configured to perform processing and operations embodying features of the present invention, as described in greater detail below, particularly with reference to Figures 4 to 7.

[0032] The signal processing apparatus 110 further includes a display 122 interfaced with the microprocessor 112, enabling information to be communicated to an operator of the apparatus 110. As shown in Figure 1, the display 122 is integrated with the apparatus 110, and may be, for example, a touchscreen display of a smartphone or similar device. Alternatively, the display 122 may be incorporated into a custom- or semi-custom-designed signal processing apparatus 110. In still other configurations, the display 122 may be the display of a personal computer, a tablet device, a notebook, or other similar conventional computing platform.

[0033] Figure 2 is a schematic diagram of a motion sensor suitable for use in the system of Figure 1. The exemplary motion sensor 108 comprises an accelerometer 202, a local processor 204, and a communications interface 206.

[0034] Various commercially available components may be used in the design and implementation of the motion sensor 108. For example, the motion sensor 202 may be the part number KXTJ9 Tri-axis Accelerometer, available from Kionix Incorporated of Ithaca, New York. The local processor 204 may comprise part number CC2541, Bluetooth Low Energy and Proprietary System-on-Chip, available from Texas Instruments Incorporated of Dallas Texas. The CC2541 part comprises all of the components necessary to communicate with the accelerometer 202, and to implement a wireless BLE communications interface for transmitting accelerometer data to the signal processor 110. The CC2541 part requires only an external antenna 206 for transmitting and receiving BLE signals at 2.4 GHz.

[0035] A more complete off-the-shelf motion sensor 108 may be obtained from corporations such as Texas Instruments, in the form of components such as the CC2541 SensorTag reference design, which is a 'development kit' comprising the CC2541 System-on-Chip, the KXTJ9 Accelerometer, along with additional sensors for detecting and recording movement, magnetic fields, humidity, pressure and temperature. The SensorTag reference design is configured for communications with smartphones and other mobile devices executing either the android or iOS operating systems. It is supported by software development tools, and the ability to download software / firmware updates directly from a smartphone or other device. Accordingly, although the SensorTag reference design incorporates sensors that may not be required in a basic implementation of the invention, it provides a convenient development platform for motion sensors 108 embodying the invention.

[0036] Figure 3 is a schematic diagram illustrating exemplary position and support of a leg 300 of a subject, arranged for testing using the system 100 of Figure 1. Supports 302, 304 are provided for the upper leg and foot, respectively, and one or more motion sensors 108 are fixed, e.g. via a strap, tape or other suitable fixing mechanism, to the lower leg 102. The arrangement of supports 302, 304 is provided by way of example only, and other arrangements are possible, e.g. the provision of upper leg support 302 may be optional. In this position, the subject may relax his or her calf muscle, such that a trained technician, clinician, or other healthcare worker may apply a mechanical stimulus, such as a percussive impulse (represented by arrow 124 in Figure 1) resulting in an oscillatory mechanical response of the subject's calf muscles. The oscillatory mechanical response causes corresponding voltages to be generated along the tri-axis detectors within the accelerometer 202. In the case of the KXTJ9 accelerometer part, the voltages are digitised on-chip, at a selectable resolution of 8, 12 or 14 bits, and transferred to the local processor 204 via an inter-integrated circuit (I 2< C) standard communications interface. The local processor 204 is programmed to collect, format and transmit the data set corresponding with the digitally sampled accelerometer readings via the transmitter 206 to the receiver 118 of the signal processor 110. In an exemplary configuration, the BLE communications channel supports a sampling rate of about 800 samples per second, using a 12-bit resolution. Increasing the bandwidth of the communications channel, and / or improving the efficiency of transmission (e.g. by compressing the transmitted data) may enable higher resolution data acquisition, such as one kilosamples / s at up to 14 bits per sample.

[0037] As noted above, in the exemplary configuration a tri-axis accelerometer is employed. This provides motion information, in the form of voltage levels representing acceleration, along three orthogonal axes. Exemplary configurations may utilise acquired data from a single axis (e.g. the axis providing the strongest signal), or may combine data from multiple axes by computing a resultant magnitude of the acceleration vector. Computation of the magnitude of the acceleration vector may be implemented by software executing on the local processor 204, or on the signal processing apparatus 110. Advantageously, computing the resultant acceleration vector magnitude at the local processor 204 reduces the volume of information that must be transmitted via the BLE communications link from the motion sensor 108 to the signal processor 110.

[0038] Figure 4 shows a graph 400 of a motion sensor data set representing an exemplary oscillatory mechanical response obtained using the arrangement and apparatus described above with reference to Figures 1 to 3. Time is represented on the horizontal axis 402, while the accelerometer voltage level corresponding with the resultant acceleration magnitude is represented on the vertical axis 404. As can be seen from the resulting acquired data trace 406, a mechanical percussive impulse is applied after approximately 0.17 seconds, resulting in a generally oscillatory motion of the subject's calf, apparent in the portion 408 of the trace 406. This motion substantially settles over a period of approximately 0.8 seconds. A low level of additive noise is also visible, particularly on the steady state portions of the trace 406. According to the invention, the goal of analysis of the oscillatory mechanical response of the subject's calf can therefore be defined as extracting a relevant parameter set, useful for assessing the possible presence of DVT, from the acquired data, in the presence of additive noise.

[0039] According to the invention, a starting assumption for the analysis, which may be performed using suitably configured software, firmware and / or hardware implemented on the signal processing apparatus 110, is that the acquired data set 406 may be represented as a superposition of two or more exponentially decaying sinusoidal waveforms. On this assumption, a waveform such as the trace 406 may generally be represented in the following form: V t = ∑ i = 1 N A i e − γ i t cos ω i t + ϕ i + n t

[0040] In the above equation, N represents the number of exponentially decaying sinusoidal components to be used in fitting the acquired data set 406. Experiments conducted by the inventors have indicated that N=3 is generally a suitable choice. Each component also has an amplitude A, a characteristic frequency ω, an associated phase ø, and a damping coefficient γ. The function n(t) represents the additive noise. It should also be noted that, for the purposes of analysis, the signal is a discrete time sequence of samples of the motion sensor voltage.

[0041] Various techniques are available for analysing signals such as that shown in the graph 400 of Figure 4. For example, Fourier analysis could be employed, based upon the Fast Fourier transform (FFT). However, the FFT may suffer from limited frequency resolution, spectral leakage, picket fence effect, and scalloping loss. While these may be mitigated through choice of proper apodisation, averaging, zero padding, up-sampling, and other well-known approaches, the use of FFT-based methods for analysis of short, aperiodic, signals is likely to be suboptimal.

[0042] An alternative technique would be to use a nonlinear regression technique, such as nonlinear least-squares fitting, in order to estimate the unknown parameters of the above equation. Such general techniques may be unstable or unreliable in the presence of additive noise, and may perform poorly when fitting large numbers of unknown parameters. Accordingly, it may be preferable to employ more-targeted techniques for fitting or estimating the parameters of a sum of exponentially decaying sinusoids in the presence of additive background noise. One such targeted technique is Prony's method, however it has been shown that better performance can be obtained using a Matrix Pencil Method (MPM) as described in Hua and Sarkar, 'Matrix Pencil Method for estimating parameters of exponentially damped / undamped sinusoids in noise', IEEE Transactions on Acoustics, Speech and Signal Processing, Volume 38, No. 5, May 1990.

[0043] Using the MPM technique, a generally damped oscillatory response, such as the trace 406 shown in the graph 400, is represented as a superposition of a plurality of exponentially-damped sinusoids, which may be in accordance with the above equation, as illustrated by the graph 500 shown in Figure 5. Again, the horizontal axis 502 represents time, while the vertical axis 504 represents the equivalent acceleration amplitude in volts, relative to the steady state reading. The acquired data set has been decomposed into three oscillatory components, comprising a dominant component 506, and two smaller components, 508, 510. By design, the MPM technique minimises error due to noise, and thus any further exponentially decaying sinusoidal components would necessarily be of smaller magnitude relative to the three components 506, 508, 510 shown in the graph 500. It is, accordingly, apparent that the inclusion of additional components in the analysis would be of minimal benefit. Indeed, increasing the number of parameters to be estimated may impact upon the stability and reliability of the analysis.

[0044] A key insight of the present inventors is that the mechanical response of a muscle is altered in the presence of DVT. While there is considerable variation between muscles of different subjects, the expected variation between corresponding muscles of pairs of limbs of a single subject is expected to be considerably smaller. Therefore, the possible presence of DVT in one muscle of a subject, such as the right calf muscle, is assessed by comparison with the corresponding muscle of the other limb, e.g. the left calf muscle. As described below, with reference to Figures 6 and 7, a suitable comparison may be made using parameter sets derived from analysis, such as described above, of the mechanical response of each muscle.

[0045] According to embodiments of the present invention a parameter set is derived from each data set using MPM analysis. The characteristic frequencies and damping coefficients (or, equivalently, damping factors) of the principal exponentially decaying sinusoidal component have been found to comprise useful parameter sets for the purpose of comparison. Highly statistically significant differences in these parameters have been identified in subjects experiencing DVT in one calf muscle.

[0046] By way of example, Figure 6 is a chart 600 illustrating a preliminary statistical analysis of characteristic frequency parameters obtained in accordance with an exemplary configuration. The horizontal axis 602 of the chart shows the characteristic frequency (in radians per second), while the vertical axis 604 represents the number of events (out of 10 repeated tests for the left leg, and nine for the right), falling within each identified characteristic frequency band. The result of this analysis is a histogram 606 for the left leg of the subject, to which a corresponding normal distribution curve 608 has been fitted, and a further histogram 610 for the right leg, to which a corresponding normal distribution curve 612 has been fitted. In this particular case, the subject is known, from other tests, to have DVT in the right calf muscle, but not in the left calf muscle. The reduction in the characteristic frequency resulting from the presence of the DVT is statistically significant, having a p-value of 0.01 (i.e. less than 0.05) when analysed using the statistical t-test.

[0047] Figure 7 shows a similar chart 700 resulting from a preliminary statistical analysis of damping factors. In this case, the horizontal axis 702 represents damping factor, while the vertical axis 704 once again shows the number of events within each identified damping factor range obtained from a sample of 10 repeated mechanical stimulus impacts. A histogram 706 representing damping factors for the left calf muscle has been fitted with a corresponding normal distribution curve 708, while a histogram 710 corresponding with damping factors for the right calf muscle has been fitted with a normal distribution curve 712. There is, once again, a clear difference between the left and right calf muscle response, and t-test analysis has established that the difference is statistically significant, having a p-value of 0.02. It is clear that a higher damping factor is observed in the muscle having DVT.

[0048] In accordance with exemplary configurations, the above analysis may be performed by the signal processor 110, and graphs such as the response 400 shown in Figure 4, and results of statistical analyses such as the charts 600, 700 shown in Figures 6 or 7, rendered on the display 122. An operator, i.e. a trained technician, clinician or other healthcare worker, will thus be able to assess the possible presence of DVT in the subject of the test by visual inspection of the displayed results.

[0049] According to some configurations, the statistical analysis may be used in further comparisons performed by the signal processor 110 in order to offer the operator a preliminary assessment of likelihood of the presence of DVT. For example, based upon the analysis of aggregate data using computational and statistical methods including, but not confined to, logistic regression, the results may be used to assess the differences in characteristic frequency and / or damping factor between a series of tests performed on left and right calf muscles of the subject. If the likelihood or probability of the differences is above a clinically and empirically determined threshold (or any other desired criterion, depending upon requirements) then the signal processor 110 may highlight to the operator the likelihood or probability that DVT is present, and identify the relevant muscle.

[0050] Advantageously, therefore, exemplary configurations provide a useful diagnostic tool that may be used by healthcare workers to identify patients having a high likelihood of DVT. Such patients may then be referred for further testing and diagnosis. Used as a preliminary test, exemplary configurations may be able to eliminate the likelihood that DVT is present in individual subjects, thus avoiding the need for unnecessary further testing, which may be costly, invasive and inconvenient.

[0051] While particular embodiments and variations of the invention have been described herein, further modifications and alternatives will be apparent to persons skilled in the relevant art. For example, while the system described with reference to Figure 1 and 2 employs a motion sensor 108 which performs data acquisition and digital transmission, along with a signal processing apparatus 110 that receives the digital signals and performs further processing and analysis, other arrangements are also possible. For example, additional analysis and processing could be performed on-board the motion sensor unit 108. Furthermore, a wired, rather than a wireless, connection may be provided between the motion sensor unit 108 and the signal processing apparatus 110. Furthermore, analog signals may be obtained by the motion sensor, e.g. voltage representing acceleration or other motion parameters, and transmitted in analog form to the signal processing apparatus 110, which would receive and digitise the analog signals. Various combinations and arrangements of the above alternatives are also possible, and fall within the scope of the present invention.

[0052] In other variations, alternative forms of analysis may be employed to determine parameter sets characterising muscle response to mechanical stimuli, such as modal analysis.

[0053] Accordingly, the particular embodiments described in detail should be understood as provided by way of example, for the purpose of teaching the general features of the invention, but should be understood as not limiting of the scope of the invention, which is defined in the following claims.

Claims

1. An apparatus (110) for assessing a subject at risk of a soft tissue abnormality such as deep vein thrombosis, DVT, comprising: a receiver (118), configured to receive a signal from a transmitter (206) associated with a motion sensor (108), the motion sensor being adapted to be fixed, in use, to a muscle of the subject, and configured to transmit a signal representing motion of the sensor; and a signal processor (112), configured to analyse data received via the receiver from the motion sensor, characterised in that: the signal processor is for receiving first and second data sets via the receiver from the motion sensor, the first data set captured by the motion sensor when fixed to a calf muscle of a first leg of the subject in response to a first percussive mechanical impulse applied to the calf muscle of the first leg while in a relaxed state, and the second data set captured by the motion sensor when fixed to a calf muscle of a second leg of the subject in response to a second percussive mechanical impulse applied to the calf muscle of the second leg while in a relaxed state; and the signal processor is configured to perform an analysis on a basis that the first data set is represented as a first superposition of two or more exponentially decaying sinusoidal waveforms and the second data set is represented as a second superposition of two or more exponentially decaying sinusoidal waveforms, wherein the analysis comprises: determining a first parameter set (608, 708) characterising the first superposition of two or more exponentially decaying sinusoidal waveforms, determining a second parameter set (612, 712) characterising the second superposition of two or more exponentially decaying sinusoidal waveforms, and comparing the first parameter set with the second parameter set to identify differences therebetween, indicative of a possible presence of DVT in the subject.

2. The apparatus of claim 1 wherein the analysis comprises decomposing the first superposition of two or more exponentially decaying sinusoidal waveforms and the second superposition of two or more exponentially decaying sinusoidal waveforms, such that the first and second parameter sets each comprise at least one frequency parameter and at least one corresponding damping parameter characterising the corresponding superposition of two or more exponentially decaying sinusoidal waveforms.

3. The apparatus of claim 2 wherein the first parameter set comprises frequency parameters and corresponding damping parameters of two or more oscillatory components of the first superposition of two or more exponentially decaying sinusoidal waveforms and the second parameter set comprises frequency parameters and corresponding damping parameters of two or more oscillatory components of the second superposition of two or more exponentially decaying sinusoidal waveforms.

4. The apparatus of claim 3 wherein the first and second data sets are each assumed to represent a signal having the form: V t = ∑ i = 1 N A i e − γ i t cos ω i i + ϕ ι + n t wherein N represents a number of exponentially decaying sinusoidal components, Ai represents an amplitude of each component, γi represents a damping coefficient of each component, ωi represents a characteristic frequency of each component, ϕi represents an associated phase of each component and n(t) represents additive noise, and wherein the damping coefficient of each component is a damping parameter of the damping parameters and the characteristic frequency of each component is a frequency parameter of the frequency parameters.

5. The apparatus of claim 2 wherein a possible presence of DVT in the subject is assessed based on a difference between the frequency parameters and / or the damping parameters of the first and second parameter sets.

6. The apparatus of claim 1 wherein, for the purpose of comparing the first parameter set with the second parameter set, the signal processor is configured to present information relating to the first and second parameter sets on a display.

7. The apparatus of claim 1 wherein the signal processor is configured to perform further statistical analysis of the first and second parameter sets, obtained over multiple responses of the calf muscle of the first leg and the calf muscle of the second leg of the subject to multiple mechanical stimuli, in order to assess a statistical significance of differences between the first and second parameter sets, and to provide a specific indication of the presence of DVT or other soft tissue abnormality in the subject.

8. The apparatus of claim 1 wherein a communications channel between the transmitter and the receiver comprises a wired connection.

9. The apparatus of claim 1 wherein a communications channel between the transmitter and the receiver comprises a wireless connection.

10. A system (100) comprising the apparatus of claim 1 and the motion sensor.

11. The system of claim 10 wherein the motion sensor comprises an accelerometer.