WO2009144461A2

WO2009144461A2 - Magnetic induction tomography

Info

Publication number: WO2009144461A2
Application number: PCT/GB2009/001335
Authority: WO
Inventors: Stuart Watson
Original assignee: Ugcs (University Of Glamorgan Commercial Services) Ltd.
Priority date: 2008-05-28
Filing date: 2009-05-28
Publication date: 2009-12-03
Also published as: WO2009144461A3; GB2462243A; GB0809679D0

Abstract

A magnetic induction tomography (MIT) apparatus comprises an excitation signal generator (70) for generating an excitation signal; a primary excitation coil (50) arranged to receive the excitation signal from the excitation signal generator (70) and to convert the excitation signal into electromagnetic radiation and to emit said radiation to excite a sample having at least one of an electrical conductivity distribution, an electrical permittivity distribution or a magnetic permeability distribution; a primary receiver coil (60) arranged to receive electromagnetic radiation from the excited sample and to convert the received radiation into a detection signal; and a signal distribution network (115) arranged to receive the detection signal from the primary receiver coil (60). The apparatus further comprises a passive reference detector arranged to detect the excitation signal and to convert the detected signal into a passive reference signal. The apparatus further comprises an active reference signal generator (230) for generating an active reference signal; and an active reference source (175) arranged to receive the active reference signal from the active reference signal generator (230) and to supply the active reference signal to the signal distribution network (115).

Description

Magnetic Induction Tomography-

Field of the Invention

This invention relates to the field of magnetic induction tomography.

Background Art

Magnetic Induction Tomography (MIT, also known as Electromagnetic Induction Tomography (EMT) ) is a technique for imaging the passive electromagnetic properties of an object. MIT applies a magnetic field from a current-carrying coil to induce eddy currents in the object which are then sensed by an array of other coils. Compared with the related technique, Electromagnetic

Impedance Tomography (EIT) , MIT has the advantage that it does not require direct contact with the object but operates through an air gap. Whereas in EIT, electrodes are employed for current injection and voltage sensing, MIT employs coils for current induction and magnetic field sensing. Another technique

Electrical Capacitance Tomography (ECT) , can operate through an air gap but is unsuitable for conductive materials because the dielectric properties of the air gap dominate the response. The contactless nature of MIT means that errors due to electrode contact impedances are avoided, and that the positions of the coil are known and can remain fixed, which is advantageous in image construction.

In principle, additional coils can be added to an MIT array to increase the spatial resolution without any loss of practicality. Alternatively, the array can be translated along an object to provide cross-sectional images at different levels. Magnetic coupling allows eddy currents easily to be induced in samples surrounded by high resistivity barriers, for example in the brain, through the skull, and into pipes or vessels constructed from non-conducting materials. Initial progress in MIT was most rapid in imaging metals, because their high conductivities produced large eddy current signals, and because the metals industry is very familiar with inductive sensors. The development of MIT for biomedical use has been much more difficult because the electrical conductivities of biological tissues (0.02 Sm^"1 to 2.0 Sm^"1) are many orders of magnitude lower than those of metals and give much weaker signals. Nevertheless, the signals are measurable and images can be reconstructed. Applications include early detection of odema, cardiac and respiratory monitoring, detection and classification of cerebral stroke, imaging body composition and wound healing, and bioelectrical spectroscopy. Industrial applications involving conductivities similar to those of biological tissues have also been suggested, such as the imaging of ionized water in pipelines and separators, and imaging food stuffs. All of those applications, biomedical and industrial, where the conductivity is less than 10 Sm^-1Zm, are conveniently termed "low conductivity" applications and raise similar issues with regard to signal capture and image construction.

A brief explanation of the operation of an MIT system will now be described. An MIT system utilizes interrogating eddy currents that are introduced into a sample by the primary field B produced by the system's excitation coils. Those eddy currents then produce a secondary magnetic field ΔB which are detected by the system's detection coils. If the sample is not magnetic and if its dimensions are small compared with the skin depth of the applied electromagnetic field then (assuming μ_r=l and σ>>ωe_oe_r where μ_r is the relative permeability of the sample, σ is the conductivity of the sample, ω is the angular frequency of the excitation field, e₀ is the permittivity of free space and e_r is the relative permittivity of the sample) then,

ΔB

— ∞ω(ωε_oε_r-jσ) (i: B

The secondary magnetic field will therefore consist of two components, an in-phase component proportional to the permittivity and an in-quadrature component proportional to the conductivity of the sample. In principle, phase-sensitive measurements of the detected secondary field can therefore provide data on the distributions of both electrical parameters (permittivity e_r and conductivity σ) within the sample.- From equation (1) it can be seen that both the in-phase and in- quadrature components are proportional to frequency.

Construction of an image from the ΔB/B information (or rather from the corresponding voltages induced in a receiver coil) - the inverse eddy current problem - is nonlinear and ill- posed, but may be carried out using numerical methods (see e.g. R.Merwa, P.Brunner, A.Missner, J.Rosell, K.Hollaus, H. Scharfetter "Solution of the inverse problem of magnetic induction tomography (MIT) in silicio and in vitro", Proc . 6th Conference on Biomedical Applications of Electrical Impedance Tomography, University College London, UK, 2005) . The expected perturbation of the received signal due to conduction eddy currents within biological tissues I/wfΔB/Bj, even with the use of HF excitation fields, is small. Modelling and single channel measurements have shown that MIT systems require a measurement precision of Iw(ΔB/B) of the order of < 0.005% to resolve a 1% variation in the B- field perturbation produced by a typical biological sample at 10MHz. Achieving such precision requires either that (i) the system possesses very high phase stability to ensure that the noise and drift in Im(ABZB), is kept to a minimum or (ii) a method is employed to reduce the sensitivity of the system to the primary field B, thereby reducing the impact of errors in the measurement of phase between the received signal and the reference signal on the precision of measurements of Im(ABZB) .

WO2008/011649 (Technische Universitaet Graz et al.) describes a method and device for magnetic induction tomography in which a measurement is conducted at at least two different frequencies and with an introduced disturbance of coil and/or field geometry in order to determine a correction factor, with the aid of which the noise signals introduced during the object measurement by changes in the geometry are said to be substantially eliminated. The document suggests using a passive RLC circuit to provide phase correction, the circuit being provided between the gradiometer coils and the preamplifiers and/or after the preamplifiers in the described system.

WO2007/072343 (Koninklijke Philips Electronics N. V.) describes an MIT system including means for providing a relative movement between one or more generator coils and/or one or more sensor coils on the one hand and the object to be studied on the other hand.

The operation of an example prior-art MIT system will now be described in more detail with reference to Figures 1 to 4 of the accompanying drawings, in order to help the reader to better understand the invention as described below.

Watson, S., R. J. Williams, et al . describe (in "A magnetic induction tomography system for samples with conductivities below 10 S m(-l) ." Measurement Science & Technology 19(4) (2008)) an example of a prior-art MIT system comprising four subsystems: the front end, consisting of the chassis/screen, the 16 coil-modules and the transmitter (excitation) and receiver (detection) circuitry; the signal -distribution and control system; the measurement system; and the image-reconstruction system comprising the data storage, forward-modelling software, image- reconstruction algorithm and image-visualization software. The operating principle is as follows. One of the 16 channels is selected as the active transmitter channel, and the oscillator and power amplifier on the transmitter circuit board are enabled. The transmitter circuits of all other 15 channels are disabled. The receiver circuits of each of these 15 channels are then sequentially enabled.

The signals are de-multiplexed to the measurement system which measures each signal as a complex number, i.e. with in- phase and in-quadrature (real and imaginary) components relative to a synchronous reference waveform. One. frame of data consists of 240 such measurements covering all the transmitter/receiver coil combinations (16 x 15) . For imaging, two frames are measured, a data frame with a sample placed within the detector space and the other, a reference frame for any empty detector space or some other reference condition. An image of the conductivity or permittivity distribution of the sample is then reconstructed .

The MIT system front end 10 consists of a cylindrical electromagnetic screen 20 of aluminium (350 mm in diameter, 250 mm in height) within which are attached 16 coil modules 30 (see Fig. 1) Sixteen corresponding transmitter and receiver circuit modules are housed in metal enclosures 40 attached to the outside of the screen. The electromagnetic screen 20 (i) provides a rigid chassis to support the coils 30, (ii) reduces interference from external electromagnetic fields, (iii) confines the excitation field within the imaging volume, thereby removing the potential for interference by external conductive or magnetic objects and (iv) acts as a ground plane that helps to reduce undesirable electric- field (capacitive) coupling between the excitation and detection coils. The capacitive coupling in this system is very small indeed.

The constructional details of the modules are shown in Fig. 2. The transmitter coils are mounted on a Perspex former 25 and consist of two turns 50 of 0.6 mm diameter, PVC-insulated, solid, copper wire; this number of turns was found to provide the best combination of magnetic field strength ( oc current x number of turns) and load impedance at 10 MHz. The receiver coils also had two turns 60 which set the self-resonance frequency above 50 MHz. This reduced the potential for phase instability which was found to increase when operating near the resonant frequency. It also avoided resonance at harmonics of the operating frequency; with four turns, for instance, the coil exhibited resonances at approximately 30 MHz and 50 MHz, causing significant distortion of the 10 MHz signals. The design of the transmitter module is show in Fig. 3. The excitation signal source is a 10 MHz, temperature-compensated, crystal -oscillator module 70 which is enabled when the channel is to be used as the transmitter. When the oscillator is disabled, its power supply is also disconnected by means of a relay (not shown) . The output of the oscillator is passed to a buffer 80 and then to a coil-driver amplifier 90 with a balanced output which provides an excitation current of approximately 100 mA rms . The output amplifiers 100 (OPA2682) have a disable function which places their outputs into a high- impedance state. This function was used to minimize the loading of the transmitter coil 50 when not in use and hence minimizing its interaction with any of the other coils 50, 60 by inductive coupling. The specifications of the OPA2682, for the disabled state, quote an impedance frδm-άts output to ground of 800 Ω in parallel with 4 pF for the inverting configuration and 1 MΩ in parallel with 4 pF for the non- inverting. Including also the two 10 Ω resistors in series with the amplifier outputs (Fig. 3), this means that when not _^in use, the coil is shunted by an impedance of (805-4133i) Q^ The receiver module circuitry (Fig. 4) consists of a wideband differential amplifier with unity gain, formed from two devices (OPA2682 and OPA682) . This allows conversion of the received signal from balanced to unbalanced. The balanced input stage helps to provide further rejection of unwanted interference due to capacitive coupling with the transmitter coil . The signal is passed to a mixer 120 (Minicircuits TUFI-H) where it is mixed with a 9.99 MHz waveform from a local oscillator 130, thus downconverting it to 10 kHz. The mixer output then undergoes low-pass filtering (single pole f_t = 30 kHz) to remove the sum frequency, and three stages of amplification in a circuit 140 which exhibited a low phase drift and phase skew.

The downconverted signals from the receivers are distributed to two analogue multiplexers (ADG406) (not shown in Figs 1 to 4) , one of which selects the received signal, and the other the reference signal (2.1 V rms) derived from the receiver coil of the channel that is active as the transmitter. It is assumed here that the signal derived from the receiver coil on the same coil former as the active transmitter coil can be employed as a reference since, due to its proximity to the transmitter coil, it will be sensitive almost entirely to the primary field and will be hardly affected by a sample placed within the array.

The measurement system is a digital lock- in amplifier (Stanford Research Systems 830) controlled by a PC via a GPIB interface. The lock-in amplifier operates as a vector voltmeter, providing the real and imaginary components of the received signal relative to the phase of its internal reference signal. Once the reference frame (empty detector space) and data frame (sample in place) have been acquired as sets of complex numbers, the differences between the data and reference measurements, Δ V₁ are divided by the reference measurements, V, to form the quantities Re ( Δ V/V) and Im ( Δ V/V) , compatible with equation (1) . Since these quantities contain ratios of signals, the influence of any variations in the receiver gains between channels and over time is reduced. The total time taken to acquire one frame of data (240 complex numbers) is 90 s.

Images of conductivity are reconstructed from the measured values of Im(ΔV/V) . First, the sensitivity matrix, S, is computed using a finite-difference forward model of the system. The volume is divided into cuboidal voxels and S computed using two different methods. For perturbations in conductivity relative to empty space, S is computed by setting the conductivity of each voxel in turn to 1 S rrf¹. For perturbations relative to a conductive background, S is computed using the reciprocity principle (Geselowitz D 1971 "An application of electrocardiographic lead theory to impedence plethysmography" JEEE Trans. Biomed. Eng. 18 38-41) . For each " transmitter/receiver coil combination, the sensitivity Si of voxel i, with conductivity Oi, is given by

Si = J^"τi • JRJ , σ±

where J_Ti is the eddy current density induced within voxel i by the transmitter coil, and J_Ri is the eddy current density which would be induced by the receiver coil if it were employed as the transmitter. S then has as many rows as there are voxels in the volume and as many columns as there are coil combinations. Expressing the forward problem as:

Sσ = b,

where a is the unknown conductivity distribution and b is the measurement vector, the conductivity distribution, a X , reconstructed using Tikhonov regularization, is given by

σ, = min {)||||SSσσ--bb£||_₊+A²|L(σ-σ₀)|j,

where λ is the regularization parameter, L is the regularisation matrix and σ₀ is an a priori estimate of σ. L may be set to the identity matrix and no a priori information incorporated, i.e. σ₀ = 0. Images may be reconstructed using a MATLAB regularization toolbox (Hansen P C 1994 Regularization tools: a Matlab package for analysis and solution of discrete ill-posed problems Numer. Algorithms 6 1-35) . λ is selected by a subjective assessment of the smallest value, consequently giving the highest spatial resolution, which did not introduce significant noise artefacts.

To determine the noise of the system, a reference frame is collected. A second 'data' frame is then collected after an interval of 10 s; both frames are collected with no sample present in the array. Im (ΔV/ V) is computed for each transmitter/ receiver combination, . This is repeated ten times and the average noise for each channel is calculated as the standard deviation of the ten measurements. The system's average noise is calculated by averaging over all 240 channel combinations. As discussed above, the signal to be measured containing the information of interest is a small secondary signal which is at 90° in phase to the large signal due to the primary interrogating magnetic field. MIT measurement is therefore inherently phase- sensitive and phase stability is a major limiting factor in the obtainable measurement precision of MIT systems. The phase instabilities in MIT systems are due primarily to (i) phase errors produced by excitation coils and amplifiers, (ii) phase errors produced by receiver coils and amplifiers and (iii) phase errors produced by signal distribution and signal measurement systems .

The present invention seeks to ameliorate at least some of the abovementioned problems. Alternatively, or additionally, the present invention seeks to provide an improved MIT apparatus and/or MIT method.

Disclosure of the Invention

A first aspect of the invention provides an MIT apparatus as claimed in claim 1. A second aspect of the invention provides a method as claimed in claim 13. A third aspect of the invention provides a method as claimed in claim 17. Preferred (but optional) features of the invention are set out in the dependent claims. It will be appreciated that aspects of the present invention described in relation to the apparatus of the present invention are equally applicable to the method of the present invention and vice versa .

Brief Description of the Drawings Certain illustrative embodiments of the invention will now be described in detail, by way of example only, with reference to the accompanying schematic drawings, in which:

Fig. 1 is an image of the front end of prior-art MIT apparatus ;

Fig. 2 is a schematic showing in detail an individual coil module of the apparatus of Fig 1;

Fig. 3 is a diagram of the circuit of a transmitter module from the apparatus of Fig. 1; Fig. 4 is a diagram of the circuit of a receiver module from the apparatus of Fig. 1; and

Fig. 5 is a schematic diagram of MIT apparatus according to an example embodiment of the invention.

Detailed Description

Fig. 5 shows a schematic diagram of an- MIT apparatus according to an example embodiment of the invention. Components of the apparatus that are functionally equivalent to those components described with reference to Figs. 1 to 4 are labelled with the same reference numbers .

The primary MIT apparatus comprises an excitation signal generator 70 for generating an excitation signal: The excitation signal is passed from the excitation signal generator 70 through transmitter module circuitry 75 (represented schematically in Fig. 5 for ease of illustration) to a primary excitation coil, which converts the excitation signal into electromagnetic radiation and emits the radiation to excite a sample 150. The sample 150 has an electrical conductivity distribution of interest.

Excitation of the sample 150 causes emission of secondary magnetic fields, and a primary receiver coil 60 is arranged to receive that electromagnetic radiation from the excited sample 150 and to convert the received radiation into a detection signal. The detection signal passes through a signal distribution network in the form of receiver module circuitry 115 (again represented schematically in Fig. 5 for ease of illustration) .

The MIT apparatus of Fig. 5 also comprises a passive reference detector in the form of coil 200 positioned adjacent to the primary excitation coil 50 and arranged to receive the electromagnetic radiation emitted by the primary excitation coil 50 and to convert the received radiation into a passive reference signal. The passive reference signal is transmitted over a passive reference network 210 to measurement equipment 160. In this example, the two coils 50 and 60 are each used alternately as primary excitation and primary receiver poils, as different pairs of coils are addressed (it will be clear from Fig. 1 that different excitation fields can be produced using different pairs of coils 30) ; consequently, a second passive reference coil 200 is provided adjacent to coil 60, for use when coil 60 is acting as an excitation coil (of course, in that case the circuitry associated with the coil will be switched to be excitation circuitry, as associated with coil 50 in Fig. 5, and the circuitry associated with coil 50 will be switched to be receiver circuitry as associated with coil 60 in Fig. 5) . The MIT apparatus of Fig. 5 further comprises an active reference signal generator 230 for generating an active reference signal, and an active reference source arranged to receive the active reference signal from the active reference signal generator 230, via active reference signal distribution network 170, and to supply the active reference signal to the signal distribution network 115. In this example, the active reference source is a coil 175 positioned adjacent to the primary receiver coil 60, and arranged to receive a first active reference signal, to convert the first active reference signal into electromagnetic radiation and to emit that radiation. (Again, a second active reference coil 175 is provided adjacent to coil 50, for use when the roles of the coils are reversed.) The primary receiver coil 60 is arranged to receive said electromagnetic radiation from the active reference coil 175 and to convert the received radiation into a second active reference signal . The signal distribution network 115 is arranged to receive the second active reference signal (in addition to the detection signal) from the primary- receiver coil 60.

The MIT apparatus also comprises measurement apparatus 160 arranged to receive the detection signal and the active reference signal from the signal distribution network 115 and also to receive the passive reference signal and to combine the signals to provide a net phase shift or resultant vector voltage signal representative of the electrical conductivity distribution of the sample .

As discussed above, the phase instabilities in MIT systems are due primarily to (i) phase errors produced by excitation coils and amplifiers, (ii) phase errors produced by receiver coils and amplifiers and (iii) phase errors produced by signal distribution and signal measurement systems. In this example method, the phase instabilities introduced by (ii) and (iii) are reduced through the use of an active -reference signal.

¹ Assuming an excitation signal with a frequency f_e then the measured signal will have an accumulated phase change Φ as it passes through the entire MIT signal chain (see Fig. 5) described by (2) below

φ = φ_sig + Φ_e + ΔΦ + Φ_r (2) where Φ_si_g is the arbitrary phase offset of the excitation signal generator 70, Φ_e is the variable phase error produced by the excitation coil 50 and amplifiers 75, ΔΦ is the phase shift of interest produced by the conductivity distribution of the object 150 under investigation, and Φ_r is the total variable phase error produced by the receiver coil 60, receiver amplifiers 115, received- signal distribution network and measurement system 160. In the method according to an example of the invention, a further signal is introduced, termed the active reference. If the active reference signal is applied directly to the receiver coil 60, the signal will undergo a phase change described by (3)

where Φ'_Si_g is the arbitrary phase offset of the active reference signal generator 230 Φ_AR is the variable phase error introduced by the active reference signal distribution network 170, and Φ_r is the variable phase error on the active reference signal introduced by the receiver coil 60, receiver amplifiers 115, received signal distribution and measurement system 160.

If the active reference signal has a frequency f_a close to the excitation signal frequency such that f_e-f_a<<fe, then it may be assumed that Φ_r ^' « Φ_r. On subtracting the total phase offset of the active reference signal (3) from the phase offset of the received signal (2) we obtain

ΔΦ + O₆ - O_M (4 ) where the arbitrary phase offsets of the signal generators 70, 230 are ignored.

The phase error Φ_e may now be removed through the use of a second network of coils, termed the passive reference coils, which are small coils 200 placed in close proximity to the excitation coils 50. Those coils 200 pick up a signal directly from the primary excitation magnetic field and have a total phase shift described by

where Φ_PR is the variable phase error introduced by the passive reference signal distribution network 210. If the passive reference signal (5) is subtracted from the received signal - active reference signal (4) then we obtain

ΔΦ - Φ_AR - Φ_PR (6)

again ignoring the arbitrary phase offsets of the signal generators 70, 230.

The signal -processed signal described by (6) contains the phase change of interest ΔΦ and a term Φ_AR + Φ_PR which represents the phase error introduced by the combined active/passive reference networks 170, 210. Advantageously, the phase drift on ΦA_R + Φ_PR is very low in magnitude. Advantageously, the phase drift on Φ_AR + Φ_PR is relatively slowly changing. Those characteristics may be achieved by constructing both the active and passive reference networks 170, 210 from passive components - in this example, from phase stable coaxial cables, precision resistors and small inductors. Such passive networks may be constructed to be highly phase stable and to be much less affected by signal amplitude/power fluctuations than would be the case for active components e.g. amplifiers and mixers .

For the passive reference network 210, the signals produced by each of the excitation coils are detected by small coils 200 placed such that they are sensitive to the primary field while being relatively insensitive to the secondary fields, such that the detected signal is dominated by the primary field signal; that may achieved by placing the reference coils 200 in close proximity to the excitations coils 50, or by positioning and/or orienting the reference coils 200 for low sensitivity to the secondary magnetic fields. Alternatively, the reference signal may be obtained via a voltage signal tapped from some point on the excitation coil 50 or from the detected current passing through the excitation coil 50 using current -voltage conversion. Each excitation channel may be allocated its own reference measurement channel or any number of the passive reference channels may be combined before measurement. Ultimately all the passive reference signals are combined into a single reference signal. Decoupling resistors 220 are used to reduce cross- channel coupling via the passive reference coils. Signals are distributed via a cable system with suitable phase stability (e.g. rigid coaxial) and low temperature coefficient resistors should be employed. The reference coils 200 may be clamped in place to a rigid coil former to enhance mechanical stability.

For the active reference, the signal may be introduced to the receiver channels by (a) induction using a coil 175 or (b) injection of the signal into the receiver coil 50 or amplifier 75 input. A single signal is derived from a signal generator 160 andamplifier and is then split and distributed to the input of each receiver channel using a network 170 of resistors and cables. As for the passive reference network 210, the signals should be distributed via a cable system with suitable phase stability (e.g. semi-rigid coaxial or "phase-stable" coaxial) and low temperature coefficient resistors and the reference coils 175 may be clamped in place to a rigid coil former to enhance mechanical stability.

MIT systems may employ more than one frequency, either for channel frequency encoding or for spectroscopy (Rosell-Ferrer, J., R. Merwa, et al . (2006) . "A multifrequency magnetic induction tomography system using planar gradiometers : data collection and calibration." Physiological Measurement 27(5) : S271-S280) . For channel frequency encoding, a single reference frequency may be employed for all receiver channels if the channel frequency separation is small such that Φ_r ^' « Φ_r for all signals. For spectroscopic measurements the active reference may either be a composite of multiple frequencies if simultaneous multi- frequency excitation is employed or a single reference frequency may be changed sequentially with the excitation frequency.

The following is an example of an algorithm for eliminating the phase error. If we assume a single excitation frequency fi then for the implementation of the example method for one of the receiver channels we will measure the amplitude and phase of three signals:

S_PR: passive reference signal due to primary field only, at frequency f_x

S_r : MIT signal detected at receiver, at frequency fi S_AR: active reference signal detected at receiver, at frequency f_AR

Signal S_PR is measured using one channel (assume chO) of the measurement system while the composite signal detected at the receiver and comprised of S_r + S_AR is measured using another channel (assume chl) of the measurement system.

A signal processing method which can measure separately the amplitude and phase of each of the frequency components of the signal is then applied to the signal from cho and chl. The applied signal processing methods may include FFT, digital phase- locked loops or digital filtering followed either by phase- sensitive detection, curve fitting or zero-crossing detection.

The measured signal components are then processed as follows:

φ[Sres] =φ [ Sr] - φ[ SPR] - φ[ SAR]

where φ[Sres] is the phase of the resultant signal Sres and φ[ Sr], φ[ SPR] and φ[ SAR] are the measured phases of the received, passive reference and active reference signals respectively. From equation (6) above the resultant signal Sres will contain terms due to the object placed within the detection volume and to the phase shift produced by the active and passive reference networks 170, 210. The phase errors introduced by the excitation and receiver circuitry, and the received signal distribution systems and measurement system will be removed to a high approximation. The above process is repeated for all receiver channels and for all excitation frequencies (sequentially or simultaneously excited) .

MIT systems are capable of measuring conductivity, permittivity and permeability; the method helps separate conductivity (which primarily affect the imaginary component andhence phase of the detected signal) from the other two material properties (which primarily affect the real component) . The detected signal is typically changed in both amplitude and phase by a sample. For systems in which the primary field is cancelled the amplitude variation is very small and may be ignored - the phase is used as a good approximation of the signal attributable to the conductivity. In either (i) gradiometer systems or (ii) for samples in which the conductivity is very high or the sample large, one may need to consider both amplitude and phase, and hence need to use a resultant vector voltage.

The measured amplitudes of the passive reference signal are in this example also employed to provide a 2nd order compensation and tracking of phase errors. Here it is assumed that a small phase skew φ_sk is produced by the receiver circuitry. The phase error will be a function of the detected signals magnitude, φ_Sk ( I Sr I ) . Measurements of the detected signal phase vs input amplitude for the full signal chain of the receiver circuit, signal distribution and measurement system are undertaken and the results stored in a look-up table to allow compensation of any phase skew produced by variation in the received signal magnitude, either due to fluctuations in the gain of the receiver, power output of the excitation signal or large signal magnitude variations produced by objects placed within the detector.

The measured amplitudes of the active reference are in this example employed to monitor the power output of the excitation signal, including all frequency components, allowing either a phase skew compensation algorithm to be applied or a closed- loop control of the frequency component amplitudes input to the excitation power amplifier to reduce fluctuations of the output power over time. In other example embodiments of the invention, the details of the implementation of the method vary in the following aspects .

One or more excitation frequencies may be employed, either for channel frequency encoding or for spectroscopy. For channel frequency encoding a single reference frequency may be employed for all receiver channels if the channel frequency separation is small such that Φ_r ^' « Φ_r for all signals. For spectroscopic measurements the active reference may be composed of multiple frequencies with either single frequencies changed sequentially, using chirp or employing a signal of combined multiple frequencies .

For the passive reference the excitation signal may be detected using (a) coils, (b) tapping the voltage directly from the coil signal or (c) detecting the current flowing to the excitation coil.

For the active reference the signal may be introduced via

(a) a coil, (b) injection of the signal into the receiver coil or amplifier input. For the passive reference network each excitation channel may be allocated its own reference measurement channel or any number of the passive reference channels may be combined before measurement. Ultimately all the passive reference signals may be combined into a single reference signal . The method may be implemented using (a) coil arrays in which each of the receiver coils are sensitive to the primary field and

(b) coil arrays in which a method of reducing the sensitivity of the receiver coils to the primary field is employed e.g. gradiometry or coil orientation methods . The passive reference signal may also be employed to monitor the amplitude fluctuations of the primary field. This may then be employed in signal processing to further stabilise the performance of the system.

The active reference signal may also be employed to monitor the amplitude fluctuations of the received signal. This may then be employed, in combination with (6) above, in signal processing to further stabilise the performance of the system.

In an MIT system in which each of the receiver coils are sensitive to the primary field, the method will improve both short term and long term measurement precision by reducing phase drift and allowing longer signal measurement integration times without limitation by phase drift. The method may also be applied in an MIT system in which a method of reducing the sensitivity of the receiver coils to the primary field is employed e.g. gradiometry or coil orientation methods. The method may be used to (a) provide a reliable and stable measurement of the phase of the primary field, (b) allow variations in the primary field cancellation to be monitored and compensated for and (c) improve measurement accuracy further by reducing phase drift. In some MIT system designs there may a need to apply the active reference signal non-simultaneously with the excitation signals. Cases where this might be necessary include low power MIT systems (e.g. battery based), and systems which employ a measurement system unable to effectively discriminate two very near frequencies (e.g. zero-crossing detection phase measurement systems) . In such cases the active reference may be switched on alternate with the excitation signal. A measurement cycle would then proceed in a fashion similar to that described below:

- excitation signal of one or more channels activated alternately and sequentially and measurement undertaken

- all excitation channels disabled - active reference enabled and reference measurement set collected on all channels (simultaneously or sequentially)

- reference channel disabled.

In this case the active reference frequency need not differ from the excitation signal, e.g. fl = f2. A single signal generator, and even a single power amplifier for excitation and reference, may be employed.

Whilst the present invention has been described and illustrated with reference to particular embodiments, it will be appreciated by those of ordinary skill in the art that the invention lends itself to many different variations not specifically illustrated herein. Some examples of such variations and alternatives have been described above.

Where in the foregoing description, integers or elements are mentioned which have known, obvious or foreseeable equivalents, then such equivalents are herein incorporated as if individually set forth. Reference should be made to the claims for determining the true scope of the present invention, which should be construed so as to encompass any such equivalents. It will also be appreciated by the reader that integers or features of the invention that are described as preferable, advantageous, convenient or the like are optional and do not limit the scope of the independent claims .

Claims

1. Magnetic induction tomography (MIT) apparatus comprising:

(a) an excitation signal generator for generating an excitation signal;

(b) a primary excitation coil arranged to receive the excitation signal from the excitation signal generator and to convert the excitation signal into electromagnetic radiation and to emit said radiation to excite a sample having at least one of an electrical conductivity distribution, an electrical permittivity distribution or a magnetic permeability distribution;

(c) a primary receiver coil arranged to receive electromagnetic radiation from the excited sample and to convert the received radiation into a detection signal; (d) a signal distribution network arranged to receive the detection signal from the primary receiver coil;

(e) a passive reference detector arranged to detect the excitation signal and to convert the detected signal into a passive reference signal; (f) an active reference signal generator for generating an active reference signal; and

(g) an active reference source arranged to receive the active reference signal from the active reference signal generator and to supply the active reference signal to the signal distribution network.

2. MIT apparatus as claimed in claim 1, further comprising measurement apparatus arranged to receive the detection signal and the active reference signal from the signal distribution network and also to receive the passive reference signal and to combine the signals to provide a net phase shift or resultant vector voltage signal representative of the electrical conductivity distribution, the electrical permittivity distribution or the magnetic permeability distribution of the sample .

3. MIT apparatus as claimed in claim 1 or claim 2, wherein the passive reference detector comprises a coil positioned adjacent to the primary excitation coil and arranged to receive the electromagnetic radiation emitted by the primary excitation coil and to convert the received radiation into the passive reference signal .

4. MIT apparatus as claimed in claim 3, wherein the passive reference coil is oriented at an orientation in which its sensitivity to electromagnetic radiation from the excited sample is substantially minimised.

5. MIT apparatus as claimed in claim 1 or claim 2, wherein the passive reference detector is arranged to detect the excitation signal by measuring a voltage across the primary excitation coil .

6. MIT apparatus as claimed in claim 1 or claim 2, wherein the passive reference detector is arranged to detect the excitation signal by measuring a current flowing to the primary excitation coil .

7. MIT apparatus as claimed in any preceding claim, wherein the active reference source comprises a coil positioned adjacent to the primary receiver coil and arranged to receive the active reference signal in the form of a first active reference signal, and to convert the first active reference signal into electromagnetic radiation and to emit said radiation, and wherein the primary receiver coil is arranged to receive said electromagnetic radiation from the active reference coil and to convert the received radiation into a second active reference signal, and wherein the active reference signal supplied to the signal distribution network is the second active reference signal, which is received by the signal distribution network from the from the .primary receiver coil.

8. MIT apparatus as claimed in any of claims 1 to 6, wherein the active reference source is arranged to supply the active reference signal to the signal distribution network as an electrical signal applied to the primary receiver coil or directly to the signal distribution network.

9. MIT apparatus as claimed in any preceding claim, wherein the active reference signal is transmitted from the active reference signal generator- to the active reference source via an active reference network constructed exclusively from passive components:

10. MIT apparatus as claimed in any preceding claim, wherein the passive reference signal is transmitted from the passive reference detector to the measurement apparatus via a passive reference network constructed exclusively from passive components.

11. MIT apparatus as claimed in any preceding claim, wherein the excitation signal generated by the excitation signal generator comprises simultaneously or successively a plurality of frequencies.

12. MIT apparatus as claimed in claim 11, wherein the active reference signal is at a frequency that is the same for all of the excitation signal frequencies.

13. MIT apparatus as claimed in any preceding claim, comprising a plurality of the primary excitation coils and the primary receiver coils, and a plurality of the passive reference detectors and the active reference sources .

14. A method of magnetic induction tomography (MIT), comprising:

(a) generating an excitation signal;

(b) converting the excitation signal into electromagnetic radiation and emitting said radiation to excite a sample having an electrical conductivity distribution, an electrical permittivity distribution or a magnetic permeability distribution;

(c) receiving electromagnetic radiation from the excited sample and converting the received radiation into a detection signal; (e) detecting the excitation signal and converting the detected signal into a passive reference signal;

(f) generating an active reference signal;

(g) supplying the active reference signal to a detection signal distribution network arranged to receive the detection signal; and

(h) measuring the detection signal, the active reference signal, and the passive reference signal and combining the measured signals to provide a net phase shift or resultant vector voltage signal representative of the electrical conductivity distribution, the electrical permittivity distribution or the magnetic permeability distribution of the sample.

15. A method as claimed in claim 14, further comprising monitoring amplitude fluctuations in the excitation signal by monitoring the passive reference signal .

16. A method as claimed in claim 15, further comprising monitoring amplitude fluctuations in the detection signal by monitoring the active reference signal .

17. A method of magnetic induction tomography (MIT), comprising: (a) generating an excitation signal; (b) using a primary excitation coil to convert the excitation signal into electromagnetic radiation and to emit said radiation to excite a sample having an electrical conductivity distribution, an electrical permittivity distribution or a magnetic permeability distribution; (c) receiving electromagnetic radiation from the excited sample and converting the received radiation into a detection signal; (e) detecting a signal transmitted to the primary excitation coil and converting the detected signal into a first reference signal;

(f) generating a second reference signal;

(g) supplying the second reference signal to a detection signal distribution network arranged to receive the detection signal; and

(h) measuring the detection signal, the second reference signal, and the first reference signal and combining the measured signals to provide a net phase shift or resultant vector voltage signal representative of the electrical conductivity distribution, the electrical permittivity distribution or the magnetic permeability distribution of the sample.

18. A method as claimed in claim 17, in which the signal converted into the first reference signal is the excitation signal.

19. A method as claimed in claim 17, in which the signal converted into the first reference signal is an active reference signal .

20. Apparatus substantially as herein described, with reference to Fig. 5 of the accompanying drawings.

21. A method substantially as herein described with reference to Fig. 5 of the accompanying drawings.