AU2008229986A1

AU2008229986A1 - Electrical Impedance Device And Method

Info

Publication number: AU2008229986A1
Application number: AU2008229986A
Authority: AU
Inventors: Vladimir Alexeyevich Cherepenin; Joanne Firth; Alexander Korjenevsky
Original assignee: Health Screening Tech Pty Ltd
Current assignee: HEALTH SCREENING TECHNOLOGIES Pty Ltd
Priority date: 2007-10-17
Filing date: 2008-10-17
Publication date: 2009-05-07
Also published as: CN101444422A

Description

Pool Secton 29 Regulaton 3.2(2) AUSTRALIA Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT Application Number: Lodged: Invention Title: Electrical Impedance Device And Method The following statement is a full description of this invention, including the best method of performing it known to us: 1 ELECTRICAL IMPEDANCE DEVICE AND METHOD FIELD OF THE INVENTION The present invention relates to electrical impedance (EI) screening, such as for health screening. 5 BACKGROUND TO THE INVENTION X-Ray mammography is currently the standard imaging modality for detection of breast cancer. As well as sending a radiation dose through the breast, the x-ray mammography procedure is generally uncomfortable and/or painful for many women. The process generally involves flattening the breast 10 against a screen in order to obtain suitable radiograph images. Also, the relatively high financial cost of x-ray mammography equipment prevents widespread adoption in many developing countries. Furthermore, exposure to the x-ray ionizing radiation can ultimately damage breast tissue, particularly through cumulative effects. In addition x-ray mammograph has a relatively high 15 percentage of misdiagnoses and false detections resulting in either missed cancers or unnecessary biopsies, surgery, chemotherapy and/or radiotherapy treatment. Recently, electrical impedance tomography (EIT) has been developed for non-invasive detection of signs of breast pathology including breast cancer. The 20 EIT procedure is suitable for all ages, comfortable for the patient, and provides both anatomical and physiological information. Capital investment in EIT equipment is much lower than for x-ray equipment, making it affordable and acceptable in many countries. The EIT procedure poses no safety hazards to either the patient or the clinician, and results have been found to be positive in 25 detecting early signs of cancer and early stage small tumours, as well as pre cancer pathology. Generally, the technology behind EIT involves an array of electrodes placed on the target site with a single electrode placed remote therefrom, usually placed on the patient's wrist. The electrodes in the array are driven sequentially 30 with a predetermined electrical frequency while the resulting current through each electrode is measured. This results in a conductivity contour map of the target area.

2 Resulting images derived from the EIT system demonstrate relatively low conductivity regions for organs such as the lungs, and higher conductivity regions for more solid tissue. The EIT system demonstrates a difference in conductivity for cancerous tissues compared with normal tissue, thereby providing a contrast 5 ratio in the resulting conductivity map derived from the signals received from the array of electrodes. In this way, signs of cancer can be determined from the conductivity map image. This method is particularly effective for malignant tumours in the breast which differ significantly from the properties of the surrounding normal tissue. 10 Disadvantages of early EIT systems are low resolution of the images, partly because resolution decreases significantly with increased depth into the patient's body, as well as images could only be derived from differences in conductivity between consecutive electrode measurements. Consequently, the aforementioned 2-dimensional (2-D) EIT electrode array has more recently been 15 supplemented by applying reconstruction software to the 2-D array data acquisition results in order to reconstruct a 3-D image of the target area. This involves a reconstruction algorithm which essentially derives data in the form of multiple cross-sections of the target area parallel to the electrode array and at different depths within the patient. Whilst this procedure has been found to 20 provide enhanced image resolution for some circumstances, there is seen a need for further improvements in image resolution and/or detection of early signs of cancer or actual tumours which do not readily resolve in the image for the electrical frequency used by the EIT system. A known 3-D EIT system discussed in a UK Institute of Physics publishing 25 journal on physiological measurements 22 (2001) 9218, discusses a system incorporating an array of electrodes positioned over the tissue to be measured, two additional electrodes placed remotely from the array of electrodes, a source of alternating (AC) current, a means to measure potential difference and computing means to reconstruct an visualise the conductivity distribution as a 30 stack of tomographic images. The joumal paper discusses a comparative analysis of electrical impedance images obtained with the disclosed system in normal and cancerous breasts. The system is driven by a 0.2 mA current at a frequency of 10 kHz. For a given input electrode, an output multiplexer 3 sequences through all of the other electrodes in the array whilst a potential difference detects measurements. The input electrode is switched and the output multiplexer sequence repeated. This paper discusses switching a single lead at a time to reduce spurious couplings which arise from cross-talk in the multiplexer, 5 and also to simplify the device and reduce costs. The current injection and voltage measurements circuitry is stated as being similar to one in an electrical impedance tomograph for static imaging by the same inventors (Cherepenin et al 1995, Korjenevsky et al 1997) a four measurement cycle of 65,280 voltage measurements (256 electrodes X 255 sequences) is used to derive the image 10 through reconstructive software again, only a single measurement frequency is utilised. It has been realised that image quality/resolution and/or accurate detection of early signs of cancer or tumours, or more positive determination of no presence of early signs of cancer or actual tumours, can be achieved over and 15 above the aforementioned systems. With this in mind, it is an object of one form of the present invention to provide improved electrical impedance tomography with enhanced image resolution or improved likelihood of positively detecting early signs of cancer or that no cancer is present. 20 Whilst electrical impedance tomography enables visualisation of special distribution of electrical properties (the real and/or imaginary parts of the impedance or admittance, absolute value of the impedance, phase of the impedance or admittance, capacitance or conductivity, or some function of the impedance or admittance components) of the biological tissues of the subject, 25 local changes to the electrical conductivity enables detection of tumours, inflammations and/or cysts, or the like. However, in order to obtain pertinent data, a significant proportion, if not all, of the electrodes used for applying the electrical current and measuring the signals obtained must be in reliable contact with the surface of the target region of the subject. Electrical impedance 30 tomography currently used for imaging and diagnostics utilises a planar array of electrodes attached to a rigid surface . The electrode array on the rigid surface must be applied to the target region, such as the breast, with sufficient force to flatten the target region so that sufficient electrodes make contact. Whilst this can 4 be very uncomfortable or painful for a subject, reasonable contact can be achieved for large, pliable regions of the subject, such as a large breast. However, if the target region is of a small, low compliance curve, such as a small breast, such as typically in south east Asia or younger women, and also in 5 women with breast augmentation, obtaining reliable measurements, or any measurements of significance, becomes difficult for the technician as well as highly uncomfortable or painful for the subject. Furthermore, any data obtained can lack validity due to the large number of poor or non-contact electrodes. There have been experimental set-ups previously where the electrodes 10 have been positioned on motorised holders. These are then adjusted to conform to the shape of the target region, such as the breast. However, such an approach is unrealistic for a 3-D measuring system where 256 measuring electrodes need to be optimally placed in contact with the breast, (IEEE transactions on medical imaging, Col. 21, No. 6, June 2002, PP 638 - 245). 15 One known device is described in US patent no. 6,308,097. This device obtains information about spatial distribution of electrical properties that are inside the human body using multi-frequency measurements. The device realizes 2D mapping of conductivity on the body surface using an array of electrodes. The value of current through each electrode directly defines the brightness of 20 corresponding image pixel. This current depends on full resistance of the body along the current pass. Depth resolution into the body is not possible because sensitivity decreases fast with depth. With this system, only 2D imaging is possible and the image quality is rigidly limited by the physics of the current flow. The collected data set is poor, so advanced data processing is not applicable. 25 Using this device, a voltage source is connected simultaneously to all electrodes and the current through each electrode is measured. So the same electrode is used both for applying voltage and for measuring current that makes the system extremely sensitive to the skin-electrode contact conditions. Dampness, skin scratches, moles, etc can change the resulting image dramatically. Furthermore, 30 the collected data set from this device consists of N values, where N is the number of electrodes. The lack of information limits possible data processing by simple 2D mapping of the data on a PC screen.

5 With these issues in mind, it would be advantageous to provide improvements to patient scanning practices and technology. For example, provision of methodology to improve scanning techniques and improved assessment, and/or electrical impedance tomography that provides 5 improved imaging and/or adapts to a shape of the target region. SUMMARY OF THE INVENTION With the aforementioned in view, in one aspect the present invention provides a method of electrical impedance screening of a body region of a subject including the steps of positioning at least one multi-element probe on at least one 10 respective surface of the body region; applying an electrical current to elements of the multi-element probe at at least two different frequencies; sensing signals from the elements derived from the different frequencies; and 15 generating at least one impedance map based on the sensed signals. Multiple frequency electrical impedance imaging advantageously permits enhanced image results for improved determination of image data. The electrical current may be applied at a first frequency to elements of the multi-element probe and subsequently switched to a second frequency to be 20 applied to the elements of the multi-element probe. Thus, data for an impedance map depending on a first frequency can be determined and used to find abnormal tissue areas at that frequency, and then compared with the impedance map data relating to the second frequency. In this way, different types of tissues that demonstrate different dependence or independence with the frequency of the 25 electrical signal may be identified. For example, benign cysts and malignant tumours may demonstrate different characteristics for a given frequency, and therefore an be identified one from the other by comparison of the impedance map data. Preferably, measurement at a first frequency is taken and then the applied 30 signal switched to another frequency, preferably through a software program interface, to conduct a subsequent measurement at the second frequency. It is envisaged that frequencies of 10 kHz and 50 kHz may be preferable; however, any multiple frequencies sufficiently differentiated between 1 kHz and 1 MHz is 6 envisaged, though 10 kHz to 100 kHz is a preferred range due to beta-dispersion of the electrical impedance in soft body tissues. According to a preferred form of the present invention, the multiple frequencies may be applied simultaneously to the elements of the multi-element 5 probe. Thus, the multiple frequencies may be applied first one and then the other (consecutive frequencies) or at the same time (simultaneous frequencies). Whilst a minimum of two frequencies provides improved results, the present invention should not be considered to be limited to only two frequencies. 10 For example, any number of multiple frequencies may be employed, two different frequencies being a minimum, the greater the number of frequencies (provided they are sufficiently separated in the frequency spectrum) may be applied. Using more than two multiple frequencies may provide enhanced resolution, and thus perceived enhanced impedance map results and diagnosis. 15 A further aspect of the present invention provides apparatus for electrical impedance imaging of a region of a subject, including: a multi-element probe including a plurality of sensing elements and adapted for mounting on a first side of the region, a second probe including at least one element for mounting on the subject remote from the first multi-element probe, electrical signal device for 20 applying electrical current at at least two different frequencies to elements of the multi-element probe, and means for deriving an electrical impedance image map from data derived from the applied current at the different electrical frequencies. The multi-element probe may comprise 256 electrodes. An alternative form of the present invention provides a multi-element probe 25 device for electrical impedance tomography, including: a flexible material and a plurality of electrodes arrayed at a surface of the material. In this way, a comfortable array of electrodes of a multi-element probe is provided which can adapt the shape of the array to permit contact with a larger number of the electrodes for a given shape of target region of a subject i.e. a 30 conformable array. The increased number of electrodes being in good contact with the surface of the target region can provide higher quality reconstructed images (impedance image map), thus improving the diagnostic value of the data obtained.

7 Preferably the flexible material provides a waterproof or water resistant membrane carrying the electrodes to prevent or at least reduce penetration of moistening agents used during the measurements. For example, the membrane may be made of or include rubberised or polymeric fabric or plastic film. 5 The flexible material may be supported on a base material, such as a fibrous or spongy material allowing the electrodes to be pressed against the surface of the target region whilst permitting the flexible material to conform to the shape of the target region of the subject. At least some of the electrodes may penetrate through the flexible material 10 and/or some may be arrayed on the outer surface of the flexible material. Similarly, the covering membrane, such as the waterproof/water resistant membrane may include apertures through which at least one of the electrodes passes. It is envisaged that the array of electrodes may be covered in a material 15 that is electrically conductive, preferably with a conductivity matched to the conductivity of the surface of the target region, such as matched to the conductivity of biological tissue, e.g., skin. In such a case, the electrodes may be attached to an interior surface of the cover material e.g., membrane, thereby keeping the outer surface of the device clean and undamaged. 20 An important advantage of the multi-frequency electrical impedance technology is particularly useful in mammography, whereby benefits are provided over the known mono-frequency technique, in than it makes it possible not only to diagnose mastopathy but also helps exactly determine cystless form from the absence of difference between indices of mean electrical conductivity in the 25 corresponding age groups during all phases of the menstruation cycle. According to morphologists, occurrence rate of hyperplasia and atypia that accompany dysplasia is higher than with cystic form, which makes it possible to ascertain a group of higher risk for special medical care to prevent breast cancer development. 30 One or more forms of the present invention realizes solving the so-called inverse problem for Maxwell's equations in quasi-static approximation. This method has no theoretical limitations in reconstruction of 3D distribution and quality of images, but suitably requires enough calculation and an accuracy of 8 measurement. The number of electrodes, the accuracy of measurements, and the complexity of the reconstruction algorithm are the general limiting factors. The collected data set is full for the given geometry of the electrodes array. Embodiments are able to reconstruct 3D static conductivity from inside the human 5 body. A large data set can be gathered and can then reconstructed into parallel tomographic planes. The present invention can realize four-electrode measurements when the voltage readings are never performed on the electrodes through which the current is injected. Such measurements are almost invariant to the contact conditions on the skin surface. Software utilised in embodiments of 10 the present invention can provide very fast "preliminary" imaging without resolution in depth and is used for mapping of electrodes with improper contacts mainly. One or more embodiments of the present invention may obtain a collected data set containing N(N-1) values. For each injecting electrode which originates 15 the current, measurements can be made sequentially on the remaining N-1 electrodes. This cycle may be repeated N times, for example, so that each electrode serves as the injecting electrode. The resulting data set may contain a maximum information possible and a 3D inverse problem solving may be effected therefrom. A fast and robust back projection algorithm may used to reconstruct a 20 few tomographic planes through the breast tissue. 3D tomographic reconstruction from the same data set can produce detailed images showing the presence or absence of problematic tissue compared with normal tissue. A further aspect of the present invention provides a method of screening a patient, including the steps of; 25 a) obtaining electrical impedance (El) measurement data of a patient; b) applying an algorithm to the measurement data to produce image data; c) assessing the image data to make a first determination of the presence or absence of problematic tissue. 30 Preferably the measurement data may be obtained as depth data values from a plurality of data planes through the tissue. For example, the measured values may derive as depth data in multiple (horizontal) planes or "slices" through the patient's tissue eg breast tissue.

9 Initial assessment may provide a first opinion to determine whether or not problematic tissue is clearly present. If nothing is detected, the patient can be assessed at a future period eg 12 months after. If a definite anomaly is detected, the patient may be referred for further, more detailed screening. Thus a Yes/No 5 screening practice can be provided. This initial assessment can be conducted by low qualified people eg in remote areas or areas of poverty where detailed and expensive assessment is difficult, not cost effective, or unavailable. Where a small anomaly is initially detected, the patient may be referred for a further assessment or may be required to undergo a further initial assessment after a 10 given period, say 6 months. The image data and/or measurement data may be stored remotely and/or (further) assessed remotely. The method may preferably include determining a probability P of malignancy being present in scanned tissue. His may include utilizing the equation P =1 /(1 + exp(-a), where, preferably, 15 a = bo +bx +b 2

X

2 +b 3 x 3 +b 4 x 4 +b x51 x, is patient age, x 2 is mean over raw measurements, x 3 is median over raw measurements, x 4 is mean over image data, 20 x 5 is standard deviation over image data, This gives a value between 0 and 1, with 0 indicating an absence of malignancy and 1 indicating a presence of malignancy. Thus, 0 values may result in the patient not requiring referral for further assessment, 1 requiring the patient to be re-assessed. Values falling between 0 and 1 may be used to 25 determine whether or not a patient is clear, requires further assessment or returns for another scan within a given shortened time frame. A further aspect of the present invention provides a method of screening a patient utilising electrical impedance for a presence of abnormal tissue, including the steps of; 30 a) connecting an array of electrodes to a patient, one of the electrodes acting as a receiver and a plurality of the remaining electrodes each acting as a transmitter; 10 b) introducing an electrical current signal to the patient through each of the transmitter electrodes; c) receiving said current signals at the receiver electrode; d) transferring a transmitter electrode to be the receiver electrode and 5 the receiver electrode to be a transmitter electrode; e) repeating steps b) to d) for each of the transmitter and receiver electrodes; f) generating image data from the multiple received signals; g) utilising the image data to assess presence or absence of abnormal 10 tissue. The term "transferring" refers to operatively (electrically) changing the relevant electrodes. The image data may be utilised to construct multiple depth layer images from the image data. Multiple depth "maps" showing layers of tissue provide enhanced visualisation of the underlying tissue, and permit comparison of 15 adjacent layers to assist in more accurately assessing whether there is abnormal tissue present or not. Initial assessment may give a positive-negative assessment, and if positive or even if the assessment is borderline, a second or further assessment may be used to obtained a more definitive determination of the presence/absence and 20 preferably extent and virulency of any abnormal tissue. Preferably the method may include generating multiple depth image layers from the measurement data, each image layer being a successive depth image into the patient. This enables multiple El images, such as slices, of the patient to be assessed against one another to determine, for example, whether a tumour is 25 present in 3D, if so, to what extent the tumour extends into the patient, and the type of tumour may be determined from the 3D shape visualisation. A further aspect of the present invention provides a device for electrical impedance screening of a patient, including means to output a plurality of electrical signals of different frequencies to the patient, at least one input to 30 receive electrical impedance signals from the patient arising from the input electrical signals, and processing means to generate image data from the received electrical impedance signals.

11 The device may be releasably connected to an array of elements, wherein each element is arranged to convey at least one of the electrical signals to the patient. The processing means may include means to generate multi layer (slice) image data of the patient based on the received signals. 5 BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows a cross-section through a complied electrode array device according to an embodiment of the present invention. Figure 2 shows an embodiment of the present invention applied to a breast. 10 Figure 3 shows an alternative embodiment of the present invention with an outer conductive membrane and interior electrodes. Figure 4 is a flow diagram of an assessment protocol according to an embodiment of the present invention. DESCRIPTION OF PREFERRED EMBODIMENTS 15 At least one embodiment of a device and related method of use employing the present invention will hereinafter be described. However, this should not be taken as limiting the generality of the invention, and any mention of prior art in this specification, whether in this section or elsewhere, should not be taken as an admission that such art or other prior art belongs to the common general 20 knowledge. The device and method of at least one embodiment of the present invention utilises an array of 256 electrodes per breast scan. One of the electrodes is connected to a voltmeter (receiver) input through a multiplexer (switch). The remaining 255 electrodes are successively each connected to a 25 current source through another multiplexer. The receiver measures voltages at a pre-selected electrode for each of the transmitting electrodes (255 measurements). The receiver is connected to the next electrode of the array and the process is repeated again. Thus, there are 256 variants of the receiver connection, and for each such connection there are 255 variants of the 30 transmitter connection. In total, there are 256*255=65280 individual data values obtained during one data acquisition. This is the maximum possible data set which can be collected using a given array of 256 electrodes.

12 This extensive data set enables 3D inverse problems to be solved for a predetermined electrical potential equation, and thus to reconstruct in a non invasive manner 3D distribution of the conductivity inside the patient's body. Such imaging is not otherwise possible in present devices and methods due to both a 5 lack of measured data and absence of a reconstruction stage of data processing. Utilising one or more embodiments of the present invention, it is possible to obtain image depth data for substantially the entire breast. The maximum depth of 3D reconstruction from the obtained data set is limited by the radius of the electrodes array arising from the geometry of the electric field. Thus, to have 10 reasonable depth of imaging, the size of the array can be optimised to correspond to the breast size. That is, the array of 256 electrodes covers substantially all of the breast surface. Embodiments of the present invention utilise filtered, weighted, back projection along the equipotential surfaces of the electrical field to solve the 15 inverse problem i.e., to calculate spatial distribution of the medium being imaged from the results of electrical measurements carried out from its surface. Thus, embodiments of the present invention enable reconstruction of conductivity or impedance values in any spatial point (voxel) within the region covered by the array (spatial mapping) e.g., of a "hemisphere" formed by the 20 breast lying behind the array with a radius equal to that of the radius of electrode array. To represent this 3D information on-a display, say a computer screen, the present invention can provide a series or set of cross-sectional 2D images representing different depth layers (depth data) within the breast tissue. These layers can be several millimetres apart, for example, 5-10mm spacing has been 25 found to be effective, with a preference for around 7mm spacing. The patient's surface skin and areas immediately below the surface exhibit the largest electrical resistance, and thus can have significant influence on the measurement results while providing zero valuable information about the condition of the breast tissue itself. The lack of depth data in prior art devices and methods results in skin 30 properties being visible in obtained image data, and also image defects or image artefacts can be present, rather than the underlying breast tissue. The present invention can utilise actual 3D distribution of the breast tissue properties in the 13 form of multiple cross-sections or "slices" of image data obtained with increasing breast tissue depth. The electrical impedance (Z) or conductivity (which is the inverse of impedance, i.e., 1/Z) has been found to correlate with the physiological state of 5 the tissue being imaged. For example, it has been found that cancerous tissue may demonstrate 2 - 20 times higher conductivity than the normal tissue of the same type. So obtaining the data about conductivity distribution inside the body provides an advantageous and effective (and importantly non-invasive) way to diagnose various diseases. 10 Utilising the present invention it is possible to reconstruct conductivity/impedance related values in any chosen spatial point (voxel) within the tissue being imaged e.g., within the hemisphere of the breast with a radius substantially equal to the radius of electrode array. To represent this 3D information on a display it has been found advantageous to use a set of cross 15 sectional 2D images obtained from different tissue depths. It is thereby possible to estimate the extent of any abnormal tissue (tumour, cyst, etc) not only in a horizontal plane but also in depth. Comparing images at different cross-sections enables to distinguish artefacts and errors from real objects. It has been realised that medical practitioners and others, especially low 20 skilled personnel or personnel operating remotely or "in the field", performing screening investigations require a screening tool to at least be able to make an initial assessment to separate image results requiring additional attention from those results for image data corresponding to normal tissue. The present invention in one or more forms can provide an algorithm which 25 enables estimation of the probability of disease or "non-normal" tissue being present for a given set of measurement results. The algorithm can be utilised in combination with training data and/or results can be confirmed by comparison with alternative investigations. A set of statistical parameters is calculated from the results of 30 measurements and reconstructed image data. These values are linearly combined using a set of coefficients preliminary obtained using the training data and/or least squares method. The obtained linear combination is used in logistic formula to have all values within 0 - 1 (i.e., probability) interval.

14 The obtained images and image data can be used to determine particular values from that information, such as average electrical conductivity in the region of interest, peak conductivities in the focal areas, and so on. Results derived from the obtained image data allow at least an initial 5 assessment of the (breast) tissue to be obtained, such as OK-not OK, or clear-not clear, refer for further assessment. Results can be formalized in written form including images if necessary and recommendations for further assessment, if any. The 3D imaging technique and device of the present invention enables a 10 user to distinguish artefacts from real objects due to enhanced depth data images i.e. appearance at the multiple cross-sections. Another tool to filter artefacts is looking on the map for poor electrical contacts. Any objects visible near poor contacts should be considered as possible artefacts, and the measurements can be repeated. 15 One or more forms of the present invention are directed to diagnostics of dyshormonal breast diseases with multifrequency electrical impedance mammography. It has been found that more than around 34,000 new cases of breast can cer are registered annually in Russia with patient age demographic becoming 20 significantly younger. 25% of women under thirty and 60% of those above forty are diagnosed as dyshormonal breast (mastopathy) cases. Though mastopathy is not considered to be an obligatory cancer prelude, breast cancer occurs 3 - 5 times more often in patients with diffuse dyshormonal non-cancerous conditions and 30 - 40 times more often with nodular forms of mastopathy aggravated by 25 proliferation of lactic gland epithelium .This evidence has heightened the general interest in non-malignant growth opening the way to incidence reduction in breast cancer through effective curing of mastopathy. The existing methods of mastopathy diagnosis have certain applicability restrictions of both objective and subjective nature. Results of a comprehensive examination of 166 women (92 30 without breast pathology and 74 with mastopathy) with four methods: multifrequency electrical impedance mammography, ultrasonic investigation, x ray mammography and needle biopsy follows. The analysis of the obtained data was based on visual estimation of the electrical impedance images, estimation of 15 indices and histograms of the conductivity distribution in breasts under multifrequency investigation, comparison of measurements in different groups according to Student's criterion and comparison with ultrasonic and x-ray data. The report presents criteria of diagnostics of various forms of mastopathy using 5 the multifrequency electrical impedance mammography method. 1. INTRODUCTION With this in mind, one or more forms of the present invention provides multi frequency electrical impedance mammography to diagnose diseases, such as dyshormonal breast diseases and breast cancer. 10 II.SUBJECTS The following presents main results test data of a comprehensive ex amination of 166 women (92 without breast pathology in the 1st group and 74 patients with mastopathy in the 2 n group). In both groups there were subgroups according to the degree of age change of the breast tissue: 15 below 35 years - with normal stable state of breast tissue; 35-40 years - with gradual loss of glandular structures; 41-45 years - with noticeable thickening of duct cylindrical epithelium, thickening of basal membrane and fibrous change of connective tissue; 46-50 years - with dilatation and occasional cystic widening of lactic ducts 20 surrounded by fibrous tissue; above 50 years - with slow obliteration of lactic ducts and small vessels accompanied by adipose tissue formation. Ill. METHODS The following diagnostic methods were applied: 25 ultrasonic examination of breast tissue was applied to all patients on the 5-9* days of their menstrual cycle; the apparatus used was ultrasonic Combison 530TM with electronic linear 7.5 MHz probe; women of 35 upwards were subjected to x-ray mammography on the 5 -9th days of their menstrual cycle; 'Mammodiagnost DC X-ray apparatus and ACF A 30 mammoray HDR film in KodakTM min-R cassette were used; 16 The test subjects were subjected to multi frequency electrical impedance mammography; the apparatus used was a 256-electrode electrical impedance mammograph, developed by the Institute of Radio-Engineering and Electronics of the Russian Academy of Science; the frequencies used were 10kHz and 50 kHz; 5 the timing was chosen 3

-

10 th and 17

-

28 th days of the menstrual cycle; after ultrasonic and X-ray examinations, a puncture biopsy was carried out with a puncture needle by standard procedure; all examination results were statistically processed by alternative variation method to obtain a mean value and standard deviation. To check the data 10 reliability we resorted to student's criterion and the appropriate tables. In addition the so called 'least square method' (LSM or Gauss' method) was used from MATLAB 7.0 applications. IV. RESULTS All dyshormonal hyperplastic processes in milk glands are known under 15 the term 'mastopathy'. Mastopathy is defined as a fibrous and cystic disease with a wide spectrum of proliferative and regressive changes of the breast with abnormal ratio of epitelic and conjunctive components that are mingled in various combinations. To find out potentiality of multi frequency electrical impedance 20 mammography in diagnostics of dyshormonal breast diseases, checkups followed the usual procedure of ultrasonic investigation and x-ray mammography (with women of 35 upwards). 92 women with ultrasonic resulting data composed the 1st clinical group [2]: 25 ultrasonic type of the breasts was in conformity with their age (juvenile, reproductive, premenopausal); ultrasonic visualizations corresponded to their physiological periods (1st or 2nd phase of the menstrual cycle); distinct differentiation of the breast tissue; 30 the parenchyma thickness did not exceed 14 mm; absence of focal symptomatology, ductectasia, duct wall fibrosis; absence of changes in regional zones of lymph outflow.

17 The 2 nd clinical group composed 74 women revealing the following ultrasonic information [21: incongruity between ultrasonic type of the breast tissue and their age; glandular tissue layer thickening for more than 14 mm in all varieties of 5 diffuse mastopathy; fibrous changes in walls of ducts; changes in echo density of glandular tissue indices that they are not typical for the patient's age; ductectasia, wall thickening, lumen increase, 10 irregularity of duct circuit, dilations along the main duct; presence of multiple cysts. Mastopathy women in Group 2 were subdivided into 2 subgroups according to the mastopathy type: 1 st type mastopathy subgroup - without cystic component (41 patients); 15 2nd type mastopathy subgroup - with one or numerous cysts (33 patients). Women of 35 upwards underwent X-ray examination to confirm the diagnosis. Mastopathy patients underwent anticancer and antiatypia puncture thin-needle biopsy. The analysis of the multi frequency electrical impedance mammography 20 began with consideration of electrical conductivity indices, given in conventional units. There is no statistically valid difference in electrical conductivity indices depending on vertical or horizontal body position and on left or right breast. Mean indices from the second scanning level have been used. The obtained results are 25 shown in Tables 1 and 2. The following analysis of the results is noted: electrical conductivity indices proportionally increase with the patients' age irrespective of their groups and menstruation phases; according to the statistics obtained, mastopathy of both types has been 30 found to reduce electrical conductivity from norm in the corresponding age groups of menstruating women (aged up to 50) on frequency 50 kHz as well as 10 kHz during both menstrual cycle phases (in all cases p<0.05).

18 As for post-menopause (51-55 years of age), mastopathy does not influence electrical conductivity or has just a slight excess over the norm (disparity is statistically inadequate p>0.05). In the 2 nd type mastopathy subgroup, all in menstruation phases and in 5 post-menopause at the frequency of 50 kHz the electrical conductivity is higher than in mastopathy of the 1st type; In norm and mastopathy of the 2 nd type, we see a clear, statistically valid decrease in electrical conductivity in both phases of the menstruation cycle and in post-menopause in the corresponding age groups at the frequency of 10 kHz (in 10 all cases p<0.05); In the 1st type mastopathy subgroup, the electrical conductivity indices are the same at all frequencies and in all groups. The results of the comparison of the electro impedance images are as follows (Fig.1): 15 Abnormal architecture of images caused by changes in proportion of breast tissue which results in discrepancy between electro impedance type (juvenile, reproductive, postmenopausal) and age; An increase of hypo-impendance areas owing to fibrous changes of adipose tissue, fibrosis duct walls and Kuper's ligament; 20 Appearance in mastopathy mammograms of the 2 nd type hyperimpedance inclusions with clear outline that correspond to breast cysts or clear widening of the ducts. Tomographic changes of breast structure are confirmed by results of ultrasonic examination. 25 V. CONCLUSIONS From the investigation, the following conclusions follow: Breast tissue electrical conductivity depends on the woman's age irrespective of the presence or absence of structural changes in the gland tissue. To avoid incidental anomalies and abnormalities about dependence of electrical 30 conductivity due to age were processed according to the "least squares method" (LSM or Gauss's method). The sought for parameters of this dependence were 19 determined. Regression lines in all clinical groups and at the frequency of 50 kHz are described with the same polynomial of degree three (f(x) = pl.x 3 + p2.x 2 + p3.x + p4), and the coefficients (p1, p2, p3, p4) of this polynomial are found from the 5 system of linear equations. Norm: 50 kHz (pl=0.00 2 5, p2=-0.01893, p3=0.06857, p4=0.38); 10 kHz (pl=0.005, P2=-0.04357, P3=0.1414, P4 =0.272). Mastopathy 1 type: 10 50 kHz (p1=0.009167, p2=-0.06464, p3=0.166 2 , p4=0.184); 10 kHz (p1=0.0075, p2=-0.05179, p3=0.1407, p4=0.188). Mastopathy 2 type: 50 kHz (p1=0.0 116 7 , p2=-0.095, p3=0.2733, p4=0.136); 10kHz (p1 =0.01167, p2=-0.1 007, p3=0.28 76 , p4=0.088). 15 In spite of the equal regularity in the changes of electrical conductivity in norm and during age dysplasia, the method of electrical impedance mammography of the present invention ensures reliable diagnosis of mastopathy by the fall in electrical conductivity indices in the corresponding age groups and at different frequencies in women during menstruation phase. The absence of 20 electrical conductivity difference between norm and mastopathy in post menopausal women confirms the fact that dysplasia appears due to dysfunction in the hormonal system "ovary - breast" and not as an independent disease. The major advantage of the multi frequency electrical impedance mammography over the mono-frequency lies in the fact than it makes it possible 25 not only to diagnose mastopathy but also helps determine cystless form from the absence of difference between indices of mean electrical conductivity in the corresponding age groups during all phases of the menstruation cycle. According to morphologists, occurrence rate of hyperplasia and atypia that accompany dysplasia is higher than with cystic form, which makes it possible to ascertain a 30 group of higher risk for special medical care to prevent breast cancer development.

20 Confirmation of visual changes in tomograms during mastopathy by quantitative characteristics of the electro conductivity excludes any diagnostic mistake. Safety of the electrical impedance mammography method makes it 5 possible to use it as a breast screening examination method for all ages, including young women. Table 1: shows mean electrical conductivity indices of breast tissue in normal and mastopathy pathology in the 1st phase of the menstruation cycle and in postmenopause with women of different age groups (10kHz, 50kHz). The 10 values given with deviations: Table 1 19 - 34 years 35 - 39 years 40 - 44 years 45 - 50 years 51-55 years Postmenopause 50 kHz 10kHz 50 kHz 10kHz 50 kHz 10kHz 50 kHz 10kHz 50 kHz 10kHz Norm 0.43± 0.39± 0.48± 0.43± 0.48± 0.43± 0.53± 0.48± 0.56± 0.51± 0.09 0.08 0.03 0.04 0.03 0.03 0.05 0.05 0.06 0.07 Mastopathy 0.28± 0.27± 0.30± 0.29± 0.35± 0.36± 0.38± 0.37± 0.55± 0.54± I type 0.05 0.07 0.07 0.07 0.05 0.06 0.09 0.09 0.03 0.05 Mastopathy 0.33 0.29± 0.36± 0.31± 0.4- 0.34: 0.46 0.37* 0.59+ 0.47: 2 type 0.07 0.04 0.06 0.05 0.09 0.06 0.04 0.05 0.02 0.03 Table 2 shows results for the mean electrical conductivity indices of breast tissue in normal and mastopathy pathology in the 2 nd phase of menstruation cycle and in postmenopause with women of different age groups (10 kHz, 50 kHz). 15 The values are given with deviations: Table 2 19 - 34 years 35 - 39 years 40 - 44 years 45 - 50 years 51-55 years Postmenopause 50 kHz IOkHz 50 kHz 10kHz 50 kHz 10kHz 50 kHz 10kHz 50 kHz 10kHz Nonn 0.43± 0.39± 0.48+ 0.43± 0.48± 0.43± 0.53± 0.48± 0.56± 0.51± 0.09 -0.08 0.03 0.04 0.03 0.03 0.05 0.05 0.06 0.07 Mastopathy 0.28± 0.27± 0.30± 0.29± 0.35± 0.36± 0.38± 0.37± 0.55± 0.54± type 1 0.05 0.07 0.07 0.07 0.05 0.06 0.09 0.09 0.03 0.05 Mastopathy . 0.33 0.29± 0.36± 0.31± 0.4± 0.34± 0.46± 0.37± 0.59± 0.47± type 2 0.07 0.04 0.06 0.05 0.09 0.06 0.04 0.05 0.02 0.03 21 One or more forms of the present invention provide or incorporate a method for 3D EIT mammogram analysis: 1. Bad contacts, including corners, and diagonal elements (i.e. when the 5 transmitter electrode coincides with the receiver) can be removed from the raw measurement results. If all contacts were good, 256*255=65280 values for the 256-electrode array are obtainable. The mean and median values are calculated from the obtained values. 2. Reconstructed image data (e.g. 7 sections with 36x36 pixels in each) can be 10 combined in a single array. Average and standard deviation values are calculated from such data. The last is calculated using the formula: 1 1/2 s = Y (y - Y) 2 (n-1 j=1' where s is standard deviation, n is total number of values (pixels), y, is value from the array, yi is mean value for the array. 15 3. A probability of malignancy P can be calculated according to the equation: P=1/(1+ exp(-a), where a =bo +btx +b 2 x 2 +b 3 x 3 + b 4 x 4 +b 5 x 5 , 20 x, is patient age in years (with 3 digits after decimal point), x 2 is mean over raw measurements according to paragraph 1, x 3 is median over raw measurements according to paragraph 1, x 4 is mean over images according to paragraph 2, x 5 is standard deviation over images according to paragraph 2, 25 The coefficients bk are calculated to fit training data set in least squares sense. Below are examples of the coefficient obtained with limited training data set and providing sensitivity and specificity above 75% for each. bo = -10.6011 22 b, = 0.213103 b2 = -0.0056579 b3 = 0.0055994 b4 = 10.5179 5 b5 = -12.2619 Further forms of the present invention are described with reference to the accompanying figures. Figure 1 shows a cross-section of a device according to an embodiment of the present invention with an array of electrodes 3 on an outer surface of a flexible material 5. The flexible material is supported by a 10 conformable base material 2, such as a fibrous or spongy material e.g., foam. A rigid support 1 for the conformable portion is provided. Figure 2, an embodiment of the device is shown applied to a breast 4, with the flexible material and base material conforming generally to the shape of the breast 4. Whilst some flattening of the breast occurs in this embodiment, reduced 15 pressure and discomfort is experienced by the patient because of the conformability of the device. Furthermore, increased contact of a greater number of electrodes is experienced. In Figure 3, an alternative embodiment is shown with the array of electrodes 3 covered by an electrically conductive covering or membrane. 20 In use, the device is applied to the target region and some pressure applied to permit the conformable array to adapt to the shape of the target region. Some deformation of the target region may be expected but this is minimal compared with a known rigid planar array of the prior design. The conformable array of the embodiment provides enhanced contact of multiple electrodes with 25 the surface of the subject. An electrically conductive liquid may be applied between the device and the surface of the subject, which can help match impedance characteristics between the array of electrodes and the outer surface of the subject. In such cases, or where other liquids are present, the array may include an outer waterproof or water resistant conductive membrane to prevent 30 the contact portions of the electrodes becoming wet, unclean or damaged. Furthermore, a smooth membrane allows the device to be cleaned/sterilised more readily than outer multiple electrodes. The base material 2 may be a sponge 23 material, such as open or closed cell sponge, allowing conformability of the device to the target region. In use, and as set with reference to figure 4, an array of electrodes is attached to a patient 100, one of the electrodes acting as a receiver and the 5 remaining electrodes each acting as a transmitter. An electrical current is applied to each of the transmitting electrodes of the array 102, and a receiver electrode of the array receives each current signal 104. This is repeated 106 by making the receiver a transmitter electrode and, in turn, each transmitter electrode a receiver electrode, to build a pattern of received signals n electrodes x n-1 electrodes e.g. 10 256x255 for an array of 256 electrodes. This generates measurement data which is used to derive images from. The images can be created as depth "slices" 108 through the patient extending width-wise approximately the width of the array. An algorithm is used to calculate a value for each voxel from the data, and thus images at depth can be created. A further assessment of the probability of the 15 presence of abnormal tissue can be calculated 108. An initial assessment for the presence or absence of abnormal tissue is conducted 110. If no abnormal tissue is detected 112, the patient can be referred to return for routine screening after a set period eg 12 months, 1 year, 2 years etc. Borderline cases can be referred to return 116 for screening at a 20 shorter interval, say 3-6 months or only a few days/weeks. If abnormal tissue is detected 114, further assessment can be recommended. Thus, screening of patients for signs or presence of abnormal tissue, such as breast cancer or other tumours, can be conducted in rural or low technology regions and/or by minimally qualified staff to give at least an initial screening. 25

Claims

1. A method of screening a patient at a location utilising electrical impedance (El) for a presence of abnormal tissue, including the steps of; a) obtaining El measurement data from a patient; 5 b) producing image data from the measurement data; and c) assessing the image data to make a first determination of the presence or absence of abnormal tissue.

2. A method according to claim 1, including the steps of; d) connecting an array of electrodes to a patient, one of the electrodes 10 acting as a receiver and a plurality of the remaining electrodes each acting as a transmitter; e) introducing an electrical current signal to the patient through each of the transmitter electrodes; f) receiving said current signals at the receiver electrode; 15 g) swapping a transmitter electrode to be the receiver electrode and the receiver electrode to be a transmitter electrode; h) repeating steps b) to d) for each of the transmitter and receiver electrodes; and i) generating the image data from the multiple received signals. 20

3. A method according to claim 1 or 2, including applying an algorithm to the measurement data to produce the image data.

4. A method according to claim 3, including the further step of utilising the formula P=1/(1+exp(-a) with the algorithm to determine a probability of the presence or absence of abnormal tissue. 25

5. A method according to claim 4, wherein; a = b o + bx,+ b 2 x 2 + b 3 x 3 + b 4 x 4 + b 5 x 5 , where x 1 is patient age, x 2 is mean over raw measurements, 25 x 3 is median over raw measurements, x 4 is mean over image data, x, is standard deviation over image data.

6. A method according to any one of the preceding claims, including the step 5 of; generating multiple depth image layers from the measurement data, each image layer being a successive depth image into the patient.

7. A method according to claim 6, including the step of comparing the depth image layers for indications of the presence or absence of abnormal tissue 10 between adjacent layers.

8. A method of impedance imaging of a body region of a subject including the steps of positioning at least one multi-element probe on at least one respective surface of the body region; applying an electrical current to elements of the multi-element probe at two 15 different frequencies or more; sensing signals from the elements derived from the different frequencies; and generating at least one impedance map based on the sensed signals.

9. An electrical impedance (El) system, including a device with output means 20 for applying consecutive electrical signals to a plurality of transmitter electrodes, the device including input means to receive a plurality of received ones of said electrical signals as measurement data, a processor for generating image data from said measurement data, and a display to display images from said image data. 25

10. A system according to claim 9, wherein the device includes means to generate multiple depth images from said measurement data. 26

11. A system according to claim 9 or 10, including processing means used to generate a probability of the presence or absence of abnormal tissue from the measurement data.

12. A device for electrical impedance screening of a patient, including means 5 to output a plurality of electrical signals of different frequencies to the patient, at least one input to receive electrical impedance signals from the patient arising from the input electrical signals, and processing means to generate image data from the received electrical impedance signals.

13. The device of claim 12, including a releasably connected array of 10 elements, wherein each element is arranged to convey at least one of the electrical signals to the patient.

14. The device of claim 12 or 13 wherein the processing means includes means to generate multi layer image data of the patient based on the received signals.

15 15. A device for electrical impedance screening substantially as hereinbefore described.

16. A method of screening a patient substantially as hereinbefore described. HEALTH SCREENING TECHNOLOGIES PTY LTD WATERMARK PATENT & TRADE MARK ATTORNEYS