The invention will now be described by way of example only, with reference to the following drawings, of which:
FIG. 1 shows the light transmission and detection system according to the invention;
FIG. 2 shows a block diagram of the transducers in FIG. 1;
FIGS. 3
a, b, c are schematic views of a preferred embodiment of the invention in FIG. 1 applied to different sites on a patient;
FIG. 4 is a signal output from the embodiment as applied in FIG. 3a;
FIG. 5 shows another preferred embodiment of the invention;
FIG. 6 shows the output from the embodiment as shown in FIG. 5 from the various sites of the legs of a patient; and
FIG. 7 shows the signal response to increased breathing impedance and hand grip.
FIG. 8 shows the vasomotor signal and extraction of the heart rate variation.
Referring to FIGS. 1 and 2, the invention comprises a light transmission and detection system including transducers 1, 2 comprising an LED and photo-detector with suitable amplifiers 3, 4 as shown in FIG. 2. Once the transducers 1, 2 are attached to the skin the central control unit 5 calibrates them by driving the LED 1 with a voltage appropriate to detect a mid-scale voltage from the photo-detector 2. The photo-detector 2 signals are digitised by A/D1 and A/D2. The drive voltages for the LEDS are produced from the output of D/A1 and D/A2. Once the calibration process is complete the central control unit 5 collects data from the photo-detector 4 (FIG. 2) at a sampling rate appropriate for the application. For DVT detection a sample rate of 6 Hz is used. A user input device 6 such as a keypad and a display for output, for example an LCD screen or LED indicators or similar is used. There is also provided an input/output port for PC connection, printer or other form of data logging device.
FIGS. 3
a to c show a preferred embodiment of the invention using a two channel system using two transducers 1, 2 for differential signal analysis. For the purpose of DVT detection, the transducers 1, 2 are positioned on the soles of the feet of a patient as shown in FIG. 3a. The configuration of 3b can give an indication of the approximate location of DVT. If the vasomotor signals are similar the DVT will be located in the thigh whereas if the vasomotor signals are dissimilar the DVT will be located in the calf. The arrangement in FIG. 3c indicates the pulse transit time between the upper and lower extremities and thus an indication of arterial stiffness. FIG. 4 shows the signal derived from the soles of the feet of a healthy subject using a two channel system. The signal from each transducer is similar if not identical. The presence of a unilateral DVT is detected by measuring any dissimilarity between the two signals.
The output presented to the user can take the form of a detailed display of vasomotor signals collected from the transducers 1, 2 as shown in FIG. 4 to a simple indication of a condition being present or absent. The display can be configured to the application.
The sampling rate of the transducer 1, 2 signals is such that the heart rate component can be resolved to within +/−1 ms or better if the heart rate is of interest in the assessment being performed, for example in autonomic function testing. Otherwise sampling frequencies that meet the Nyquist requirements are adequate.
The signals acquired from each transducer 1, 2 are subject to appropriate analytical algorithms. The signals are subject to amongst others complex demodulation a mathematical technique used for investigating the vasomotor activity centred at specific frequencies with a bandwidth chosen in accordance with the application, for example DVT detection. The output of the complex demodulation algorithm consists of an amplitude signal and a phase signal which when combined, produce a time varying signal modulated by both amplitude and phase with limited bandwidth, all centred on the demodulating frequency.
As well as the arrangements shown in FIGS. 3a to c, another preferred embodiment has two further transducers 7, 8 applied behind the knees for a four channel system as shown in FIG. 5. The signals are passed through the stages of signal pre-processing including filtering and DC removal followed by complex demodulation at a set of chosen frequencies, for example 8 to 30 cycles per minute. The mean absolute phase differences (MAPD) from the right foot (RF) and the left foot (LF) are calculated for each frequency to produce a spectrum RFLF(MAPD) and the RFLF(MAPD) is then used by a pattern classifier such as a pre-trained artificial neural network to provide an output on a screen that there is either “DVT PRESENT” or “DVT NOT PRESENT”.
For a four channel system as shown in FIG. 5, there will be six MAPDs as shown in FIG. 6:
Right Foot Left Foot: RFLF=mean(abs(RF(Φ)−LF(Φ))),
Right Knee Left Knee: RKLK=mean(abs(RK(Φ)−LK(Φ))),
Right Foot Right Knee: RFRK=mean(abs(RF(Φ)−RK(Φ))),
Left Foot Left Knee: LFLK=mean(abs(LF(Φ)−LK(Φ))),
Right Foot Left Knee: RFLK=mean(abs(RF(Φ)−LK(Φ))),
Right Knee Left Foot: RKLF=mean(abs(RK(Φ)−LF(Φ))),
giving six times the diagnostic information of the two channel system, described above.
In addition to detecting DVT, the present invention can monitor and assess a range of clinical conditions including diabetic peripheral neuropathy, critical limb ischaemia, autonomic neural function and arterial and venous disease.
In each of these conditions the vasomotor activity of the micro circulation possesses a unique signature which is extracted and assessed using the appropriate signal processing algorithms. These algorithms are tuned to the appropriate frequency bands determined by the clinical condition of interest. The algorithms exploit the property of vasomotor symmetry between the left and right feet and also use the similarity between the low frequency components of the vasomotor activity and the low frequency components of heart rate variation. As shown in FIG. 8, the device according to the invention, extracts from the vasomotor signal the heart rate variation and direct comparison of the simultaneous low frequency heart rate variation and the low frequency vasomotor variation provides information relating to diabetic sympathetic neuropathy, any dissimilarity between the two components indicating diabetic sympathetic neuropathy.
FIG. 7 shows the changes in vasomotor activity related to increased breathing resistance and the hand grip test of a healthy person. These tests affect systemic blood pressure and cardiac output which in turn cause neurologically mediated responses in heart rate and peripheral vasomotor activity as observed with the transducers on the soles of the feet. Any changes from the signals in FIG. 7 between the resting phase and the increased breathing resistance and the hand grip test will indicate diabetic sympathetic neuropathy since the pathology of the sympathetic nerve fibres which innovate the micro-blood vessels within the feet will cause significant change in vasomotor behaviour.
Patent Application
20080076984
